Idea Cruncher Version: 1.3.1 serial key or number

Idea Cruncher Version: 1.3.1 serial key or number

Idea Cruncher Version: 1.3.1 serial key or number

Idea Cruncher Version: 1.3.1 serial key or number

Mindwipe (G1)

The name or term "Mindwipe" refers to more than one character or idea. For a list of other meanings, see Mindwipe (disambiguation).
Mindwipe is a DecepticonHeadmaster from the Generation 1continuity family. Blah!
"You're feeling very sleepy..."

No one knows the dead quite as well as the Decepticon hypnotist, Mindwipe. A mysterious and mystical one, Mindwipe claims to have the powers to commune with Transformers who have long since left this plane of existence. From Black Sabbaths to seances to super-conductive antennae, Mindwipe knows of many methods to reach those who now rest in peace.

Well... not really. He desperately wants to speak with the departed sparks of dead Decepticons, but all his labors to create antennae and receivers for the departed have had... "limited" success. In reality, all he's picking up are radio broadcasts of old Earth sitcoms, which he's actually become quite fond of. Still, some day, Mindwipe hopes to finally bridge the gap between the realm of the living and the realm of the dead.

Despite those shortcomings, Mindwipe is a formidable opponent, able to hypnotize and control both organic and mechanical lifeforms just by making eye contact, and boasting an eerie and precognitive "sixth sense". But sometimes he forgets this and, when the odds look stacked against him, Mindwipe's more than happy to turn tail and get while the getting's good. His loyalty lasts exactly as long as his bravery.

He is binary bonded to the NebulanVorath, a scientist. Yes, a scientist. They disagree on pretty much everything. Sometimes he works with the Partner DroneServant.

The powers of darkness are a more powerful weapon than all the toys your science can muster.

—Mindwipe, talking all spooky to Vorath, "The Rebirth, Part 2"

Fiction

Marvel Comics continuity

Marvel The Transformers comics

Events from the UK-only comic stories are in italics.
Bring me the lamp! The rest of the treasure is yours, but the lamp is mine.

Mindwipe was one of Scorponok's soldiers on Cybertron. Upon storming the Autobots' base, the Decepticons were dumbfounded in finding it already evacuated. Mindwipe was responsible for detecting a series of "unintelligible sounds" coming from a nearby console, which turned out to be a signal coming from the planet Nebulos. Once translated, the transmission revealed that the Decepticons' old enemies, the unit led by Fortress Maximus, had settled on the planet, and Scorponok immediately took a strike team—of which Mindwipe was part—to Nebulos to exact revenge. Broken Glass!

Upon arrival, the Decepticons kidnapped the Nebulans who had summoned them, and Scorponok turned them over to Mindwipe for use in his experiments. Using his hypnotic powers, Mindwipe purged the fear from Monzo's mind and encased him in a polymer alloy bubble which was then sent floating off into the sky by a shot from Skullcruncher's anti-gravity gun. The alloy malfunctioned and the bubble popped, prompting Mindwipe to catch Monzo so as to not waste a "good laboratory animal". Before he could carry out any further experiments, Lord Zarak, one of their Nebulan prisoners, suggested that the Decepticons binary bond with them to become Headmasters, so that they could match the power of the Autobots who had recently gone under this process themselves. When the Decepticons next met the Autobots in battle, Mindwipe was now bonded with Vorath, and the two sides were evenly matched. To tip the odds, Mindwipe hypnotized members of the Council of Peers and Zarak's daughter, Llyra, into taking refuge in the Decepticon base, where he enclosed them in more polymer alloy bubbles. With the Autobots distracted by having to same the Nebulans, Mindwipe and the other Headmaster Decepticons shot them in the backs. Love and Steel!

Surprise, Mindwipe hypnotizes a lot.

Mindwipe and the Decepticon Headmasters struck next at the Mercury Gardens of Melanossus, where they battled and defeated the Monsterbots. Brothers in Armor!!A little later, during another battle with the Autobot counterparts, Mindwipe hypnotized Brainstorm, causing him to crash into Hardhead. He, Weirdwolf, and Skullcruncher then surrounded the two Autobots, but a volley of firepower from Highbrow and Chromedome totalled Mindewipe's comrades, and he quickly retreated. The panicked withdrawal was merely a ruse, however; the Decepticons were attempting to lure the Autobots Headmasters into a trap so they could capture and study them, but the gambit ultimately failed.Worlds Apart! Eventually, the Autobots decided to depart Nebulos to protect it from the Transformers' conflict, and Mindwipe and the Decepticons pursued them. Brothers in Armor!!

Well, that's randomly toyetic.

Following the Autobots to Earth, Mindwipe detected the signal that had brought their enemies to the planet, and discovered that it originated from an abandoned Autobot headquarters inside a volcano cavern. But Fortress Maximus's Autobots weren't far, and the Decepticons fought them once again. Mindwipe and the others were stampeded out of the volcano when Fortress Maximus picked up a new Headmaster partner, the Earth human Spike Witwicky. Trial by Fire! The Decepticons proceeded to establish themselves on Earth, with Mindqipe's partner Vorath helping Scorponok to create Pretender technology, as Mindwipe watched silently in bat mode. Pretender to the Throne!

Scorponok put Mindwipe's hypnotic powers to use setting a trap for Highbrow, by having him mesmerize some humans who then led the Autobot Headmaster into a deep wood. The trap sprung, Highbrow attempted to flee, but Mindwipe pursued and used his powers to give him a severe case of amnesia. Scorponok tried to convince the brainwashed Highbrow that he was really a Decepticon, but Highbrow remained ill at ease, so the Decepticons tried to defuse the tense atmosphere by transforming their heads into Nebulan mode. Unfortunately, it was at this point that they discovered Mindwipe's fatal mistake: He had neglected to also blank the mind of Highbrow's partner, Gort. Gort knocked out Vorath, taking Mindwipe out of the fight, and then recombined with Highbrow, restoring his memory. He proceeded to rip Scorponok's head from his shoulders, the mental shock knocking the Decepticon leader offline.All in the Minds!Not long thereafter, amidst a time storm that threatened to end all reality, Mindwipe helped hunt down Highbrow in order to reclaim Scorponok's stolen head. Once Mindwipe retrieved the head and restored it to Scorponok's shoulders, the rest of his team was shunted to Limbo and replaced by a group of Decepticons from 2009 led by Soundwave. After some misunderstandings were ironed out, Mindwipe and the two other Decepticons not limboed away formed a temporary alliance with the Autobots, the future Autobots, and the future Decepticons against the time-traveling Galvatron. But the alliance was short-lived; when the future Ultra Magnus accidentally shot Scorponok, Mindwipe joined his fellow present-day Decepticons in tearing Ultra Magnus apart. They retreated with Scorponok's wounded body just before Galvatron was thrown into the time storm, sating it and saving the universe.Time Wars

And then I conveniently disappear for that whole Underbase massacre.

Inevitably, Scorponok's band of Earthbound Decepticons met with Ratbat's band of Earthbound Decepticons, and a group including Mindwipe was given a really boring tour by Starscream of Ratbat's island headquarters. When the tour's star attraction, the kidnapped Buster Witwicky, unwittingly let slip that Ratbat was searching for a powerful artifact called the Underbase, Scorponok was enraged. Scorponok took Mindwipe and the others to confront Ratbat for hording such information, and a civil war broke out between the two Decepticon camps. Cold War! In time, they discovered that this fight was in line with Starscream's plan to seize the Underbase's power for himself, and a cease-fire was agreed upon. But this reprieve was short, as the Autobots were drawn to the island base by Starscream's manipulations as well. Mindwipe was quick to maul the hapless Beachcomber before another cease-fire was brokered, this time between the two opposing faction. Starscream went on a rampage, armed with the Underbase's cosmic energy, but partially-organic Transformers were immune to its power. The Decepticons suffered many casualties, but Mindwipe was not among them. Dark Star

In the wake of the Underbase Saga, Scorponok took command of the remnants of the Decepticon army. Mindwipe and Weirdwolf accompanied their leader to a rendezvous with the mercenary Dealer in order to purchase Autobot battle plans the merc had stolen, but before they could depart, they were attacked by Optimus Prime. "Dealer" was actually Doubledealer, and had sold them out to their enemies!(Double) Deal of the Century! The Decepticons soon established a headquarters under a garbage dump in New Jersey, across the river from New York City. Mindwipe was charged with engineering an energy conversion machine to convert power into Energon, and all he needed was for a cable to be laid across the river, connected to an electrostator placed at the top of the Empire State Building. Unfortunately, Iguanus failed at this and seemingly perished. King Con!

The Decepticons scrambled once again when the MicromasterAir Strike Patrol summoned them to MacDill Air Force Base after the wayward new recruits, facing Optimus Prime and the entire Autobot army, were seemingly over their heads. Though both Optimus Prime and Scorponok privately suspected someone had arranged for this altercation, a fight broke out regardless. Mindwipe got punched in the batface by Blurr, while the Air Strike Patrol made their quiet retreat. The Resurrection Gambit! Mindwipe recuperated and joined Scorponok in advancing on the Autobots, just as Optimus Prime, strangely, called all his troops other than himself to return to the Ark. Scorponok was suspicious, and Mindwipe suggested that some powerful weapon was hidden inside the Autobots' shuttlecraft. Mindwipe and the others rushed Optimus Prime as he tried to stop Scorponok from firing on the shuttlecraft, but Prime, who was quickly becoming frustrated with the day's events, was able to overpower all of them. As Optimus pleaded to Scorponok to stop fighting and determine the mastermind who was pulling their strings, they were interrupted by a huge hint: Starscream had arrived, reborn as a Pretender. All the Familiar Faces! Curiouser still, Mindwipe was shocked as Starscream claimed he had been sent by Megatron, also believed to be dead! As Starscream's Pretender shell trounced Optimus Prime, Scorponok half-heartedly ordered Mindwipe and Weirdwolf to help him. Mindwipe learned the hard way that his hypnotic power had no effect on the new Starscream, and was taken out of the fight with a signle blast. Shortly thereafter, Starscream suddenly reverted to his classic cowardly self and begged for mercy. Skin Deep

And though you can't see the bottom, believe me it's a long way down.

Scorponok reluctantly allowed Starscream back into their ranks, a decision that didn't sit well with his troops. Mindwipe and most of the other Decepticons gave Starscream a wide berth, fearing the power he had displayed during his debut battle as a Pretender,Fallen Star! and soon began to turn their anger on Scorponok. Mindwipe was one of the most vocal members of this retaliation, and pointed to the hole Starscream had put in his chest. He reminded Scorponok that Starscream was a chronic traitor, but Scorponok rebuked him, noting that Starscream was a powerful warrior needed to bolster their ranks, and was better kept close so they could monitor his actions. Primal Scream But when Starscream went missing, Mindwipe and Triggerhappy decided they'd had enough and left Scorponok's command. Ironically enough, they ran into Starscream himself and were strong-armed into joining his small band of Decepticons against Scorponok. Eye of the Storm Mindwipe and Triggerhappy returned with Starscream, Shockwave, and Ravage to Shockwave's underwater headquarters, where they learned that their forces also included the Battlechargers. With only seven on their team, Mindwipe was not impressed and expressed doubts, but when Starscream told him that they'd either fight for them or be executed, he quickly changed his mind.

How much more hypnotism can we cram into this page?

Immediately, the seven flew to Scorponok's base and fired into it from the air, collapsing it on an Autobot/Decepticon alliance that had been forming. Surrender! Once again, Mindwipe was the designated worrier. Though the collapsed base showed no signs of life, if Scorponok had survived, he said their best bet was to get while the getting was good. Triggerhappy agreed, and Starscream silently pitied their cowardice. Unfortunately for Starscream, Mindwipe was right, and Scorponok burst out of the ground to deal with them. Shockwave ordered Mindwipe and the others to rush Scorponok, an idea that was not popular with Mindwipe, and for good reason. Scorponok shrugged off their attacks. Mindwipe and Starscream subsequently concerned themselves with Scorponok's troops while Scorponok battled his equal, Shockwave. ...All This and Civil War 2

Mindwipe put his magic powers to good use and hypnotized Field Commander Soundwave with the intention of controlling the troops through him. It was a good plan! And Kup agreed! At gunpoint, Mindwipe told Soundwave to tell all the other Decepticons to lay down their arms in service of the Autobot/Decepticon alliance against Unicron—just in time for every Transformer on Earth to be warped to Cybertron by their creator, Primus. Out of Time! Mindwipe watched with the other warped Transformers as Primus explained to them the history between himself and Unicron, and that the Transformers' destiny was to defeat him. The Void!

He doesn't look that bad, he's—oh, he's not in bat mode? Oh. Oh.

Once the massive planet-sized Unicron arrived, Mindwipe and the other Transformers froze in terror. After watching Brainstorm get eaten by the Chaos Bringer, Mindwipe regained enough of his faculties to be able to run away. The Transformers did eventually come around and face their destined adversary, but it did not end well for Mindwipe. As Unicron tore an avatar of the Creation Matrix to pieces, the resulting shockwave sent what was left of Mindwipe's body flying. On the Edge of Extinction!

Earthforce
These stories do not fit into the normal Marvel continuity. See Earthforce for details.

After having been thrown out of the Decepticon army for trying to hypnotize Scorponok into making him leader, Mindwipe devised a master scheme of his own. He manufactured thousands of thought-sensitive Sleepfast™ night lights, selling them to humans on the novelty that they actually turned on or off just by thinking at them. Little did the humans know, however, that repetitive use of the night lights had a subliminal mind-control effect. Over time, Mindwipe would have gained an army of human slaves! Luckily for Humanity, the Autobots Wheeljack and Snarl came to investigate, and they confronted Mindwipe in his storage facility. He was beaten by his own night lights when the Autobots suddenly thought "Lights on!" in a darkened room, turning on thousands of Sleepfasts™ at once and temporarily overloading the Decepticon's neural circuitry with the sudden flash.

Mindwipe was, to coin a phrase, batshit nuts. The Living Nightlights!

Marvel Generation 2 comic

Mindwipe: All magic, no pluck.

Mindwipe was still a part of the Decepticons after Megatron took back control of the army from Bludgeon. He was involved in the disastrous first encounter between Megatron's troops and Jhiaxus's Cybertronians. After Megatron was literally knocked out of orbit by Jhiaxus and plummeted to the planet below, Mindwipe joined Soundwave, Drag Strip and Weirdwolf in looking for their fallen leader. He was in the midst of voicing the selfish but sensible idea of abandoning Megatron to his fate and gettin' the slag outta there when Drag Strip found their commander. New Dawn

The Transformers cartoon

Voice actor:Stephen Keener (English), Eiji Itō (Japanese), Hans-Rainer Müller (German, The Rebirth, Part 2), Georges Atlas (European French)
I'd do anything this picture told me to.

After an all-too-brief period of peace following the resurrection of Optimus Prime, Mindwipe was one of the many Decepticons who attacked Autobot City in 2007, creating a huge distraction to cover for Pounce and Wingspan's theft of the key to the Plasma Energy Chamber. During the battle, Mindwipe used his hypnotic powers to seize control of Lightspeed's mind, forcing him to back away, over the edge of a nearby cliff and into the lake below. The Autobots were able to steal back the key, but not before it was used to briefly open the Plasma Energy Chamber, releasing a bolt of energy that hurled a group of Autobots, and the key, across the galaxy to the planet Nebulos. Mindwipe was among the Decepticons sent to Nebulos in pursuit of the key, and he participated in the first battle on the planet. When the Decepticons were forced to retreat, Scourge instructed them to gather the fallen Autobots so they could torture them for information on the key's location, but Mindwipe went one better, seizing the not-fallen-at-all Crosshairs and dragging him off. The Rebirth, Part 1

And we wonder why he hates science.

In order to rescue their allies, the Autobots became binary bonded to a group of Nebulan rebels as powerful "Headmasters". Defeated in battle by this new breed of Autobot, Mindwipe and the Decepticons were sent fleeing into the Nebulan wilderness, where they were then accosted by the mentally-controlled machines of Nebulos's evil ruling council, the Hive. The Decepticons were brought to the Hive's underground city, where their ruler, Lord Zarak, proposed that they join to become Headmasters as well. Naturally, the Decepticons were not responsive, until Zarak forced them to agree by attacking them with more of their machine servants. Mindwipe's head was subsequently removed and re-engineered to become an exo-suit for the Hive scientist Vorath, but the partnership had immediate teething troubles when Vorath bemoaned that fact that someone of his rank had been paired with a "mangy night-scavenger". Warning him that his mystic powers could trump anything science had to offer, Mindwipe mesmerized Vorath and forced him to combine with him.

Sure, make fun of Vorath's tiny guns, but, unlike magic, at least they're real.

The newly empowered Decepticons proceeded to attack the Autobots, and in the midst of the fighting, Mindwipe used his extra-sensory powers to deduce that the Autobot Brainstorm was carrying the Plasma Energy Chamber key. Mindwipe tried asking forcefully, but when that did work, he just let Vorath shoot the Autobot so they could steal the key from his fallen body. The Rebirth, Part 2

With the key recovered, the Decepticons and their new Hive allies all returned to Cybertron to help Galvatron overcome the Autobots on the planet. Mindwipe personally took down Ultra Magnus in the fight, with a little help from Weirdwolf and Apeface. Galvatron then used the key to open the chamber with the intention of destroy Cybertron and Earth; although his plan was foiled, as Mindwipe and the Decepticons were escaping aboard Scorponok, a stray tendril of plasma energy sent them hurtling off into the depths of space. The Rebirth, Part 3

Japanese Generation 1 cartoon continuity

In Japan, Mindwipe is a small robot who transforms into a head and controls a larger, lifeless transtector body. The Headmasters toyline referred to the small robot as "Vorath", with "Wipe" (the character's Japanese name) being the name of the combined small robot/transtector form, though this distinction of nomenclature is very rarely made in fiction.

The Headmasters cartoon

Voice actor:Naoki Tatsuta (Japanese), John Culkin (English, Omni Productions dub)
"Blah! Try my new chocolatey cereal! Blah!"

Four million years ago, Mindwipe was among a group of diminutive Cybertronians who departed Cybertron under the leadership of Fortress, seeking to escape the eternal war that had consumed the planet. The Mystery of Planet Master Alas, engine trouble caused their craft to crash land on the planet Master, an inhospitable world whose murderous environments and wildly changeable atmospheric conditions threatened the lives of the Cybertronians. Fighting against the odds, the Cybertronians made several technological quantum leaps to survive on Master, including the creation of larger, lifeless Transformer bodies called transtectors. Mindwipe and many of his comrades underwent rigorous physical and mental training to master the art of transformation, learning to transform into heads that connected to the transtectors and placed them under their control. The Four-Million-Year-Old Veil of Mystery Mindwipe was one of ten warriors who became one of these "Headmasters", but he and several other warriors soon began to chafe under Fortress's leadership, and ultimately rebelled against him, siding with a traitor named Scorponok. Via the Decepticon ninja Sixshot, Scorponok soon forged an alliance with the Decepticon leader, Galvatron. The Mystery of Planet Master

"Kōmori amamori oritatande WIPE!"

In the year 2011, Mindwipe and his fellow Headmasters Weirdwolf and Skullcruncher joined Galvatron's Decepticons in a massive attack on Cybertron, seeking to seize control of the destabilized Vector Sigma and establish a Decepticon hold over the planet through the computer. Mindwipe's hypnotic powers proved instrumental in the battle, as he put Grimlock to sleep, but Galvatron was less than impressed, since he also accidentally knocked out Skullcruncher with the technique. Undeterred, Mindwipe put his powers into play again a short time later, taking out Grimlock (again!), Slag and the Throttlebots, and leaving the entrance to the computer base wide open. Before the villains could enter, however, Fortress's Autobot Headmasters arrived on Cybertron to stop them. Four Warriors Come out of the Sky In ensuing battle, Mindwipe tried using his powers on Chromedome and Brainstorm, but the Decepticons were soon forced to retreat when faced with the power of Fortress's Battleship Maximus. The Mystery of Planet Master

Soon after, Mindwipe accompanied Galvatron to Earth to help interfere in the Autobots' search for the missing Matrix of Leadership, which could restore balance to Vector Sigma. The villains caught up to Hot Rod just as he located the Matrix, and Mindwipe tried using his hypnotism to stop him, only to be beaten down by Hardhead. Hot Rod and the Autobot Headmasters quickly pulled out, heading for Cybertron with the Matrix, and Mindwipe and the Decepticons gave chase, leading to a showdown in Vector Sigma's chamber. When Hot Rod used the Matrix to become Rodimus Prime again, he and Optimus Prime proceeded to double-team Galvatron, and Mindwipe and his fellow Decepticons fled rather than face the combined might of the two Primes. Birth of the Fantastic Double Prime

Mindwipe and his fellow Headmasters accompanied the Decepticons' newest weapon, the Madmachine, when it was unleashed upon Cybertron, and used his hypnosis to make Wideload and Rollbar fight each other. Goldbug helped his team-mates over come the effect, and he and Rollbar tried to haul Mindwipe out of the sky. The Decepticon Headmasters retreated when the Madmachine was destroyed by the Autobot Headmasters. The Great Cassette Operation After helping the Decepticons enslave planet Beast, Rebellion on Planet Beast Mindwipe again clashed with the Throttlebots, this time on Earth, as the Decepticons kept the Autobots busy there while their newest evil creation, Metamorphose, headed towards Athenia. Approach of the Demon Meteorite

Why talk funny I do now?

Under the orders of Scorponok, Mindwipe and the Decepticon Headmasters attacked an Autobot energy outpost on Earth and killed all those stationed there. Before slaying the last one, Mindwipe hypnotised him into sending a message to Rodimus Prime, summoning him and Ultra Magnus there with news of a new energy discovery. The two Autobots discovered the deception immediately upon arrival, but it was too late: Mindwipe and his partners in crime had already departed to the planet Praum, where they watched as Scorponok sprang a trap on the Autobot Headmasters. The Autobots overcame the trap by switching heads to boost their power, so the Decepticons tried doing the same (with Mindwipe taking over Weirdwolf's transtector), but they still got thrashed. The Four-Million-Year-Old Veil of Mystery After this defeat, Mindwipe and his chums decided to sit out the next mission, and played cards instead. Terror! The Six Shadows

I did it! I programmed a VCR!

When Vector Sigma developed a powerful new alloy named cybertonuron, Mindwipe was among the Decepticons who invaded the planet seeking to take possession of it for Galvatron. Scorponok, however, disliked the idea of seeing the alloy in Galvatron's hands, and so resolved instead to destroy Cybertron, and the cybertonuron with it. Cybertron Is in Grave Danger, Part 1 Mindwipe was given the unenviable task of transporting a clutch of bombs into Cybertron's depths, but even with the Autobot Headmaster nipping at his heels, he succeeded in his task, setting up the explosives in Vector Sigma's chamber and activating their timers. He taunted Chromedome with the fact that the bombs could not be deactivated, then transformed to bat mode and made himself scarce with all due speed. The Transformers all abandoned Cybertron with minutes to spare, save for Galvatron, who attempted to remove the bombs, and was caught at ground-zero of the planet-shattering explosion. Cybertron Is in Grave Danger, Part 2

For the last time, you little brat, stop peeing on my roses!

With Galvatron apparently destroyed, Scorponok seized command of the Decepticons, and as his first order of business, had Mindwipe, Weirdwolf and Skullcruncher steal the Sol 1 satellite. The Shadow Emperor, Scorponok As part of Scorponok's plan to keep the Autobots' attentions focused on Earth, Mindwipe and his fellow Headmasters accompanied the Predacons in menacing a small Peruvian village. To defeat the Autobots, Mindwipe threatened the life of a local boy named Pipiro, but was thwarted by the heroic actions of his sister Alyssa's pet donkey Dodo. Embarrassing. The Dormant Volcano Mysteriously Erupts A little later, still distracting the Autobots from Scorponok's bigger scheme with attacks on Earth, Mindwipe joined in a Decepticon attack on San Francisco with a giant carnivorous plant. Head On!! Fortress Maximus

Scorponok's plans came to a head when he plotted to destroy Mars, and sent the Decepticon Headmasters to the planet to place a series of plasma energy bombs beneath its crust. After doing so, the villainous trio engaged the Autobot Headmasters on Mars's surface, and fooled the heroes with an elaborate trick: during the fight, Hardhead managed to get Mindwipe to "accidentally" hypnotize Skullcruncher, so that he would reveal the location of the bombs. Following Skullcruncher's co-ordinates, the Autobots were actually caught in a pre-prepared trap, and when Fortress arrived to save them, Mindwipe and his fellows got particularly crafty, blasting him with a plasma energy bomb in mid-transformation! Explosion on Mars!! Maximus Is in Danger Mindwipe and the Decepticons then withdrew to the planet Zarak, where Scorponok unveiled his secret project: a giant transtector of his own! The Headmasters boarded Scorponok's giant new body and returned with him to Mars, putting the finishing touches to the bombing set-up while he battled Fortress. Mindwipe and his pals completed their task quickly, and withdrew with Scorponok as Mars was destroyed. Explosion on Mars!! MegaZarak Appears

Ooooh, right in the belfry!

Mindwipe subsequently attended the coronation of Scorponok on Earth's moon, but during the event, Galvatron abruptly returned, alive and well, and retook leadership. Return of the Immortal Emperor Soon after, when Galvatron had him try to interfere with an Autobot energy run to the planet Sandra, Mindwipe got kicked around by the Autobots pretty badly. Literally. SOS from Planet Sandra

This picture requires no caption....

The appearance of two unknown robots in the vicinity of Chaar set off alarm bells for Mindwipe and the Decepticon Headmasters, and they immediately attacked the newcomers. Mindwipe tried hypnotizing the one with a dragon alternate mode, but was promptly swatted out of the sky by its tail, at which point Galvatron arrived to inform them that these were actually their old allies from Master, Apeface and Snapdragon. Mindwipe, Weirdwolf and Skullcruncher subsequently went to Earth with the two new arrivals to serve as bodyguards for Galvatron while he negotiated a potential peace with Spike Witwicky. When the negotiation was discovered to be a Decepticon ruse to cover their preparations for an interplanetary energy raid, a fight broke out, and Mindwipe got smacked around by Kaen. The Decepticons were able to escape and join up with their fellows to carry out the planned raid, destroying the space bridge to delay the Autobots' pursuit. Daniel Faces His Biggest Crisis Ever!!

The Decepticons' started their energy gathering by attacking the planets HiveFight to the Death on Planet Hive!! and Twin Star for energy, but Mindwipe didn't really have much to get up to, and entertained himself on the journey between planets by watching asteroids fly by. Battle for Defense of the False Planet When the Decepticons set down on the planet Daros, however, Mindwipe and the Headmasters were given a special task by Scorponok: hunt down the BeastformerTekna, who had helped build his transtector and who was currently imprisoned in a penal colony on the planet, and kill him. Mindwipe and his foul fellows took pleasure in massacring every inmate in the prison, but Tekna managed to escape their wrath long enough to give the Autobots information on the secret weak spot of Scorponok's transtector. Find MegaZarak's Weak Spot!!

On the Decepticons' next target planet, Paradise, Mindwipe was able to get the drop on Highbrow and tried to subdue him with his hypnotic powers. Highbrow fought off the effect long enough to take to the air, so Mindwipe simply shot him down and allowed Weirdwolf and Skullcruncher to beat him into submission. When Highbrow's fellow Autobots came to rescue him, Skullcruncher seized Brainstorm so Mindwipe could punch him, but of course, Brainstorm ducked, and Mindwipe wound up clocking Skullcrunching instead. Head Formation of Friendship Mindwipe then participated in the battle on Pirate Planet. Mystery of the Space Pirate Ship

WhY mY knEEs hUrt?

Upon returning to Earth, Mindwipe and the Decepticon Headmasters were assigned a secret mission by Scorponok, sneaking into an Autobot outpost in their small robot mode and planting explosives that wiped out the base personnel. With the base now in their clutches, the trio planet one of a series of explosives beneath it. Ultra Magnus Dies!! They then helped Scorponok plant another explosive in Alaska, though their activities were briefly interrupted by a wide-ranging battle between Galvatron and the Autobot Headmasters that concluded with the Decepticon leader's destruction. The Emperor of Destruction Vanishes on an Iceberg This allowed Scorponok to take control of the Decepticons once again, and he dispatched Mindwipe and the Headmasters to plant the final explosive on the Kamchatka Peninsula. When detonated, this string of explosives would trigger every volcano on the Pacific Ring of Fire and tear the Earth apart. Luckily, before the bombs could be triggered, Chromedome deduced the location of Scorponok's weak spot, and Fortress Maximus exploited it to critically damage the Decepticon leader. Scorponok withdrew, and Mindwipe and all the other Decepticons fled Earth with him. I Risk My Life for Earth

Mindwipe and the Decepticons withdrew with Scorponok to their old home of planet Master, where they terrorized the locals until the Autobots arrived to stop them. Mindwipe tried hitting Highbrow with his hypnotic waves again, but Brainstorm stepped in to take him down. The Decepticons sought revenge by capturing Chromedome's friend Jack, planting a bomb in him, and having Mindwipe hypnotize him into walking into the heart of the Autobot camp. Chromedome was able to break Mindwipe's control over his friend, but alas, it did not save Jack's life. Miraculous Warriors, Targetmasters (Part 2)

Mindwipe, Weirdwolf and Skullcruncher observed as Scorponok processed G-Metal from ore found beneath Master's surface, which he would use to forge a weapon to counter Fortress's Master Sword. The three Headmasters then captured Wheelie and ransomed for information on the Master Sword, allowing Scorponok to counter its powers. The Master Sword Is in Danger!!

"Upff...hot dog burp."

With this data in their hands, the Decepticons briefly withdrew from Master to throw the Autobots off their tail, hiding in a field of space debris where Mindwipe and Weirdwolf teased Skullcruncher for confusing the space rocks with aliens. Their trail suitably cold, they then returned to Master and headed underground so Scorponok could build his "Zarak Shield". The Zarak Shield Turns the Tide A mighty battle against the Decepticons followed, with the Autobots backed up by the young warriors of Master, during which Mindwipe took a shot at hypnotizing Chromedome, only to once again accidentally put Skullcruncher to sleep. Luckily, Mindwipe's mistake did not adversely affect the fight, and with the battle won, the Decepticons quickly departed Master, heading back to Earth. Operation: Destroy the Decepticons On the way, they were separated from SixshotMy Friend Sixshot! and when they set down on the asteroid Pallas for repairs, Mindwipe opined that they wouldn't be seeing the ninja again. No sooner had he voiced this opinion, of course, than Sixshot returned. Duel on the Asteroid

The Brementown Mutations

Once back on Earth, Scorponok immediately had the Decepticons begin setting up "Death Towers" at strategic points around the globe, forged from the crysmagnetal of the lost continent of Lemuria. Attempting to defend the tower set up in Egypt, Mindwipe, Weirdwolf and Skullcruncher unveiled their newest technique: the "Trio Formation", an odd move that involved them stacking on top of each other like a totem pole. The mode was not exactly a success, as Apeface and Snapdragon thought to get in on the action and wound up toppling it over. The Headmasters' tomfoolery notwithstanding, the Egyptian Death Tower was successfully raised, creating an energy-channelling circuit of towers around the Earth that threatened to rip the planet apart. To stop the Decepticons' plan, the Autobots faced their whole army down in a massive final battle at the North Pole, during which Mindwipe turned his hypnotic powers on Chromedome and Hardhead, but was stopped by Twincast and his cassettes. The conflict came to a close when the Autobots all channeled their energy into Fortress Maximus, enabling him to destroy Scorponok's transtector and the circuit, saving the planet and forcing Mindwipe and all the other Decepticons to flee Earth for good. The Final Showdown on Earth (Part 2)

The Great Transformer War

In the closing days of 2010, while significant Transformer battles were being waged on Earth and across space, Mindwipe and the other Decepticon Headmasters were running from their Autobot counterparts on the planet Master. The Decepticons appeared to flee by leaping into a chasm, but moments later, they burst up through the ground to catch their pursuers unaware. Unfortunately for the villains, the Autobots were able to sneak away under cover of the dust cloud their emergence created, leaving Weirdwolf and Skullcruncher to mistakenly beat up Mindwipe. The Decepticons soon realized their error, but the Autobots then crashed into them in vehicle mode, ending the fight. At that point, the Decepticon ninja Sixshot arrived and instructed them to follow him back to Chaar. The Decepticons fled into space, but the Autobot Headmasters quickly gave chase aboard the battleship Maximus, heading for Cybertron rejoin the larger Transformer war. The Great Transformer War #3

The Headmasters manga

That's some real nice work, there, Mindwipe.

Shortly after their failure to seize control of Vector Sigma, Galvatron led Mindwipe and the Decepticon Headmasters in a rather direct plan to destroy Battleship Maximus: They would lure it into space, and then Mindwipe would, uh, just throw a bomb inside. Unsurprisingly, this plan met with failure when the speedy Chromedome was able to snatch the bomb right out of Mindwipe's hands before he even had the chance to hurl it. The Headmasters #1

Mindwipe and the Decepticons later made a return to planet Beast, where they were able to capture Highbrow and Brainstorm. Mindwipe and his partners wanted to take the two Autobots' transtectors for themselves, but their evil scheme was foiled by the Beastformers' human ally Kane, who threw rocks at them and got them to chase him, thereby giving the Autobots time to escape. During his flight, Kane tripped, and Mindwipe swooped into finish him off, but the Decepticon was abruptly blasted out of the sky by Hardhead. Great Decisive Battle of Planet Beast!

After attacking an Alaskan power plant, Mindwipe, Weirdwolf, and Skullcruncher were prepared for the arrival of the Autobot Headmasters. Catching their heroic foes unawares, the three villains were able to stop them from combining with their transtectors, leaving them seemingly helpless before the vicious animal modes of the trio's own bestial transtectors. Mindwipe and his allies proceeded to hunt the Headmasters through the Alaskan wilderness, until they were tricked into making a pounce that allowed the Autobots to leap over them and rejoin with their transtectors. Mindwipe and company could only watch and hug each other for support as the four Autobots blasted the everlovin' bejeezus out of the Decepticons' unmanned transtectors. The Headmasters #2

In addition to all their other faults, Mindwipe and the Decepticon Headmasters proved themselves to be a bunch of Scrooges when they attacked a Christmas party the Autobots were attending at a Japanese children's hospital. Mindwipe viciously blasted Highbrow as the Autobot was attempting to protect a crippled boy named Masato, and while he did injure Highbrow, his evil deed did have a good side-effect: Masato was so consumed with emotion at the sight of Highbrow's wounding, he found the strength of will to walk again! This didn't mean anything to Mindwipe, who came zeroing back in to finish the fleshie off, but Chromedome quickly leapt into action and shot Mindwipe down. The three Decepticons were then sent packing, hurled off into the snow-flecked night by the united power of the Autobot Headmasters' "Hell's Great Wheel" technique. The Headmasters #6

Super-God Masterforce cartoon continuity

Super-God Masterforce cartoon

Archive footage of a battle in space involving Mindwipe was used to educate the future Headmaster Juniors about the Master Wars. Birth! Headmaster Jrs

Legends comic

Brainstorm explained to the Legends World's Rattrap that he and other diminutive Transformers, including Vorath before he became Mindwipe, left Cybertron to escape the war and settled on Master. Vorath was among those seen suffering in the deadly environment before they were able to tame the planet. Bonus Edition Vol. 9

In 2021, Mindwipe and the rest of Scorponok's troops battled Fortress Maximus's Autobots near a black hole when Scorponok detonated it with no concern for his own soldiers, causing miniature black holes to appear within them. Their bodies doomed by the phenomenon, the Headmasters disconnected from their transtectors and traveled to the Legends World to have new ones built for them. Headmaster Chapter Prologue They allied with the local Tera-Kura Co., with Mindwipe joining Tarantulas in the partnership's science department. At some point, he also hypnotized Skullcruncher so badly he developed chronic narcolepsy. Bonus Edition Vol. 22 Mindwipe was present when the chairman of the company, Megatron, was manipulated by Tarantulas and Weirdwolf into turning into Galvatron. Bonus Edition Vol. 23 After their Galvatron transtector failed to properly control Megatron, the two schemers had Mindwipe hypnotize Megatron into "willingly" merging with it. Bonus Edition Vol. 25

Mindwipe was present at Tera-Kura when a mob mass-produced Nightbeat-type Headmasters showed up and demanded their own transtectors. Bonus Edition Vol. 28 Following a failed bid by Weirdwolf to take over Fortress Maximus, Mindwipe and Skullcruncher carried their knocked out comrade to safety. Bonus Edition Vol. 31 With Tarantulas in jail, Mindwipe decided to make a deal with the devil for a new transtector by using dark magic to resurrect Devil Z. The entity granted him a new body and a Servant partner drone that boosted his hypnotic powers to the point where he was able to take over not just the Decepticon company, but the world! Unfortunately, he found it unsatisfying to be a leader when his subjects were just brainwashed and didn't actually have any love for him. When Devil Z asked for repayment in the form of the life force of his hypnotized victims, Mindwipe refused on the grounds that some of them were his friends, but Devil Z wouldn't take no for an answer and took Mindwipe's body for himself. Aided by Megatron, who was immune to his hypnotism by now, Mindwipe was able to neutralize Devil Z with a spell before any harm was done. Bonus Edition Vol. 34

However, Devil Z soon returned, took control of Decepticon Co., and ordered the three Headmasters to destroy a trailer that had appeared in the city. The vehicle turned out to be a Godmaster transtector belonging to Ginrai, who swiftly defeated Mindwipe and his pals with Chōkon Power, forcing Devil Z to power them up with his own Devil Power. The trio appeared to defeat Ginrai with their Trio Formation Devil Attack, only for the Autobot to transform into a base mode and send them fleeing with a barrage of blasts. Bonus Edition Vol. 35 When Devil Z was enraged, he began emitting lighting blasts that Mindwipe and the others had to dodge. Bonus Edition Vol. 36 His living transtector was later lured away by the former Decepticon Headmaster Juniors and briefly used by Cancer before it was retrieved. Bonus Edition Vol. 37 He and Weirdwolf were shocked when Apeface developed superpowers and an unhealthy dose of ambition. Bonus Edition Vol. 38

Mindwipe was later seen loading energon cubes at work. Bonus Edition Vol. 40 Tired of Megatron's command, he and the other Headmasters began plotting to revive Trypticon and use him to overthrow Megatron. LG-43 Trypticon Project Chapter For this purpose, they worked with Nightbeat to steal Godbomber from the Autobots, though they were defeated before they could really use its power. Bonus Edition Vol. 42 They then had to sit tight for a while when the Dimensional Patrol sealed away everyone's transtectors. Bonus Edition Black Convoy 2 Mindwipe and his two buds tried to take revenge on Ginrai when he turned into a normal robot, but fled when he transformed into the giant Super Magna Convoy. Bonus Edition Magna Convoy As they neared completion on Trypticon, the Scorponok loyalists rebelled against Tera-Kura and attacked Metroplex for his transformation cog. During the battle with Megatron's troops, Mindwipe confused Ultra Magnus with hypnotism. Decepticon Civil War He was stationed inside Trypticon during his rampage across Tokyo, though the beast was ultimately defeated. Bonus Edition Vol. 43

The Decepticon Headmasters gave up on their scheming and instead established the Wolf Circus, where Mindwipe was seen when Megatron visited to borrow Noble. Bonus Edition Vol. 63 When Devil Z resurfaced and took control of Scorponok in the form of BlackZarak, Mindwipe attempted to banish him again but was sent flying by a strike from BlackZarak's spear. Bonus Edition Grand Maximus

Scorponok was at the Wolf Circus when Scorponok returned and announced his plans to destroy the Legends World and claim its energy. He joined Scorponok on the bridge of his bestial new creation, Majin arak, as it prepared to lay waste to New Akihibara City. Mindwipe helped gather the six Targetmaster partners, whom Scorponok had modified into immensely powerful plasma energy bombs. Legends World in Imminent Danger! Part One He prepared the Targetmaster bombs to detonate at the site of the proton energy reservoir. Even after Scorponok and Majin Zarak were defeated, Mindwipe continued prepping the bombs. It was revealed that Devil Z was inhabiting and directing Mindwipe's transtector. Slugslinger and the Targetmasters arrived, blowing the transtector apart and separating Mindwipe from Devil Z's influence.

Источник: [https://torrent-igruha.org/3551-portal.html]
, Idea Cruncher Version: 1.3.1 serial key or number

31984D0157

+++++

( 1 ) OJ NO L 67 , 9 . 3 . 1984 , P . 54 .

COUNCIL DECISION OF 28 FEBRUARY 1984 ADOPTING THE 1984 WORK PROGRAMME FOR A EUROPEAN PROGRAMME FOR RESEARCH AND DEVELOPMENT IN INFORMATION TECHNOLOGIES ( ESPRIT ) ( 84/157/EEC )

THE COUNCIL OF THE EUROPEAN COMMUNITIES ,

HAVING REGARD TO THE TREATY ESTABLISHING THE EUROPEAN ECONOMIC COMMUNITY ,

HAVING REGARD TO COUNCIL DECISION 84/130/EEC OF 28 FEBRUARY 1984 CONCERNING A EUROPEAN PROGRAMME FOR RESEARCH AND DEVELOPMENT IN INFORMATION TECHNOLOGIES ( ESPRIT ) ( 1 ) , AND IN PARTICULAR ARTICLE 3 ( 2 ) THEREOF ,

HAVING REGARD TO THE DRAFT WORK PROGRAMME SUBMITTED BY THE COMMISSION ,

WHEREAS , AT TALKS ORGANIZED BY THE COMMISSION SERVICES , INDUSTRY AND THE ACADEMIC WORLD HAVE GIVEN THEIR OPINIONS ON THE CONTENT OF THE PROJECTS TO BE LAUNCHED IN 1984 AND THEIR PRIORITY ;

WHEREAS THE PROJECTS FORMING PART OF THE WORK PROGRAMME FALL WITHIN THE AREAS DESCRIBED IN THE ANNEX TO THE DECISION ON THE ESPRIT PROGRAMME ,

HAS DECIDED AS FOLLOWS :

ARTICLE 1

THE ESPRIT WORK PROGRAMME AS SET OUT IN THE ANNEX IS HEREBY ADOPTED FOR 1984 .

ARTICLE 2

THIS DECISION SHALL TAKE EFFECT ON THE DAY OF ITS PUBLICATION IN THE OFFICIAL JOURNAL OF THE EUROPEAN COMMUNITIES .

DONE AT BRUSSELS , 28 FEBRUARY 1984 .

FOR THE COUNCIL

THE PRESIDENT

L . FABIUS

ANNEX

ESPRIT 1984 WORKPLAN

INTRODUCTION AND OVERALL RESOURCES BREAKDOWN

1 . ADVANCED MICROELECTRONICS

1.1 . SUBMICRON MOS

1.2 . SUBMICRON BIPOLAR

1.3 . CAD

1.3.1 . CAD MANAGEMENT

1.3.2 . HIGH LEVEL DESIGN AND LAYOUT

1.3.3 . RESEARCH : ADVANCED / INNOVATIVE CAD

1.3.4 . ESTABLISHMENT OF CAD CENTRES

1.4 . COMPOUND SEMICONDUCTOR MATERIALS AND INTEGRATED CIRCUITS

1.5 . OPTOELECTRONICS

2 . SOFTWARE TECHNOLOGY

2.1 . PROCESS - UNDERSTANDING AND IMPLEMENTING

2.1.1 . METHODS , TECHNIQUES AND TOOLS

2.1.2 . INTEGRATION OF MANAGEMENT AND TECHNICAL ASPECTS

2.1.3 . SOFTWARE METHODOLOGY

2.2 . MANAGEMENT - UNDERSTANDING AND IMPLEMENTING

2.2.1 . SOFTWARE PRODUCTION AND MAINTENANCE MANAGEMENT SUPPORT

2.2.2 . QUALITY AND RELIABILITY ASSURANCE

2.3 . ENVIRONMENT - UNDERSTANDING AND IMPLEMENTING

2.3.1 . COMMON TOOL ENVIRONMENT

2.4 . PROCESS , MANAGEMENT AND ENVIRONMENT - EVOLVING

2.4.1 . ADVANCED INTERACTIVE SOFTWARE DEVELOPMENT

2.5 . DEMONSTRATION PROJECTS

3 . ADVANCED INFORMATION PROCESSING ( AIP )

3.1 . KNOWLEDGE ENGINEERING

3.1.1 . KNOWLEDGE-BASED SYSTEMS AND THEIR METRICATION

3.1.2 . DIALOGUE AND NATURAL LANGUAGE

3.1.3 . KNOWLEDGE REPRESENTATION AND INFERENCE TECHNIQUES

3.1.4 . PREPARATORY WORK TOWARDS ADVANCED KBS

3.2 . EXTERNAL INTERFACES

3.2.1 . BASIC SIGNAL ANALYSIS AND RECOGNITION

3.2.2 . RECOGNITION OF SPECIFIC SIGNALS

3.3 . INFORMATION AND KNOWLEDGE STORAGE

3.3.1 . INTERFACE BETWEEN STORAGE AND ENVIRONMENT

3.3.2 . DATA AND KNOWLEDGE BASES

3.3.3 . STORAGE STRUCTURES AND ARCHITECTURES

3.3.4 . NEW GENERATION KNOWLEDGE MACHINE

3.3.5 . MEDIUM-TERM RESEARCH ON STORAGE MEDIA

3.3.6 . LONG-TERM RESEARCH ON OPTICAL DISCS

3.3.7 . LONG-TERM RESEARCH ON BIOLOGICAL STORAGE

3.4 . COMPUTER ARCHITECTURES

3.4.1 . ULTRACOMPUTER , MULTIPROCESSOR MACHINES

3.4.2 . NON-VON NEUMANN ARCHITECTURES

3.5 . DESIGN AND SYSTEM ASPECTS

3.6 . FOCUSING PROJECTS

4 . OFFICE SYSTEMS

4.1 . OFFICE SYSTEMS SCIENCE AND HUMAN FACTORS

4.1.1 . OFFICE SYSTEMS ANALYSIS

4.1.2 . OFFICE SYSTEMS DESIGN

4.1.3 . HUMAN FACTORS

4.2 . ADVANCED WORKSTATIONS

4.2.1 . SYSTEM ASPECTS OF WORKSTATION DESIGN

4.2.2 . VISION INTERFACE

4.2.3 . PAPER INTERFACE

4.2.4 . SPEECH INTERFACE

4.2.5 . OFFICE LANGUAGES

4.3 . COMMUNICATION SYSTEMS

4.3.1 . COMMUNICATION SYSTEM ARCHITECTURE

4.3.2 . OPTICAL WIDEBAND LAN

4.3.3 . MULTI-MODE MESSAGING

4.3.4 . ISDN-BASED ADVANCED VIDEOTEX

4.4 . ADVANCED FILING AND RETRIEVAL SYSTEMS

4.4.1 . OFFICE INFORMATION SERVER DESIGN AND EVALUATION

4.4.2 . SYSTEM ISSUES

4.4.3 . USAGE AND NEEDS

4.4.4 . COMPONENTS

4.5 . INTEGRATED OFFICE SYSTEM DESIGN AND EVALUATION

4.5.1 . ADVANCED INTEGRATED OFFICE SYSTEM PROTOTYPES

4.5.2 . OFFICE SYSTEM TEST AND EVALUATION ENVIRONMENTS

5 . COMPUTER INTEGRATED MANUFACTURE

5.1 . INTEGRATED SYSTEM ARCHITECTURE

5.1.1 . SYSTEM ARCHITECTURES

5.1.2 . GRAPHICS SUBSYSTEMS

5.2 . CAD/CAE

5.2.1 . CAD/CAE FACILITIES FOR PRODUCT AND PROCESS DESIGN

5.2.2 . USE OF AI TECHNIQUES IN CAD/CAE

5.3 . COMPUTER-AIDED MANUFACTURING ( CAM )

5.4 . MACHINE CONTROL SYSTEMS

5.4.1 . FLEXIBLE MACHINING SYSTEMS

5.4.2 . AUTOMATED ASSEMBLY AND ROBOTICS

5.4.3 . PLANT AVAILABILITY AND QUALITY OPTIMIZATION

5.5 . SUBSYSTEMS AND COMPONENTS

5.5.1 . IMAGE PROCESSING

5.5.2 . MICROELECTRONIC SUBSYSTEMS

5.5.3 . SENSOR PROGRAMMING AND STANDARDS

5.6 . CIM SYSTEMS APPLICATIONS

5.6.1 . CIM APPLICATION AND DEVELOPMENT CENTRES

5.6.2 . MISCELLANEOUS SUPPORT FOR CIM

INTRODUCTION AND OVERALL RESOURCES BREAKDOWN

THIS SECTION IS AN INTEGRAL PART OF THE ESPRIT 1984 WORKPLAN . IT IS NECESSARY FOR A PROPER READING AND INTERPRETATION OF THE PLAN

1 . BACKGROUND

IN SUCH A FAST-MOVING SECTOR AS IT , WHERE THE AVERAGE LIFE OF A PRODUCT IS THREE YEARS , IT WOULD BE ILLUSORY AND MISLEADING TO TRY AND DEFINE AB INITIO DETAILED ACTIVITIES , OBJECTIVES AND TIME SCHEDULES FOR THE NEXT FIVE YEARS TO COME AND PARTICULARLY SO IN THE CASE OF R AND D ACTIVITIES , ALTHOUGH OF INDUSTRIAL AND ESSENTIALLY APPLIED CHARACTER .

THIS IS WHY IT WAS DECIDED THAT A MORE DETAILED PROGRAMME OF WORK , NECESSARY FOR THE DAY-TO-DAY IMPLEMENTATION OF THE PROGRAMME BE ESTABLISHED , AND UPDATED AS REQUIRED , BY THE COMMISSION IN CONSULTATION WITH THE ESPRIT MANAGEMENT COMMITTEE , AND SUBMITTED EVERY YEAR TO THE COUNCIL FOR APPROVAL . THIS IS THE FIRST OF SUCH A " WORKPLAN " .

2 . PURPOSE AND CHARACTERISTICS

THE MAIN PURPOSE OF THE WORKPLAN IS TO PROVIDE THE ESSENTIAL GUIDING ELEMENTS FOR :

( A ) DRAFTING THE CALL FOR PROPOSALS AND SELECTING THE PROJECTS TO BE LAUNCHED ;

( B ) REVIEWING THE INDIVIDUAL PROJECTS UNDER WAY AND APPRAISING THEIR PROGRESS INDIVIDUALLY AND IN RELATIONSHIP WITH EACH OTHER ;

( C ) COORDINATING ACTIVITIES UNDER ESPRIT AND UNDER NATIONAL PROGRAMMES IN THE DOMAIN OF IT ;

( D ) APPRAISING THE RESULTS OF THE PROGRAMME AS A WHOLE AND FOR EVALUATING THE SUITABILITY OF ITS TECHNICAL OBJECTIVES AND OPTIONS TO MEET THE STRATEGIC OBJECTIVES : THIS TO BE DONE IN THE LIGHT OF THE RESULTS OBTAINED AND OF THE TECHNOLOGICAL PROGRAMMES WITHIN AS WELL AS OUTSIDE THE COMMUNITY ;

3 . TYPE A AND TYPE B PROJECTS

THE EXISTENCE AND NEED OF DIFFERENT KINDS OF PROJECTS FOR THE CHARACTER OF WORK AND THE SIZE OF EFFORT IN A LARGE R AND D PROGRAMME OF THE NATURE OF ESPRIT IS A FACT OF LIFE ; INDUSTRIAL R AND D IS ACKNOWLEDGED TO REST ESSENTIALLY ON TWO BROAD CLASSES OF PROJECTS :

( A ) PROJECTS THAT REQUIRE LARGE INFRASTRUCTURE AND RESOURCES , BOTH HUMAN AND FINANCIAL , AS WELL AS CLEAR AND CONSTANT STRATEGIC PERSPECTIVE TO ENSURE CONTINUITY OF ACTIONS AND THE BREADTH NECESSARY TO REAP THE LONG-TERM BENEFITS . SUCH MEDIUM - TO LONG-TERM " SYSTEM DRIVEN " R AND D ACTIVITIES , THAT WILL BE REFERRED TO IN THIS DOCUMENT AS " TYPE A PROJECTS " , WILL REPRESENT THE BACKBONE OF ESPRIT AND ARE EXPECTED TO ACCOUNT FOR SOME 75 % OF THE OVERALL RESOURCES .

( B ) PROJECTS THAT RELY MAINLY ON FLEXIBLE INFRASTRUCTURE AND ON INDIVIDUAL THINKING RATHER THAN ON A SYSTEM APPROACH , AND REQUIRE RELATIVELY MUCH SMALLER RESOURCES . SUCH ACTIVITIES , THAT WILL BE REFERRED TO AS " TYPE B PROJECTS " , COULD RANGE FROM VERY LONG TERM , VERY SPECULATIVE R AND D TO SPECIFICALLY ORIENTED R AND D WHICH MAY WELL BE OF SHORTER DURATION , AND ARE EXPECTED TO ACCOUNT FOR SOME 25 % OF THE OVERALL EFFORT UNDER ESPRIT .

ESSENTIALLY TYPE A PROJECTS ARE BIG , COMPLEX , SYSTEM-ORIENTED WHEREAS TYPE B ARE SMALL , IDEA-ORIENTED .

THE PURSUIT OF STRATEGIC OBJECTIVES OF COMMON INTEREST REQUIRES THAT A NUMBER OF WELL-IDENTIFIABLE MAJOR ACTIVITIES ARE UNDERTAKEN AND CARRIED OUT ACCORDING TO AGREED , AND IN SOME CASES MUTUALLY INTERRELATED , TIME SCHEDULES : TYPE A PROJECTS ARE DESIGNED TO REPRESENT THESE ACTIVITIES . IN ORDER TO OPTIMIZE RESOURCES ALLOCATION AND PROVIDE THE BEST CONDITIONS FOR SYNERGISM , PROPOSALS FOR SUCH PROJECTS WILL THEREFORE HAVE TO BE INVITED GIVING DETAILED DESCRIPTIONS PRECISELY IDENTIFYING THE OBJECTIVES AND THE EXPECTED TIMING OF THEIR ACHIEVEMENT , WHEREAS THE PARTICULAR APPROACH OR THE CHOICE OF THE TECHNOLOGICAL OPTIONS WOULD , AS A RULE , BE LEFT FREE .

SINCE B TYPE PROJECTS ARE DESIGNED TO FILL POSSIBLE GAPS BETWEEN THE A TYPES ; TACKLING ALTERNATIVE , OR COMPLEMENTARY , OR MORE SPECULATIVE ACTIVITIES , THE LEVEL OF THEIR DEFINITION WOULD HAVE TO BE MORE GENERAL IN ORDER TO MAXIMIZE FLEXIBILITY AND LEAVE MAXIMUM ROOM TO INNOVATION . THEIR TIMING , AS A RULE , COULD ALSO BE LESS CRITICAL . AS A CONSEQUENCE AND BY CONTRAST WITH THE TYPE A PROJECTS THEY ARE IDENTIFIED AS A RULE BY THEME ONLY .

THE DEFINITION OF THE TYPE A PROJECTS AND THE IDENTIFICATION OF R AND D DOMAINS WITHIN WHICH TYPE B PROJECTS MAY BE INVITED IS DONE IN THE LIGHT OF THE OBJECTIVES TO BE PURSUED . IN PRINCIPLE , TOPICS THAT ARE PART OF A SPECIFIC TYPE A PROJECT COULD ALSO BE THE SUBJECT OF A TYPE B PROJECT IF , E . G . , THIS LATTER PROPOSES AN ALTERNATIVE OR COMPETITIVE APPROACH . WHENEVER TYPE B PROJECTS WILL BE PROPOSED IN SUCH CONDITIONS , HOWEVER , CARE MUST BE TAKEN TO VERIFY THAT SUCH " SUBSET OF TYPE A PROJECT " REMAINS MEANINGFUL IN ITSELF .

THE WORKPLAN CLEARLY INDICATES THE DISTINCTION BETWEEN TYPE A AND TYPE B , SO THAT THE PROPOSERS WILL BE ABLE AND WILL HAVE TO DECIDE BEFOREHAND TO GO EITHER FOR ONE OR THE OTHER . NO AMBIGUITY SHOULD THEREFORE BE POSSIBLE AS FOR THE CHARACTER OR " TYPE " OF ANY PARTICULAR PROPOSAL DURING THE SELECTION PHASE .

BOTH TYPES OF PROJECTS WILL BE PART OF THE SAME STRATEGIC EFFORT , PLANNED AND EXECUTED AS ONE SINGLE PROGRAMME .

4 . PILOT PROJECTS

THE CONTINUATION OF THE WORK STARTED UNDER THE PILOT PHASE IS POSSIBLE WITHIN THE ESPRIT PROGRAMME AND IS , IMPLICITLY OR EXPLICITLY , INCLUDED IN THE PROGRAMMES THAT ARE PRESENTED IN THE WORKPLAN . SUCH CONTINUATION WILL NOT , AS A RULE , BE SUBJECT TO A CALL IN THE STRICT SENSE BUT THE INFORMATION AND THE PROPOSALS GATHERED IN RESPONSE TO THE GENERAL CALL FOR PROPOSAL MAY GENERATE MODIFICATIONS , WIDENING OR RESTRICTION ( DOWN TO ZERO IN EXTREME CASES ) OF THE ACTIVITIES ORIGINALLY FORESEEN IN ORDER TO MAKE THE PROJECTS BEST SUITED TO MEET THE REVISED OBJECTIVES AS THEY EMERGE FROM THE WORKPLAN .

5 . VALIDITY OF THE WORKPLAN

IN ORDER TO FULFIL ITS PURPOSE THE WORKPLAN HAS TO SPAN OVER A REASONABLY LONG TIME , SUFFICIENT TO COVER THE WHOLE DURATION OF THE LONGER PROJECTS : FIVE YEARS WAS DEEMED TO OFFER A GOOD COMPROMISE BETWEEN AN ACCURATE AIMING ( THAT WOULD SUGGEST SHORT-TERM INDICATIONS ) AND DEVELOPMENT PERSPECTIVE THAT REQUIRE A WIDER TIME FRAMEWORK . SUCH FIVE-YEAR WORKPLAN SHALL BE REVIEWED EVERY YEAR .

IN THE LIGHT OF THE ABOVE , THE INTERMEDIATE OBJECTIVES THAT ARE SET IN THE WORKPLAN ARE SPECIFIED WITH LESS ACCURACY AND RELIABILITY AS ONE LOOKS FURTHER AHEAD AT THE YEARS TO COME AND IT MUST BE STRESSED THAT SINCE WE ARE AIMING AT A MOVING TARGET EVEN THE CLOSEST OBJECTIVES MAY HAVE TO BE CHANGED IN THE COURSE OF THE YEARLY REVISION , IN THE LIGHT OF THE PROGRESS WITHIN AS WELL AS OUTSIDE ESPRIT , IN EUROPE AND ELSEWHERE IN THE WORLD .

SUCH INTERMEDIATE OBJECTIVES ARE , HOWEVER , NECESSARY TO PROVIDE THE GENERAL FRAMEWORK AND AS REFERENCE FOR EVALUATION AND FURTHER REFINEMENTS . THEY HAVE THEREFORE BEEN INTRODUCED IN THE WORKPLAN AS THE BEST ESTIMATE OF HOW , AT THIS MOMENT IN TIME , THE OVERALL END-OBJECTIVES OF THE PROJECT MAY BE ACHIEVED . IN THIS SENSE THEY ARE INDICATORS AND FIRST-INSTANCE CHECK POINTS AGAINST WHICH CONCRETE PROPOSALS WILL BE EXAMINED . THEY SHALL , HOWEVER , NOT BE REGARDED AS LEGALLY MANDATORY IN THE SENSE THAT PROPOSALS HAVING PERFECTLY ACCEPTABLE END OBJECTIVES SHOULD NOT HAVE TO BE REJECTED ONLY BECAUSE THEY MAY HAVE BEEN DESIGNED TO REST ON DIFFERENT INTERMEDIATE OBJECTIVES THAN THE ONES PROPOSED HERE .

6 . ESTIMATED RESOURCES REQUIRED AND EXISTING CAPABILITIES

THE WORKPLAN FORMULATES A NUMBER OF OBJECTIVES AND PROPOSES A NUMBER OF R AND D THEMES . EACH OF THE MAIN OBJECTIVES WOULD REQUIRE FOR THEIR ACHIEVEMENT PROJECTS OF A MINIMUM " CRITICAL SIZE " THAT WOULD AT THE SAME TIME JUSTIFY THE JOINT TRANSNATIONAL APPROACH AND GIVE IT REASONABLE CHANCES OF SUCCESS .

THE PREPARATION OF THE DRAFT WORKPLAN IN THE PRESENT FORM IS BASED ON THE WORK OF SOME 300 SPECIALISTS SELECTED AMONGST ALMOST 1 000 WHO HAD DECLARED THEIR READINESS TO CONTRIBUTE IN INDUSTRY , UNIVERSITY AND RESEARCH CENTRES .

IT WAS THE COMMON FEELING OF THESE CONTRIBUTORS WHO HAD THE OPPORTUNITY OF MEETING AND DEBATING THEIR VIEWS , THAT THE WORK THAT IS BEING PROPOSED IS BOTH ESSENTIAL FOR THE ACHIEVEMENT OF THE ESPRIT OBJECTIVES AND REALISTIC IN TERMS OF THE AVAILABILITY IN EUROPE OF THE FINANCIAL AND HUMAN RESOURCES AS WELL OF THE SCIENTIFIC CAPABILITIES TO CARRY IT OUT SUCCESSFULLY .

7 . RESOURCES BREAKDOWN PER AREA

THE OVERALL WORKPLAN BREAKDOWN OF RESOURCES PER AREA , AS RESULTS FROM THIS FIRST ITERATION IS , LIKE ALL OTHER ELEMENTS OF THE WORKPLAN , SUBJECT TO MODIFICATION IN THE LIGHT OF THE RESPONSE AS WELL AS OF THE TECHNOLOGICAL EVOLUTION ON THE WORLD SCENE THAT MIGHT INFLUENCE INDUSTRIES .

IN PARTICULAR , IN THE LIGHT OF THE RESPONSE OF INDUSTRY TO THE FIRST CALL FOR PROPOSALS , TRANSFER OF RESOURCES BETWEEN THE VARIOUS AREAS WILL HAVE TO BE POSSIBLE .

WITH RESPECT TO THIS IT IS TO BE STRESSED THAT THE PARTITIONING OF ESPRIT INTO FIVE AREAS IS INCIDENTAL AND DICTATED BY MANAGEMENT NEED AND NOT INTRINSIC TO THE NATURE OF WORK , THAT ON THE CONTRARY IS VERY HEAVILY INTERRELATED .

THIS WILL BECOME MORE APPARENT IN THE NEXT ISSUES OF THE WORKPLAN WHEN SUFFICIENT DATA CONCERNING WORK ACTUALLY UNDER WAY WILL BE AVAILABLE .

THE FOLLOWING BREAKDOWN OF RESOURCES IS NOW ENVISAGED :

( MAN-YEARS )

SUBPROGRAMME * PROJECT YEAR

* 1 * 2 * 3 * 4 * 5 * TOTALS

1 . MICROELECTRONICS * 186 * 258 * 360 * 410 * 456 * 1 670

2 . SOFTWARE * 177 * 317 * 343 * 318 * 285 * 1 440

3 . AIP * 140 * 281 * 392 * 441 * 441 * 1 695

4 . OFFICE SYSTEMS * 210 * 310 * 440 * 390 * 100 * 1 450

5 . CIM * 121 * 216 * 215 * 220 * 172 * 944

SUBPROGRAMME 1

ADVANCED MICROELECTRONICS

THE MAJOR THRUST OF THIS SUBPROGRAMME IS AIMED AT MAJOR IMPROVEMENT OF THE MOST IMPORTANT STRATEGIC TECHNOLOGICAL AREAS . HENCE :

1 . MOST OF THE RESOURCES ARE CONCENTRATED INTO A LIMITED NUMBER OF PROJECTS IN THE AREAS OF SILICON MOS AND BIPOLAR FOR VLSI AND VERY HIGH PERFORMANCE CIRCUITS . WHEN ASSOCIATED WITH CAD , CAM AND MANUFACTURING METHODS , THESE WILL PRODUCE BY 1988 A SOUNDLY BASED 1 UM INDUSTRIAL CAPABILITY IN A NUMBER OF EUROPEAN SITES , WITH SUBSTANTIAL PROGRESS TOWARDS SUBMICRON CAPABILITY . ALSO IN PLACE WILL BE THE NECESSARY CAD TO DESIGN THE VLSI CIRCUITS WHICH USE THESE TECHNOLOGIES .

2 . COMMUNICATION WITH THE OTHER SUBPROGRAMMES MUST BE ENHANCED TO ENSURE RAPID DEFINITION OF THE REQUIRED VLSI DEMONSTRATOR CIRCUITS , THEIR DESIGN AND FABRICATION , FUNDED FROM THOSE SUBPROGRAMMES .

3 . THE SUPPLY OF SCIENTIFIC AND ENGINEERING PERSONNEL TRAINED IN THE SPECIFIC SKILLS OF MICROELECTRONIC PROCESS TECHNOLOGY AND DESIGN MUST BE INCREASED .

4 . SUPPORT FOR LONGER RANGE , LOWER PRIORITY , WORK ON TWO IMPORTANT AREAS WHICH COMPLEMENT SILICON VLSI , NAMELY COMPOUND SEMICONDUCTOR INTEGRATED CIRCUITS AND OPTOELECTRONICS ARE PROVIDED FOR AT A REASONABLE LEVEL .

IN ADDITION TO THE MAIN THRUST OF THE PROGRAMME AS OUTLINED IN 1 , 3 AND 4 ABOVE AND EXPANDED IN SECTION A BELOW , PROVISION IS ALSO MADE FOR A NUMBER OF SMALLER PROJECTS . SUITABLE TOPICS FOR THESE ARE IDENTIFIED IN SECTION B .

IN SECTION A THE MOS AND BIPOLAR PROCESS PROGRAMMES HAVE BEEN STRUCTURED AS LARGE INTEGRAL PROJECTS RATHER THAN SEPARATE 1 UM AND SUBMICRON PROJECTS . THIS HAS BEEN DONE IN THE INTERESTS OF EFFICIENCY , AS EACH LEVEL OF TECHNOLOGY BUILDS ON THE KNOWLEDGE GAINED FROM PRECEDING LEVELS . MANY ASPECTS OF A GIVEN PROCESS REQUIRE CONTINUOUS IMPROVEMENT , E . G . OXIDE QUALITY IN MOS , TO MEET THE DEMANDS OF EACH SUCCESSIVE DEVELOPMENT OF GREATER CIRCUIT COMPLEXITY AND HIGHER PERFORMANCE . THIS CONTINUITY OF EFFORT WOULD BE LOST BY HAVING SEPARATE PROJECTS AT 1 UM AND SUBMICRON LEVELS . THE OTHER PART OF THE PROCESS STRATEGY IS TO INCLUDE WORK ON TECHNIQUES COMMON TO MOS AND BIPOLAR IN THE MOS PROJECT , AS MOS TENDS TO LEAD IN SCALE OF INTEGRATION AND FEATURE SIZE , AND HENCE DOES THE PIONEERING WORK IN AREAS LIKE LITHOGRAPHY .

THE IMPORTANCE OF CAD FOR VLSI TO THE OVERALL SUCCESS OF ESPRIT CANNOT BE OVEREMPHASIZED . GIVEN GOOD TECHNOLOGY AND OUTSTANDING CAD , THE HIGHLY INNOVATIVE EUROPEAN INDUSTRY WILL BE ABLE TO COMPETE SUCCESSFULLY IN WORLD MARKETS .

A TOTAL OF SOME 13 CAD PROJECTS , INVOLVING ABOUT 650 MAN-YEARS BETWEEN 1983 AND 1987 , SUPPORTED UNDER THE MICROELECTRONICS REGULATION ( EEC ) NO 3744/81 AND THE ESPRIT PILOT PHASE ARE ALREADY UNDERWAY . THE INTEGRATION OF THE RESULTS OF THIS WORK AND EXTENSION TO COVER THE DESIGN OF SEVERAL MILLION COMPONENT CIRCUITS FOR A WIDE RANGE OF APPLICATIONS IS A MAJOR TASK . A CAREFUL ANALYSIS SHOULD BE CARRIED OUT DURING THE FIRST SEMESTER OF 1984 TO ANALYZE PROGRESS AND OBJECTIVES OF SUCH WORK , IN ORDER TO HELP IN IDENTIFYING NEW PROJECTS TO BE LAUNCHED AS FROM 1985 .

THERE ARE FAR MORE ORGANIZATIONS WITH A DIRECT INTEREST AND COMPETENCE IN THIS FIELD THAN IS THE CASE WITH VLSI TECHNOLOGY . THE PROJECT PROPOSED HERE MUST BE PLANNED WITH SUFFICIENT FLEXIBILITY TO DEAL WITH THE COMPLEX TECHNICAL AND ORGANIZATIONAL ISSUES INVOLVED . THE ESPRIT INFORMATION EXCHANGE SYSTEM WILL BE A VALUABLE AID IN THE AREA .

COMPUTER CONTROL OF MANUFACTURING VLSI IS SEEN AS AN ESSENTIAL PART OF THE PROGRAMME , NOT JUST A MEANS OF INCREASING MANUFACTURING EFFICIENCY . THE REDUCTION IN FEATURE SIZE AND LAYER THICKNESS PLACES INCREASING DEMANDS ON PROCESS CONTROL IN A SITUATION WHERE THE FINAL DEVICE PARAMETERS ARE DEPENDENT ON A LARGER NUMBER OF VARIABLES . THE MINIMIZATION OF HUMAN HANDLING AND AN ATTACK ON THE SOURCES OF YIELD FALLS RAPIDLY AS FEATURE SIZE IS REDUCED . THE SUBMICRON MOS AND BIPOLAR PROJECTS WILL ADDRESS COMPUTER-AIDED MANUFACTURING ASPECTS WHERE APPROPRIATE .

SECTION A

TYPE A PROJECTS

R AND D AREA 1.1 :

SUBMICRON MOS

DESCRIPTION

THE REQUIREMENT IS TO DEVELOP ALL THE INDIVIDUAL PROCESS STEPS SUCH AS LITHOGRAPHY , ETCHING , DOPING , ETC . , TO ACHIEVE SUBMICRON FEATURE SIZE IN MOS . THE TARGET IS A PROCESS CAPABLE OF MAKING SEVERAL MILLION COMPONENTS PER CHIP OF LOGIC AND MEMORY . IT IS ENVISAGED THAT BELOW ABOUT 0,7 UM OTHER THAN OPTICAL LITHOGRAPHY TOOLS WILL BE USED . PROCESS AND DEVICE MODELLING WILL BE INCLUDED .

PREREQUISITES

- EXPERIENCE WITH STATE-OF-THE-ART PROCESSING ;

- ACTIVITIES ON ADVANCED PROCESS STEPS ;

- AVAILABILITY OF FUNCTIONING CAD/CAT TOOLS ;

- CRITICAL MASS AT LEAST 200 MAN-YEARS OVER FIVE YEARS .

PROGRAMME

YEAR 1 :

- CHOICE OF PROCESS ( CMOS OR MIXED C - AND NMOS ) AND OF DESIGN METHODOLOGY ;

- CHOICE OF OPTICAL LITHOGRAPHY EQUIPMENT ;

- ESTABLISHMENT OF ACTIVITY ON PROCESS STEPS AND MODELLING .

YEAR 2 :

- OPTIMIZATION OF INDIVIDUAL PROCESS STEPS ;

- ESTABLISHMENT OF PRELIMINARY DESIGN RULES ;

- CHARACTERIZATION OF PROCESS STEPS .

YEAR 3 :

- DEMONSTRATION OF FEASIBILITY WITH A 1 UM FEATURE SIZE , 0,5 MILLION COMPONENTS CHIP . THE TARGET IS A FIGURE OF MERIT OF 50 FEMTO-JOULES AND 1 NS DELAY TIME .

YEAR 4 :

- EVALUATION OF NOVEL PROCESS STEPS FOR SUBMICRON TECHNOLOGY ( GATE MATERIAL , ISOLATION , MULTILAYER THREE-DIMENSIONAL ACTIVE STRUCTURES , ETC . ) ;

- CHOICE OF SUBMICRON LITHOGRAPHY EQUIPMENT ;

- FLOW CHART OF A 0,7 UM PROCESS INCLUDING DESIGN RULES . YEAR 5 :

- EVALUATION OF THE 0,7 UM PROCESS ;

- DEMONSTRATION OF THE PROCESS FEASIBILITY ;

- DESIGN OF A CHIP WITH MORE THAN 1 000 000 COMPONENTS , A FIGURE OF MERIT OF 50 FEMTO-JOULES AND 1 NS GATE DELAY .

INTERMEDIATE OBJECTIVES

18TH MONTH

- DESIGN AND EVALUATION OF TEST CHIP OF MORE THAN 1 000 TRANSISTORS BASED ON 1 UM DESIGN RULES WITH A PITCH ( METAL PLUS SPACING ) OF 3 UM ;

36TH MONTH

- FIRST SAMPLES OF CIRCUITS WITH 0,5 MILLION TRANSISTORS , 1 UM DESIGN RULES , WITH DATA ON FIGURE OF MERIT AND DELAY TIME ;

48TH MONTH

- STATISTICAL DATA ON HOMOGENEITY ON A SLICE AND YIELD FOR 1 UM DESIGN RULES ; AND TEST CIRCUITS WITH MORE THAN 1 000 TRANSISTORS WITH 0,7 UM DESIGN RULES AND PITCH OF 2 UM ;

60TH MONTH

- FIRST SAMPLES OF CIRCUITS WITH MORE THAN 1 000 000 TRANSISTORS , 0,7 UM DESIGN RULES , WITH DATA ON FIGURE OF MERIT AND DELAY TIME .

ACTIVITIES DEPENDENT ON 1.1

VLSI MICROELECTRONICS FOR INFORMATION TECHNOLOGY .

R AND D AREA 1.2 :

SUBMICRON BIPOLAR ( SINGLE PROJECT )

DESCRIPTION

THE OVERALL OBJECTIVE IS TO DEVELOP SPECIFIC BIPOLAR SUBMICRON PROCESS STEPS , LEADING TO A COMPLETE PROCESSING SEQUENCE FOR VERY HIGH PERFORMANCE ICS .

TO REALIZE THIS OBJECTIVE THE FOLLOWING DEVELOPMENTS ARE NECESSARY :

- OVERALL CIRCUIT CONCEPTS WHICH MUST EVOLVE TOGETHER WITH THE TECHNOLOGY ;

- A VERTICAL DEVICE STRUCTURE APPROPRIATE TO SUBMICRON LITHOGRAPHY ;

- A CONVENIENT MULTILAYER INTERCONNECTION TECHNOLOGY ,

- GOOD OPTIMIZATION BETWEEN DEVICE PARAMETERS AND SPECIFIC LOGIC FORM USES ;

- APPROPRIATE HIGH DISSIPATION , HIGH PIN COUNT , ELECTRICALLY MATCHED LEADS , PACKAGES ;

- PROCESS AND DEVICE MODELLING .

PREREQUISITES

- EXPERIENCE WITH VLSI AND ADVANCED SILICON PROCESSING ; AVAILABILITY OF CAD/CAT TOOLS ;

- CRITICAL MASS OF AT LEAST 200 MAN-YEARS OVER FIVE YEARS .

ACTIVITIES DEPENDENT ON 1.2

FAST VLSI MICROELECTRONICS FOR INFORMATION TECHNOLOGY .

PROGRAMME

YEAR 1 :

- 1 UM TECHNOLOGY : ESTABLISH ACTIVITY ON PROCESS STEPS , CIRCUIT STRUCTURES , MODELLING . EVALUATE AND CHOOSE CRITICAL EQUIPMENT ;

- FIRST CHOICE OF PROCESS AND CIRCUIT STRUCTURES .

YEAR 2 :

- PRELIMINARY OPTIMIZATION OF PROCESS STEPS ;

- ESTABLISH DESIGN RULES 1 UM ;

- START DESIGN OF DEMONSTRATOR CHIPS ;

- START RESEARCH ON NOVEL PROCESSES AND STRUCTURES FOR SUBMICRON TECHNOLOGY ;

YEAR 3 :

- DEMONSTRATION OF 1 UM PROCESS FEASIBILITY FOR 10/20 KGATES , 100 PS GATE DELAY CHIPS ;

- START WORK ON MULTILAYER 2 UM PITCH IN ADDITION TO PILOT PROJECT CONTINUATION ;

- CONTINUE RESEARCH ON SUBMICRON TECHNOLOGY ;

- PRELIMINARY DESIGN RULES FOR 0,7 UM STRUCTURES .

YEAR 4 :

- START DESIGN OF DEMONSTRATOR CHIP AT 0,7 UM ;

- OPTIMIZATION OF SUBMICRON PROCESS STEPS ;

- CHOICE OF CRITICAL EQUIPMENT ;

- YIELD IMPROVEMENT OF 1 UM DEMONSTRATOR .

YEAR 5 :

- DEMONSTRATION OF 0,7 UM PROCESS FEASIBILITY FOR 20/50 KGATES , 50 PS GATE DELAY CHIPS .

INTERMEDIATE OBJECTIVES

12TH MONTH

- EVALUATE AND CHOOSE CRITICAL EQUIPMENT ;

24TH MONTH

- ESTABLISH DESIGN RULES AT 1 UM ;

36TH MONTH

- DEMONSTRATE 1 UM PROCESS FEASIBILITY FOR 10/20 KGATES , 100 PS GATE DELAY CHIPS ;

42ND MONTH

- ESTABLISH PRELIMINARY DESIGN RULES FOR 0,7 UM STRUCTURES ;

48TH MONTH

- DESIGN DEMONSTRATOR CHIP AT 0,7 UM .

R AND D AREA 1.3 :

CAD ( PROJECTS TO BE DEFINED )

DESCRIPTION :

GLOBAL OBJECTIVES

( A ) TO ENCOURAGE RESEARCH INTO NOVEL AND ADVANCED CAD TECHNIQUES TO MANAGE THE EVER INCREASING CIRCUIT COMPLEXITIES INVOLVED ;

( B ) TO PROVIDE A CAPABILITY FOR COMPLEX VLSI DESIGN THAT IS WIDELY ACCESSIBLE .

CAD/CAT OBJECTIVES

THE OVERALL AIM IS AN INTEGRATED DESIGN SYSTEM CAPABLE OF HANDLING SUBMICRON VLSI CIRCUITS CONTAINING SEVERAL MILLION COMPONENTS .

THE SYSTEM SHOULD :

- PROVIDE A USER FRIENDLY FACILITY THAT IS READILY ADAPTABLE TO CHANGES IN TECHNOLOGY ;

- ALLOW SYSTEMS DESIGNERS TO ACHIEVE A RAPID TURNROUND OF VALID AND TESTABLE DESIGNS AND ASSOCIATED TEST INFORMATION ;

- PROVIDE FACILITIES FOR THE OPTIMIZATION OF CIRCUITS FOR HIGH-VOLUME APPLICATIONS ;

- INCLUDE LIBRARIES OF PRIMITIVE AND COMPLEX CELLS ;

- INCORPORATE RELEVANT RESULTS OF CAD PROJECTS UNDER REGULATION ( EEC ) NO 3744/81 .

POLICY

( A ) DEVICE AND PROCESS MODELLING IS NOT CONSIDERED AS PART OF THE CAD PROJECTS AND IS INCLUDED IN THE PROJECTS CONCERNED WITH THE DEVELOPMENT OF NEW VLSI PROCESSES ( SEE 1.1 AND 1.2 ) ;

( B ) AS FAR AS POSSIBLE , COMPLEX DEMONSTRATOR CHIPS WILL BE USED AS A FOCUS FOR EACH STAGE OF THE PROJECT ;

( C ) AFTER THE FIRST FIVE YEARS CONSIDERED HERE , THERE WILL BE ANOTHER FIVE YEARS FURTHER DEVELOPMENT OF THIS PROJECT TO SUPPORT SPECIFIC TECHNOLOGY ADVANCES AND INCORPORATE WORLDWIDE CAD IMPROVEMENTS ;

( D ) CARE WILL BE TAKEN TO INTERFACE WITH , SUPPORT AND EMPLOY THE RESULTS OF OTHER ESPRIT ACTIVITIES .

BALANCE AND PRIORITIES

( A ) THE ASSIGNED PROJECTS WILL HAVE TO ENSURE OPTIMUM BALANCE BETWEEN WORK TO DEVELOP NOVEL AND ADVANCED CAD TECHNIQUES AND WORK TO CONSOLIDATE EUROPEAN CAPABILITY ;

( B ) ONE EMPHASIS WILL BE TO PROVIDE A GENERAL FRAMEWORK AND DATA INTERCHANGE BASIS TO MAKE EXISTING AND NEW TECHNIQUES GENERALLY AVAILABLE TO A WIDE RANGE OF INDUSTRY . A KEY FACTOR IS TO PROMOTE COOPERATIVE DEVELOPMENT BETWEEN UNIVERSITY AND INDUSTRIAL GROUPS ;

( C ) A SECOND EMPHASIS WILL BE TO CONCENTRATE ON THOSE CAD FACILITIES THAT ARE SPECIFICALLY REQUIRED BY THE DEMANDS OF A 1 UM AND SUBMICRON PROCESS CAPABILITY IN ORDER TO ENABLE THE DESIGNS OF VERY LARGE AND COMPLEX CHIPS CONTAINING UP TO 10 MILLION TRANSISTORS .

MAIN TOPICS

1.3.1 . CAD MANAGEMENT

PROVIDE COORDINATION AND INTELLIGENCE .

1.3.2 . HIGH LEVEL DESIGN AND LAYOUT

- HIGH LEVEL ( BEHAVIOURAL ) DESIGN AIDS ,

- CELL LIBRARY AND TECHNIQUES FOR LARGE , PARAMETERIZED CELLS ,

- LAYOUT DESIGN AND TESTING TECHNIQUES ,

- ANALOGUE AND SWITCH LEVEL VERIFICATION ,

- DESIGN MANAGEMENT .

WITH A TARGET TO ACHIEVE FULL CUSTOM DESIGN AT THE 0,5 MILLION TRANSISTOR LEVEL IN YEAR 4 .

1.3.3 . RESEARCH : ADVANCED/INNOVATIVE CAD

- USE OF AIP AND EXPERT SYSTEMS ,

- USE OF SPECIAL HARDWARE ,

- NOVEL ALGORITHMS .

COMPLETE DEVELOPMENT OF SECOND GENERATION SYSTEM IN YEAR 5 .

1.3.4 . ESTABLISHMENT OF CAD CENTRES

- IMPLEMENT FIRST GENERATION SYSTEM IN YEAR 4 .

R AND D AREA 1.4 :

COMPOUND SEMICONDUCTOR MATERIALS AND INTEGRATED CIRCUITS ( SINGLE PROJECT )

DESCRIPTION

INTEGRATED CIRCUITS IN III-V COMPOUND SEMICONDUCTOR MATERIALS OFFER POTENTIAL SPEED ADVANTAGES OVER SILICON CIRCUITS BECAUSE OF HIGHER ELECTRON MOBILITY . TECHNOLOGY COMPLEXITY IS MANY YEARS BEHIND SILICON AND CONSIDERABLE MATERIALS AND PROCESS RESEARCH ARE NECESSARY . BECAUSE OF THE INCREASING MARKET FOR HIGH-SPEED , LOW-POWER CIRCUITS THE PROBLEMS NEED TO BE TACKLED .

THE TECHNOLOGY ENCOMPASSES GAAS HOMOJUNCTION STRUCTURES BASED ON THE FET AS WELL AS GAAS/GAALAS HETEROJUNCTION STRUCTURES USING THE HIGH ELECTRON MOBILITY TRANSISTOR ( HEMT ) OR THE HETEROJUNCTION BIPOLAR . BASIC RESEARCH IS REQUIRED IN MATERIALS PREPARATION , ION IMPLANTATION , AND IC PROCESS TECHNOLOGIES .

PREREQUISITES

- SOURCE OF PURE MATERIALS ;

- CAD FACILITIES ;

- SUITABLE LABORATORY AND EXPERIENCED TEAM .

PROGRAMME

YEAR 1 :

- INITIATE MATERIALS RESEARCH IN GAAS INCLUDING ION IMPLANTATION TO PRODUCE GAAS FET-BASED ICS ;

- INITIATE WORK ON MOLECULAR BEAM EPITAXY ( MBE ) AND METALLO-ORGANIC CHEMICAL VAPOUR DEPOSITION ( MOCVD ) FOR THE GROWTH OF SUITABLE HETEROJUNCTION STRUCTURES IN GAALAS .

YEAR 2 :

- CONTINUE MATERIALS RESEARCH ;

- DEVELOP GAAS IC FABRICATION PROCESS ON 2-INCH WAFERS AT 1 UM FEATURE SIZE USING MULTILAYER METALLIZATION , ETC . ;

- FABRICATE EXPERIMENTAL DISCRETE HEMTS AND BIPOLARS IN GAAS/GAALAS .

YEAR 3 :

- CONTINUE MATERIALS RESEARCH IN GAAS AND GAALAS ;

- INITIATE RESEARCH ON NEW MATERIALS , E . G . GAINAS ;

- EXTEND GAAS TECHNOLOGY TO 3-INCH WAFERS AND SUBMICRON FEATURES ;

- CONTINUE RESEARCH INTO OPTIMIZATION OF HEMT AND BIPOLAR DISCRETES ;

- DEMONSTRATE 1 000 GATE ARRAY IN GAAS FET-BASED LOGIC .

YEAR 4 :

- CONTINUE MATERIALS RESEARCH INCLUDING NEW MATERIALS ;

- CARRY OUT YIELD IMPROVEMENT EXERCISE ON BASIC GAAS IC PROCESS ;

- ESTABLISH BASIC PROCESS STEPS FOR A GAAS/GAALAS IC PROCESS BASED ON THE HEMT OR BIPOLAR .

YEAR 5 :

- INTEGRATE PROCESS STEPS FOR HETEROJUNCTION BIPOLAR ( OR HEMT ) IC TECHNOLOGY ;

- FABRICATE EXPERIMENTAL ICS BASED IN THIS TECHNOLOGY ;

- CONTINUE NEW MATERIALS RESEARCH ( E . G . GAINAS ) ;

- FABRICATE EXPERIMENTAL DISCRETE DEVICES IN GAINAS .

INTERMEDIATE OBJECTIVES

12TH MONTH

- SINGLE-GATE DEMONSTRATION CIRCUIT BASED ON 0,5 UM DESIGN RULES ;

24TH MONTH

- 30-GATE DEMONSTRATION CIRCUIT BASED ON 0,5 UM DESIGN RULES ;

36TH MONTH

- INITIATION OF 0,3 UM DESIGN RULES ;

- 1 000-GATE DEMONSTRATION CIRCUIT BASED ON 0,5 UM DESIGN RULES ;

48TH MONTH

- DEMONSTRATION OF 100-GATE CIRCUIT BASED ON 0,3 UM DESIGN RULES ;

60TH MONTH

- DEMONSTRATION OF 1 000-GATE CIRCUIT BASED ON 0,3 UM DESIGN RULES .

R AND D AREA 1.5 :

OPTOELECTRONICS ( SINGLE PROJECT )

DESCRIPTION

OPTOELECTRONIC DEVICES WILL BE INCREASINGLY REQUIRED FOR TELECOMS TRANSMISSION , INTRA - AND EXTRA-COMPUTER CONNECTIONS , ULTRA WIDE BAND IMAGE PROCESSING AND SWITCHING . FUTURE GENERATIONS OF MONO-MODE COMMUNICATIONS SYSTEMS MAY USE COHERENT DETECTION AND MULTI-CHANNEL WAVELENGHT MULTIPLEXING AND MAY BE PHASE MODULATED . THIS WILL PROVIDE IMPROVED PERFORMANCE AND BE COMPATIBLE WITH INTEGRATED OPTICAL LOGIC . THIS WILL ALLOW PROCESSING , COMBINING AND ROUTING AT VERY HIGH SPEEDS . IN SEMICONDUCTOR FORM THEY WILL ALSO BE COMPATIBLE WITH III-V INTEGRATED CIRCUITS PROVIDING A FAST ELECTRICAL INTERFACE .

PREREQUISITES

- SOURCE OF PURE MATERIALS ;

- SUITABLE LABORATORY AND EXPERIENCED TEAM .

PROGRAMME

YEAR 1 :

- INITIATE MATERIALS RESEARCH ON OPTICAL STRUCTURES AND IMPROVED METHODS OF MATERIALS PREPARATION SUCH AS METALLO-ORGANIC CHEMICAL VAPOUR DEPOSITION AND MOLECULAR BEAM EPITAXY ;

- EXTEND MATERIALS RESEARCH TO ORGANICS INCLUDING FAST RESPONSE PHOTOCHROMIC MATERIALS ;

- FABRICATE BASIC DISCRETE DEVICES , E . G . LASERS , LEDS , DETECTORS , SWITCHES , ETC .

YEAR 2 :

- CONTINUE MATERIALS RESEARCH ;

- FABRICATE NARROW LINE LASERS AND INVESTIGATE DETECTION TECHNIQUES FOR COHERENT SYSTEMS ;

- DEVELOP BISTABLE OPTICAL DEVICES , HIGH-SPEED MODULATORS AND OPTICAL COUPLERS ;

- RESEARCH INTO MONOLITHIC OPTOELECTRONICS .

YEAR 3 :

- INVESTIGATE INTEGRATION IN OPTICAL MATERIALS USING THIN FILM WAVEGUIDES FOR INTERCONNECTION ;

- CONTINUE MATERIALS RESEARCH ;

- FABRICATE MULTI-COMPONENT STRUCTURES , E . G . LASER ARRAYS FOR WAVELENGTH MULTIPLEXING , SWITCH MATRICES AS BASIC SWITCHING ELEMENTS .

YEAR 4 :

- COMBINE WAVEGUIDE INTEGRATION WITH MONOLITHIC INTEGRATION TECHNOLOGY TO PRODUCE EXPERIMENTAL OPTICAL SYSTEMS ON A SINGLE CHIP ;

- CONTINUE TECHNOLOGY DEVELOPMENT IN LINBO3 , PHOTOCHROMICS FOR OPTICAL INTERCONNECTIONS , AND NEW MATERIALS .

YEAR 5 :

- DEVELOP HIGH-SPEED OPTICAL SWITCHING SYSTEM AND STUDY APPLICATION OF OPTICAL TECHNIQUES TO COMPUTERS .

INTERMEDIATE OBJECTIVES

12TH MONTH

- GAALAS METAL ORGANIC CHEMICAL VAPOUR DEPOSITION ( MOCVD ) ESTABLISHED FOR DISCRETE DEVICE FABRICATION ;

- OPTIMUM TI-LITHIUM NIOBATE TECHNOLOGY ESTABLISHED AND 10 GHZ MODULATION DEMONSTRATED .

24TH MONTH

- GAINAS MOCVD ESTABLISHED FOR DISCRETE DEVICES ;

- BASIC ELECTRO-OPTIC SWITCH TECHNOLOGY OPTIMIZED ;

- PROTON EXCHANGE AND ORGANIC OVERLAY WAVEGUIDE TECHNOLOGY ESTABLISHED FOR SWITCHES AND MODULATORS .

36TH MONTH

- GAALINAS MOCVD ESTABLISHED FOR DISCRETE DEVICES ;

- 100 ELEMENTS GAALAS MOCVD OPTOELECTRONIC INTEGRATED CIRCUIT DEMONSTRATED ;

- 100 ELEMENT ELECTRO-OPTIC SWITCH DEMONSTRATED ;

- HYBRID INTEGRATED ELECTRO-OPTIC TECHNOLOGY ESTABLISHED ;

48TH MONTH

- LITHIUM NIOBATE/OXIDE EPITAXIAL GROWTH ESTABLISHED ;

- 100 GHZ ELECTRO-OPTIC MODULATION DEMONSTRATED ;

- MOLECULAR BEAM EPITAXY ( MBE ) DEMONSTRATED FOR DISCRETE AND INTEGRATED COMPONENTS ;

60TH MONTH

- " OPTICAL SYSTEM ON A CHIP " DEMONSTRATED , USING MBE AND MOCVD MATERIAL ;

- ULTRAFAST HYBRID ELECTRO-OPTIC PROCESSOR DEMONSTRATED .

SECTION B

SUGGESTED THEMES FOR TYPE B PROJECTS

IN ADDITION TO THE TOPIC AREAS ALREADY IDENTIFIED IN SECTION A ( WHERE TYPE B PROJECTS MAY USEFULLY COMPLEMENT THE WORK PROGRAMMES BY PROVIDING THE REQUISITE TECHNOLOGY AND TOOLS AND/OR ANTICIPATE NEXT GENERATION REQUIREMENTS ) , THE FOLLOWING ARE SPECIFIC EXAMPLES OF TOPICS WHICH WOULD BE ELIGIBLE :

( A ) VLSI-RELATED

- LITHOGRAPHY ( E-BEAM , X-RAY , UV ) ,

- ION IMPLANTATION ,

- SEMICONDUCTOR MATERIALS ,

- RESIST TECHNOLOGY ,

- RAPID ANNEAL TECHNIQUES ,

- LAYER PROCESSING TECHNIQUES ,

- CONDUCTORS ( METALS , SILICIDES , CONTACTS ) ,

- INSULATORS ( THIN , THICK , ORGANIC ) ,

- DEVICE RELIABILITY ,

- CAD FOR GAAS ,

- COMPUTER-AIDED MANUFACTURING METHODS .

( B ) OTHERS

- INTERCONNECTION ( OFF-CHIP ) ,

- ADVANCED PHYSICAL ANALYSIS TECHNIQUES ,

- FLAT SCREEN DISPLAY TECHNOLOGY ,

- SENSORS AND TRANSDUCERS ,

- NEW TECHNOLOGIES FOR ADVANCED INFORMATION STORAGE ,

- NEW INORGANIC AND ORGANIC MATERIALS .

SUBPROGRAMME 2

SOFTWARE TECHNOLOGY

THE OBJECTIVE OF THIS SUBPROGRAMME IS TO REACH A STAGE WHERE THE PRODUCTION OF INFORMATION SYSTEMS ( I . E . PRODUCTS INCLUDING HARDWARE AND SOFTWARE SOLUTIONS ) WOULD HAVE THE CHARACTERISTICS OF AN INDUSTRIAL PROCESS , AND , TO A LARGE EXTENT , WOULD BE COMPUTER ASSISTED . THIS IMPLIES BRINGING TO THE CONSCIOUSNESS OF EVERY INDUSTRIAL ORGANIZATION THE FACT THAT SOFTWARE PRODUCTION IS A TRUE ENGINEERING DISCIPLINE , SUBJECT TO TECHNICAL , ORGANIZATIONAL AND ECONOMIC FACTORS .

THE PROGRESS TO BE MADE IN SOFTWARE TECHNOLOGY HAS BEEN DIVIDED INTO THREE GENERAL CATEGORIES . AS WITH ANY CLASSIFICATION , SUCH A DIVISION IS ARBITRARY , AND THE CATEGORIES ARE NOT INDEPENDENT . THE FIRST CATEGORY OF WORK TO BE PERFORMED IS TO CONTRIBUTE TO THE UNDERSTANDING OF THE PROCESS OF SOFTWARE PRODUCTION AND MAINTENANCE , TAKING AS A DEPARTURE POINT THE BEST OF CURRENT PRACTICES . THE SECOND CATEGORY CONCERNS ITSELF WITH IMPLEMENTING THE SUPPORT ( INCLUDING TOOLS ) THAT WILL MAKE THE RESULTS OF THE UNDERSTANDING USABLE IN PRACTICE . THE THIRD CATEGORY IS CONCERNED WITH MEDIUM - AND LONGTERM R AND D AIMED AT EVOLVING NEW SOFTWARE PRODUCTION METHODS .

FOR FURTHER STRUCTURING OF THE WORK , A SIMPLE SOFTWARE LIFE-CYCLE MODEL IS INTRODUCED . IT IS NOT INTENDED THAT THIS MODEL SHOULD LIMIT THE SCOPE OF PROJECTS ; INDEED THERE WILL BE PROJECTS SPECIFICALLY CONCERNED WITH THE DEVELOPMENT OF SUCH MODELS . IN THIS SIMPLE MODEL , SOFTWARE PRODUCTION AND MAINTENANCE CONSISTS OF A SEQUENCE OF INTERMEDIATE LEVELS OF REPRESENTATION WITH TRANSFORMATIONS BETWEEN THEM , PROCEEDING FROM INITIAL CONCEPT TO OPERATIONAL SYSTEM . IN PRACTICE , SUCH DEVELOPMENT MUST BE ITERATIVE , AND ALMOST NEVER FOLLOWS A SIMPLE LINEAR COURSE . SUCH ITERATIVE OPERATIONS ARE INDICATED IN FIGURE 2-1 WHICH SHOWS THEM EMBEDDED IN A BROAD PROJECT CONTEXT . THUS , THE WORK TO BE DONE CAN BE SUBDIVIDED INTO THREE MAJOR AREAS ( SEE FIGURE 2-1 ) : THE ( TECHNICAL ) PROCESS ITSELF ; MANAGEMENT AND CONTROL ; AND THE SUPPORT ENVIRONMENT .

COMBINING THE GENERAL CATEGORIES WITH EACH OF THE LIFE-CYCLE AREAS , A 3 BY 3 MATRIX RESULTS , GIVING A TOTAL OF NINE SUBDIVISIONS . ON THE BASIS OF THE STRATEGIC PROJECTS THAT HAVE BEEN IDENTIFIED , THESE SUBDIVISIONS MORE OR LESS NATURALLY GROUP INTO FOUR R AND D AREAS .

A FIFTH AREA IS FORMED BY DEMONSTRATION PROJECTS , WHICH WOULD NOT BE ARTIFICIAL PROJECTS BUT RATHER REAL PROJECTS USED AS EVALUATION AND DEMONSTRATION VEHICLES .

THE FIVE R AND D AREAS ARE DESCRIBED ON THE FOLLOWING SHEETS .

R AND D AREA 2.1 :

PROCESS - UNDERSTANDING AND IMPLEMENTING

DESCRIPTION

THE ACTIVITIES IN THIS AREA ADDRESS THE TRANSITION TO AN ENGINEERING APPROACH TO SYSTEMS AND SOFTWARE ENGINEERING WITHIN THE FRAMEWORK OF THE CURRENT LIFE-CYCLE VIEW . A SCENARIO IS ENVISAGED IN WHICH WELL-FOUNDED SCIENTIFICALLY-BASED METHODS ARE GRADUALLY ADOPTED TO REPLACE AD HOC TECHNIQUES CURRENTLY IN USE . SUCH METHODS SHOULD CONTRIBUTE SIGNIFICANTLY TO THE REDUCTION OF LIFE-CYCLE COSTS , PARTICULARLY IN THE AREAS OF TESTING AND MAINTENANCE , THROUGH EARLIER ERROR DETECTION AND REMOVAL , AND THE DEVELOPMENT OF MORE RELIABLE AND HIGHER QUALITY SYSTEMS . THE RESEARCH AND DEVELOPMENT PROGRAMME IN THIS AREA SHOULD CONTINUE FOR ABOUT FIVE YEARS , SO MORE SPECULATIVE AND RADICAL APPROACHES TO SYSTEM DEVELOPMENT , FOR WHICH CONSIDERABLE FUNDAMENTAL RESEARCH IS REQUIRED , ARE NOT CONSIDERED HERE ( SEE 2.4 ) .

THE EMPHASIS IS ON THE DEVELOPMENT OF IMPROVED METHODS FOR SYSTEM CONSTRUCTION . THESE METHODS MUST THEN BE SUPPORTED BY INTEGRATED SUPPORT TOOLS . IT IS ASSUMED HERE THAT THE OVERALL DEVELOPMENT APPROACH , WHICH THE TOOLS SUPPORT , WILL EMPLOY MULTIPLE LEVELS OF REPRESENTATION WITH INCREMENTAL VERIFICATION AND VALIDATION . WHERE POSSIBLE , TOOLS MAY BE PROVIDED BY RE-WORK OR ENHANCEMENT OF EXISTING PROGRAMS RATHER THAN BY COMPLETELY NEW DEVELOPMENT .

THERE IS ALSO A STRONG INTERDEPENDENCE WITH THE MANAGEMENT ACTIVITIES DESCRIBED IN AREA 2.2 . MANAGEMENT METHODS MUST TAKE ACCOUNT OF IMPROVED SYSTEMS AND SOFTWARE DEVELOPMENT METHODS AND , SIMILARLY , SYSTEM DEVELOPMENT METHODS MUST TAKE ACCOUNT OF MANAGEMENT REQUIREMENTS .

INDUSTRIAL AND ACADEMIC COLLABORATION IS PARTICULARLY RELEVANT TO THIS AREA , SINCE ACADEMIC RESEARCHERS IN EUROPE HAVE PRODUCED A SIGNIFICANT THEORETICAL BASE WHICH SHOULD PROVIDE THE FOUNDATIONS OF IMPROVED SYSTEMS AND SOFTWARE ENGINEERING METHODS .

R AND D TOPICS

THE WORK IN THIS AREA WILL ADDRESS :

- REFERENCE MODELS OF SYSTEMATIC APPROACHES TO SYSTEM DEVELOPMENT ;

- PRACTICAL AND DISCIPLINED SYSTEM DEVELOPMENT METHODS ;

- EFFECTIVE METHODS OF SOFTWARE PRODUCTION AND MAINTENANCE ;

- STUDY OF APPLICATION AREA NEEDS ;

- REPRESENTATION AND TRANSFORMATION TOOLS ;

- VERIFICATION AND VALIDATION TOOLS ;

PROCESS

* APPLICATION DOMAIN

USER INTERFACES

* MANAGEMENT

* TRANS . * REPRESENTATION * TRANS . * REPRESENTATION * MANAGEMENT AND CONTROL * TARGET SYSTEM

* VERIFICATION AND VALIDATION * VERIFICATION AND VALIDATION

* HOST SYSTEM

ENVIRONMENT

FIGURE 2-1

- SUPPORT TOOLS :

- COMPONENT LIBRARY SUPPORT TOOLS ,

- DOCUMENTATION TOOLS ,

- PERFORMANCE PREDICTION AND MEASUREMENT TOOLS ,

- RELIABILITY SPECIFICATION AND MEASUREMENT TOOLS ,

- PROCESS SUPPORT TOOLS .

THE PROGRAMME INCLUDES CONTINUATION OF WORK STARTED BY THE PILOT PROJECTS IN YEAR 0 .

TYPE A PROJECTS

2.1.1 . METHODS , TECHNIQUES AND TOOLS

THIS TOPIC SHOULD BE COVERED BY UP TO TWO PROJECTS TO BE STARTED IN YEAR 1 , BOTH ANSWERING TO THE DESCRIPTION GIVEN BELOW . THE DIFFERENTIATION BETWEEN THESE PROJECTS WILL BE ON THE BASIS OF THE MAIN APPROACH ADOPTED BY THEM , OR ON THE APPLICATION AREA THEY ADDRESS .

THE OBJECTIVES ARE :

- TO IDENTIFY AN ADEQUATE RANGE OF EFFECTIVE METHODS OF SOFTWARE PRODUCTION AND MAINTENANCE ,

- TO PROVIDE TOOL SUPPORT FOR THESE METHODS , ASSUMING MULTIPLE LEVELS OF REPRESENTATION WITH INCREMENTAL VERIFICATION AND VALIDATION .

THE METHODS MUST ACHIEVE COMPLETE COVERAGE OF THE ENTIRE SPECTRUM FROM REQUIREMENTS DEFINITION THROUGH TO THE MAINTENANCE PHASE , INCLUDING NECESSARY ITERATIONS ; METHODS MUST AIM TO MINIMIZE TOTAL LIFE-CYCLE COST , AND COMPATIBILITY BETWEEN VARIOUS METHODS AND NOTATIONS USED IN SUCCESSIVE LIFE-CYCLE PHASES IS A PRIME REQUIREMENT .

TO ESTABLISH THE OVERALL CONTEXT IN WHICH SOFTWARE DEVELOPMENT TAKES PLACE , REFERENCE MODELS OF SYSTEMATIC APPROACHES TO SYSTEM DEVELOPMENT MUST BE ADDRESSED , AND SPECIAL ATTENTION BE PAID TO THE CRUCIAL REQUIREMENTS ANALYSIS TASK .

ONE ESPECIALLY IMPORTANT TASK IS TO INVESTIGATE THE ISSUES RELATING TO THE USE AND RE-USE OF SOFTWARE COMPONENTS .

INTERMEDIATE OBJECTIVES

12TH MONTH

- PROTOTYPE VERSION OF TOOLS ;

18TH MONTH

- REFERENCE MODEL OF SYSTEM DEVELOPMENT ;

24TH MONTH

- INITIAL TOOL SET ;

36TH MONTH

- SYSTEM AND SOFTWARE DEVELOPMENT METHODS ;

60TH MONTH

- FINAL TOOL SET .

2.1.2 . INTERGRATION OF MANAGEMENT AND TECHNICAL ASPECTS

THIS PROJECT ADDRESSES THOSE MANAGEMENT ISSUES WHICH ARE STRONGLY AFFECTED BY THE TECHNICAL PROPERTIES OF THE SOLUTION EXPECTED FROM PROJECT 2.1.1 .

THE OBJECTIVES ARE :

- TO DEVELOP TOGETHER , IN ONE PROJECT , BOTH MANAGEMENT AND TECHNICAL APPROACHES TO SOFTWARE PRODUCTION AND MAINTENANCE ,

- TO EXPLORE PROBLEMS RELATED TO MANAGEMENT AND CONTROL OF THE SOFTWARE PRODUCT , ESPECIALLY CONFIGURATION MANAGEMENT AND VERSION/VARIANT CONTROL ,

- TO DEVELOP OR ABSTRACT THOSE STANDARDS THAT WILL BE EFFECTIVE ALSO FROM A MANAGEMENT POINT OF VIEW .

SPECIAL ATTENTION WILL HAVE TO BE PAID TO THE USE OF FAMILIES OF " PREFABRICATED " COMPONENTS IN INTEGRATION AND PRODUCTION MANAGEMENT .

INTERMEDIATE OBJECTIVES

24TH MONTH

- METHODS AND TECHNIQUES FOR PRODUCT MANAGEMENT AND CONTROL ;

48TH TO 60TH MONTH

- SUPPORT TOOLS FOR CONFIGURATION MANAGEMENT , INTEGRATION AND PRODUCTION MANAGEMENT .

2.1.3 . SOFTWARE METHODOLOGY

THIS PROJECT WILL INVESTIGATE THE SUITABILITY OF METHODS AND TECHNIQUES FOR VARIOUS APPLICATION AREAS , AND DEVELOP THE CRITERIA FOR SELECTING , FROM AMONG EXISTING OR ENVISAGED METHODS AND TECHNIQUES , THOSE THAT ARE MOST APPROPRIATE FOR THE PROJECT IN HAND ON THE BASIS OF :

- THE PARTICULAR APPLICATION AREA ,

- THE OVERALL DEVELOPMENT SITUATION ( TEAM SIZE , OTHER PERSONNEL AND MANAGEMENT CONSTRAINTS , CAPABILITIES OF THE TECHNICAL ENVIRONMENT , ETC . ) ,

- THE ENVISAGED PRODUCT LIFE ( FREQUENCY OF CHANGES , NUMBER OF VARIANTS , ETC . ) .

WHERE POSSIBLE THE CRITERIA SHOULD BE QUANTITATIVE , AND TO THIS END GLOBAL NUMERICAL DATA WILL BE COLLECTED AND ANALYZED FROM A NUMBER OF DEVELOPMENT PROJECTS OVER A LONG PERIOD OF TIME . THE UTILITY OF THE CRITERIA WILL REQUIRE VERIFICATION IN APPLICATION TO REAL PROJECTS AND ADJUSTED IN THE LIGHT OF PRACTICAL EXPERIENCE .

INTERMEDIATE OBJECTIVES

18TH MONTH

- STUDY OF APPLICATION AREA NEEDS ;

36TH MONTH

- ( QUALITATIVE ) CRITERIA ;

48TH MONTH

- QUANTITATIVE ANALYSIS OF DATA COLLECTED FROM DEVELOPMENT PROJECTS ;

60TH MONTH

- FULL SET OF CRITERIA .

TYPE B PROJECTS

THE FOLLOWING THEMES AT LEAST WILL HAVE TO BE ADDRESSED IN THIS R AND D AREA :

- SYSTEM DEVELOPMENT MODELS ;

- SYSTEM/SOFTWARE DEVELOPMENT METHODS , INCLUDING :

- USE OF EXISTING COMPONENTS IN NEW DEVELOPMENTS ,

- FORMAL SEMANTICS OF INTERFACES ,

- FAULT TOLERANCE ,

- VALIDATION AND VERIFICATION ,

- CAPTURING OF REQUIREMENTS ,

- DEVELOPMENT OF NOTATIONS ( INCLUDING GRAPHICAL ONES ) WHICH HAVE WELL-FOUNDED SEMANTICS ,

- RELIABILITY OF SPECIFICATIONS ,

- SPECIFICATION OF BOTH SEQUENTIAL AND CONCURRENT SYSTEMS ,

- DECOMPOSITION , INTEGRATION AND RETAINING OF COMPATIBILITY BETWEEN EVOLVING HARDWARE AND SOFTWARE SUBSYSTEMS BOTH DURING DEVELOPMENT AND DURING OPERATIONAL LIFE ,

- HARDWARE/SOFTWARE MIGRATION ,

- SYSTEM OPTIMIZATION ;

- REPRESENTATION AND TRANSFORMATION TOOLS , INCLUDING THE USE OF SEMI-FORMAL AS WELL AS GRAPHICAL LANGUAGES ;

- VERIFICATION AND VALIDATION TOOLS ;

- COMPONENT LIBRARY SUPPORT TOOLS ;

- PERFORMANCE PREDICTION AND MEASUREMENT TOOLS ;

- RELIABILITY SPECIFICATION AND MEASUREMENT TOOLS .

R AND D AREA 2.2 :

MANAGEMENT - UNDERSTANDING AND IMPLEMENTING

DESCRIPTION

PROJECTS IN THIS AREA ARE INTENDED TO CONSOLIDATE UNDERSTANDING OF , AND EFFECTIVE MEANS OF SUPPORT FOR , THE MANAGEMENT OF SOFTWARE PROJECTS . THE MANAGEMENT APPROACH WHICH IS DEVELOPED MUST BE CONSISTENT WITH THE TECHNICAL APPROACH DEVELOPED IN THE " PROCESS - UNDERSTANDING AND IMPLEMENTING " AREA ( 2.1 ) . THIS CAN BEST BE ACHIEVED IF THE MANAGEMENT AND TECHNICAL APPROACHES ARE DEVELOPED TOGETHER ( SEE 2.1.2 ) . IT IS ALSO IMPORTANT THAT THE PROJECTS IN THIS AREA ADDRESS BOTH TECHNICAL MANAGEMENT ISSUES , SUCH AS CONFIGURATION CONTROL , AND PROJECT MANAGEMENT ISSUES , SUCH AS EFFORT ESTIMATION . ANOTHER IMPORTANT TOPIC IS THE ESTABLISHMENT OF TECHNIQUES FOR THE COLLECTION OF DATA WHICH ALLOW MANAGERS TO EVALUATE AND SELECT METHODS AND TOOLS .

R AND D TOPICS

WORK IN THIS AREA WILL ADDRESS :

- MANAGEMENT MODELS OF SOFTWARE PRODUCTION AND MAINTENANCE ;

- PROJECT PLANNING AND CONTROL TOOLS AND TECHNIQUES ;

- CONFIGURATION MANAGEMENT TOOLS AND TECHNIQUES ;

- PRODUCTION MANAGEMENT , INTEGRATION AND CONTROL TOOLS AND TECHNIQUES ;

- MAINTENANCE MANAGEMENT METHODS AND TOOLS ;

- QUALITY AND RELIABILITY ASSURANCE TOOLS AND TECHNIQUES ;

- MANAGEMENT DATABASE SUPPORT ;

- DATA COLLECTION AND ANALYSIS .

THE PROGRAMME INCLUDES CONTINUATION OF WORK STARTED UNDER THE PILOT PROJECTS IN YEAR 0 .

TYPE A PROJECTS

2.2.1 . SOFTWARE PRODUCTION AND MAINTENANCE MANAGEMENT SUPPORT

THE OBJECTIVES ARE :

- TO DEVELOP QUANTITATIVE MANAGEMENT MODELS OF SOFTWARE PRODUCTION AND MAINTENANCE ;

- TO PROVIDE SUPPORT FOR THE MAJOR FUNCTIONS OF SOFTWARE PRODUCTION AND MAINTENANCE MANAGEMENT :

- PLANNING AND CONTROL OF SOFTWARE PROJECTS ,

- PLANNING AND CONTROL OF SOFTWARE PRODUCTS ,

- MAINTENANCE MANAGEMENT ;

- TO DEVELOP DATABASE SUPPORT FOR MANAGEMENT TOOLS ( E . G . BY EXPLOITING THE DATABASE FACILITIES OF THE COMMON ENVIRONMENT ) .

IT IS IMPORTANT THAT A WIDE VARIETY OF MANAGEMENT PRACTICES BE ACCOMMODATED WITH A LIMITED SET OF COMMON FACILITIES ( E . G . THROUGH EXPLOITATION OF GENERIC PROPERTIES ) .

INTERMEDIATE OBJECTIVES

12TH MONTH

- COST ESTIMATING MODELS ;

18TH MONTH

- FULL QUANTITATIVE MODELS ;

36TH MONTH

- INITIAL TOOL SET ;

42ND MONTH

- MANAGEMENT DATABASE ;

54TH MONTH

- COMPLETE TOOL SET .

2.2.2 . QUALITY AND RELIABILITY ASSURANCE

THE OBJECTIVES ARE :

- TO INVESTIGATE IN DETAIL THE EFFECTIVENESS OF TWO CONTRASTING APPROACHES TO QUALITY AND RELIABILITY ASSURANCE , NAMELY THAT OF CLOSE CONTROL OVER THE DEVELOPMENT PROCESS , AND THAT OF MEASUREMENT AND CORRECTIVE ACTION ,

- TO PROVIDE TECHNIQUES AND TOOLS THAT WILL ASSIST THE ASSURANCE CONTROLS BOTH PRE - AND POST-PRODUCTION .

THE PROPERTIES BY WHICH A PRODUCT'S QUALITY MAY BE CHARACTERIZED AND QUANTIFIED SHOULD BE IDENTIFIED AND MEANS BY WHICH A PRODUCT'S QUALITY AND RELIABILITY MAY BE ESTIMATED AND PREDICTED SHOULD BE DETERMINED .

PARTICULAR ATTENTION SHOULD BE GIVEN TO THE VERIFICATION AND VALIDATION ACTIVITY AT EACH OF THE TRANSFORMATIONS IN THE LIFE-CYCLE .

DATA COLLECTION AND ANALYSIS SHOULD BE ADDRESSED MAINLY FROM THE SHORT-TERM ASPECT ( PROGRESS MONITORING , ANALYSIS OF ERROR REPORTS , ETC . , FOR INDIVIDUAL PROJECTS ) . THE LONG-TERM ASPECT ( ASSESSMENT AND EVALUATION OF METHODS THEMSELVES ON THE BASIS OR LARGER STATISTICS ) IS ADDRESSED IN PROJECT 2.1.3 .

INTERMEDIATE OBJECTIVES

24TH MONTH

- CRITERIA FOR QUALITY AND RELIABILITY ASSURANCE ;

36TH MONTH

- DATA COLLECTION AND ANALYSIS TECHNIQUES ;

60TH MONTH

- TOOL SET AVAILABLE .

TYPE B PROJECTS

THE FOLLOWING THEMES AT LEAST WILL HAVE TO BE ADDRESSED IN THIS R AND D AREA :

- MONITORING TECHNIQUES AND PROGRESS PARAMETERS ;

- MAINTENANCE MANAGEMENT , E . G . :

- FAULT AND ERROR REPORTING ,

- CHANGE CONTROL ,

- RELEASE CONTROL ,

- HANDLING OF ENHANCEMENT REQUESTS ;

- FAILURE MODE EFFECTS AND CRITICALITY ANALYSIS .

R AND D AREA 2.3 :

ENVIRONMENT - UNDERSTANDING AND IMPLEMENTING

DESCRIPTION

THE PROJECTS IN THIS AREA WILL DEVELOP A COMMON ENVIRONMENT TO BE USED :

- AS A PRIMITIVE SOFTWARE DEVELOPMENT ENVIRONMENT ,

- AS THE BASIS FOR THE DEVELOPMENT OF A COMPLETE INTEGRATED SOFTWARE ENGINEERING ENVIRONMENT .

SINCE MULTI-NATIONAL DEVELOPMENT OF SOFTWARE WILL BE A CENTRAL FEATURE OF ESPRIT AS A WHOLE , THIS ENVIRONMENT IS AN IMPORTANT BUILDING BLOCK FOR ALL ESPRIT PROGRAMMES . HENCE IT IS IMPERATIVE THAT THERE IS GENERAL PUBLICATION OF ALL INTERFACES , THUS PROVIDING FOR GENERAL EXPLOITATION AND INTERCEPTION OF THE ENVIRONMENT .

THE COMMON ENVIRONMENT WILL CONSIST OF AN INFRASTRUCTURE , WHICH PROVIDES THE BASIC FRAMEWORK , AND A SET OF TOOLS AND COMPONENTS WHICH WILL BE OF RELEVANCE IN ALL INSTANCES OF THE ENVIRONMENT . IN PARTICULAR , THE TOOLS AND COMPONENTS ARE INTENDED TO ASSIST THE DEVELOPMENT OF NEW TOOLS , THUS PROMOTING EXTENSION TO THE COMPLETE INTEGRATED ENVIRONMENT .

THE ENVIRONMENT DEVELOPED MUST ALLOW A RANGE OF IMPLEMENTATIONS , ON HARDWARE OF DIFFERENT CAPABILITIES , AND WITH DIFFERENT AIMS ( E . G . EASE OF INSTALLATION V . EFFICIENCY ) . THEREFORE IT IS IMPORTANT THAT A SPECTRUM OF INTERFACES IS DEFINED THROUGH WHICH TOOLS OPERATE ON " OBJECTS " SUCH AS PROGRAMS , FILES , DEVICES AND OTHER TOOLS IN ORDER TO ACCOMMODATE THESE NEEDS .

R AND D TOPICS

WORK IN THIS AREA WILL ADDRESS :

- ENVIRONMENT INFRASTRUCTURE ;

- COMMON ENVIRONMENT TOOLS AND COMPONENTS ;

- COMMON ENVIRONMENT GENERAL SERVICES .

THE PROGRAMME INCLUDES CONTINUATION OF WORK STARTED UNDER PILOT PROJECTS IN YEAR 0 .

TYPE A PROJECT

2.3.1 . COMMON TOOL ENVIRONMENT

THE OBJECTIVES ARE :

- TO DEVELOP A COMMON ENVIRONMENT TO BE USED AS PRIMITIVE SOFTWARE DEVELOPMENT ENVIRONMENT , AND AS THE BASIS OF A COMPLETE INTEGRATED SOFTWARE ENGINEERING ENVIRONMENT ,

- TO DEVELOP AN OBJECT-BASED APPROACH TO THE PROVISION OF A TOOL-RICH ENVIRONMENT , WHERE " OBJECT " INCLUDES PROGRAMS , FILES , DEVICES AND OTHER TOOLS ; THE EMPHASIS WILL BE ON META-TOOLS , WHICH CAN BE USED TO ASSIST IN THE CONSTRUCTION OF SPECIFIC ENVIRONMENT TOOLS AND FACILITIES ; COMMON TOOLS AND COMPONENTS WILL PROMOTE A HOMOGENEOUS INTERFACE ACROSS ALL INSTANCES OF THE ENVIRONMENT AND REDUCE THE COST OF NEW TOOL DEVELOPMENT ,

- TO PROVIDE AN APPROPRIATE INFRASTRUCTURE ( BASIC FRAMEWORK ) WITH STANDARDS AND CONVENTIONS FOR ENVIRONMENT WORK-STATIONS , WITH SUPPORT FROM COMMON USER INTERFACING MECHANISMS , WITH BASIC MECHANISMS FOR PROGRAM EXECUTION , COMMUNICATION AND OBJECT MANAGEMENT ( ENVIRONMENT DATABASE ) AND WITH LOCAL AREA NETWORK PROTOCOLS AND INTERFACE STANDARDS APPROPRIATE FOR ENVIRONMENT FUNCTIONS ,

- TO PROVIDE A RANGE OF GENERAL SERVICES OF INTEREST TO ALL USERS ( E . G . MAIL , BULLETIN BOARD , DOCUMENT PREPARATION , ETC . ) .

INTERMEDIATE OBJECTIVES

12TH MONTH

- FIRST PROTOTYPE ENVIRONMENT ;

24TH MONTH

- INITIAL COMMON TOOL SET ;

36TH MONTH

- INITIAL ENVIRONMENT , AND FULL COMMON TOOL SET AND GENERAL SERVICES .

R AND D AREA 2.4 :

PROCESS , MANAGEMENT AND ENVIRONMENT - EVOLVING

DESCRIPTION

PROJECTS IN THIS AREA SHOULD ADDRESS THE TOTAL ACTIVITY OF SOFTWARE DEVELOPMENT FROM THE EMERGENCE OR STATEMENT OF A COMPUTER APPLICATION CONCEPT OR OF A PROBLEM TO BE SOLVED , THROUGH THE INSTALLATION AND OPERATION OF THE APPROPRIATE SOFTWARE OR SOFTWARE BASED SYSTEM , AND THROUGHOUT ITS SUBSEQUENT LIFE-TIME IN WHICH CONTINUING ADAPTATION TO CHANGING ENVIRONMENTS AND CHANGING USER PERCEPTIONS IS INEVITABLE .

THE PROGRAMME PROVIDES BOTH FOR A TOTALLY COHERENT PROCESS APPROACH AND FOR EXPLORING FUNDAMENTALLY NEW APPROACHES . THE EMPHASIS IS THEREFORE ON LONGERTERM RESEARCH .

IT IS ANTICIPATED THAT OVERALL STRUCTURAL CHANGES IN SOFTWARE MANAGEMENT WILL RESULT FROM IMPROVED UNDERSTANDING OF THE WHOLE SOFTWARE PROCESS . IN ADDITION DETAILED EVOLUTION WILL FOLLOW IMPROVEMENTS IN SUCH AREAS AS MEASUREMENTS , QUALITY ASSURANCE AND SOFTWARE TOOLS SUPPORT . PART OF THE WORK IN THIS AREA IS THEREFORE DIRECTED AT SIGNIFICANT IMPROVEMENT IN SOFTWARE MANAGEMENT TECHNIQUES .

ANOTHER PART INTENDS TO PROVIDE THE BASIS FOR THE MORE ADVANCED ENVIRONMENTS OF THE FUTURE . THIS WILL INVOLVE THE APPLICATION OF AIP TECHNIQUES ( SEE SUBPROGRAMME 3 ) TO THE PROBLEMS OF SOFTWARE ENVIRONMENTS . SUCH ENVIRONMENTS MIGHT BE EXPECTED TO BE EXTENSIBLE AND PLAY AN ACTIVE RATHER THAN A PASSIVE ROLE . ON THE OTHER HAND , REQUIREMENTS COMING FROM THE AIP DOMAIN MUST BE TAKEN INTO ACCOUNT TO PROVIDE ENVIRONMENTS SUITED TO THEIR NEEDS . MOREOVER , IT IS DESIRABLE THAT APPROPRIATE INTERFACES BETWEEN THESE AND THE INITIAL COMMON ENVIRONMENT ( SEE 2.3 ) BE PROVIDED .

R AND D TOPICS

THE TOPICS DESCRIBED HERE ARE OF LONG-TERM AND SPECULATIVE NATURE . DETAILS OF THE GROUPING OF ALL TOPICS INTO PACKAGES CANNOT BE GIVEN AT THIS STAGE AND ONLY ONE TYPE A PROJECT HAS BEEN DEFINED . HOWEVER , IT IS EXPECTED THAT PROPOSALS WILL PROVIDE OTHER SUCH GROUPINGS , AND UP TO THREE TYPE A PROJECTS COULD BE ACCOMMODATED . THEY WILL INITIALLY HAVE A PREDOMINANTLY RESEARCH ORIENTATION , BUT INDUSTRIAL TAKE-UP MUST BE ENSURED FROM THE BEGINNING , BOTH BY THE COMBINATION OF RESEARCH TOPICS AND BY THE STRUCTURE OF THE PARTICIPATION .

WORK IN THIS AREA WILL ADDRESS :

- IMPROVED MODELS FOR SOFTWARE PRODUCTION AND MAINTENANCE ( WITH SPECIAL REFERENCE TO THE COHERENCE OF THE PROCESSES MODELLED ) ; BOTH THE TECHNICAL AND THE MANAGEMENT ASPECTS NEED TO BE ADDRESSED ;

- CONSTRUCTION AND TRANSFORMATION OF REPRESENTATIONS ( INCLUDING AUTOMATIC AND INTERACTIVE PROGRAM SYNTHESIS , WITH SPECIAL ATTENTION FOR GENERIC AND RE-USABLE SOFTWARE AND FOR VALIDATION AND VERIFICATION ) ;

- GENERAL METHODS AND TOOLS FOR APPLICATION DOMAIN ANALYSIS WITH SPECIAL REFERENCE TO CAPTURING OF REQUIREMENTS ;

- HARDWARE/SOFTWARE SYNERGY AND NEW ARCHITECTURES ;

- NON-IMPERATIVE LANGUAGES ;

- CONVERGENCE OF TECHNOLOGIES IN PROGRAM SPECIFICATION , PROGRAM IMPLEMENTATION AND DATABASE DESIGN ;

- MEASUREMENT AND MODELLING ;

- QUALITY AND RELIABILITY ;

- EXPERT SYSTEM APPROACHES TO SOFTWARE PRODUCTION AND MAINTENANCE , INVOLVING ACTIVE DATABASES , MANAGEMENT EXPERT SYSTEMS ;

- ADVANCED SUPPORT ENVIRONMENTS AND HUMAN INTERFACING FOR THE DEVELOPMENT OF HIGHLY DEMANDING APPLICATIONS SUCH AS ADVANCED INFORMATION PROCESSING APPLICATIONS .

TYPE A PROJECT

2.4.1 . ADVANCED INTERACTIVE SOFTWARE DEVELOPMENT

THE OBJECTIVES ARE :

- TO INVESTIGATE THE POSSIBILITY OF DEVELOPING AN APPROACH WHICH BEGINS TO INTEGRATE THE RIGOUR OF THE " SOFTWARE ENGINEERING " APPROACH AND THE FLEXIBILITY OF THE EXPERIMENTAL AND " THROWAWAY " STYLE OF SOFTWARE DEVELOPMENT EXEMPLIFIED BY KNOWLEDGE PROCESSING WORK ;

- TO PROVIDE AN ENVIRONMENT SUPPORTING THIS FORMAL INTERACTIVE STYLE OF PROGRAM DEVELOPMENT . THIS ENVIRONMENT SHOULD PROVIDE INTEGRATED SUPPORT FOR EXISTING LANGUAGES AND METHODS OF SEVERAL CLASSES ( FUNCTIONAL , LOGIC , OBJECT-ORIENTED . . . ) ;

- TO PROVIDE APPROPRIATE INTERFACES TO ENSURE EFFECTIVE INTERWORKING BETWEEN THIS ENVIRONMENT AND THE INITIAL COMMON TOOL ENVIRONMENT ( SEE 2.3.1 ABOVE ) ;

- TO PROVIDE FACILITIES FOR RAPID INCORPORATION OF FUTURE DEVELOPMENTS IN DIFFERENT CLASSES OF LANGUAGES .

INTERMEDIATE OBJECTIVES

12TH MONTH

- COMMON APPROACH CHARACTERISTICS DESCRIPTION ;

24TH MONTH

- EXPERIMENTAL INTERACTIVE ENVIRONMENT ;

48TH MONTH

- ADVANCED INTERACTIVE ENVIRONMENT ;

60TH MONTH

- INTERFACING WITH COMMON TOOL ENVIRONMENT .

TYPE B PROJECTS

THESE CAN ADDRESS ANY OF THE TOPICS MENTIONED ABOVE .

R AND D AREA 2.5 :

DEMONSTRATION PROJECTS

THERE IS A SIGNIFICANT BARRIER TO THE INTRODUCTION OF NEW METHODS AND TOOLS INTO INDUSTRY - NO PROJECT MANAGER WANTS TO BE THE FIRST TO ATTEMPT PRACTICAL USAGE OF A NEW APPROACH . THIS BARRIER CAN BE ADDRESSED BY PROVIDING FINANCIAL SUPPORT FOR DEMONSTRATION PROJECTS . THE INTENTION IS NOT TO CREATE AN ARTIFICIAL PROJECT SOLELY TO DEMONSTRATE SOME NEW APPROACH , BUT RATHER TO USE A REAL PROJECT AS AN EVALUATION AND DEMONSTRATION VEHICLE . THE FINANCIAL SUPPORT SHOULD THEN COVER THE ADDITIONAL COSTS THAT ARE INCURRED IN PERFORMING THIS EVALUATION AND DEMONSTRATION ROLE . IN PARTICULAR , THE FUNDING MUST COVER TRAINING AND FAMILIARIZATION OVERHEADS AT THE START OF THE PROJECT , AND THE PREPARATION OF REPORTS AND CRITIQUES OF METHODS AND TOOLS FOR THE BENEFIT OF THE COMMUNITY AS A WHOLE . THE FUNDING OF DEMONSTRATION PROJECTS SHOULD BE CONDITIONAL UPON THE PRODUCTION OF SUCH REPORTS AND CRITIQUES .

IT IS IMPORTANT THAT THE RESULTS OF SUCH DEMONSTRATION PROJECTS SHOULD BE SEEN TO BE VALID AND CONCLUSIVE ; THIS REQUIRES , AMONG OTHER THINGS , THAT THE DESIGN AND PLANNING OF SUCH DEMONSTRATIONS SHOULD BE EVALUATED FROM THE PERSPECTIVE OF THE STATISTICAL THEORY OF " DESIGN OF EXPERIMENTS " .

PROJECTS OF THIS TYPE ARE EXPECTED TO START IN YEARS 2 AND 3 ( I . E . 1985/86 AND 1986/87 ) , AND CANNOT BE PLANNED IN ANY DETAIL AT THIS STAGE .

SUBPROGRAMME 3

ADVANCED INFORMATION PROCESSING ( AIP )

THE MAIN THEME IS MACHINE INTELLIGENCE . THIS EMBRACES THE PROCESS OF EXTRACTING KNOWLEDGE , STORING THAT KNOWLEDGE FOR EASY ACCESS AND USE AND TO ENHANCE THE KNOWLEDGE BASE THROUGH USAGE . A MAJOR AIM IS TO BRIDGE THE GAP BETWEEN THE COMPUTER AND THE NON-TECHNICAL USER BY USING COMPUTER-BASED INTELLIGENCE TO ASSIST THE USER AND PRESENT AN EASY , FRIENDLY INTERFACE .

INITIAL THEORETICAL WORK WILL BE ON THE PROCESS OF REASONING FROM WHICH TO BASE THE DESIGN OF ADVANCED EXPERT OR KNOWLEDGE-BASED SYSTEMS . WHILE THE MAIN EMPHASIS IN THE EARLY YEARS WILL BE KNOWLEDGE ENGINEERING AND KNOWLEDGE STORAGE AND USAGE , THE EMPHASIS LATER WILL BE ON THE EXPLORATION OF NEW ARCHITECTURES AND THE IMPLEMENTATION OF SYSTEMS . IN PARALLEL , RESEARCH WILL BE UNDERTAKEN ON EXTERNAL INTERFACES , THE INTERACTION OF MAN AND THE MACHINE . AN IMPORTANT FEATURE OF THIS WORK WILL BE SIGNAL ANALYSIS AND PROCESSING , PARTICULARLY FOR HANDLING VISUAL DATA . WORK WILL ALSO BE CARRIED OUT ON THE SPECIFICATION AND DESIGN FACETS OF AIP SYSTEMS .

EARLY IN THE PROGRAMME A NUMBER OF DEMONSTRATOR AIP SYSTEMS WILL BE BUILT USING CONVENTIONAL EQUIPMENT . THESE WILL BE BUILT AS QUICKLY AS POSSIBLE TO ACCELERATE THE CORE OF KNOWLEDGE AND EXPERIENCE IN AIP SYSTEMS AND THROUGH ACCESSIBLE DEMONSTRATORS OBTAIN CONTRIBUTIONS FROM THE WIDER RESEARCH COMMUNITY . IN ADDITION , PROVISION IS MADE FOR A NUMBER OF INTEGRATED , INTERDISCIPLINARY " FOCUS " PROJECTS , THAT SPAN NOT ONLY THE FIVE MAIN ASPECTS OF AIP BUT ALSO THE OTHER AREAS OF THE ESPRIT PROGRAMME .

R AND D AREA 3.1 :

KNOWLEDGE ENGINEERING

DESCRIPTION

THIS R AND D AREA IS CONCERNED WITH THE TOOLS AND TECHNOLOGIES WHICH WILL BE NEEDED FOR THE PRACTICE OF KNOWLEDGE ENGINEERING IN ORDER TO REALIZE COMMERCIALLY AND SOCIALLY ACCEPTABLE KNOWLEDGE-BASED SYSTEMS APPLICATIONS , OF WHICH EXPERT SYSTEMS , DECISION SUPPORT AND COMPUTER AIDED INSTRUCTION ARE EXAMPLES .

KNOWLEDGE ENGINEERING INVOLVES THE PRACTICE OF :

- BUSINESS ANALYSIS TO DETERMINE THE VIABLE DOMAINS FOR KNOWLEDGE-BASED SYSTEMS ( KBS ) APPLICATION ,

- KNOWLEDGE ACQUISITION IN THE CHOSEN DOMAIN ABOUT THE OBJECTS OF REASONING , THEIR TAXONOMY , LINES OF REASONING , HEURISTICS , FACTS , RULES AND DIALOGUE ,

- CLASSIFICATION OF THE DOMAIN CHARACTERISTICS AND SELECTION OF THE MOST APPROPRIATE CORRESPONDING KNOWLEDGE REPRESENTATIONS AND INFERENCE MODELS , LATER TO BE EXTENDED TO INCLUDE KNOWLEDGE TRANSFORMATION FACILITIES ,

- REALIZATION OF THE KBS APPLICATIONS ON THE MOST APPROPRIATE DELIVERY VEHICLE IN TERMS OF MMI , ENGINE CAPABILITY , DATA/KNOWLEDGE BASE CAPABILITIES , AND STORAGE CAPABILITIES AS WELL AS COMMUNICATION FACILITIES ,

- CUSTOMIZATION OF THE SYSTEM FOR USERS BY PROVIDING AN ADAPTIVE INTERFACE WHICH MINIMIZES COGNITIVE LOAD IN TERMS OF THE DIALOGUE WITH THE SYSTEM ,

- MANIPULATION OF KNOWLEDGE FOR MAINTENANCE OF INSTALLED SYSTEMS , AND FOR TRANSFORMATION OF KNOWLEDGE FOR THE CONSTRUCTION OF NEW KBS APPLICATION .

R AND D TOPICS

WORK WILL RESEARCH , SPECIFY , CLASSIFY AND PROVISION THE TOOLS AND TECHNIQUES OF KNOWLEDGE ENGINEERING , AND WILL COVER :

3.1.1 . ( 1 ) KNOWLEDGE-BASED SYSTEMS AND THEIR METRICATION

3.1.2 . ( 1 ) DIALOGUE AND NATURAL LANGUAGE

3.1.3 . ( 1 ) KNOWLEDGE REPRESENTATION AND INFERENCE TECHNIQUES

3.1.4 . PREPARATORY WORK TOWARDS ADVANCED KBS

3.1.4.1 . IMPLEMENTATION LANGUAGES AND ENVIRONMENTS ( 1 )

3.1.4.2 . COMPILERS AND INTERPRETERS

3.1.4.3 . KNOWLEDGE ACQUISITION AND MANIPULATION

3.1.4.4 . ADVANCED GENERATION KBS APPLICATIONS

3.1.4.5 . LEARNING TECHNIQUES ( 1 )

WORK IN THE FIRST YEARS WILL CONCENTRATE ON GAINING EXPERIENCE BUILDING KBS , PARTICULARLY EXPERT SYSTEMS , USING CONVENTIONAL EQUIPMENT , SO AS TO UNDERSTAND AND EXTEND CAPABILITIES . THIS WOULD BE FOLLOWED SUBSEQUENTLY WITH BUILDING AN ADVANCED GENERATION OF KBS , INCORPORATING THE RESULTS OF ALL ASSOCIATED AIP RESEARCH AND RESULTS FROM THE SOFTWARE TECHNOLOGY PROGRAMME .

TOPICS 3.1.1 AND 3.1.4.4 JOINTLY REPRESENT A ROLLING PROGRAMME OF DEMONSTRATORS FOR EARLY AND ADVANCED KBS . ALSO RESEARCH TOPICS 3.1.3 AND 3.1.4.1 ARE EXPECTED TO BE VERY CLOSELY RELATED , WITH A PRIME OUTPUT INTO THE ADVANCED PHASE OF DEMONSTRATORS . HOWEVER , IT IS EXPECTED DEMONSTRATORS DURING THE FIRST FIVE YEARS WILL CONTINUOUSLY FEED PROBLEMS AND WEAKNESSES INTO THE ABOVE GROUP OF RESEARCH TOPICS , AND CONVERSELY , EACH NEW ROLLING DEMONSTRATOR CAN USE EARLY RESEARCH RESULTS CURRENT AT THE TIME OF INCEPTION , WHICH IT WILL THEN KEEP FROZEN FOR ITS TWO TO THREE YEAR DURATION .

TYPE A PROJECTS

3.1.1 . KNOWLEDGE-BASED SYSTEMS AND THEIR METRICATION

THE OBJECTIVE IS AN ON-GOING PROGRAMME , OVER AT LEAST FIVE YEARS , OF CONSTRUCTION OF DEMONSTRATOR KBS APPLICATIONS , USING STATE-OF-ART FACILITIES ( HARDWARE , SOFTWARE AND LARGE DATABASES ) AVAILABLE AT THE TIME , SO AS TO MAINTAIN AN UP-TO-DATE POOL OF EXPERTISE ON KNOWLEDGE ENGINEERING SKILLS AND ON CLASSIFICATION OF DOMAIN CHARACTERISTICS AGAINST AVAILABLE TOOLS AND METHODS . IT WILL BE A VEHICLE FOR LEARNING ABOUT THE STRENGTHS AND WEAKNESSES OF THE TOOLS AND METHODS USED FOR PROCESS ANALYSIS , KNOWLEDGE REPRESENTATION AND MANIPULATION , INFERENCE AND PROCESSING , PROBLEM SOLVING PARADIGMS AND HUMAN FACTORS/MMI .

THIS PARTICULAR PROJECT COVERS THE FIRST PHASE OF THE ON-GOING DEMONSTRATOR PROGRAMME , WITH A SWITCH OVER AT RESOURCES FROM THE FIFTH YEAR INTO PROJECT 3.1.4 FOR THE SECOND PHASE OF THE DEMONSTRATOR PROGRAMME .

SINCE THE TOOLS AND METHODS VARY FOR DOMAIN CLASSES SHOWING MAJOR DIFFERENCES IN CHARACTER , DEMONSTRATORS SHOULD INCLUDE WIDELY VARIED EXAMPLES IN FIELDS OF CREATIVE DESIGN , PRODUCTION CONTROL DIAGNOSIS , SYSTEMS MODELLING , SIGNAL PROCESSING , DECISION MAKING AND SOFTWARE ENGINEERING . AS SUCH , THIS PROGRAMME WILL BE A CRUCIAL SOURCE OF EDUCATION , AND OF DIRECTION FOR ON-GOING/FUTURE RESEARCH PROGRAMMES , AND OF EXPERIENCE FOR MAKING MARKETABLE KBS APPLICATIONS .

ALMOST NOTHING IS KNOWN ABOUT THE PRACTICE OF KNOWLEDGE ENGINEERING AS A HUMAN SKILL , OR ABOUT KBS PERFORMANCE , EVALUATION , MEASUREMENT AND ACCEPTABILITY IN TERMS OF CLIENT PRODUCTIVITY . IT IS NECESSARY TO IDENTIFY THOSE METRICS WHICH ARE OF SIGNIFICANCE AND TO ESTABLISH METHODS AND TOOLS FOR THEIR MEASUREMENT - QUALITY OF KNOWLEDGE IS CRUCIAL BUT DIFFICULT TO METRICATE . THIS WORK MUST ALSO CREATE A " TOOLS AND METHODS " KIT FOR KNOWLEDGE ENGINEERS TO SIZE KBS APPLICATIONS IN TERMS OF CONSTRUCTION AND CLIENT USAGE RESOURCES , AND IN TERMS OF THE MOST APPROPRIATE SYSTEM COMPONENTS .

KNOWLEDGE ENGINEERING WILL BE A BIGGER BOTTLENECK THAN SOFTWARE PRODUCTION IF THE ISSUES OF HUMAN PRODUCTIVITY , AS WELL AS SYSTEM PERFORMANCE , ARE NOT UNDERSTOOD . AS SUCH , THIS WORK IS OF KEY IMPORTANCE IN THE TOP DOWN DESIGN OF KBS APPLICATIONS . SINCE NEW TOOLS AND TECHNIQUES ARE ADDED CONTINUALLY TO THE KNOWLEDGE ENGINEERING REPERTOIRE DURING THE ESPRIT PROGRAMME , THIS METRICATION WORK SHOULD BE ORGANIZED AS A CONTINOUSLY ROLLING PROGRAMME .

PROGRAMME AND INTERMEDIATE OBJECTIVES

YEAR 1 :

- SELECT APPLICATION DOMAIN ;

- CHOOSE APPROPRIATE TOOLS AND METHODS ;

- ESTABLISH IMPLEMENTATION ENVIRONMENT ;

- IDENTIFY METRICATION ASPECTS FOR EVALUATION ; DEVELOP CONCEPTS AND METHODS FOR SOLUTIONS .

YEAR 2 :

- CHOOSE METRICATION TOOLS AND METHODS TO BE DEVELOPED ;

- DEFINE WHAT MEASURES ARE TO BE MADE AND ANALYZED ;

- DEFINE THE APPLICATIONS TO BE INSTRUMENTED AND MEASURED .

YEARS 1 TO 3 :

- ASSEMBLE KBS DEMONSTRATOR ;

- RECORD CONSTRUCTION METHODS AND RESULTS ;

- RECORD NEW DOMAIN CHARACTERISTICS .

YEAR 3 :

- MEASUREMENT AND ANALYSIS OF PRODUCTIVITY AND PERFORMANCE FOR THE TARGET APPLICATIONS .

YEARS 2 TO 4 :

- CARRY OUT EXPERIMENTS ON DEMONSTRATORS TO DETERMINE ACCEPTABILITY AND IMPACT ON " CLIENT " PRODUCTIVITY ;

- REPORT ON CONSTRUCTION PRODUCTIVITY AND RECOMMEND NEW TOOLS AND METHODS .

YEAR 4 :

- PUBLISH PRODUCTIVITY AND PERFORMANCE RECOMMENDATIONS , AND GUIDELINES FOR KNOWLEDGE ENGINEERS ;

- PROPOSE NEW METRICATION METHODS , ESPECIALLY THOSE WHICH ARE MORE AUTOMATED .

3.1.2 . DIALOGUE AND NATURAL LANGUAGE

DESCRIPTION

A MAJOR WORK PROGRAMME IS REQUIRED ON THE TOOLS AND METHODS FOR DIALOGUE AND COMMUNICATION LANGUAGES WHICH ARE NATURAL TO THE DOMAINS IN WHICH KBS APPLICATIONS ARE TO BE COMPETENT . INCLUDED IN THIS IS THE APPLICATION OF KNOWLEDGE AND CONTEXT TO THE CONDUCT AND UNDERSTANDING OF TWO-WAY DIALOGUE , AND OF " INTELLIGENCE " TO THE INTERCHANGE OF DIALOGUE INITIATIVE BETWEEN SYSTEMS AND CLIENT USERS .

THE DIALOGUE ITSELF SHOULD ENHANCE THE QUALITY OF SYSTEM BEHAVIOUR BY OFFERING COOPERATION , UNDERSTANDING OF PERSONAL BEHAVIOUR AND PREFERENCE AND PARAPHRASE ( ALL OF WHICH ARE KNOWLEDGE BASED ) ; IT SHOULD ALSO ALLOW CLIENTS TO INTERACT WITH THE SYSTEM AT VARIOUS META-LEVELS WHILST KEEPING TRACK OF ALL THE CONTEXT WHICH MAY HAVE TO BE RETURNED TO BY SYSTEM OR CLIENT , FOLLOWING ERROR OR CONFUSION . EXPLANATION OF KBS BEHAVIOUR IS ABSOLUTELY CRITICAL TO ACCEPTANCE OF SUCH SYSTEMS ; THIS CAN BE ENHANCED BY INTERWORKING OF THE KNOWLEDGE ABOUT THE USER WITH THAT OF THE APPLICATION .

DIALOGUE IS MORE GENERAL THAN CHARACTER STRING COMMUNICATION ; IT EXTENDS TO USE OF HIGH BANDWIDTH GRAPHICS , EXPLANATORY PICTURES AND FILMS , AND TO THE USE OF AUDIO SUPPORT .

PROGRAMME AND INTERMEDIATE OBJECTIVES

YEAR 1 :

- STUDY OF REQUIRED LANGUAGE STYLES AND CAPABILITIES ;

- SPECIFY FRAMEWORK FOR EXPLOITATION OF HIGH BANDWIDTH MMI .

YEARS 1 AND 2 :

- UNDERSTAND DIALOGUE REQUIREMENTS CLASSIFIED BY DOMAIN OF APPLICATION .

YEARS 2 TO 4 :

- CREATE GENERATORS AND PARSERS FOR A VARIETY OF DIALOGUE LANGUAGES .

YEARS 3 AND 4 :

- SPECIFY KNOWLEDGE-BASED LANGUAGE PROCESSING TOOLS .

YEARS 4 TO 7 :

- IMPLEMENT KNOWLEDGE-BASED LANGUAGE PROCESSING/DIALOGUE TOOLS FOR INTEGRATION INTO KBS APPLICATIONS .

3.1.3 . KNOWLEDGE REPRESENTATION AND INFERENCE TECHNIQUES

DESCRIPTION

KNOWLEDGE REPRESENTATION IS A KEY FACTOR ANALOGOUS TO DATA MODELLING IN CONVENTIONAL PROCESSING . BECAUSE OF ITS COMPLEXITY , IT REQUIRES ENGINEERING , MODELLING AND TRANSFORMATION TOOLS OF HIGH SOPHISTICATION . IT IS NOT EXPECTED THAT ANY ONE LANGUAGE WILL MERGE - RATHER THAT THERE WILL BE A WIDE RANGE OF STANDARDS AND TECHNIQUES , WHICH VARY ACCORDING TO DOMAIN CLASSIFICATION , PARADIGMS EMPLOYED , AND TO INFERENCE/PROCESSING MODELS USED . A COMPREHENSIVE EFFORT IS NEEDED TO DIFFERENTIATE , CLASSIFY , AND EXTEND KNOWLEDGE REPRESENTATIONS FOR THESE REQUIREMENTS . IT IS EXPECTED THAT MORE SOPHISTICATED APPLICATIONS WILL NEED TO WORK WITH MULTIPLE DIFFERING , BUT COOPERATING , KNOWLEDGE DOMAINS , AND THUS RESEARCH ON INTERWORKING KRS AND INFERENCE MODELS WILL BE ESSENTIAL .

EXISTING KBS APPLICATIONS USE TOOLS AND METHODS WHICH LARGELY INCORPORATE SIMPLE RULES OF INFERENCE - ESPECIALLY , LITTLE ( META ) KNOWLEDGE IS APPLIED TO THE USE OF KNOWLEDGE AND TO CONTROL OF INFERENCE . THIS LEAVES A LOT TO BE DESIRED IN EXPLOITING HIGH QUALITY KNOWLEDGE AND HEURISTICS , WHICH ARE OFTEN THE KEY TO KBS EFFECTIVENESS RATHER THAN BRUTE FORCE HARDWARE SPEED . BASIC RESEARCH SHOULD INCLUDE :

- NATURAL DEDUCTION TECHNIQUE NOT REQUIRING SPECIAL IMPLICATIVE FORM ;

- NON-CLASSICAL LOGIC SUCH AS TEMPORAL AND FUZZY LOGIC , WHICH ARE CLOSER TO REAL-WORLD TIME RELATED AND QUALITATIVE BEHAVIOUR , BUT WHICH LACK PRACTICAL PROOF ALGORITHMS ;

- KNOWLEDGE-BASED CONTROL OF THE INFERENCE PROCESS ITSELF ;

- CONSTRAINT PROPAGATION AND TRUTH MAINTENANCE ;

- EXPLOITATION OF CONTEXTUAL KNOWLEDGE ;

- COST/QUALITY DIRECTED PARADIGMS ;

- CLASSIFICATION OF INFERENCE TOOLS AND METHODS AGAINST DOMAINS , KRS AND PARADIGMS ;

- RECOGNITION OF THE NEED FOR REAL TIME TRANSITION OF SYSTEM STATE ( AND THUS IMPLICATED KNOWLEDGE ) AS A RESULT OF REAL-WORLD CHANGE .

PROGRAMME AND INTERMEDIATE OBJECTIVES :

KNOWLEDGE REPRESENTATIONS :

YEARS 1 AND 2 :

- COMPILE ALL EXISTING KRS PROPOSALS ;

- CROSS-CHECK AND FERTILIZE SO AS TO DIFFERENTIATE AND CLASSIFY KRS ;

- DETERMINE WHAT IS SPECIFICALLY AND GENERALLY NEEDED ;

- IMPACT OF PSYCHOLOGY ON KRS .

YEARS 2 AND 3 :

- EXTEND AND CONSOLIDATE - REFERRING TO MODELS SUCH AS DATA DICTIONARIES , DATABASES AND PROGRAMMING LANGUAGES .

YEARS 3 TO 5 :

- PROPOSE NEW KRS TOOL KITS AND METHODS , AND INTRODUCE THESE INTO PROJECT 3.1.1 FOR " ROLLING " EXPLOITATION .

YEARS 4 TO 6 :

- PUBLISH EUROPEAN GUIDELINES FOR COMPONENTS OF KRS TOOL KITS ;

- REFERENCE " PRODUCTION " IMPLEMENTATION OF NEW KRS .

INFERENCE TECHNIQUES :

YEARS 1 AND 2 :

- DETERMINE STATE-OF-ART IN TOOLS AND METHODS , AND CLASSIFY AGAINST DOMAINS , KRS AND PARADIGMS ;

- SET UP SUBPROJECTS FOR RESEARCH .

YEARS 3 AND 4 :

- UNDERTAKE BASIC RESEARCH ON NATURAL DEDUCTION , TEMPORAL LOGIC , FUZZY LOGIC , INFERENCE PRIMITIVES , AND KNOWLEDGE BASED META-LEVEL CONTROL OF INFERENCE .

YEARS 5 AND 6 :

- CONSTRUCT INFERENCE MODELS FOR OUTPUT TO THE DEMONSTRATOR AND METRICS PROGRAMMES ;

- CONTINUE RESEARCH ON APPROXIMATE REASONING AND META-LEVEL CONTROL .

3.1.4 . PREPARATORY WORK TOWARDS ADVANCED KBS

ALTHOUGH THIS IS CATEGORIZED AS AN A-PROJECT , WORK IN THE FIRST FEW YEARS WILL BE AT A COMPARATIVELY LOW LEVEL AND THEN BUILD UP AS ADVANCED GENERATION KBS BECOME FEASIBLE . THE PREPARATORY WORK WILL INCLUDE :

3.1.4.1 . IMPLEMENTATION LANGUAGES AND ENVIRONMENTS ;

3.1.4.2 . COMPILERS AND INTERPRETERS ;

3.1.4.3 . KNOWLEDGE ACQUISITION AND MANIPULATION ;

3.1.4.4 . ADVANCED GENERATION KBS APPLICATIONS ;

3.1.4.5 . LEARNING TECHNIQUES ,

THOUGH IT IS POSSIBLE THAT OTHER TOPICS IN THIS AREA WILL BECOME IDENTIFIED AS WORK PROCEEDS .

ALL FIVE OF THE ABOVE TOPIC AREAS ARE BRIEFLY DESCRIBED BELOW , BUT ONLY THE TWO WHICH ARE PLANNED TO START IN YEAR 1 ( NAMELY 3.1.4.1 AND 3.1.4.5 ) HAVE PROGRAMME OUTLINES INCLUDED .

3.1.4.1 . IMPLEMENTATION LANGUAGES AND ENVIRONMENTS

DESCRIPTION

EVALUATION OF , RECOMMENDATIONS ON , AND EXTENSION OF , CANDIDATE IMPLEMENTATION LANGUAGES FOR TOOLS OF THE RESEARCH PROGRAMME AND AS TOOLS FOR KNOWLEDGE ENGINEERS ENGAGED IN THE CONSTRUCTION OF KBS APPLICATIONS . THIS SHOULD INCLUDE LANGUAGES FOR KNOWLEDGE MODELLING AND FOR SPECIFICATION/DESIGN PROTOTYPING OF KBS COMPONENTS . LATER WORK SHOULD CONCENTRATE ON ESTABLISHING STANDARDS , WHICH MANUFACTURERS WILL THEN TREAT AS A SAFE BASE FOR SIGNIFICANT HARDWARE INVESTMENT . AN IMPORTANT ASPECT OF LANGUAGE EVALUATION IS THE RICHNESS AND UTILITY OF THEIR EMBEDDED DEVELOPMENT ENVIRONMENTS . IT IS EXPECTED THAT THE SOFTWARE ENGINEERING COMMON TOOLS PROGRAMME WILL ADDRESS PORTABILITY .

PROGRAMME AND INTERMEDIATE OBJECTIVES

YEARS 1 TO 3 :

- IDENTIFY AND EVALUATE CANDIDATE LANGUAGES ( FOR EXAMPLE , SMALLTALK , LISP , PROLOG , POPLOG ) ENVIRONMENTS AND ARCHITECTURES .

YEARS 4 AND 5 :

- PUBLISH CRITIQUES AND RECOMMEND CHANGES E . G . LANGUAGE EXTENSIONS TO EXPLOIT PARALLELISM ;

- PROCURE EXTENSIONS/CHANGES .

YEARS 5 AND 6 :

- ISSUE GUIDELINES TO ESPRIT ;

- MAKE REFERENCE LANGUAGE SYSTEMS AVAILABLE .

3.1.4.2 . COMPILERS AND INTERPRETERS

THE OBJECTIVE IS THE SOFTWARE IMPLEMENTATION OF INTERPRETERS/COMPILERS FOR THE VARIOUS KNOWLEDGE REPRESENTATIONS AND MANIPULATION METHODS DEVELOPED FROM 3.1.3 AND 3.1.4.1 .

3.1.4.3 . KNOWLEDGE ACQUISITION AND MANIPULATION

FOLLOWING EARLY WORK ON KNOWLEDGE REPRESENTATION ( 3.1.3 ) , EXPERIENCE FROM EARLY APPLICATIONS AND STUDIES OF KNOWLEDGE ENGINEERING MEASURES ( 3.1.1 ) , AND STUDIES OF DIALOGUE ( 3.1.2 ) , THE NEEDS FOR METHODS AND TOOLS ASSOCIATED WITH THE PRACTICE OF THE KNOWLEDGE ENGINEERING SKILL WILL BECOME CLEAR . THE ENVIRONMENT FOR KNOWLEDGE ACQUISITION AND MANIPULATION WILL OF COURSE BE COMPLETELY INTERACTIVE , AND BE AIDED WITH GRAPHICS . WORK IN THIS AREA WILL INCLUDE :

- KNOWLEDGE EDITING , UPDATING AND RETRIEVAL ,

- KRS PROOF CHECKERS ,

- KNOWLEDGE DICTIONARY SYSTEMS ,

- KNOWLEDGE TRANSFORMATION ( DIFFERENT VIEWS OF COMMON KNOWLEDGE ) ,

- KNOWLEDGE ABSTRACTION FROM EXTERNAL SOURCES ,

- KNOWLEDGE ACQUISITION/PROCESS ANALYSIS ,

- GUIDE AND TUTOR FOR KNOWLEDGE ENGINEERING ITSELF ,

- KBS DISTRIBUTION AND AUTOMATED KNOWLEDGE MAINTENANCE ,

- MMI TOOL KITS .

THIS DEVELOPING SET OF CAPABILITIES SHOULD BE INTEGRATED INTO RECOMMENDED " EUROPEAN " TOOL KITS FOR HANDLING KNOWLEDGE .

3.1.4.4 . ADVANCED GENERATION KBS APPLICATIONS

THIS PROJECT WILL COVER THE DEVELOPMENT OF ADVANCED EXPERT AND KNOWLEDGE-BASED SYSTEMS INCORPORATING LEARNING , APPROXIMATE AND META-REASONING WITH COMMON-SENSE CAPABILITIES . KBS APPLICATIONS WILL HAVE TO BE EXPLOITED BY NON-SPECIALISTS WHO WILL EXPECT FROM THESE SYSTEMS NOT ONLY FORMAL AND APPROXIMATE REASONING PROCESSES , BUT ALSO COMMON-SENSE REASONING PROCESSES SIMILAR IN EFFECT TO THOSE OF HUMANS .

MANY APPLICATIONS WHERE REASONING IS NEEDED INVOLVE SOME KIND OF APPROXIMATE REASONING . FOR INSTANCE , APPLICATIONS OF MODELLING WHERE QUALITATIVE HEURISTICS CHARACTERIZE MOST LEARNED HUMAN EXPERIENCE .

MANY TIMES A PROBLEM HAS NO SIMPLE SOLUTION AT A GIVEN REASONING LEVEL ; ABSTRACTING FROM THAT LEVEL ALLOWS ONE TO ARGUE ABOUT THE REASONING PROCESS ITSELF AND GIVES RISE TO SO-CALLED META-REASONING . META-REASONING DEALS WITH TWO TYPES OF PROBLEMS : EXTENDING INFERENCE TECHNIQUES FOR FINDING CONCEPTUAL SOLUTIONS TO PROBLEMS , AND REASONING ABOUT INFERENCE TECHNIQUES OR RULES FOR MORE EFFICIENT EXPLOITATION OF THESE INFERENCE RULES .

THE ADVANCED GENERATION OF KBS APPLICATIONS WILL BE EXPECTED TO EXPLOIT THE LESSONS OF FIRST GENERATION ATTEMPTS ( 3.1.1 ) , THE FRUITS OF KRS AND INFERENCE RESEARCH ( 3.1.3 ) , THE TOOLS OF KNOWLEDGE AND ACQUISITION MANIPULATION ( 3.1.4.3 ) , THE LATEST LANGUAGES AND ENGINES ( 3.1.4.1 AND 3.1.4.2 ) , THE FRUITS OF NATURAL LANGUAGE PROCESSING RESEARCH ( 3.1.2 ) , AND THE FRUITS OF THE EXTERNAL INTERFACES PROGRAMME ( 3.2 ) FOR HANDLING SOUND AND IMAGES .

3.1.4.5 . LEARNING TECHNIQUES

DESCRIPTION

THE KNOWLEDGE ACQUISITION PROCESS IS COMPLEX AND THEREFORE IT IS MANDATORY TO DEVELOP CONCEPTS , ALGORITHMS AND TECHNIQUES TO SUPPORT THE DEVELOPMENT OF PROGRAMS AND MACHINES WHICH LEARN . THIS TASK COVERS THE IDENTIFICATION OF APPROPRIATE MODELS OF LEARNING AND THE INVESTIGATION OF LEARNING TECHNIQUES APPROPRIATE FOR AIP SYSTEMS .

LEARNING TECHNIQUES MIGHT BE BASED ON THE CAPABILITY OF THE ARTIFICIAL SYSTEM TO MEASURE ITS PERFORMANCE AND TO DEVELOP AN EVOLUTIONARY PROCESS WHICH MIGHT IMPROVE SUCH PERFORMANCE .

PROGRAMME AND INTERMEDIATE OBJECTIVES

YEARS 1 AND 2 :

- IDENTIFICATION OF MODELS ;

- DESIGN OF INTERNAL DATA STRUCTURES AND OPERATIONS FOR A PROGRAM WHICH LEARNS .

YEARS 3 AND 4 :

- INVESTIGATION OF LEARNING TECHNIQUES SUCH AS WRITING CODE OF PROGRAMS WHICH LEARN ;

- PRELIMINARY TESTING .

YEARS 5 AND 6 :

- FULL-SCALE COMMERCIALLY VALUABLE LEARNING EXPERIMENTS WITH THE PROGRAM .

TYPE B RESEARCH THEMES

RELATED TO 3.1.1 :

- APPLICATION OF EXISTING AND DEVELOPING TECHNIQUES IN COGNITIVE PSYCHOLOGY TO KBS ,

- REPRESENTATION AND USE OF GENERAL AND SPECIFIC REAL-WORLD KNOWLEDGE ,

- COMPLEXITY METRICS FOR KBS ,

- MEASURES OF COMPLETENESS , CONSISTENCY ,

- APPLICABILITY OF GENERAL MEASUREMENT THEORY ,

- WHOLE LIFE-CYCLE OF KBS ; HUMAN ACCEPTABILITY AND PRODUCTIVITY ASPECTS FOR BOTH DEVELOPERS AND CLIENTS AS WELL AS TECHNICAL PERFORMANCE ISSUES .

RELATED TO 3.1.2 :

- DIALOGUE SPECIFICATION AND ANALYSIS TOOLS ,

- NATURAL LANGUAGE SYSTEMS ARCHITECTURE ,

- APPLICATION OF EXISTING AND DEVELOPING TECHNIQUES IN COGNITIVE PSYCHOLOGY AND PSYCHOLINGUISTICS TO DIALOGUE AND NATURAL LANGUAGE PROCESSING ,

- SEMANTICS AND THE ROLE OF CONTEXT .

RELATED TO 3.1.3 :

- DIFFERENTIATION , CLASSIFICATION AND EXTENSION OF KNOWLEDGE REPRESENTATIONS FOR VARIOUS DOMAIN CLASSIFICATIONS , PARADIGMS EMPLOYED AND INFERENCE/PROCESSING MODELS USED ,

- EXPERIMENTAL APPROACHES TO KNOWLEDGE REPRESENTATION , THEIR SPECIFIC RANGES OF APPLICATION AND THEIR INTEGRATION ,

- THE RELATIONSHIPS BETWEEN KNOWLEDGE REPRESENTATION AND INFERENCE TECHNIQUES , PROBLEMS OF SELF-REFERENCE AND SYSTEM " SELF-AWARENESS " ,

- REAL-TIME INFERENCE TECHNIQUES ,

- THE ROLE OF META-KNOWLEDGE IN INFERENCE PROCESSES ,

- TECHNIQUES FOR PLANNING AND PROGRAM SYNTHESIS ,

- STRUCTURING OF REFERENCE TECHNIQUES ,

- AUTOMATIC ADAPTATION AND ENHANCEMENT ,

- INFERENCE UNDER CONDITIONS OF CONFLICTING AND UNCERTAIN KNOWLEDGE ,

- THE IMPACT OF KNOWLEDGE REPRESENTATION ON INFERENCE TECHNIQUES .

RELATED TO 3.1.4.1 :

- HIGHER LEVEL IMPLEMENTATION LANGUAGES FOR KNOWLEDGE REPRESENTATION AND INFERENCE ,

- ADAPTATION OF FORMALISMS FOR REAL-WORLD KNOWLEDGE TO PRACTICAL IMPLEMENTATION LANGUAGES FOR KBS ,

- LANGUAGES AND ENVIRONMENTS FOR CONCURRENT PROCESSING AND FOR THE INTEGRATION OF FORMAL , STRUCTURED " SOFTWARE ENGINEERING " WITH KBS REQUIREMENTS .

RELATED TO 3.1.4.5 :

- CONCEPTS , ALGORITHMS AND TECHNIQUES TO SUPPORT THE DEVELOPMENT OF PROGRAMS AND MACHINES WHICH LEARN ,

- IDENTIFICATION OF APPROPRIATE MODELS OF LEARNING ,

- INVESTIGATION OF LEARNING TECHNIQUES APPROPRIATE FOR AIP SYSTEMS ,

- CAPABILITIES OF AN ARTIFICIAL SYSTEM TO MEASURE ITS OWN PERFORMANCE AND TO IMPROVE SUCH PERFORMANCE .

R AND D AREA 3.2 :

EXTERNAL INTERFACES

DESCRIPTION

IN THE PAST , FACTS ABOUT THE REAL WORLD HAVE LARGELY BEEN REPRESENTED TO A COMPUTER SYSTEM AT SECOND-HAND BY PEOPLE . THE DATA FED IN HAD TO POSSESS A FORMAL STRUCTURE THAT NEEDED TO BE CAREFULLY SPECIFIED TO THE SYSTEM . THIS SITUATION WILL CHANGE .

TWO TYPES OF EXTERNAL INTERFACE WILL OCCUR . THOSE THAT EXTRACT THEIR INFORMATION BY DIRECT COMMUNICATION WITH HUMANS , AND THOSE THAT ACQUIRE KNOWLEDGE BY MEANS OF THEIR SENSORS . IN EITHER CASE THE SYSTEM WILL NEED TO PROCESS THE SIGNALS AND INTERPRET THEIR SIGNIFICANCE . OTHER IMPORTANT ASPECTS CONCERN THE PRESENTATION OF THE RESULTS BY THE SYSTEM TO A HUMAN , AND THE USER-FRIENDLINESS OF THE SYSTEM FROM A HUMAN USERS POINT OF VIEW .

R AND D TOPICS

WORK IN THIS AREA FALLS UNDER TWO MAIN HEADINGS :

3.2.1 . : BASIC SIGNAL ANALYSIS AND RECOGNITION

- ADVANCED ALGORITHMS AND ARCHITECTURES FOR SIGNAL PROCESSING ,

- STUDY OF PATTERN ANALYSIS TECHNIQUES ,

- OPTICAL SIGNAL PROCESSING ,

- MULTI-SENSOR SIGNAL PROCESSING .

3.2.2 . : RECOGNITION OF SPECIFIC SIGNALS

- SIGNAL UNDERSTANDING ,

- HANDWRITING ,

- SPEECH ,

- TEXT-TO-SPEECH SYSTEMS WITH NATURAL QUALITY ,

- OBJECT AND MOVEMENT ANALYSIS ,

- HUMAN PERCEPTION ,

- PICTURE SYNTHESIS .

TYPE A PROJECTS

3.2.1 . BASIC SIGNAL ANALYSIS AND RECOGNITION

THE FIRST LAYERS OF A SIGNAL UNDERSTANDING SYSTEM WILL CONSIST OF PREPROCESSING AND FEATURE EXTRACTION ( E . G . FILTERING , IDENTIFICATION , ETC . ) . IN MANY CASES THIS WILL HAVE TO BE CARRIED OUT IN REAL-TIME . IN ORDER TO ACHIEVE THIS , NEW ALGORITHMIC STRUCTURES AND HARDWARE WILL NEED TO BE DEVELOPED . SUCH APPLICATIONS AS IMAGE PROCESSING WILL REQUIRE LARGE BANDWIDTHS AND HIGHLY PARALLEL ARCHITECTURES , AND HENCE VLSI PLAYS AN IMPORTANT PART IN THE PROGRAMME .

PATTERN ANALYSIS IS THE BASIS OF SIGNAL RECOGNITION FOLLOWING FEATURE EXTRACTION AND THEREFORE IS A VITAL PART OF A SIGNAL UNDERSTANDING SYSTEM . THERE IS A NEED TO IMPROVE AND EXTEND THE TECHNIQUES FOR PATTERN ANALYSIS . WORK IS PROPOSED IN THREE TOPICS :

- CLUSTERING AND TEMPLATE FORMATION ,

- DYNAMIC PROGRAMMING ,

- RELAXATION - AN INTERACTIVE PROCEDURE TO ASSIST IN THE PROVISION OF OPTIMAL DECISIONS WHEN ONE HAS A SET OF COOPERATIVE PHENOMENA .

OPTICAL PROCESSING OFFERS A POTENTIALLY ENORMOUS BANDWIDTH WITH PROCESSING AT THE SPEED OF LIGHT . IT IS PARTICULARLY SUITABLE FOR GLOBAL OPERATIONS INVOLVED IN PATTERN RECOGNITION E . G . FILTERING , CORRELATION , CONVOLUTION . WORK IS REQUIRED TO DEVELOP A HIGH PERFORMANCE ELECTRO-OPTIC INTERFACE AND TO DEVELOP REAL-TIME 2-D PROCESSING TECHNIQUES . A NONLINEAR PROCESS KNOWN AS DEGENERATE FOUR WAVE MIXING ( DFWM ) ALSO HAS CONSIDERABLE POTENTIAL . ALSO RESEARCH INTO OPTICAL PROCESSING ARCHITECTURES AND SYSTEM DESIGN IS NEEDED .

THE FINAL ASPECT OF THE PROJECT IS CONCERNED WITH THE PROCESSING AND CORRELATION OF MULTIPLE STREAMS OF SIGNAL INFORMATION , A PROBLEM OF INCREASING IMPORTANCE IN MANY AREAS WHERE MULTIPLE SENSORS ARE USED TO ACQUIRE INFORMATION AT DIFFERENT POINTS OR ABOUT DIFFERENT ASPECTS . EXAMPLES OCCUR IN THE AREAS OF PROCESS CONTROL , CIM , EARTH RESOURCES AND BIOMEDICAL SIGNALS .

PROGRAMME AND INTERMEDIATE OBJECTIVES

( A ) ADVANCED ALGORITHMS

YEAR 0 :

- CHOOSE TARGET APPLICATION , E . G . EARTH RESOURCE ANALYSIS , MEDICAL IMAGING , SPEECH RECOGNITION ;

- DEVELOP ALGORITHMS AND THROUGH ANALYSIS AND SIMULATION PROVIDE A SYSTEM PACKAGE FOR ARCHITECTURAL ANALYSIS .

YEAR 1 :

- SELECT AMONG DIFFERENT PARALLEL ARCHITECTURES USING SIMULATION AND EXTENSIVE MODELLING .

- INITIATE DESIGN OF VLSI PACKAGE IMPLEMENTATION OF CHOSEN ALORITHMS .

YEAR 2 :

- CONTINUE ANALYSIS AND SIMULATION ;

- CONTINUE DESIGN OF VLSI PACKAGE OF CHIPS .

YEAR 3 :

- DEVELOP A PROGRAMMING ENVIRONMENT ADAPTED TO THE PROPOSED ARCHITECTURES ;

- COMPLETE VLSI DESIGN AND FABRICATE CHIPS ;

- CONSTRUCT PARTIAL BREADBOARD OF PROPOSED ARCHITECTURE WITH LIMITED NUMBER OF PROCESSORS .

YEAR 4 :

- COMPLETE PROGRAMMING ENVIRONMENT ;

- CONSTRUCT DEMONSTRATOR MODEL AND EVALUATE .

( B ) PATTERN ANALYSIS TECHNIQUES

YEAR 1 :

- SURVEY STATE-OF-THE-ART ;

- PLAN PROGRAMME OF RESEARCH AND SELECT TARGET APPLICATIONS ;

- INITIATE RESEARCH ON CLUSTERING AND TEMPLATE FORMATION ;

- INITIATE RESEARCH ON DYNAMIC PROGRAMMING .

YEAR 2 :

- CONSIDER METHOD OF IMPLEMENTATION SUCH AS SYSTOLIC ARRAYS ;

- INITIATE WORK ON RELAXATION IN RELATION TO EDGE DETECTION .

YEAR 3 :

- CONTINUE RESEARCH ;

- DEMONSTRATE CLUSTERING AND TEMPLATE TECHNIQUES ;

- CONSTRUCT SYSTOLIC ARRAY MODEL FOR DYNAMIC PROGRAMMING DEMONSTRATORS .

YEAR 4 :

- EVALUATE PERFORMANCE OF DEMONSTRATORS ;

- DEMONSTRATE RELAXATION TECHNIQUES .

( C ) OPTICAL SIGNAL PROCESSING

YEAR 1 :

- IDENTIFY APPLICATIONS . INITIATE , DESIGN STUDIES ;

- SPECIFY PERFORMANCE REQUIREMENTS FOR SPATIAL LIGHT MODULATOR ( SLM ) AND DFWM .

YEAR 2 :

- EVALUATE AND IMPLEMENT INTO SYSTEM :

- SLM ,

- DFWM MATERIALS AND ARCHITECTURES .

YEARS 3 :

- EXTEND SYSTEMS E . G . INCLUDE MELLIN-FOURIER TRANSFORMS . IMPLEMENT IMPROVED SLM AND DFWM MATERIALS/ARCHITECTURES .

YEAR 4 :

- CONSTRUCT COMPACT PRACTICAL PROCESSORS FOR IDENTIFIED APPLICATIONS . HIGHLIGHT AREAS FOR FUTURE RESEARCH AND DEVELOPMENT .

( D ) MULTI-SENSOR SIGNAL PROCESSING

YEAR 2 :

- STUDY REQUIREMENTS AND PLAN PROGRAMME .

YEAR 3 :

- SIMULATE SYSTEM AND CONSTRUCTION OF MODEL .

YEAR 4 :

- CONSTRUCT DEMONSTRATOR .

YEAR 5 :

- COMPLETE AND EVALUATE DEMONSTRATOR .

TYPE B RESEARCH THEMES

RELATED TO ( B ) ABOVE :

- SECTOR QUANTIZATION AND MARKER MODELLING ,

- TECHNIQUES FOR DESCRIPTION OF TIME VARYING PATTERNS ,

- LEARNING TECHNIQUES IN PATTERN ANALYSIS .

RELATED TO ( D ) ABOVE :

- SENSOR TASKING ; MODIFICATION OF SENSOR OPERATION AFTER PROCESSING .

3.2.2 . RECOGNITION OF SPECIFIC SIGNALS

DESCRIPTION

TO UNDERSTAND THE MEANING AND CONTENT OF REAL-WORLD SIGNALS SUCH AS SPEECH AND IMAGES IS A STAGE BEYOND MERE RECOGNITION OF INDIVIDUAL ITEMS SUCH AS WORDS . THE PROJECT WILL INVESTIGATE METHODS OF HANDLING THE INFORMATION CONTENT OF THE SIGNALS , AND HENCE USE OF KNOWLEDGE AND INFERENCE TECHNIQUES IS ESSENTIAL . USE OF NON-NUMBER-CRUNCHING PROCESSOR STRUCTURES WILL BE NECESSARY . IN ADDITION , ATTENTION MUST BE DIRECTED TOWARDS DESCRIPTION LANGUAGES FOR EXTERNAL DOMAINS ( E . G . 2-D AND 3-D SCENES ) .

DIRECT INPUT OF MANUALLY WRITTEN INFORMATION WOULD ENHANCE MAN-MACHINE COMMUNICATION . TO ACHIEVE THIS , APPROPRIATE SCANNING TECHNIQUES MUST BE SELECTED AND DEVELOPED , THE OUTPUT FROM WHICH IS SUBJECTED TO SUITABLE FEATURE EXTRACTION AND PREPROCESSING TECHNIQUES . DIFFERENT METHODS OF CLASSIFICATION MUST BE INVESTIGATED . THE PROJECT WILL LOOK INITIALLY AT HAND-WRITTEN CHARACTERS , AND SUBSEQUENTLY AT CURSIVE SCRIPT .

FLEXIBLE INPUT OF DATA INTO SYSTEMS BY SPEECH IS NEEDED AS THIS IS MAN'S MOST NATURAL SIGNAL . BECAUSE SPEECH IS NORMALLY PRODUCED IN CONNECTED FORM IT IS NOT ENOUGH TO RECOGNIZE ONLY ISOLATED SPOKEN COMMANDS . SPEAKER INDEPENDENCE AND INSENSITIVITY TO NOISE ARE OTHER IMPORTANT REQUIREMENTS . THE FIRST PHASE WILL CONSIDER CONNECTED SPEECH FROM LIMITED BUT LARGE VOCABULARIES . ALGORITHMS ARE REQUIRED FOR FEATURE EXTRACTION AND PREPROCESSING . ACCOUNT MUST BE TAKEN OF THE CONDITIONS OF SPEECH INPUT SUCH AS THE NOISE ENVIRONMENT , NUMBER OF USERS AND VOCABULARY . THE CLASSIFICATION TECHNIQUE SHOULD BE ECONOMICAL IN HARDWARE . PHASE 2 ( AFTER YEAR 5 ) WILL EXTEND THE WORK TO CONTINUOUS SPEECH RECOGNITION FROM GENERAL VOCABULARIES . THIS WILL NECESSITATE RESEARCH INTO RECOGNITION TECHNIQUES BASED UPON SPEECH AND LINGUISTIC KNOWLEDGE .

THE MOST CONVENIENT MEDIUM FOR THE OUTPUT OF INFORMATION FROM COMPUTERS FOR HUMANS IS BY SPEECH SYMBOLS . THE QUALITY OF SUCH A SPEECH SIGNAL MUST BE AS NATURAL AND INTELLIGIBLE AS THE NORMAL SPEECH SIGNAL . THE DIFFERENT RESEARCH TOPICS MUST COVER THE LINGUISTIC AND PHONETIC ASPECT AS WELL AS THE PRODUCTION OF NATURALLY SOUNDING SPEECH SIGNALS BY ELECTRONIC MEANS . THE RESULTING SYSTEM WILL DEPEND ON THE LANGUAGE CHOSEN , HENCE MORE THAN ONE PROJECT IS NEEDED TO COVER THE DIFFERENT LANGUAGES .

THE OBJECT AND MOVEMENT ANALYSIS PART OF THE PROJECT WILL START BY LOOKING AT STEREO AND OPTIC FLOW TECHNIQUES FOR DEVELOPING DEPTH MAPS FROM 2-D IMAGES . SUBSEQUENT STUDIES WILL CONCERN THE INTERPRETATION AND SYMBOLIC DESCRIPTION OF 3-D OBJECTS AND THEIR DEFORMATION AND MOVEMENTS . THE PROJECT BRIDGES THE GAP BETWEEN IMAGE ANALYSIS/FEATURE EXTRACTION AND SCENE UNDERSTANDING . AS SUCH IT CONTRIBUTES TO THE GOAL OF PROVIDING A COMPUTER OR MACHINE WITH EYES .

WORK ON PERCEPTION TO OBTAIN A BETTER UNDERSTANDING OF NATURAL SYSTEMS SUCH AS THE EYE/BRAIN AND EAR/BRAIN WILL BE OF GREAT ASSISTANCE IN THE DESIGN OF SIGNAL PROCESSING SYSTEMS . SPEECH ANALYSIS , BASED ON A MODEL OF THE HUMAN SPEECH PRODUCTION PROCESS , IS AN EXAMPLE WHERE LESSONS FROM NATURAL SYSTEMS HAVE BEEN SUCCESSFULLY APPLIED . THE OBJECTIVE OF PHASE 1 IS TO STUDY LOW-LEVEL IMAGE AND SPEECH PROCESSING . EDGE EXTRACTION IS KNOWN TO BE A FUNDAMENTAL PART OF THE EARLY VISUAL PROCESSING IN HUMANS , AND CELLS HAVE BEEN IDENTIFIED IN THE VISUAL CORTEX THAT ARE SENSITIVE TO EDGE SEGMENTS OF SPECIFIC DURATION . THE STUDY SHOULD LEAD TO NEW ALGORITHMS FOR IMAGE AND SPEECH ANALYSIS . THE OBJECTIVE OF PHASE 2 IS TO STUDY HIGHER LEVEL PROCESSES INVOLVED IN HUMAN SIGNAL UNDERSTANDING ; THIS PHASE BEGINS AFTER YEAR 5 .

PICTURE SYNTHESIS IS ESPECIALLY IMPORTANT IN A SCENARIO WHERE COMPUTERS SIMULATE PARTS OF THE REAL WORLD E . G . IN CAE OR TRAINING APPLICATIONS . VERY FAST GRAPHICS PROCESSING TECHNIQUES ARE AS IMPORTANT AS FLEXIBLE STORAGE TECHNIQUES FOR IMAGES . ALSO IMPORTANT IN THIS PROJECT IS THE DEVELOPMENT OF SYSTEMS WITH A RESOLUTION OF MORE THAN 1 000 BY 1 000 PIXELS .

PROGRAMME AND INTERMEDIATE OBJECTIVES

( A ) SIGNAL UNDERSTANDING

YEAR 1 :

- CHOOSE A SIMPLE APPLICATION WITH VERY RESTRICTED SIGNAL VARIATIONS .

YEAR 2 :

- DEFINE GRAMMAR OF THE EXAMPLE , BUILD THE BASIC EXPERIMENTAL SYSTEM .

YEAR 3 :

- EVALUATE THE CHOSEN EXAMPLE .

YEAR 4 :

- INCLUDE A SECOND AND MORE COMPLEX SIGNAL DOMAIN TO TEST THE ADOPTION OF SIGNAL UNDERSTANDING TO NEW PROBLEMS .

YEAR 5 :

- EVALUATE THE TOTAL SYSTEM .

( B ) HANDWRITING

YEAR 1 :

- SURVEY SCANNING TECHNIQUES TO INITIATE DEVELOPMENT ;

- SELECT ALGORITHMS FOR FEATURE EXTRACTION FOR CHARACTERS .

YEAR 2 :

- SPECIFY A DEMONSTRATOR FOR CHARACTER RECOGNITION ;

- CHOOSE METHOD OF CLASSIFICATION .

YEAR 3 :

- COMPLETE DESIGN OF DEMONSTRATOR AND COMMENCE CONSTRUCTION ;

- INITIATE CONTEXTUAL ANALYSIS SURVEY ;

- INITIATE STUDY OF TECHNIQUES FOR CURSIVE SCRIPT RECOGNITION .

YEAR 4 :

- COMPLETE CHARACTER RECOGNITION DEMONSTRATOR ;

- DEFINE RESEARCH PROGRAMME FOR CURSIVE SCRIPT RECOGNITION .

YEAR 5 :

- EVALUATE CHARACTER RECOGNITION DEMONSTRATOR ;

- START DESIGN OF CURSIVE SCRIPT RECOGNIZER .

( C ) SPEECH

YEARS 2 AND 3 :

- REVIEW ALGORITHMS FOR FEATURE EXTRACTION AND PREPROCESSING . RELATE TO NOISE DYNAMICS . IDENTIFY THE INFORMATION CONTENT OF THE FEATURE TO OPTIMIZE CHOICE . STUDY CLASSIFICATION TECHNIQUES .

YEAR 4 :

- SIMULATE RECOGNITION PROCESS ON FAST ARRAY PROCESSOR .

YEAR 5 :

- DESIGN AND BUILD DEMONSTRATOR .

( D ) TEXT-TO-SPEECH

YEAR 1 :

- BASIC ANALYSIS OF SPOKEN SPEECH , DESIGN OF FLEXIBLY CONTROLLABLE SPEECH SYNTHESIZER , IDENTIFY THE PARAMETERS NECESSARY FOR NATURAL SOUND .

YEAR 2 :

- EVALUATE LINGUISTIC RULES FOR SPEECH PRODUCTION INCLUDING PROSODIC PARAMETERS , PHONETICS OF SPEECH PRODUCTION .

YEAR 3 :

- FURTHER EVALUATION OF PHONETIC AND LINGUISTIC RULES , SIMULATIONS OF SYNTHETIC SPEECH INCLUDING ARTICULATORY RULES .

YEAR 4 :

- REALIZATION OF FLEXIBLE SPEECH SYNTHESIZER ;

- FURTHER EVALUATION OF PROSODIC PARAMETERS ;

- DESIGN OF A LETTER CRUNCHER MACHINE FOR THE LINGUISTIC AND PHONETIC PROCESSES .

YEAR 5 :

- REALIZING THE TOTAL TEXT-TO-SPEECH SYSTEM .

( E ) MOVEMENT ANALYSIS

YEARS 1 AND 2 :

- RESEARCH INTO BINOCULAR STEREO ;

- RESEARCH INTO OPTIC FLOW .

YEAR 3 :

- GENERATION OF 2 1/2-D SKETCH ;

- GENERATION OF DEPTH MAPS FROM OPTIC FLOW ;

- SYMBOLIC DESCRIPTIONS OF OBJECTS .

YEAR 4 :

- STUDY DESCRIPTIONS OF MOVEMENT , ANALYSIS TECHNIQUES ;

- STUDY OF TRANSFORMATION OF 3-D OBJECTS .

YEAR 5 :

- INVESTIGATE AMORPHIC CHANGES OF OBJECTS .

( F ) HUMAN PERCEPTION

YEAR 2 :

- STUDY OF SPEECH PERCEPTION IN THE EAR , INCLUDING SIMULATION OF DIFFERENT MODELS ;

- EVALUATION OF HUMAN EDGE DETECTION MODEL .

YEAR 3 :

- EXTEND WORK TO MORE GENERAL FEATURE EXTRACTION .

YEAR 4 :

- EXTEND MODEL TO DEPTH PERCEPTION .

YEAR 5 :

- STUDY ROLE OF RELAXATION IN HUMAN VISUAL PROCESSING ;

- STUDY PERCEPTION OF TEXTURE .

( G ) PICTURE SYNTHESIS

YEAR 1 :

- DESIGN NEW POSSIBILITIES FOR SEQUENTIAL MEMORIES , PICTURE TRANSACTIONS IN THE MEMORY .

YEAR 2 :

- DESIGN OF A HIGH-SPEED PROCESSOR FOR FAST ACCESS TO MEMORY .

YEAR 3 :

- INTERPRETATION OF MEMORY AND PROCESSOR .

YEAR 4 :

- REVISED VERSION OF PROCESSOR , ESPECIALLY CONCENTRATING ON THE COOPERATION WITH FAST MEMORY STRUCTURES .

YEAR 5 :

- INTERPRETATION OF THE PICTURE SYSTEMS INTO THE DISPLAY SYSTEM .

TYPE B RESEARCH THEMES

RELATED TO ( A ) ABOVE :

- COMPUTATIONAL LINGUISTICS ,

- HIGH-LEVEL VISUAL PERCEPTION ,

- SCENE DESCRIPTION LANGUAGES ,

- APPLICATION OF RELEVANT TECHNIQUES FROM COGNITIVE PSYCHOLOGY TO SIGNAL UNDERSTANDING .

RELATED TO ( B ) ABOVE :

- AUTOMATIC DICTIONARY GENERATION ,

- MACHINE LEARNING PROCESSES .

RELATED TO ( C ) ABOVE :

- PHONETICS ,

- SYNTAX AND GRAMMAR ,

- PSYCHO-ACOUSTICS .

RELATED TO ( D ) ABOVE :

- LINGUISTICS ,

- PHONETICS ,

- LINGUISTIC RULE COMPILERS ,

- TEXT COMPREHENSION ,

- STUDY OF MORE REALISTIC SPEECH MODELS .

RELATED TO ( E ) ABOVE :

- GEOMETRIC REASONING ,

- GENERATION OF SPARSE RELIABLE IMAGE REPRESENTATIONS SUITABLE FOR CORRELATION IN STEREOPSIS AND OPTIC FLOW .

RELATED TO ( F ) ABOVE :

- RELEVANT RESULTS AND TECHNIQUES FROM COGNITIVE PSYCHOLOGY ,

- PHYSIOLOGY OF EYES AND EARS ,

- MODELS OF HUMAN INFORMATION PROCESSING .

RELATED TO ( G ) ABOVE :

- PICTURE DESCRIPTION LANGUAGES .

R AND D AREA 3.3 :

INFORMATION AND KNOWLEDGE STORAGE

DESCRIPTION

KNOWLEDGE-BASED SYSTEMS ARE DESIGNED TO DERIVE INFORMATION FROM THEIR KNOWLEDGE BASE AND TO INDICATE THEIR LINE OF REASONING IN ARRIVING AT A CONCLUSION . THE EFFECTIVENESS OF ANY KNOWLEDGE-BASED SYSTEM IS DETERMINED BY THE QUALITY OF ITS KNOWLEDGE BASE ( ITS COMPLETENESS , VALIDITY AND ACCESSIBILITY ) AND OF ITS DEDUCTION MECHANISM .

THIS R AND D AREA INVESTIGATES THE NEW FORMS OF ORGANIZATION WHICH WILL BE NECESSARY TO HOLD AND PROCESS REPRESENTATION OF DATA , KNOWLEDGE , AND META-KNOWLEDGE .

THE MAIN OBJECTIVES OF THIS AREA ARE :

- THE DEFINITION OF INFORMATION AND KNOWLEDGE STORAGE ORGANIZATIONS INCLUDING MACHINE ASPECTS .

- THE IDENTIFICATION OF FUNCTIONAL PARTITIONS , STRUCTURES AND MECHANISMS INCLUDING DISTRIBUTED AND VERY LARGE KNOWLEDGE BASES ,

- THE DEVELOPMENT OF NEW STORAGE STRUCTURES ,

- THE INTRODUCTION OF NEW PHYSICAL PRINCIPLES TO SUPPORT NEW STORAGE DEVICES ( E . G . OPTICAL , BIOLOGICAL ) ,

- THE REALIZATION OF ADEQUATE DIALOGUE TECHNIQUES ESPECIALLY FOR THE NON-EXPERT USER ,

- AUTOMATIC OR SEMI-AUTOMATIC CONSTRUCTION OF VERY LARGE KNOWLEDGE SYSTEMS .

WORK IN THIS AREA WILL STRONGLY DEPEND ON ADVANCES IN KNOWLEDGE ENGINEERING AND MEMORY TECHNOLOGY .

R AND D TOPICS

THE TOPICS TO BE COVERED BY THIS AREA ARE AS FOLLOWS :

3.3.1 . ( 2 ) INTERFACE BETWEEN STORAGE AND ENVIRONMENT

3.3.2 . ( 2 ) DATA AND KNOWLEDGE BASES :

- KNOWLEDGE BASE STUDIES FOR AIP ,

- DISTRIBUTED DATA BASES AND KNOWLEDGE BASES ,

- ADVANCED KBMS .

3.3.3 . STORAGE STRUCTURES AND ARCHITECTURES :

- DATABASE ORIENTED ARCHITECTURES ,

- PERFORMANCE MODELS FOR STORAGE STRUCTURES ,

- STORAGE ARCHITECTURE

3.3.4 . NEW GENERATION KNOWLEDGE MACHINE

3.3.5 . MEDIUM-TERM RESEARCH ON STORAGE MEDIA

3.3.6 . LONG-TERM RESEARCH ON OPTICAL DISCS

3.3.7 . LONG-TERM RESEARCH ON BIOLOGICAL STORAGE .

THE WORK IN THE FIRST FIVE YEARS WILL CONCENTRATE ON GAINING EXPERIENCE IN KNOWLEDGE STORAGE AND CORRESPONDING ACCESS/DIALOGUE TECHNIQUES .

TOPIC 3.3.1 INCORPORATES CONTINUATION OF WORK ON INTERACTIVE QUERY SYSTEMS STARTED UNDER THE PILOT PROJECTS IN YEAR 0 ( 1983/84 ) .

TYPE A PROJECT

3.3.2 . DATA AND KNOWLEDGE BASES

DESCRIPTION

STUDIES ARE PROPOSED TO ESTABLISH THE INTERFACES , FORMALISMS , LANGUAGES , HARDWARE AND SOFTWARE COMPONENTS WHICH ARE REQUIRED FOR THE CONSTRUCTION , DISTRIBUTION , FUNCTIONAL PARTITIONING AND HIERARCHIC STRUCTURING OF DB AND KB , INCLUDING FORMULATION OF INFERENCE AND DATA QUERY ACCESS AT LEVELS RANGING FROM HUMAN VISIBLE TO THOSE INTERNAL TO THE NEW GENERATION OF SYSTEMS . THE FOCUS OF THIS WORK IS TO INVESTIGATE COMMONALITIES AND SPECIFICS IN THE ARCHITECTURAL MODELS OF DATABASE AND KNOWLEDGE MANAGEMENT SYSTEMS .

THERE WILL BE A PARALLEL STUDY AIMED AT ESTABLISHING ADDITIONAL " ACCESS " FORMULATION LANGUAGES , FORMALISMS , INTERFACES AND TECHNOLOGY WHICH ARE REQUIRED FOR DISTRIBUTED KNOWLEDGE-BASED INFORMATION SYSTEMS . THE STUDY INCLUDES INVESTIGATIONS ON VERY LARGE KNOWLEDGE BASES AND ITS IMPACT ON ( E . G . PARALLEL ) STORAGE ARCHITECTURES . DISTRIBUTED STRUCTURES WILL BE MODELLED AND STANDARDS DEVELOPED .

Источник: [https://torrent-igruha.org/3551-portal.html]
Idea Cruncher Version: 1.3.1 serial key or number

PdivdivdivИсточник: [https:torrent-igruha. org3551-portal. html]div divh2ZD Soft Screen Recorder 11. 0 Crack Free Key Downloadh2divh3ZD Soft Screen Recorder 11. 0 Full Version Crack Free Downloadh3pimg src"https:i2.

.

What’s New in the Idea Cruncher Version: 1.3.1 serial key or number?

Screen Shot

System Requirements for Idea Cruncher Version: 1.3.1 serial key or number

Add a Comment

Your email address will not be published. Required fields are marked *