1st > March > 1999 Archive

The Register breaking news

AMD thrashes Intel at low end

A survey released by PC Data late Friday has shown the extent to which Intel underestimated the nature of the PC market. But Intel senior executives know they miscalculated. At Intel's Developer Forum last week, both Paul Otellini, senior VP of IA architecture, and Pat Gelsinger, GM of Intel's desktop products division, said they had miscalculated at the low end. According to PC Data, AMD's share from November to January rose to 33 per cent in the $1,000 to $1,500 market. It had previously held around a third of the sub-$1,000 sector. Now AMD is positioning its K6-III processor against the recently released Pentium III, with benchmarks, such as those posted on Tom's Hardware Page last week, showing there is merit to its argument. Compaq will use AMD's K6-III, as will Gateway. Intel says the sweet spot for performance machines is currently just under $2,000, but that is likely to change. That will mean Intel is likely to step up future product announcements. Its 550MHz Pentium III part will be released in May, and price cuts on the new Pentium III processors is likely to be accelerated, sources said. (Story: Intel ships PIII, slashes PII prices) Later this year, AMD will introduce its K7 slot processor, which is also aimed at capturing Intel market share. Intel still owns the lion's share of the x.86 market, with an estimated 75 per cent market share. But it is also under pressure from the Alpha platform. Last week, the Alpha consortium cut its prices to compete with x.86 parts. ®
Mike Magee, 01 Mar 1999
The Register breaking news

Intel employee saves UK journalist's life

Graham Palmer, one of a limited number of quick thinking Intel PR representatives, saved the life of a UK journalist last week at the Palm Springs Intel Development Forum. Gordon Thorn, Labs Manager at VNU Business Publications in the UK, suffered an acute allergic reaction after being bitten by a desert nasty last week. Palmer prevented blocking of Gordon's airways until a 911 paramedic team was able to provide full resuscitation. Gordon has now made a full recovery and told The Register he was very grateful. After recuperating for two additional days in Palm Springs, Gordon was able to return to the UK yesterday. ®
Mike Magee, 01 Mar 1999
The Register breaking news

Compaq promotes Alpha as high-end solution

A slide presentation Compaq has shown its business partners and now in the hands of The Register, has demonstrated its serious intent to use the Alpha microprocessor at the high end. The presentation, delivered by Compaq 64-bit Win NT program and product manager Mont Francisco, is titled Win64 and Alpha, the future is now. According to Francisco, Alpha has demonstrated its performance leadership in the 64-bit market since September 1995. The presentation says that TPC-C results show that an eight-way Alpha system holds the industry record for absolute performance. Further, the document gives details of the Win64 OS. Applications must comply with Win64's API, and two platforms are supported, the IA-64 and the Alpha chip. Compaq told its ISVs: "You can either do nothing and wait for IA-64 in the shape of Merced, mid-2000 or McKinley, in late 2001 or prepare for the future and convert your applications to 64-bit today." Compaq, said Francisco, has platform SDKs, and a developer advice package, with a Win64 porting centre at Bellevue and a Solutions Alliance. "Alpha is ready now," Francisco told the ISVs. The logic for moving apps to 64-bit now is that end users will be able to build bigger databases and perform faster transactions. Compaq is pursuing a dual strategy with Merced and the Alpha. ® Related Stories Intel anxious about Alpha threat An insider tells us stuff Merced only slightly undead We talk to Stephen Smith, Intel Merced manager, and hold the package in our hands. Hopefully, we'll have the pictures up on our site today. For more detailed information on Compaq and Alpha, see Shannon knows Compaq
Mike Magee, 01 Mar 1999
The Register breaking news

Intel makes The Register sweat IV

Intel does not allow any cameras inside its fab besides from its own for obvious reasons. We did point out to them that these days you can get really good cameras that will fit in your tie if you're wearing one, but we didn't have one of those... Diffusion area A technician carries a lot box of wafers down an alleyway of diffusion surfaces. He or she is coming from the photographic area (see colour at back). The furnaces grow layers of oxide on the wafers and the oxide is then etched using a pattern photographically imprinted on the surface in a later step. See the robots on each of the technician loading and unloading the furnaces. Etch area Here a technician unloads a boat of wafers from the lot box for processing in acid baths in the tanks behind him or her. The acid tanks etch unwanted portions of the surface of the wafers. The robot arm in the background carries the boats from tank to tank, so the human being can avoid being near the very nasty acids. Intel has new Hitachi machines which are completely closed off from the bay. Thin Film area The human being in bunny suit loads a metal sputtering tool with a boat of wafers. The tools deposit layers of metal that will make connections in the minute circuits. The floor is a vent, with filtered air travelling from the ceiling's HEPA filters. The air system is on the next floor above the clean room. This is to avoid human and other particulates spoiling the wafers Mike Magee area Here we see Mike Magee in the garb of an Intel technician but without his crash helmet. The strange red eyes are due to the Intel photographer not being a very good one. And here's Mike Magee in bunny suit in Fab 11. Note the arrays of $700 suits behind him. Now there's something better inside.
Mike Magee, 01 Mar 1999
The Register breaking news

Intel decks more chips

Chip giant Intel has discontinued manufacture of the Celeron 300a desktop processor, the Pentium II 333MHz part and Pentium mobile with MMX 300, 266 and 266MHz low power chips. It explained the move by saying the company is experiencing a rapid market conversion to higher performance chips in both those segments. Intel is notifying all of its customers, and they will have four months up until the 25th of June this year to place final orders for the mobile Pentium parts. Customers will have twelve months to place orders for the 300a desktop part, and eighteen months for the PII/333 chip. ®
Mike Magee, 01 Mar 1999
The Register breaking news

How MS destabilised Netscape and Marimba

MS on Trial During his evidence at the Microsoft trial, Eric Engstrom had some excuses ready about Microsoft's acquisition of DimensionX in 1997, which he engineered. He had wanted support from various people in Microsoft for the acquisition "so I put some statements in there that, you know, I was not fully briefed on, nor do I necessarily completely understand, but that represented the opinion of other people whose consensus I wanted in order to go through with the acquisition". There were however some more sinister reasons. The draft email advocating the acquisition to Gates and Maritz noted that DimensionX was about to be acquired by Javasoft. One of the statements that Engstrom referred to was that "Netscape is planning a large announce around this strategy [the key authoring tool for Marimba] and partners in the next few weeks. We could achieve a big tactical win here by removing their key tool." It was a clear case of a strategic acquisition that had an anti-competitive motivation. Engstrom defended this by claiming that Carl Jacobs, the CEO of Liquid Motion (acquired with DimensionX), had told him that the company was not working with Netscape. The Liquid Motion runtime was in many ways the equivalent of some multimedia controls that Microsoft was trying to develop. Engstrom had also written that "'derailing' what Netscape wanted would be a "HUGE HUGE thing". Brad Silverberg had asked Engstrom how the acquisition would take away the authoring tool from Marimba, so Engstrom (who confessed during his evidence that he did not understand what he was talking about) emailed Silverberg: "The authoring tool only works on the DimensionX runtime. It wouldn't be hard at all to make it (a) require our browser; (b) require DirectX runtime; (c) some other option to make Marimba's life difficult." This was not a good statement for a company accused of monopolisation to be seen to be making. Microsoft was considering a technical tie, so that users had to have IE in order to author with the tool. This was the same kind of "deep integration" that Microsoft had used with Windows and IE, and Malone's cross-examination exposed the motivation very well. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

The MS campaign to rein-in Intel multimedia development

MS on Trial In the Microsoft trial, Eric Engstrom is being used to challenge the account of Microsoft's Intel relationship, as told by Steve McGeady and Intel's documents. Strangely, McGeady's name was never mentioned: he has evidently become a non-person. An Intel internal email from Gerald Holzhammer on 13 April 1995 summarised a face-to-face meeting with Engstrom, Carl Stork and Marshall Brumer of Microsoft. He concluded that Microsoft wanted to own the drivers in Windows 98; that nobody but Microsoft was qualified to do good driver software; that Microsoft would not collaborate on NSP; and that Microsoft had "completely missed the boat" on developing a compelling state of the art media subsystem for Windows 95", according to a confession by Carl Stork. Engstrom sent a status report email on 26 May 1997 about the "collaborative" multimedia efforts of Microsoft and Intel. It was very revealing: "The goal for our interactions with the Intel Architecture Labs is that all their efforts are neutral or positive for our strategic mm [multimedia] initiatives. I think it is unlikely that we will achieve 100 per cent of this goal". It turned out that Microsoft was obsessed at this particular time with collaboration between Sun and Intel over Java-based competing multimedia standards - the MPEG4 specification. Engstrom's goal: "Intel to stop helping Sun create Java Multimedia APIs, especially ones that run well (ie native applications) on Windows". Engstrom was unhappy that every IAL partner slide referred to Sun. His proposed actions included Microsoft stopping giving any help to Javasoft wherever Microsoft agrees to ship Intel technology in DirectX media. Microsoft was concerned at Intel's efforts "to do multimedia on Intel-based Windows platforms" such as RSX (3D sound), RD (sprite engine), video playback extensions, and procedural textures generally. So far as large model management was concerned, this was competitive with "our" HP Jupiter technology. Engstrom ruefully noted that Intel had "a rather large evangelist force that is willing to use monetary incentives." Whatever next? It could not have helped Microsoft that Chrome (the multimedia browser) was projected to Intel as "the big carrot", in view of what subsequently happened to Chrome. Engstrom said in his testimony that Intel wanted some software that would produce special effects to make its processors look good, but Microsoft just pressed on with pressuring Intel to stop software development. Rather than admit that Sun and others were important Intel partners, Engstrom puts Intel's behaviour down to Microsoft "not paying enough attention to Intel" rather than recognising that Microsoft's lack of expertise and arrogance might be part of the problem. Microsoft needed "to own all the driver software down to the metal" Engstrom agreed, because in 1995 "we were trying to move the games market forward. They had been basically stuck using very similar technologies for ten years because all of the advances in Windows hadn't helped games at all. So, the purpose of the DirectX technology was to make the Windows environment and Windows hardware acceleration accessible to games. At that time, given the speed of the processors and the hardware state, you wanted the game to know exactly what was happening. If there was a bunch of code running in there that the application didn't know about and couldn't test against, you would get bizarre behaviour in the game. It would stop being realtime." Surely there is no relationship between Intel's friendliness towards Gates' arch-enemy Sun and Microsoft's violent reaction against Intel producing software that Sun could use? ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

MS OEM boss shocked by Compaq-Netscape deal – allegedly

MS on Trial Just what could have been in the movie that Joachim Kempin, Microsoft's penultimate witness and longtime OEM sales vp, did not play to the court? The one that was played was only made on 11 February 1999, and Kempin was there throughout the day. Microsoft decided to cut some of the video demonstrations, and therein lies a separate tale. The parts of Kempin II that Microsoft decided to leave out were of course those that were less favourable to Microsoft's case. The clue to there being Kempin I and II was in Kempin's sworn-to-be-true direct testimony, signed on 6 January 1999, which referred to what must have been a previous video, Kempin I. It was clear that Microsoft had had second thoughts about the contents, in view of the blunders with previous videos, and had remade it. Boies elicited this from Kempin, who at first was not averse to producing it. The hapless Steve Holley, who was Kempin's legal minder, was less sure. Judge Jackson said he would return to this, but alas it was overlooked in the rush to finish the business. The point of Kempin II, it was said, was to show just how easily Windows 98 could be customised if OEMs wanted to do this. That may have been the idea, but it did not work that way. Even Kempin II was marred: Microsoft bought two of each PC to use, and one was used the previous day to check that everything was as Microsoft would wish for the demos, so there were no surprises the next day. We reported earlier that Compaq had had a change of heart about including a Netscape icon on the desktop, and when a Presario was booted, there it was. Bizarrely, Kempin said he was "totally surprised". As it's generally held that very little moves in Microsoft's OEM business, at least at top customer level, his surprise was surprising. Later, Boies asked Kempin about the addition of Netscape to Compaq PCs from January 1999: How much did Netscape have to pay? [That's an interesting thought.] How long was the commitment? Kempin only said: "I just know they put it on; that's the only thing I'm saying." Whether he was being tight-lipped because he was annoyed at what had happened, or whether he knew more than he said, is hard to determine. It did seem as though Kempin might have known about Compaq and the Netscape icon all along. So far as terminating Compaq's Windows licence was concerned, this was little short of being an act of blackmail, since it was very clear that Compaq had no option but to cave in. Judge Jackson became very short with Kempin when he noted that Kempin had been very reluctant to estimate numbers when he claimed he had no data, but suddenly was not averse to using anecdotal evidence. A bench conference discussed Kempin's manner of giving evidence. Boies said he was reluctant to allow Kempin's direct testimony to remain on the record unchallenged. There was considerable cross-examination by Boies about OEMs that had been granted special concessions to modify the boot-up sequence, with Microsoft exhibiting a particularly nasty kind of favouritism by only hinting to selected OEMs that any modification was allowed ("We have sent a letter to some PC manufacturers ..."). It was revealed that only Compaq has a licence that allows it to customise the Windows start-up. It was also confirmed that Compaq gets the best prices from Microsoft. "A handful or more OEMs have a source licence" from Microsoft, but Kempin could only think of Compaq, HP and IBM. The source code must be kept in escrow and the sacred tablets only examined when there is a need -- so let's hope all that handling does not wear down the tablets. Even so, the tablet keepers cannot modify the sacred carving: only Microsoft can do this if it so desires. Kempin tied himself in knots trying to defend Microsoft's reluctance to allow OEMs to interfere with the start screen process. It was not surprising that the greatest desire was to keep Netscape off the desktop, by covert means rather than overt ones. Some of the reasons for this paranoia probably include making it difficult to substitute components from rival software developers; and protecting against illegal software development. Kempin used as an excuse for Netscape not being allowed to be added during a boot-up sequence that it was not an ISP. Gateway was allowed to let users have a choice if browser. In addition, the persistence of IE4 for certain tasks (like browsing HELP) makes it almost impossible for users to use an alternative browser without IE4 popping up all the time. Boies taxed Kempin with the finding that response is very slow if Netscape and IE are both on the PC. The ignorance - or maybe feigned ignorance - that Kempin displayed was laughable: Boies: Is IE4 uninstallable from Windows 95? Kempin: I don't know. I have never tried it. I just don't know. It was not therefore a surprise that Kempin said he had never used a browser other than IE. Boies suggested that Microsoft paid money to OEMs to preserve the standard Windows interface. Kempin denied it, but the difference between a sack of fivers and a Microsoft marketing development agreement is pretty slim. Boies -- and the DoJ team --has been doing a very good job of coming up with documents to contradict a statement made by a Microsoft witness. A particular gem was an email from Microsoft vp Brad Silverberg, now dubbed a consultant and who went walk-about when he was passed over for promotion in favour of Jim Allchin. Silverberg said in an email to Gates on 15 December 1995 that "[IE3] is a standalone browser that runs on Windows 95". This is damning for Microsoft's case. In Gates' response, he said "I think it is a little strange to ship IE3 for free ...". In the OEM stakes, Kempin admitted that Compaq was first, Dell second, and Gateway third so far as licensing fees are concerned. IBM was not in the same league, because as Gates noted in a 30 October 1997 email, IBM had software ambitions. Gates continued in his mature, leadership style: "I can deal with this just fine if IBM wasn't such rabid Java backers [sic]." Some OEMs still with per processor licences have a five per cent exclusion, although Compaq apparently has three per cent. An interesting claim: Kempin had emailed Gates at the end of 1997 suggesting "even lower prices". An ironic situation arose when Judge Jackson asked about the licensing terms defined by the consent decree that he had signed after the folk-hero Judge Sporkin was elbowed aside by the appeals court. Kempin did not do a good job for Microsoft: he was a semantic games player with a selective memory that miraculously recovered from time to time when David Boies for the DoJ tempted him with a question into giving some detail when only minutes earlier he could not remember the most general details. Holley's redirect examination was unexceptional, but managed to bring out Kempin's deep ignorance of the marketplace: he thought that there were "maybe 30 to 40 people in the world" creating Linux, and what's worse from his perspective, they "hardly do any marketing for it". Prompted by the fear of overrunning The Register's timetable, Judge Jackson steamed out of court on Friday afternoon exactly on schedule. His court did sit longer hours towards the end, and five days last week. The judge did not specify the length of the recess, but current estimates are six weeks while he and David Boies are busy with other cases, and other trial lawyers work on their refutations and closing arguments. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

K6-2 PCs outsold Pentium II, Celeron machines in January

US retailers sold more desktops PCs based on AMD's K6-2 processor than Intel-based machines in January, according to the latest data from market researcher PC Data. Overall, AMD-based PCs accounted for 43.9 per cent of the market. Intel was a gnat's whisker behind it with 40.3 per cent. NatSemi's Cyrix subsidiary accounted for 15.8 per cent. In the sub-$1000 category, AMD took 50 per cent of sales, Intel 25.4 per cent and Cyrix 24 per cent. Narrowing the price focus even further, to the sub-$800 market, and Intel's share fell to just 14.6 per cent -- AMD accounted for 45.1 per cent, Cyrix for 40.3 per cent. PC Data's release highlighting AMD's success didn't include statistics for the over-$1000 range, suggesting the Intel came out on top in that category. The researcher's figures also only cover the US retail channel, and it's likely that AMD's showing won't be so high when sales through other channels are taken into account. Equally, the figures go some way to confirm the emerging picture of a market dominated by Intel at the high-end and mid-range, but the Great Satan showing a weak hand further down. Even so, PC Data's figures are an impressive statistic that AMD marcoms people will be justified in shouting about. ®
Team Register, 01 Mar 1999
The Register breaking news

K7 chipset reported delayed

Jonathan Hou, over at his new site at Fullon3D has posted an interview with Shuttle which suggests AMD is re-thinking its slot strategy for the K7. According to the unnamed executive at Shuttle, AMD's K7 chipset is still in its Alpha phase. Although other chipset vendors are backing AMD's K7, the Shuttle executive said that a "good guess" for the product launch was likely to be in Q4 of this year. If true, that represents a delay of some months for AMD. Intel told The Register last week that it will put the Pentium III into socket form. (Story: Intel says new PIII socket on way for iMac look-alikes) According to the Shuttle exec, it is unlikely that AMD, Cyrix, IDT or Rise will release chips using a 133MHz frontside bus. ®
Mike Magee, 01 Mar 1999
The Register breaking news

Tetra accepts Sage offer

Sage is pumping up its mid-range capability with the takeover of UK rival Tetra. The Tetra board is recommending the all-share offer which values the company at £78.1 million. Tetra effectively came into play last month when it received an unsolicited bid approach from Lynx Group, a diversified computer services conglomerate. The 305p per share offer from Sage is a 63 per cent premium on Tetra’s share price on the day it announced it was in bid talks with Lynx. With the backing of the Tetra board, and the muscle of Sage behind the bid, it is hard to envisage a counter-bid from Lynx -- or other parties. Sage’s offer coincides with the publication of a good set of interims from Tetra. For the six months to 30 November, 1998. Sales jumped 26.2 per cent ot £16.2 million (1998: £12.8 million), and PBT edging up to £1.53 million (1998: £1.24 million). Software licence sales through the channel grew 21.2 per cent to £7.3 million (1998: £6.1 million); and consultancy revenue leapt 62 per cent to £4.5 million). Sage dominates the European low-end accountancy software market, but it is weaker in mid-range -- and non-existent -- in high-end markets. Sage has the money and the distribution channels to take forward CS/3, Tetra’s new Microsoft- platform client-server offering, much further than Tetra could do by itself. Last week, the company extended its accountancy software portfolio with the £10.1 million cash acquisition of London-based TaxSoft. Following the Tetra takeover, there will effectively be only two UK accountancy software vendors with significant reseller channels, privately-held Systems Union, and AIM-quoted Pegasus. In recent years, the Pegasus board has rebuffed several bid approaches, including a serious offer to buy the company from Sage. ®
Drew Cullen, 01 Mar 1999
The Register breaking news

MS exec wriggles over Hardcore Bill's Java support

MS on Trial Like Kempin's Muglia's video was also clearly Muglia II, filmed on 19 February 1999 in Washington DC while Muglia was hanging around waiting for his moment of fame. Having made the point with Kempin, Boies was not inclined to spend time quizzing Muglia on this. A document that Microsoft dearly wished did not exist, by Farana Ahmed of Visual J++ marketing, noted in a heading that "Kill cross-platform Java by grow [sic] the polluted Java makers. Muglia tried to pass off "polluted" as a humorous antidote to Sun's "pure" Java, but Microsoft was too serious for this to be funny. In the semantics stakes, Muglia probably wins the prize for keeping a straight face when he redefined fragmentation: "if Sun and we disagree on this, and diverge in these areas, as long as Netscape doesn't buddy up with Sun, that is a super outcome for us (more fragmentation)." This turned out to be a great blunder, and henceforth he will surely be known as Muglia the Fragmenter. Sun's San Jose case loomed in the background, with plucky Microsoft doing its best to be brave, but failing. Muglia claimed that the dispute was centred on Sun not delivering JDK 1.2 to Microsoft, but he omitted to mention that Microsoft had very publicly said it would no support JDK 1.2, thereby putting Microsoft in breach of contract. The colloquy that gained some headlines is really pretty innocent, but following The Register's practice of bringing its excellent and well-informed readers the full facts, we include most of it: Boies: I want to direct attention to the email at the bottom of the page from Mr Gates, dated May 14, 1997, in particular to the next-to-the-last line where he says, 'I am hardcore about NOT supporting JDK 1.2.' Now, when somebody in Microsoft says they are hardcore about something, that means they really, really mean it; right, sir? Muglia: Perhaps. However, in this particular case, it's interesting--no, in this particular case, if you look at the next sentence in the next paragraph it says, 'if you think we should support JDK 1.2, it's okay.' Boies: Now, when Mr Gates says to somebody in your company that he is hardcore about not supporting something, that's generally taken pretty seriously; would you agree with that? Muglia: We take Bill seriously in general ... Boies: Have you seen any documents from Mr Gates in which Mr Gates says or suggests in any way that he would really like to support JDK 1.2 but he can't because Sun won't give it to him? Muglia: I have had conversations with Bill that he's unhappy that Sun hasn't given us JDK 1.2. No, I've never seen him say he would like to support it, but yes, I have seen him say he would like to get it from Sun. And frankly, really the reason is because we would like to be able to move forward with our contract that we have with them and deliver this technology to the marketplace... In this particular case, exactly what Bill meant by the word 'support' isn't really clear, and we have, as a part of our contract with Sun, a set of obligations to deliver this technology to the marketplace if we choose to include it in our products, so I'm not certain exactly what Bill was referring to when he said 'support'. Let me give you an example of how he could have been meaning that. The Court: Mr Muglia, there is no question pending [meaning Muglia was not allowed to carry on fragmenting, but should only answer questions]. I read it as saying he doesn't like the idea of supporting it. I don't think it could be read any other way. Maybe he changed his mind, but that's what he is saying here. Muglia: I understand that, your honour. If I could clarify one thing, please, though, which is that our contract with Sun allows us to deliver the Java technology as a part of products like Windows, and also it lets us put it on things like our web site. Now, when Bill says 'support', what he means is build it into Windows. Up to this point, in JDK 1.1, we actually took the technology from Sun. We took our competitor's APIs and put it [sic] in Windows. We have the ability as a part of our contract- The court: No, no. Stop. There is no question pending. When the time comes for your counsel to take you on redirect examination, you may then offer this information. It was pretty mild stuff really. Muglia put his foot in it rather more seriously when he said that "Office is an integrated package, and within it are programs that are also sold separately". The analogy with IE and Windows was just too close for Muglia's legal minder Tom Burt, who jumped up to protest that the question was beyond the scope. Judge Jackson told him that if Boies had not asked, he would have done so. In the miscellaneous facts department, Muglia said that Microsoft pays Sun $3.5 million/year for the Java licence. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

Update: Computerland UK issues profit warning

Computerland UK has pumped out a profit warning, after seeing its product sales fall unexpectedly in December and January. The Nottingham-based company says it expects sales to continue at this new lower level for the rest of the financial year, ending 30 April 1999. This means results will be "materially below market expectations". Computerland UK’s share price fell 60p to 100p on the news. According to the Evening Standard, analysts have slashed 1998 profits forecasts from £1.7 million to £1 million. The Evening Standard says the problem lies in sales of "prepackaged software" -- in other words Microsoft licences -- which are likely to be down 15 per cent in the second half. In an interview with the paper, Graham Gilbert, chief executive, said restructuring would take "months, not years". Computerland says it wants to be a one-stop shop for its corporate customers. This means less emphasis on price-driven, low-margin product-only sales However, product sales slumped, before the fruits of the one-stop shop strategy came through. Computerland UK is a classic corporate reseller – the company flogs value-added services to UK companies on the back of tin shifting – in much the same fashion as a Computacenter . Only it is a fortieth of the size of the European market leader The company make the majority of its profits from networking, installation and maintenance. It has also invested heavily in a remote support centre – which may make money one day. ®
Drew Cullen, 01 Mar 1999
The Register breaking news

Quake III Arena leaked to Web

Id Software's Quake III Arena beta programme has sprung a leak. Yesterday, an early test version of the highly-anticipated 3D network shoot-'em-up appeared on the Quake fan Web site ClanForums. The 16MB download is believed to have been developed for hardware manufacturers to test compatibility, released to them by Id under a non-disclosure agreement. However, according to Register sources, the Clanfest download contains no licensing information or other notification that it is a restricted release. With previous versions of Quake, Id has released such compatibility software to the public to enable users with unusual system configurations to test the game's graphics engine and report glitches. It seems likely then that this leak is the Quake III public test release that has simply emerged from Id's HQ a little too early. Our sources also report that the download contains three basic levels that simply show off some of the game engine's features, including curved walls, animated architecture and environmental elements, such as fog. The release also contain three of the game's weapons, including the machine gun and the rail gun. The levels sound like those shown by Id's John Carmack when he ran the Mac version of the game during the keynote at MacWorld Expo in January. ®
Tony Smith, 01 Mar 1999
The Register breaking news

MS licences block Be's bid for PC market

Be CEO Jean-Louis Gassee says he has had no takers for his offer (see Be boss offers free OS) to license BeOS free for a year to any OEM prepared to install it on their PCs as an option to Windows. But that's scarcely surprising, because the way he's suggesting they do it would be a breach of their licensing agreements with Microsoft. As we said last week, Be has a problem on the Intel platform because there are as yet few drivers for its OS. But the real long-term problem, as Gassee would know if he was keeping up with his Registers, is that the OEMs just plain can't do it. Gassee suggests that any free licensee could put BeOS onto its machines alongside Windows, and give buyers the option of selecting it or Windows when they first turn on the machine. But Microsoft's licensing agreements make this practically impossible (see What MS licences really say). The agreements sportingly allow OEMs to do whatever they like as the BIOS kicks in, but as soon as software starts loading off the hard disk on first switch-on, that's Microsoft's turf. Hence OEM VP Joachim Kempin's bizarre suggestions last week that OEMs could advertise their wares by running DR-DOS from the BIOS before the MS software starts loading. Clearly, unless MS is going to vary the OEM agreement in order to let them give the option of a non-MS operating system before the loading procedure starts, would-Be customers can't offer BeOS in the way Gassee suggests. Microsoft did, under OEM pressure, offer an initial choice between Windows 3.11 and Windows 95 when the latter first shipped, but that was then, and they were both MS operating systems anyway. It's also worth noting that the licensing agreements rule out shipping machines with a boot manager preinstalled (non-MS code kicking in after the BIOS), so that's not a route either. Gassee claims he's discussing the notion with one major PC manufacturer, and that he has one unnamed customer (small, presumably) which does ship Be alongside Windows, but keeps very quiet about it because it interprets the licence as meaning it can't display non-Microsoft products during the start-up. The company is of course correct -- you can't run non-MS code, and you can't publicise third-party products. So this company is indeed in breach of its agreement, if it's really shipping BeOS. But the ever-helpful Register has a suggestion (you'll recall we said "practically impossible" above). You could do that DR-DOS BIOS thing, and offer the alternative before the MS software started loading. The tricky thing to do next would be to run a boot manager from the BIOS, but failing this, bundle a BeOS CD, and if they select BeOS, boot from the CD and vape the Windows installation. We think that's clean, so give it a go, Jean-Louis. ®
John Lettice, 01 Mar 1999
The Register breaking news

How MS ‘publishes’ Java code by burying it

MS on Trial As the first phase of the antitrust trial drew to a close on Friday DoJ attorney David Boies contrived what appeared to be a neat exposure of the Microsoft dirty tricks department's Java activities. At least two Microsoft employees, it appears, took the view that Microsoft could fulfill its contractual obligations to ship software by burying it somewhere on a Web site, and making it practically impossible to find. "There will be no entry in the index for this file," one says. "They'll have to stumble across it to know it's there." The document was produced by Boies in his redirect examination after Microsoft's attorney's had helped company senior VP Robert Muglia explain what Bill Gates had meant when he wrote: "I am hardcore about NOT shipping JDK 1.2." Muglia had said: "I did also remind Bill that as a part of that contract [the Java one with Sun] we don't have to ship the Sun APIs inside Windows, inside our products. We have the ability to put them on our Web site." So Bill had been worried about putting JDK 1.2 in Microsoft's products, whereas in Muglia's view, posting on the Web constitutes shipping. Note, however, that Muglia accepts that Microsoft was contractually obliged to ship JDK 1.2, but says that Microsoft had not ever had to decide whether or not it will support it, because "Sun, in violation of our contract, has not delivered it to us." Sun of course hasn't shipped it because it says Microsoft is in violation of the contract. But then Boies came on to help demonstrate how Microsoft was 'complying' with the contract. Boies brought up the subject of Russ Arun; Muglia confirmed he knew him , and that he was a development and test manager who'd been involved with Microsoft's JVM. Boies noted that the notorious "pollute Java" document had been found in Arun's files, although Muglia insisted that "Mr Arun doesn't establish Java policy. That's not really his job." Boies then produced a series of emails relating to the publication of RMI bits, and while Muglia didn't absolutely accept that, pending Sun's legal action, Microsoft was obliged to publish these under its contract, he confirmed that Sun takes the position that this is so. In the emails, in late August Arun sends one concerning RMI bits, and Brad Abrams responds with a description of where he'd put them, and "There will be no entry in the index for this file. They'll have to stumble across it to know it's there." Boies: "And then he's got another location, FTP colon two reverse slashes and some more symbols, and microsoft.com slash development dash MSDN and some more letters. Then he says, 'Is where I put it on the Internet with 37 other old files in this directory, I'd say it's pretty buried.'" Arun mails back three minutes later and says, "Awesome, thanks." But it got worse. Muglia still tried to argue that this didn't mean Arun was involved in deciding what to post on the Internet regarding the Java contract, and dug himself in further with: "This was before we actually shipped our Java virtual machine … Before we actually ship JDK 1.1, and our Java contract obligations, as I understand them, for posting on the Internet are with final products. So in this time frame I believe it was still a beta version of it. Now, when the final version was released… I directed that it be put at a place which would be very straightforward for a developer to find." JDK 1.1 was released in October of 1997, so at this point Boies was able to cast some more doubts. One of the emails (late August, remember) says: "I will put it back October 1," and "We should not post this before September 30. Can you sit on it until then." It's not entirely clear what all this means. Boies was able to make the point that the time frame suggested the emails were talking about shipping code, rather than beta, but Muglia insisted it was about beta code, and that the shipping code had on his say-so gone up prominently. But he was forced to concede the point that it was being at least proposed to put the code in a place where people couldn't find it: "For in a beta release time frame, yes." Then Boies finishes up with the coup de grace: "OK, and were you aware, prior to my showing you this, that there had been this proposal, in your language, to hide it?" Muglia: Yes I was. Boies: If you were aware of it when I asked you whether Mr Arun was at all involved in deciding where to post things on the web in connection with your obligations to the contract, why did you tell me no? Muglia: I don't remember that that was the question that you asked. For the record, Muglia didn't entirely say no when Boies first asked the question. What he said was: "I don't know if he had any responsibility involved in that. He certainly didn't make any strategic decisions about it, but he may or may not have decided what particular files were put on the web site at various points in time." ® Complete Register trial coverage
John Lettice, 01 Mar 1999
The Register breaking news

MS rebuttal exec deletes his email every two months

MS on Trial When not cruising on his Harley and "enjoying world travel adventures" [that's how Microsoft PR puts it], self-taught programmer Eric Engstrom is currently trying to make a success of MSN (again), after an earlier career as Microsoft's games man and DirectX developer. Engstrom was one of the architects of DirectX, but it was six weeks too late for Windows 95, so it had to ship along with games, ostensibly its main raison d'etre anyway. Note however that widespread distribution of DirectX with games furthers Microsoft's campaign to disseminate its own multimedia standards greatly. DirectX is one of the latest instances of Microsoft being able to use muscle to force Redmond standards onto the world. Engstrom said "it ended up being the most successful API we ever shipped." To the surprise of court watchers in the Washington trial, Engstrom turned out to be Microsoft's straight man. He was a little more forthcoming than previous Microsoft witnesses, and avoided any amateur debate on semantics. He did not even claim the usual Microsoft quota of memory lapses. Originally, Microsoft had planned to use Thomas Reardon, Microsoft's program manager, interactive media group, as a witness, but when the DoJ said that Avie Tevanian of Apple would be called, Microsoft asked that Engstrom could be substituted. In his direct testimony, crafted by Microsoft lawyers, Engstrom was in denial: Microsoft hadn't sabotaged QuickTime, or "knifed the baby", nor told Compaq not to distribute QuickTime. Of course. But there was no denial that Microsoft had been caught with key parts of QuickTime code in its software, and had had to pay the price. Engstrom was put up by Microsoft principally to answer allegations by Tevanian that Microsoft sought to carve up the market for software relating to the creation, display, and playback of multimedia content; and had taken steps to impede usage of Apple's QuickTime software (or more simply, sabotaged QuickTime). Engstrom claimed in his lawyer-guided direct testimony there was a mischaracterisation of proposals made by Microsoft to Apple. He also was pressed on the relationship with Intel, and the acquisition of DimensionX. Phillip Malone of the DoJ San Francisco office and essentially the DoJ Microsoft case officer, conducted the cross-examination. He lacked Boies' exquisite style of trapping a witness by contrasting answers to questions with the witness' own email, but perhaps that was because Engstrom was in the habit of deleting his emails from his laptop after two months, or so he claimed. "What I do is I delete mail that is two months old on a regular basis because I work on a hard disk on a laptop. The machine is fairly old, the reason for that being I tend to test the software my group is producing, and I like to make sure that it runs on what a customer's machine is typically to be." It's a good story, but the destruction of records in this way when the company is operating under a consent decree, or being investigated during a new case is both unwise and potentially a criminal act. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

Will MS roll out Gates to shore-up the defence?

MS on Trial The Microsoft trial has now taken 17 weeks (twice as long as initial estimates) and there are probably at least three more weeks to go with rebuttal witnesses and closing arguments. There remains an outside chance that Gates will be roped in as a third rebuttal witness, to rebut his own non-testimony so-to-speak. But with Mrs G expecting again, the odds are that he's none too keen on the idea, especially as he could expect to be in the box for at least a week, and possibly two weeks. Gates will know where to put the blame however: stand up Bill Neukom, the man who got Microsoft off the hook with a toothless consent decree, who wrote his own press release when he was promoted to senior vice president afterwards, and who foolishly dated a Gates' girlfriend (although her deposition about alleged Microsoft document destruction has not yet surfaced in this trial, but could be a serious factor n Caldera versus Microsoft). Microsoft's defence in the trial was disastrous. The company lacks true friends in useful places, and the chums that did come along were hopelessly beholden to the Microsoft shilling: if he said anything important, nobody is likely to remember it. Microsoft's executives did not show any of the real expertise seen in "Java" Gosling, "OS/2" Soyring, "QuickTime" Tevanian or "Air Supply" McGeady. Microsoft's witnesses came across as semi-entrepreneurial wheeler dealers of mediocre merit, gluing together irregularly-shaped components and covering them with shrinkwrap. Microsoft is not focussed on software engineering, it is focussed on survival: but its lifeboats have holes. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

Did MS break QuickTime? It didn't help fix it…

MS on Trial Eric Engstrom, Microsoft's witness about its relationship with Apple, wiped key email from his hard disk, he claimed, which meant there was less documentary evidence to contradict his account what had happened at the meetings with Apple - and very little to support it of course. So whatever actually happened, this gross stupidity has lost Microsoft most of its credibility in its Apple relationship. Microsoft is also trying a new excuse to minimise the importance of meetings. Unless "senior executives" are present at meetings with competitors, everything is informal: "All of the communications were preliminary and exploratory, undertaken to determine whether there was any basis for Microsoft and Apple to work together on multimedia technologies for the benefit of both parties and of computer users". Microsoft claims that only a meeting on 15 June 1998 meeting with Apple could be counted. It sounded like an argument to provide a future excuse for the Court of Appeals to cite in reversing the District Court's decision. The issue of Microsoft's alleged technical sabotage of QuickTime rumbles on. Before Engstrom was tendered as a witness for cross-examination by Microsoft lawyer Theodore Edelman, Microsoft was at pains to stress that Microsoft had provided a patch to fix the "problems" reported by Apple, although there were still some minor problems with the patch, Engstrom admitted in court. The core of the argument was that Microsoft did not get its own way, and persuade Apple to dump QuickTime. There was a counter-proposal from Steve Jobs for Microsoft to use QuickTime, but this was "rejected out of hand". Engstrom's premise for collaboration was that "currently-available multimedia technology is characterised by a mass of divergent and incompatible formats and protocols that inevitably result in consumer confusion and frustration." Avie Tevanian complained to Gates on 8 August 1997 about Microsoft-induced problems with QuickTime: "I've learned that IE4 on Windows disables QuickTime and QuickTime VR. My understanding is that IE4 sets the default for .MOV files to be ActiveMovie instead of QuickTime. As you know, .MOV files have long been QuickTime files. Can you look into this and get it reversed? There seems to be a lot of ill-will generated as a result and there is a perception tat Microsoft is trying to lock-out QuickTime from Windows. This is the last thing we need as it could sour our now good relationship." Gates emailed the same day to Maritz and others at Microsoft: "I want to get as much mileage as possible out of our browser and JAVA relationship here. In other words a real advantage against SUN and Netscape. Who should Avie be working with? Do we have a clear plan what we want Apple to do to undermine SUN." The capitalisation of Sun and Java shows the extent to which Gates was obsessed at the time. In the event, nothing happened for a year about Tevanian's request, and Microsoft took no action. Engstrom claimed that Apple had provided Microsoft with insufficient data to analyse the problem, but strangely Microsoft had not thought to call Apple to ask for more. Microsoft also did not tell Microsoft about a flag that could be set that would help in overcoming the problem. Engstrom admitted that Microsoft had found two bugs in its software and fixed these. Microsoft wanted Apple to use an ActiveX control for QuickTime, which would of course have forced Apple to write a new interface for QuickTime. Tim Schaaff, senior director of Apple's interactive media group, emailed Engstrom on 21 July 1998 asking about fixes that Microsoft had said it had for problems that Apple was experiencing with QuickTime on Microsoft's Media Player. The fixes had not worked. Schaaff included considerable detail about the problems, and it would seem unlikely that it is just an Apple problem, much as Microsoft would like to characterise it that way. Engstrom had a confession, although it sounded like a sales message: Malone: Are you saying now that there essentially is no practical difference, no qualitative difference, between the multimedia runtimes, at least for audio and video content, as offered by Microsoft's DirectX media, Apple's QuickTime and RealNetworks? Engstrom: ... it has been my opinion for a very long time, that there is no substantive difference there. It's strange how experiences and interpretations of "substantive" can differ: perhaps those who had found that QuickTime and RealNetworks to be better than Microsoft's offerings were just being picky. Engstrom described one bad day he had had when Apple complained in a story in the WSJ about Microsoft sabotaging QuickTime, and he had also heard that Rob Glaser of RealNetworks ("I am friendly with his wife," Engstrom confessed) was going to testify to the Senate committee. When Malone urged Engstrom to be more precise in his answers, Judge Jackson nodded agreement, but remarked that he was "sympathetic with his bad day". In redirect examination, Edelman brought out how Microsoft had developed some code that Microsoft said should have been in Apple's installation program to make it work, and had this "verified by three organisation". The problem for Microsoft is that it has lost all credibility for the reliability of commissioned reports and market studies. Engstrom said: "It's amazing to me that Dr Tevanian could come in here with the data he had and make a claim of sabotage against Microsoft." Engstrom continued: "We built the enable plug-in flag to allow QuickTime, as it existed at the time we created the plug-in, to work. We set the flag for them so it would work. We tried to answer their questions at every point. The thing that is not probably clear here from using the word 'sabotage' in this court case -- though I, again, can't believe anyone would do that without data -- is that at Microsoft, our company would never do something like that. So, you know, this allegation reflects -- well, while it reflects on Microsoft in this court case, it reflects on me personally and my colleagues at work. I spent, you know, a fair amount of time trying to explain to my management that I had not done something, you know, this egregious, because they, you know, at first went, man, we can't believe this happened. As reports came in, you know, I believe I was vindicated, but this, you know, could have had -- it ended up it didn't, because the reports were so clear -- significant impacts to me personally and, you know, my team that worked on this project. So I have got to tell you that, you know, Microsoft doesn't do this. This would -- this was just, you know, out of our scope of understanding in this space." Well, that's his story, you know. ® Complete Register trial coverage
Graham Lea, 01 Mar 1999
The Register breaking news

Java to drive home-Net link standard

A consortium of 15 major IT and telecoms companies today launched the Open Service Gateway (OSG), a Java-based interface standard for connecting consumer and small business appliances and systems to the Internet. The companies involved -- IBM, Sun, Toshiba, Oracle, Network Computer (so Larry Ellison is represented twice, Sybase, Motorola, Lucent, Nortel, Cable & Wireless, Ericsson, Alcatel, Philips, Electricite de France and Enron Communications -- will jointly draw up the OSG spec., which will, they say, "allow the consolidation and management of voice, data and multimedia communications too and from the home". The specification will also provide for sufficiently secure connections to be made to allow utility metering devices, burglar alarms, fire alarms medical alert kit and the like to be linked to service providers and emergency services via the Net. It's that LAN-to-WAN connectivity angle that separates the OSG from the numerous emerging home networking, network infrastructure, and device communication and co-operation technologies, such as Microsoft's HomePnP, Sun's own Jini, the consumer electronics industry's HAVi and Bluetooth. While these standards are fighting among themselves to see how devices on a network talk to each other, the OSG will get on with defining how those devices communicate with the outside world. It will also establish the idea that such devices can and will communicate with service providers directly, rather than via a Windows PC, which is how Microsoft would do (and probably is, through its Millennium project). Microsoft is, of course, the most notable absence in the OSG member list, but given who is there and the technology's foundation on Java, it's not surprising the Beast of Redmond wasn't invited to the party. And indeed the real winner here is Sun. Not only does it gets to earn open source brownie points -- all the OSG specs. and APIs will be released under Sun's open(ish) software licence -- but it will strengthen Java's position and improve Jini's chances of winning the home and office LAN war. It should be good for most of the other players too, though more through the scope it gives them to define the spec. than for any direct financial rewards. Oracle, for instance, will probably ensure the OSG is primed for its next attempt to flog video on-demand. Toshiba and Philips will probably use the spec. to pave the way for delivering digital TV via the Web, and connecting hi-fi kit to the Net to download music. Motorola will look forward to selling the embedded CPUs that will power all those Java Virtual Machines being built into all these networked appliances. IBM might submit its Madison Project digital music delivery system. In any case, it gets a buzz out of still being invited to sit on technology standards committees. This particular committee is set to post its draft OSG spec. for public scrutiny by the middle of the year, once the 15 have made their own recommendations. ®
Tony Smith, 01 Mar 1999
The Register breaking news

Mobile phones cause memory loss

Mobile phones cause memory loss, according to today’s Daily Mail.
Al Z Heimer, 01 Mar 1999
The Register breaking news

MP3 developer to head SDMI

The Secure Digital Music Initiative (SDMI), the music industry-led attempt to define a standard for digital music delivery, kicked off its first major meeting, held in Los Angeles last Friday. While the meeting was little more than a chance to discuss the SDMI's aims and organisational structure, it did begin with a hand been tentatively extended to the MP3 community. First, the organisation announced it had appointed Dr Leonardo Chiariglione as executive director. Chiariglione isn't well known in Net circles, but as one of leaders of the MPEG development process, he brings a close understanding of MP3 (or MPEG 1 Audio Layer 3, as it's more formally known). His proximity to the standards process should help to calm the fears of those in the MP3 community who believe the SDMI will ultimately become the play-thing of the music industry, to be used to dominate Internet-based music business. Chiariglione is still involved in the evolution of the MPEG standard, which is beginning to incorporate the kind of rights-protection technology that bodies like the Recording Industry Association of America (RIAA) have always slammed MP3 for lacking. Specifically, MPEG 4 will add content delivery tracking technology. Later during the meeting, the RIAA's executive VP and general counsel Cary Sherman, discussed the SDMI's role in detail, which now seems to be considering a more ecumenical approach than its early publicity as an MP3-smasher suggested. "We intend to achieve an infrastructure that will support every kind of consumer transaction involving music in the future," he said, adding that that infrastructure would "enable all music to be read and reacted to in a common way." And a spokeswoman added: "The SDMI is not attempting to select from a field of competing technologies, but to look for a way to marry multiple approaches." Essentially, the SDMI is seeking to develop a rights management specification that can work alongside existing technologies and format. MP3 can clearly play a part in that as much as formats like a2b and Liquid Audio that build in rights management data -- it's simply a matter of tying it specific levels of access and duplication freedom, as set by artists and publishers. Still, SDMI does have interests beyond the MP3 debate. It was also revealed that it is looking at ways of controlling access to music stored on existing formats, including CDs. However, the success of that programme will depend on hardware manufacturers adding the appropriate firmware to their drive, and the SDMI recognises this will take some negotiation to be made to work. ®
Tony Smith, 01 Mar 1999
The Register breaking news

US telecomms giant homes in on UK ISP

Reports in yesterday’s Mail On Sunday claim Internet Network Services (INS) - one of the UK’s largest service providers - is the target of a takeover bid by Level 3. Level 3 is a US telecomms company, which is said to have £4 billion set aside with which it can go shopping. It has been around for only a year and has already upset the telecomms apple cart with an extensive price cutting regime. The report valued INS at around £130 million. The UK company was set up in 1995 and has offices in London, Manchester and Coventry. It claims to be the third largest UK ISP, according to the volume of its traffic transmitted across the London Internet Exchange. It is privately owned and backed by Baring Communications Equity and Spectrum Equity Investors, among others. It provides Web access and hosting services to large corporate companies, including Barclays and Reuters. The Mail On Sunday said Level 3 was planning to cut the cost of Web access calls by around 60 per cent. In January this year, INS bought another UK ISP, Wisper, taking some of the assets ad contracts of the outfit. No one from INS was available to comment at press time. ®
Sean Fleming, 01 Mar 1999
The Register breaking news

One-2-One to IPO

Cable & Wireless (C&W) is gearing up to spin off its mobile phone division, One-2-One. The mobile phone company is the smallest of the UK’s four mobile operators and is owned 50/50 by C&W and MediaOne, the US cable company. It has around 2 million subscribers. MediaOne has a 30 per cent stake in UK cable company Telewest. If floated, One-2-One would be worth an estimated £7 billion, £2 billion of which would be taken care of by existing debt. Analysts believe C&W has been under performing, compared to other telecomms companies, and that spinning off One-2-One would free up untapped commercial potential. ®
Team Register, 01 Mar 1999
The Register breaking news

Samsung pumps up the DRAM volume

What a difference a price upturn makes. Samsung, the world’s biggest DRAM producer, is pumping up its 1999 capital spending by 20 per cent, taking the pot up to $1.2 billion. Samsung said the investment would enable it to increase DRAM sales by 10 per cent over the year. The new money will boost Korean production of 128MB and higher DRAMs to more than 10 million units per month by the end of 1999, according to Taiwanese computer magazine Eurotrade. Samsung will hold 64MB DRAM output steady at 15 million chips per month. Last year, Samsung conducted a series of production "holidays" between April and October, in response to structural overcapacity and depressed prices in the DRAM market. The company began increasing output in November, to take advantage of a rebound in the PC market worldwide. ®
Drew Cullen, 01 Mar 1999
The Register breaking news

Sensaura parent prepares IPO

British technology development company Scipher is planning an IPO, to be completed later this year, the Financial Times reported today. Scipher is perhaps best known in the PC world for its Sensaura subsidiary which develops and licenses its eponymous 3D sound technology. Scipher claims 70 per cent of the 50 million multimedia PCs sold every year contain Sensaura chips. Supporters include Dell, NEC, Fujitsu and Compaq. Other Sensaura licensees, including ESS, Cirrus and Yamaha, account for over 65 per cent of the single-chip PC sound market. In the year to 31 March 1998, the company made a profit of £43,000 on revenues of £12.6 million. This year, its revenue is expected to have increased by 30 per cent, but a 20 per cent of turnover commitment to R&D will probably ensure Scipher's profits won't rise by anything like that figure. Still, according to the FT, Scipher is valued at around £100 million. The IPO would raise about £20 million and be used to fund new developments. The company was formed in 1996 after a management buyout made during electrical retailer Thorn's split from EMI. EMI continues to hold a 20 per cent stake in Scipher. ®
Team Register, 01 Mar 1999
The Register breaking news

Intel NDA found on floor

At the Intel Developer Forum, The Register was asked to sign a non-disclosure agreement. We never do so, partly for legal reasons and partly because we believe journalists should not and need not sign NDAs. We prefer to operate on trust and contacts and absolutely hate burning our contracts. For example, if an Intel representative said he or she trusted us not to publish before a certain date, we might or might not do so. There is another issue here. NDAs were invented by computer companies so they could tell their partners and customers something ahead of time. Journalists are neither partners nor customers of Intel, nor of any other computer company. Finally, the legal nature of NDAs is decidedly dodgy. Why, for example, should a Scottish journalist be bound by the Laws of Delaware? Is there a reciprocal agreement between the UK government and the State of Delaware? We don’t think so. As far as we aware, these pieces of paper are untested in courts. A few years ago, The Register obtained a future Intel roadmap from one of its OEMs and published the lot. At that time, the staffer worked for a big publishing house, and Intel’s European lawyers put the screws on us, big time. If we had ever signed an NDA with Intel, we would probably be, as the saying goes, bang to rights. We found this NDA lying around at the Intel Developer Forum and thought readers might be interested in its contents, so we scanned it in. We already knew that Intel had an eight-way system but didn’t know it was called Saber. What is codename Holmdel all about? Here it is:- SINGLE-USE NON-DISCLOSURE AGREEMENT This Single Use Non-Disclosure Agreement (‘Agreement') is entered into and made effective as of the date set forth above, by and between lntel Corporation ("Intel"), and the Participant identified below ("Participant"). The parties as specified below agree on the following terms: 1. Confidential Information. The confidential, proprietary and trade secret information being disclosed by the disclosing party ("ConfidentialInformation"), is that information marked with a "confidential","proprietary", or similar legend, and is described as: (Be specific.Include subject or product, any document title, drawing/document number, date, rev. etc.) (Use additional sheets if necessary.) (a) Intel Confidential Information): Future Roadmaps, Potential Partnerships, Timelines, Functionality and ArchitecturalConsiderations regarding the following Intel Lab Technologies: 8 Way Saber Enterprise System, Pentium® I11 processor enabled Intel Video Phone, "Athens" project, Pentium® III processor enabled Fan Duct Cooling System, Merced SDK, Advanced Libraries for Graphics Tools "Geyeserville' technology, New. 18 micron 400 & 433 notebooks, StrongARM technology utilizingnew Speech Recognition algorithms, Multi-Resolution MeshTechnology Format Conversion Tools, "Holmdel' technology,and Next Generation Content Protection Technology. (b) Participants Confidential Information: All Confidential Information received from the disclosing party will be in tangible form. Nontangible disclosures must be identified as confidential prior to disclosure and reduced to writing marked as provided above and delivered to the receiving party within thirty (30) days to be considered Confidential Information. 2. Disclosure will start on: Tuesday, February 23rd 1999. Disclosure may continue for up to thirty (30) days thereafter. The parties' representatives for disclosing or receiving Confidential Information are: Intel: Michelle Klein, Bill Hammond, Jay Gilbert, Scott Trotter, Andrew Liu, Bill Colson, Gary Carleton, Mark Holier, Monique Hayward, Craig Farrer, Rusty Schafer, Derek Everett, Patrick Bohardt, Debashis Chowdury, Will Schreiber, Bart Kessler, Dan James, & Brendan Traw. Participant: All information exchanged by the parties will be made by/to these representatives. 4. Obligations of Receiving Party. The receiving party will maintain the confidentiality of the Confidential Information of the disclosing party with at least the same degree of care that it uses to protect its own confidential and proprietary information, but no less than a reasonable degree or care under the circumstances. The receiving party will not disclose any of the disclosing party's Confidential Information to employees or to any third parties except to the receiving party's employees, parent company and majority owned subsidiaries who have a need to know and who agree to abide by nondlsclosure terms at least as comprehensive as those set forth herein; provided that the receiving party will be liable for breach by any such entity. The receiving party will not make any copies of Confidential Information received from the disclosing party except as necessary for its employees, parent company and majority. owned subsidiaries with a need to know. Any copies which are made will be identified as belonging to the disclosing party and marked "confidential", "proprietary" or with a similar legend. 5.Period of Non-Assertion. Unless a shorter period is indicated below, the disclosing party will not assert any claims for breach of this Agreement or misappropriation of trade secrets against the receiving party arising out of the receiving party's disclosure of disclosing party's Confidential Information made more than five (5) years from the date of receipt of the Confidential Information by the receiving party. However, unless at least one of the exceptions set forth in Section 6 below has occurred, the receiving party will continue to treat such Confidential Information as the confidential information of the disclosing party and only disclose any such Confidential Information to third parties under the terms of a non-disclosure agreement. If initialed and filled in below, the period after which the disclosing party agrees not to assert claims against the receiving party with respect to the Confidential Information disclosed under this Agreement will be 24 months (not less than 24 nor more than 60 months), (/) 6. Termination of Obligation of Confidentiality. The receiving party will not be liable for the disclosure of any Confidential Information which is: (a) rightfully in the public domain other than by a breach of this Agreement of a duty to the disclosing party; (b) rightfully received from a third party without any obligation of confidentiality; (c) rightfully known to the receiving party without any limitation on use or disclosure prior to its receipt from the disclosing party (d) independently developed by employees of the receiving party; or (e) generally made available to third parties by the disclosing party without restriction on disclosure. 7. Title. Title or the fight to possess Confidential Information as between the parties will remain in the disclosing party. 8. No Obligation of Disclosure; Termination. Neither party has any obligation to disclose Confidential Information to the other. Either party may terminate this Agreement at any time without cause upon written notice to the other party; provided that each party's obligations with respect to information disclosed during the term of this Agreement will survive any such termination. Either party may, at any time: (a) cease giving Confidential Information to the other party without any liability, and/or (b) request in writing the return or destruction of all or part of its Confidential Information previously disclosed, and all copies thereof, and the receiving party will promptly comply with such request, and certify in writing its compliance. 9.Residuals. Notwithstanding anything herein to the contrary, eitherparty may use Residuals for any purpose, including without limitation use in development, manufacture. promotion, sale and maintenance ofits products and services; provided that this right to Residuals does not represent a license under any valid patents, copyrights or other intellectual property rights of the disclosing party. The term "Residuals" means any information that are retained in the unaided memories of the receiving party's employees who have had access to the disclosing party's information pursuant to the terms of thisAgreement. An employee's memory is unaided if the employee has not intentionally memorized the information for the purpose of retaining and subsequently using or disclosing it. 10. General. (a) This Agreement is neither intended to nor will it be construed as creating a joint venture. partnership or other form of business association between the parties, nor an obligation to buy or sell products using or incorporating the Confidential Information. (b) Both parties understand and acknowledge that no license under any patent, copyright, trade secret or other intellectual property right is granted to or conferred upon either party in this Agreement or by the transfer of any information by one party to the other party as contemplated hereunder, either expressly, by implication, inducement, estoppel or otherwise, and that any license under any such intellectual property rights must be express and in writing. (c) The failure of either party to enforce any right resulting from breach of any provision of this Agreement by the other party will not be deemed a waiver of any right minting to a subsequent breach of such provision or of any other right hereunder. (d) This Agreement will be governed by the laws of the State of Delaware without reference to conflict of laws principles. (e) This Agreement constitutes the entire agreement between the parties with respect to the disclosure(s) of Confidential Information described herein, and may not be amended except in a writing signed by a duly authorized representative of the respective parties. Any other agreements between the parties, including non-disclosure agreements, will not be affected by this Agreement. SIGNATURE BY AN AUTHORIZED REPRESENTATIVE OF EACH PARTY INTEL CORPORATION 2200 MISSION COLLEGE BOULEVARD SANTA CLARA, CA 95052-g 119 [Space for signatures etc.] ®
Mike Magee, 01 Mar 1999
The Register breaking news

Hacker infiltrates military satellite

The UK Ministry of Defence has come under attack from a hacker who is allegedly threatening to target military satellites unless a £3 million ransom is handed over. According to a story in today’s Daily Mail, the hacker has already seized control of one satellite, altering its course. The satellite in question is said to be involved in co-ordinating bombing raids on Iraq. Other targets for the hacker have been GCHQ - the spying operation that listens in on telephone calls and other communications - and a number of UK operations overseas. Officers from the Metropolitan Police Computer Crime Unit are said to be engaged in tracking down the source of the attacks. The authorities are said to have been so concerned about the attack on the satellite that the prime minister, Tony Blair, was informed. High profile hackings are becoming more common. One of the most well known was involved two UK hackers, Datastream Cowboy (Richard Pryce) and Kuji (Mathew Bevan), who caught the CIA’s attention in 1994 after the Pentagon’s computer was broken into. The South Korean atomic research institute was also hacked, provoking fears that World War III might be started by a teenage computer hacker sitting in his bedroom. ®
Sean Fleming, 01 Mar 1999
The Register breaking news

3Dfx to promote Linux Voodoo development

3Dfx today hopped aboard the Linux and open source software bandwagon as part of the company's attempts to shore up its leadership of the 3D acceleration market against aggressive rivals nVidia, ATI and S3. On the Linux front, 3Dfx announced a programme whose goal is to encourage developers to release titles on Linux and other species of Unix that can utilise the company's Voodoo technology. The scheme includes the release of a Linux version of 3Dfx's Glide graphics library plus drivers, and the formation of a Usenet newsgroup to "raise Linux awareness in the 3Dfx community". The Open Source element of the 3Dfx announcement centres on the company's decision to release the 2D acceleration spec. for its Voodoo2-based Banshee chip-set to the open source community. In effect, all this is simply putting 3Dfx's official seal on work already underway in the Linux world, particularly as more games begin to be ported to Linux, particularly those based on the OpenGL graphics library. Earlier this year, Id Software said it would release a Linux version of Quake III Arena simultaneously with the PC and Mac versions. The game requires a 3D accelerator card. Still, it's a cunning move. As more 3D gamers -- and Linux users tend to be big fans of 3D games -- shift their allegiance to technologies from nVidia and other, 3Dfx stands to see its 70-odd per cent share of the market narrow. It's latest scheme could well see increased development of 3D games on Linux, most of it utilising 3Dfx technology. Giving the Banshee 2D spec. to the open source community seems less advantageous since promoting it as the de facto standard for Linux 2D graphics ultimately depends on how many Banshee cards are out there. ®
Tony Smith, 01 Mar 1999
The Register breaking news

AMD recruits IBM for 450MHz launch

IBM has selected the spanking gorgeous 450MHz AMD 6-2 to power its latest Aptiva, the model 520. Priced at $1,299, the Aptiva 520 will be available "soon" in US retail outlets. The machine is decked out with 64MB of memory, 10GB hard drive, 32X CD-ROM and bundled with Windows 98 and Lotus SmartSuite. For good measure, IBM is throwing in ViaVoice 98 Home Edition. This software has been optimised for AMD’s 3D Now! technology. In OEM quantities of 1,000, the AMD K6-2/450 costs $203 per chip. ®
A staffer, 01 Mar 1999
The Register breaking news

Merced – those pictures

Developer Forum - Updated See our original story Merced only slightly undead. We have also since published another story: Merced chieftain outlines chip futures, which is an interview with Stephen Smith, pictured below. And if you want to see a schematic of the Merced cartridge, go to Merced Solutions Overview Our full coverage of the Intel Developer Forum Feb 99 is here. The Merced package is 5 x 3. Pictured here is the connector side of the packaging. It has approximately 560 connections and can be hung from a motherboard and attached on top. To give an idea of scale, here are the hands of Stephen Smith, who runs the Merced programme and is holding a special Merced mug, next to a cup of coffee (his) and to a glass of red wine (ours) The heatsink side of the Merced packaging. You can see the clip mounts on the side but the shiny nature of the alloy didn't come out well in our photo. Merced held in a journalist's hand. This is the hand of Gordon Thorn, VNU's lab manager, whose life was saved later in the week by an Intel employee (see story). A shot of Stephen Smith, Merced's programme manager, in profile ®
Mike Magee, 01 Mar 1999
The Register breaking news

SAP confirms Linux interest

A reader has pointed our attention to a SAP Web site which shows how much the German company is interested in the Linux OS. We reported that SAP was on the Linux trail last month. If you tune your browser or micro browser to this site, and you have a smattering of German, you'll get the idea. The piece was posted today. SAP will show this stuff at CeBIT 99. The Register has a team in Hannover on the spot. ®
Mike Magee, 01 Mar 1999
The Register breaking news

Intel completes Shiva deal

Intel’s purchase of networking vendor, Shiva, has at last been completed. The deal will see Shiva rolled into the Intel Networks division to form a new company Intel Network Systems. Shiva shareholders will be offered $6.00 in cash for each outstanding share of common stock. Announced back in October,the deal soon encountered opposition from Shiva shareholders who alleged Shiva and its directors had not acted with shareholders’ best interests in mind when negotiating with Intel. ®
Team Register, 01 Mar 1999