PC giant Compaq managed to turn in a profit of $140 million on turnover of $9.21 billion as its CEO Michael Capellas reported the company's third quarter earnings. That was a result better than analysts on Wall Street had expected but only represents a growth rate of two per cent in the European and US geographies. As market research over the weekend showed, Compaq suffered in the PC sector, losing market share to its chief rival Dell. Capellas blamed competition and the earthquake in Taiwan for this, showing how much Compaq relies on the island manufacturers to OEM its products. Capellas announced himself satisfied with the small profit, which he attributed to the changes, including cost cutting, that Compaq has undertaken over the last three months. He said: "We began to articulate a clear strategy to enable the next-generation Internet infrastructure, a key to accelerating profitable growth for Compaq. We do face some significant challenges, but are confident we will continue to make good progress on our core business objectives and drive increased value for our shareholders." Compaq's share price stood at $19.25 dollars on Wall Street at close of play. Changes Capellas has wrought included a restructuring of the company into enterprise solutions and services, consumer and commercial groups, which Capellas broke out separately. The company has not yet made all the job cuts it promised. He said the enterprise and services group turned over $4.9 billion, representing 54 per cent of Compaq's total turnover for the quarter. Services revenues only grew by seven percent. The consumer group turnover was $1.5 billion, rising 15 per cent from Q3 last year. The group "significantly" grew year on year in Europe, Latin America, Asia Pacific and China, but equally significantly Compaq did not give figures for growth in its home market. But in the lucrative and formerly strong commercial PC group, turnover for Q's Q3 at $2.7 billion was a drop of 12 per cent, year and year and turned in an operating loss of $169 million, compared to a profit of $116 million this time last year. The commercial group represents a third of Compaq's turnover in the quarter. Capellas said that its Alpha and Himalaya (Tandem) division saw a good 27 per cent growth compared to this time last year, amounting to turnover of $3.3 billion. He was cautious about how Compaq will perform in the fourth quarter. ®
Microsoft yesterday re-filed its IPO for Expedia, hoping to cash in on the Internet stock craze. A 13.6 per cent stake, amounting to 5.2 million shares, will be offered, with the share price expected to be in the 10 to $12 range. This was not disclosed in the first SEC filing in September. Assuming an $11 initial price and the sale of all the shares offered (a safe assumption), the 38.2 million outstanding shares would give an implied market value of $420 million. The CEO is Richard Barton and the CFO Gregory Stanger. In baring its soul (an assumption here) with the SEC in the revised initial registration statement, Microsoft had to file some interesting information on the financial viability of Expedia, which should make potential investors think again before parting with their greenbacks - except that they won't. For the financial year ending 30 June, Expedia lost $20 million on revenue of only $38 million. Yes, it's a small, unprofitable business. For the last quarter, losses were $5 million on revenue of $15 million, which means that profitability is not going up, although the revenue is. This suggests that the jolly forces of competition are alive and well, with Travelocity and Preview Travel being significant contenders in the airline ticket part of the game. Of course, nearly all the action for online booking is in the USA, where people evidently do not mind spending more time online to find a bargain flight than it would take to go by bicycle. All ten of Expedia's top ten destinations are in the USA, and the inclusion of leisure destinations like Orlando and Las Vegas suggest that the likes of Expedia is for consumers spending as little of their own cash as possible. Good on them - but it doesn't make for a really big business. There are localised versions for Canada, The UK and Germany. The registration statement has not yet become effective, so there is not yet a prospectus. ®
Microsoft finally gave in to pressure from The Register yesterday and fixed the release date of Windows 2000 for February, as we've been predicting for some time. The date - specifically, 17th February - doesn't entirely abolish the frenzied debate (Date prediction industry goes into overdrive) over when the software will go to beta RC3 and when it will be released to manufacturing, but it certainly takes the pressure off. Microsoft was intending to release RC3 prior to Comdex, but that's probably no longer seen as urgent. The company does however say it still intends to RTM before the end of the year. But that again needn't be too vital. Microsoft's party line now is that it always meant RTM by year end when it said it intended to ship Win2k by the end of the year, and so long as it does that, Win2k isn't late - honest. There's a fair bit of flexibility there too, however. You could interpret RTM as meaning just freezing the code prior to shipping it to the PC OEMs, and if you claimed you'd done that a day or two before Christmas, your luckless Microserfs would still have a couple of weeks for final twiddling before any of the OEMs actually wanted it. But that's an unworthy Register thought - if Microsoft is confident enough to set a firm date, then it ought to be in a position to have Win2k done before 31st December. We also incidentally hear some interesting things about the focus of the Win2k push Microsoft and its happy campers in the PC companies will be mounting from February. Microsoft's strategists, we're told, have decided an early push of Win2k in the server sector will be futile and damaging, and are therefore going bald-headed for client upgrades. This is actually a sensible idea, at least from Microsoft's point of view (Full story follows shortly). ®
Bright and shiny PCs in interesting shapes and intended to use Coppermine Socket 370 processors may not now be in the shops by Christmas. That will be a further blow to Intel, which has spent much of this year promoting the designs, which use highly unusual form factors, not suited to Slot 1 cartridges. Some of the designs were intended to make use of the flip chip S370 design, which has been delayed for up to five weeks, because of problems with the fan mechanism, as reported here yesterday. Although Celeron processors now come in the Socket 370 form factor, they do not use the faster Coppermine design that Intel hopes will be the AMD Athlon killer. At the Intel Developer Forum (IDF) in February, senior VP Pat Gelsinger introduced the designs, intended to compete against the Macintosh iMac, to a fashion fanfare where bunny people strutted their stuff. And again, at the last IDF in September, CEO Craig Barrett rolled out a whole new set of form factors, while some vendors, including NEC and Toshiba, even showed prototype machines. In February, senior VP Paul Otellini and other executives told The Register that he hoped designs would be in the shops by this Christmas. That now looks increasingly unlikely. Meanwhile, Intel tells us that the OR840 motherboard, which supports Rambus memory and Coppermine processors, will ship next week. However, as CuMine chips are somewhat constrained in supply, it could be quite some time before we see machines up and running. ®
The Cyrix technology acquired by chipset company Via earlier this year is now poised to appear in the first machines. Iwill has announced it will introduce a motherboard called the WA133-S, aimed at supporting Socket 370 flip chip Celerons and the Cyrix (Via) Joshua chip. The mobo also comes with onboard Ultra SCSI, the company said and will support the 133MHz Front Side Bus. The mobo will also have Ethernet built in. At the same time, Iwill announced a Socket 370/Slot 1 converter. Thanks to Japanese reader Battlax for pointing us to the information. See also Via poised to ship 500MHz+ Socket 370 Cyrix chip
Interesting reports in the UK press about people wearing suits at work. According to both the Daily Mail and The Guardian, a European Union ruling may mean that people can claim unfair dismissal if they're forced to wear a suit. Intel has a fab in Ireland... ®
The hunting season for stray ducks is officially over after Cheshire-based telco, Telinco, finally let fly the details behind its ornithological ISP. What's more, it's not so bird-brained as it seems. For although Strayduck.com does not offer 24/7 0800 access to the Net, it does promise to cut access bills by around a third. Strayduck.com offers free, unmetered calls every third week based on the average amount of time users spend on line during the previous fortnight. If someone uses the service for 10 hours during the first week, and eight hours during the second, they can expect to get nine hours of free time online during the third week. In the future it's also planned that Beenz can be traded for 0800 access, which would make it by the best commodity available to those who deal in this Web currency. According to Simon Preston, marketing director at Telinco, it doesn't matter how much people use the Net, they will all benefit from reduced Net access costs. Telinco's model and equation might seem tortuously difficult to grasp at first but Strayduck.com maintains the maths behind this is all done automatically so strayduckers don't have to worry about it. That, at least, is reassuring. Pluck the feathers from Strayduck.com, go beyond the inventive marketing and consumer-friendly packaging, and Telinco has simply come up with a different way of handing out cheaper Net access to its consumers. It could have charged a subscription in return for 0800 access, such as the struggling 08004U; it could have given away free time in return for users switching telco, such as Screaming.net. Instead, it's taken this cleverly disguised option in the hope to attract users. Since there's little doubt that it can deliver cheaper Net access, the real question lies in whether it can deliver a quality service that lets users access the Net first time, with uninterrupted sessions and good download speeds. For that, time will only tell. But if it gets this wrong, Strayduck might just become a dead duck. ®
City-based Net venture capitalists netvest.com Plc has pumped £340,000 into five Net companies as part of its first tranche of investments in the Net-related industry. The move comes less than two months after netvest.com floated its business on the Alternative Investment Market (AIM). Online price comparison service, shopgenie.com, is among the clutch of companies to benefit from the cash injection of £150,000. Netvest.com also paid £50,000 for a stake in Powernet, an independent reseller of free phone numbers in the UK. "These investments are in line with our stated strategy of enabling individuals who want to buy into emerging technologies to spread their risk by investing in a range of companies," said Chairman Andrew Balcombe. "We have invested in an Internet price comparison system, a telecom carrier/ISP, an Internet authentication software and hardware developer and an online music, entertainment and media group. "This is a broad range of companies but all of them have a sound proposition, a unique market position in each of their areas and exciting potential for future growth," he said. So how can fledgling companies contact netvest.com to see if this investor wants a slice of their action? It's anyone's guess -- they don't even have their own Web site up and running yet. ® Why not dangle your financial toes in Cash Register -- got to be better than a kick in the teeth
Fixing the Win2k rollout date for February (Earlier story) is only the first stage of the battle for Microsoft; now the company has to make sure the new OS is a success from day one. Information received by The Register over the past couple of days indicates that Microsoft will be specifically aiming for widespread desktop client upgrades from the off, and has been busily forging alliances with key PC OEMs to accelerate this process. The old enemy, IBM, is one of the most important of these - and now Microsoft and IBM are once again the firmest of friends, it turns out that Big Blue is going to be one of Microsoft's best and biggest Win2k early adopters as well. According to IBM VP of integrated solutions marketing Dick Sullivan, the company has already adopted Win2k as its company standard desktop OS, and intends to have it in place in the majority of IBM offices before the end of 2000. That's 300,000 seats, which positions IBM as virtually the ideal corporate customer for Microsoft. It's buying stacks, and is going for wholesale Win2k adoption earlier than most of the analysts and most major corporations think decent. Wait for Service Pack 1, did we hear? We'll get back to that. Aside from raising some interesting questions about why IBM should expect its own customers to buy, say, thin client solutions when IBM itself is buying something completely different, the accelerated rollout begs some questions about marketing alliances between MS and IBM. Is there a quid pro quo? Joint push on Win2k Pro An - ahem - entirely coincidental one, assuredly. Register sources say that IBM and Microsoft are to kick-off an intensive joint marketing programme for Win2k in February. Note that this will have been signed-off prior to Microsoft's announcement of the February rollout date, so it's clear that February has been fixed as the real target for some time. Which is how come Dell accidentally leaked it a while back (See story). Now, the interesting thing about the planned IBM-MS marketing push is that its aims seem to be a close match with those of IBM's own internal rollout, and views on initial markets for Win2k, as explained to us by Dick Sullivan. Our sources indicate that the campaign will be playing to Win2k's major strengths, which at the moment are largely at the client level. Sullivan meanwhile tells us that the IBM client rollout will not be matched by a server rollout. "The server side people are going to be cautious," he says. They'll quite possibly wait until Service Pack 2, and even then IBM isn't going to be exactly gagging to run its mission critical systems on Win2k. But a client push, both as a customer and a vendor, makes sense. Corporate customers have in the past few years been steered towards NT as the client OS of choice, but more recently the direction's been Win2k, and the protracted gestation period means there's plenty of pent-up demand, and plenty of upgrade programmes ready and waiting to be dusted down. Microsoft has also, you'll have noted in the past few months, made bullish predictions about upgrade levels (70 per cent of you will upgrade and made moves to turn off the air supply for NT Workstation by programming in early expiration dates for MS-approved qualifications (See story). There are also reasonable arguments that Win2k is a more stable and secure client OS, and on top of that there's the small matter of Win2k advanced features not being available on other MS operating systems. Basically, so long as it doesn't screw-up big-time, Microsoft can get an intensive upgrade wave going at client level. Stony ground for servers Server, on the other hand, is a different matter. Microsoft missed the Y2K upgrade boat by miles, corporate customers have spent money ahead of Y2K, and will largely be in lock-down mode through Q1 2000. After that they won't be spending heavily on servers, and will be far more likely to take a cautious approach to moving to Win2k server. In addition they've read the stuff about waiting for Service Pack 1, and they also know Microsoft has more Win2k server product scheduled for the middle of next year. So wait, wait - Microsoft won't rule the world via Win2k servers in H1 2000, and Microsoft knows that. Our sources suggest that Microsoft intends to roll out Win2k Professional first, then follow-up with server. We think that might turn out to be the effective truth, but it certainly won't look like that. Microsoft will shift a lot of client product in 2000 and not much server, but it'll be a question of focus rather than staggered rollouts. That said, we should bear in mind that a staggered rollout was once planned. Earlier this year when Microsoft started hinting at October, the intention was to launch Win2k Professional on 6th October and follow up with server later on. Aspects of this plan quite probably remain embedded in the latest strategy. We also hear that Microsoft has a cunning answer for the people who're going to be waiting for SP1 - ship it within 90 days of Win2k itself shipping. And actually, this isn't a particularly daft notion. While Win2k has been heading for RTM bit of code and features have been pulled, and there will be a sizeable initial wave of bugs to be fixed after the code's frozen. Some of these will be known before Win2k RTMs, but won't be fixed in order to make the schedule. There's nothing particularly new or nefarious in this, but effectively Microsoft will have the makings of a decent-sized service pack on the stocks by the time Win2k ship, so going to SP1 fast won't be difficult. But if Microsoft does go early, it will also mean a swift follow-up with SP2, to deal with the stuff that doesn't get fixed in time for SP1. ®
Further to an article by Graham Lea entitled Microsoft piracy losses claims don’t stack up’ appearing on The Register on 19 October, 1999, I wish to draw your attention to a number of points in response to the accusations directed at Microsoft, namely "trying to scare consumers, businesses, governments and institutions into having licensed software". The anti-piracy message on our Web site that Mr Lea refers to does not refer simply to a Microsoft issue but to an industry wide concern. Other organisations care just as deeply about the rightful protection of licensed software and the figures used are in line with figures provided by several other parties. Although Microsoft runs a number of global campaigns to highlight the piracy issue, these are in fact just one part of our marketing activity through which we are trying to educate our various audiences as to the seriousness of software piracy. Even today many people do not realise that they are committing a civil or criminal act when using illegal software. To educate them includes, amongst other things, clearly stating the consequences that software piracy is having on businesses and the general public. Mr. Lea may view our actions in condemning software piracy as being purely driven by financial concerns. However, our business and that of every software company is completely dependent on the protection of the intellectual property rights. Without this fundamental protection, the software side of the IT industry simply wouldn’t evolve. Software vendors of all sizes have to invest large amounts of money in research and development. In a highly competitive industry, such a commitment is vital in every sense of the word. The goal of anti-piracy campaigning is intended to help keep this industry as strong and vibrant as possible. We therefore have to question Mr. Lea’s logic in finding fault with this approach. International Planning & Research (IPR) should be directly questioned as to the validity of their estimates and Mr. Lea should not just attempt to discredit Microsoft because he does not accept these figures. If anything, the figures questioned by Mr. Lea are more likely to be an understatement as to the extent of software piracy, rather than an inflation of the issue. Finally, if Mr. Lea believes that Microsoft could be doing a better job at fighting piracy, we would be delighted to hear his views on the subject and welcome any constructive suggestions. Tackling software piracy is a delicate issue and we would like to think that we have been considerate and tactful in dealing with its distasteful criminal effects. I hope this clarifies our position with regards to Mr. Lea’s article. The theft of software is a criminal offence and this is a position taken by the majority of governments across the world. ® Well, what do you think? Email us here.
Analysis Well - the fight over one of Europe's key mobile markets, the UK, has suddenly turned decidedly teutonic. Not long after Deutsche Telekom paid a remarkable sum to acquire One-2-One, its homegrown nemesis, Mannesmann, hit back by gobbling up Orange. The stakes are now getting dizzyingly high; Mannesmann ended up paying about $7,000 dollars per subscriber... double the amount DT paid for its foothold. A couple of caveats: of course Mannesmann is not paying that much for the existing subcriber base - it's paying for the future customers Orange is likely to sign up in short order. After all, England is now expected to hit the 50 per cent mobile penetration mark by the end of next year and Orange has been remarkably successful in grabbing market share from the established British mobile champions, Vodafone and Cellnet. Nevertheless, it's hard to avoid the impression that the M&A boom in the mobile telecom industry is now starting to boil over. Mannesmann had earlier pulled off an enviable coup in acquiring Omnitel, the Italian mobile operator that is one of the continent's hottest properties. Deutsche Telekom retaliated by getting a foothold in the UK via One-2-One - and as soon as Mannesmann crowded into the British market, DT announced an ambitious acquisition bouquet in Eastern Europe, giving it stakes in mobile operators all over east of Danube. There aren't many obvious growth market plays left in Europe, which raises the question about the second stage of the Mannesmann - DT rivalry. A sign of the times: the ultimate takeover bait in the US industry, Voicestream, has now zoomed from its May share price of $20 to over $90 during a span of a few dizzy summer months. If markets are already concerned that Mannesmann overpaid for Orange - just wait for the Voicestream take-over price. Actually, that's exactly what the markets have been doing. They have been pricing in the takeover premium into the Voicestream share price, since it looks like such an obvious target. Which leaves the putative acquirers in a tough situation - will they pay a premium over the premium already built into Voicestream price? Can they afford not to? Voicestream is the stand-out in the current US mobile operator third quarter reporting season. By delivering a 135 per cent growth in subscriber additions compared to the same period during 1998, the company was able to demonstrate far more robust growth than the sluggish AT&T and managed to clearly outperform Nextel and Sprint, the other two leading national players. But that's just the growth advantage - in absolute terms, Voicestream added just 122,000 subscribers, far less than the national leaders. The aggressive corporate strategy of the company has led to recent acquisitions like Omnipoint and Aerial, which will add some bulk to Voicestream. Ambitious new build-ups in major markets are expanding Voicestream's grasp. And Omnipoint's recent move to acquire rights to build networks in Los Angeles and Washington DC will turn the Voicestream/Omnipoint entity into a genuinely national player. All this frantic activity combined with superior growth would seem to make Voicestream an irresistible target for European operators who have now more or less divided the major European markets among themselves. The fact that Voicestream uses GSM like its European counterparts adds another seductive lure -all that international mobile roaming income that could be unlocked with the right marketing and packaging! Is this the best acquisition prospect ever or what? Well - no. Not necessarily - and not at a substantial premium. Because behind the glitter and glamour of a US acquisition lurks a dark menace to putative European acquirers. Voicestream is probably going to turn into the mother of all money pits in the short term. Building new networks in places like Los Angeles to attack AT&T and Sprint is extremely expensive - the American urban sprawl has already defeated the quality aspirations of the deep-pocketed Sprint and AT&T, leading to patchy coverage and high churn rates as irritated customers sign off. Voicestream needs an expensive, national marketing campaign to turn it from a successful regional operator into a major force. And as good as the roaming angle sounds, it also demands a massive overhaul of roaming agreements with European partners, a big ad campaign to familiarise consumers with the concept and price cuts in roaming rates to make people embrace the idea. So the potential pay-off is massive - but this pot of gold lies years in the future and demands a very strong investment commitment. All this probably means that Mannesmann simply can't afford the venture. Its recent acquisitions have already strained the investor-relations of the company. Deutsche Telekom is a far more likely candidate - the company has lost a few high-profile bidding wars and is criticised constantly for being too staid. Apparently France Telecom is now content with bulking up its European mobile network to become as cohesive and extensive as possible - and Telefonica has probably wisely decided to focus on gaining South American supremacy and leveraging its latin brand to the max. Which may mean that Deutsche Telekom will compete with Hutchison or some American company for gaining the privilege to start pouring billions into Voicestream. Both have the capital - DT landed a windfall for selling its stake in Sprint for an exorbitant price and Hutchison pulled off more or less same trick with Orange. But Hutchison's real interests probably lie in the mainland China; getting the right to invest in China Telecom or China Unicom may be the dream ticket for Hutchison with its corporate history elaborately intertwined with Hong Kong's. But if China decides to pull out of WTO negotiations, it's extremely unlikely that the draconian mainland rules on foreign telecom investments will be relaxed. In this case, Hutchison's global ambitions might turn to North America. The stake Hutchison retained in Orange leaves the door open for some sort of a global branding deal with Mannesmann/Orange. If that happens, the fight for Voicestream may turn into another teutonic grudge match after all. ® Tips? Rants? Raves? Inside info? Give a piece of your mind to Tero. Confidentially: firstname.lastname@example.org Tero Kuittinen is the Vice President of Wireless Telecommunications, an investment firm based in New York. The firm may hold positions in companies featured in his columns. The opinions expressed in the columns are personal views of Mr. Kuittinen and they should not be interpreted as investment advice.
The MD of Scottish-based ISP 08004u -- which offers unmetered 24/7 access to the Net in return for a flat fee of £49.99 -- has admitted that his service is "absolutely appalling" but has promised that things will improve in the future. "I am not happy with the service," said David Banks. "At the moment, it's absolutely appalling... I'm not proud of it. "But we are about to upgrade the service in the next two weeks and once those changes are made we'll be able to offer a good service at a good price," he said. Unfortunately, Banks' frank and honest public admission about a service that has been dogged by a constant stream of criticism will be of little comfort to those who have paid up front for 0800 access to the Net but have received little or no service. One London-based Register reader, who asked not to be named, said that he'd been unable to access the Net via 08004u at all since he signed up last month, and his attempts to contact the ISP to rectify the situation had proved fruitless. To make matters worse, he found his credit card had been debited $80 from a company called CCBill.com. After contacting CCBill.com to query the transaction, he found that the online billing company had been contracted by 08004u to collect revenue from subscribers. To add insult to injury, his first correspondence from CCbill.com seemed to suggest that he had been billed $80 because he had subscribed to a pornographic Web site -- not because he'd subscribed to an ISP. "CCBill does billing for Web sites, most of them are adult sites with monthly memberships," explained CCBill. The disgruntled user maintains he was never told about the change in billing procedure or that it would be made in a foreign currency. What's more, he's livid that he's been charged for something that he has not been able to use and has reported the company to Trading Standards officers in Dundee. ®
Earlier this week Intel and ARM agreed a licence deal (see story), and ARM's share price went crazy. The deal means future development of Intel's StrongARM processors is assured, and the endorsement of Chipzilla is of course mightily important for ARM. But the funny thing is, Intel had an ARM licence already. Exactly why Monday's licensing deal was hugely important while the one announced on 23rd February 1998 isn't is shrouded in some considerable murk, as the terms are confidential. The 1998 announcement said that the licence allowed Intel "to produce, sell, and enhance the StrongARM microprocessor family." It also said that there was a cross-licensing deal involved. This week's announcement says the two "have finalised a licensing agreement which will enable Intel to develop a full range of solutions based on current and future versions of the ARM architecture beginning with version 5TE." The specific mention of the 5TE is one possible clue to what's different this time around. ARM tells us that the previous licence was effectively inherited from Digital, and that it covered ARM4 and SA-1 StrongARM. The latest licence covers ARM5 and SA-2 StrongARM, so although Intel already had a licence, this is the first time it's actually asked for one. That explains (perhaps) the curious quote from ARM worldwide marketing VP Reynette Au, which suggests that Intel just came on board the ARM team this week: "Intel has joined the growing ranks of companies that have embraced the ARM architecture and the ARM Partnership Model," said Au, without referring to the earlier club membership. Mind you, it's strange that Intel started talking about SA-2 in March, but didn't actually secure the licence it needs to build it until October. A lot of wrangling, perhaps? Well, we hear a funny story, which is sourced to a very highly-placed ARM executive. The original StrongARM licence was with Digital, and Digital's lawyers, we're told, wanted to assure the future of the line in the event of ARM being taken over. So they suggested a clause which assigned the architecture entirely to Digital if this happened. ARM agreed, and suggested a quid pro quo - if Digital was taken over, then ARM should get to own StrongARM. Agreed, and what happened next? Well, if there's truth to the above, what happened next is at least a bit debatable. StrongARM was transferred to Intel as part of the settlement arrangements to Digital's lawsuit against the company. But after that started rolling, Compaq bought Digital. You can see how the timing of this could be rather important if the story of the ARM-Digital contra clause is true. Could we consider the Intel-Digital deal a done one prior to the Compaq takeover, even though the takeover threw the whole thing up in the air? Was the Digital StrongARM licence transferable in the first place? To what extent was the 1998 Intel licence a new one, rather than a transfer? How many lawyers and millions of bucks would it take to sort this little lot out, if Intel and ARM couldn't come to a mutually agreed solution? Given the circumstances of the original Intel-Digital deal it's perfectly plausible that the IP could have gone a bit haywire. Intel didn't actually want StrongARM in the first place, but was essentially shipped it on Digital's insistence as part of a big package of agreed measures. Intel then took a while to figure out what to do with StrongARM, at which point it would have started to figure out whether it actually owned it or not. Even if the contra clause story is true the timing of the various signings will have had an effect on this, and matters will have been confused enough for it to be difficult to say with any certainty who owned StrongARM, and what the courts might have said about who owned it. But the thought of Intel's attorneys combing the licence agreements and discovering that Digital had transferred nothing much at all to Intel is one to treasure. ®
The fall-out from the collapse of CHS Electronics into the hands of the administrative receivers has extended to some of the group's European divisions. German-based CHS Frank & Walter, CHS Germany and CHS Austria have all voluntarily filed for court protection from creditors, according to a statement issued on Reuters last night. The filing buys the European companies 30 days' protection from their creditors, during which time an insolvency administrator will decide what action needs to be taken. It's a far cry from the optimism that surrounded CHS's $37 million acquisition of Frank & Walters back in January 1997. The deal made CHS the biggest name in German distribution, knocking C2000 off the number one spot. Back in the UK, sources close to the scramble to salvage as much of CHS as possible have told The Register they expected to see an end to uncertainty regarding the group's future by the end of this week. ®
Java virtual machines, where "bragging rights rest on speed, speed, speed" according to an article in JavaWorld, have made some very significant progress in the past few months. In the latest round of tests by Volcano using their benchmarks, raw server speed shows significant improvement since May, but in most cases, stability and scalability have some way to go. JVM top honours went to Tower Technology's Tower J3.1.4 on Red Hat's Linux 6.0 Intel, which was just 1.3 per cent faster than IBM's JDK 1.1.8 on Windows NT. This in turn pipped the same JDK on OS/2 by 6.7 per cent. Microsoft's VM 3229, as it now has to call its non-standard JVM, only managed fourth place in the speed trials, so that's the end of that bragging about speed by Microsoft that punctuated the Microsoft trial. The performance of Sun's HotSpot (sixth) and JDKs (seventh and eighth) on NT were middling, but better than on Solaris or Linux. For ease of comparison, all tests were performed on 200 MHz Pentium Pros, which could in part account for Sun's results. We shall have to see whether NT continues to outperform Windows 2000 JVMs. IBM had three of the top five slots, which suggests that its massive investment in Java does at least bring some prizes. Tower has the only static compiler, which converts class files to C source code which it then compiles to a native executable program. The men and boys were sorted out however when it came to the stability and scalability tests, with only Tower and the Sun JDK 1.2.2_03 Solaris passing the 4,000 concurrent connections test. With the speed issue essentially solved, most developer attention is now on keeping machines running under varying and high loads. The poor scalability of many JVMs is put down to the difference in the threading model for Java on Linux. Sun's solution for the threads and sockets dilemma is seen to be the more promising. There is increasing pressure on Tower to offer Tower J without charge, but the problem is of course that with a commercial rather than a volunteer development, it is not easy to cover the development costs. Maybe Sun or IBM will snuggle up to Tower, but at this particular moment, it is unlikely that Microsoft would try to catch Tower's eye. ®
Motorola has bought a majority stake in Digianswer from stricken networking vendor Olicom. Terms were undisclosed. The phone-to-chips giant will embed Digianswer's Bluetooth technology into the grandly named its DigitalDNA product portfolio for wireless connectivity, portable computing and home networking platforms. Danish-based Digianswer will also develop a Home RF wireless solution for its new master. ®
Opinion It's awfully hip and trendy these days to spout on about how great it is that AMD has finally got its act together with a credible threat to Intel'sdominance. Athlon is winning plaudits from reviewers and Intel has shot itself in the foot yet again (how many feet does Chipzilla have? -- Ed) with the Caminogate cockup. And yes, it is good that Chimpzilla has a good product. It is also vastly amusing that the top banana has fallen flat on its face (that's enough metaphors -- Ed) with the i820. But in the cold light of day, sheer weight of statistics would seem to indicate that it's pretty much business as usual in the Wonderful World of Semiconductors. AMD has one fab capable of producing chips at 0.18 micron. Intel has four (and will soon have five). That's an awful lot more Coppermines than Athlons rolling off the production line, and at considerably better margins to boot. AMD has copper interconnects. Big deal. Intel says it doesn't need 'em until it gets down to 0.13 and Foster/Itanium. Chipzilla's revenues last quarter were a measly $7 billion. AMD managed a tenth of this. Intel needs a credible rival -- or at least what appears to be a credible rival -- to keep the FTC off its back. It amuses Intel to allow AMD to survive, but never forget this means AMD is effectively in the intensive care ward, under the care of Dr Chipzilla MD. And if AMD ever even looks like posing a genuine threat to Intel, Chipzilla's pricing will mysteriously plummet, effectively turning off the life support machine. This is fighting talk where we come from. ® Flame Pete Sherriff, and we'll publish your replies
The term Skunk works originated in a satirical American comic strip "Li'l Abner" in the 1950's. Li'l Abner Yokum was a hillbilly who lived in Dogpatch somewhere in the mountains.
Compaq has developed a technology capable of tracking a consumer's every move.
The Register apologises for being the bearer of bad news… but BT is not about to cut charges for Net access next Monday. Despite reports to contrary, the monster telco has denied it is about to undertake a "dramatic U-turn" in pricing policy. Sorry. A spokeswoman at BT dismissed a story that appeared yesterday on ZDNet UK that said the telco was set to offer "cheaper data calls" as being simply untrue. "We are aware of the interest in this subject, and are in active talks with the regulator, industry and government about what might happen," she said. "We have not told anyone that BT is to make any cuts to its charges on Monday," she continued. Her views were echoed by a rather bemused spokesman at the Department of Trade and Industry (DTI). Someone there supposedly told ZDNet UK that BT was set to make an announcement next week. "We would never presume to speak on behalf of BT," he said. The news that BT has not cracked under pressure from industry groups and campaigners will, no doubt, come as a bitter disappointment to those who may have already put the bubbly on ice to celebrate Monday's groundbreaking news. But it's far better to face up the disappointment now, rather than to come down to earth on Monday with a painful, arse-thumping, face-losing jolt. ® See also: Cheaper BT net access story looks like a dud
It's time to grab hold of your input device (Oi, we mean mouse you mucky-minded bugger – Ed) and get ready for the next live Web cast from top US lingerie firm Victoria's Secret. Famous for causing servers to go down all over the place during its last live Web cast at the NFL Superbowl, the lace-clad lovelies of Victoria's Secret are set to do it again in May 2000 at Cannes. We first spotted this story on ZDNet UK but when we clicked on it we couldn't get it up. Victoria's Secret is, we are led to understand, planning to push into ecommerce in a big way next year. It already has a Web site, which allows you to order from its catalogue. It has a number of existing sections such as Bare Essentials and the Bra salon. If you want to look at pictures of young women in their underwear, you could click here, but don't come crying to us if you get into hot water for inappropriate use of company IT systems. ®
There will be four way Merced systems available for its customers in June next year, a Hewlett Packard executive said today. But whether all of the hardware vendors who originally committed to the platform are still with it is a trickier question. Last week, Dr Josef Reger, a senior executive at Fujitsu-Siemens in Germany, said his company will port its BS2000 mainframe OS to Merced. When asked why Fujitsu Siemens did not seem to be part of Intel's porting system, he said: "With all respect to Intel, we do not need their expertise to help us with our plans". Compaq, with Tru64 (D/UX), has of course already dropped out of the software porting race, while yesterday Sun confirmed it had managed to port the Solaris OS to Merced silicon. Hugh Jenkins, enterprise product marketing manager at HP UK, said today his company was on target to produce a Merced-Itanium "very shortly after" June next year. He said HP will sell a four-way system initially. HP has its own Merced chipset, based on its But doubt still remains over which memory technology Merced-Itanium will employ. At the Intel Developer Forum last September, there was no sign of Rambus technology on Intel's own prototype. Further, IBM has already committed to using its own chipset on its flavour of Merced when it launches. That is likely to use double data rate (DDR) synchronous memory, which IBM is promoting heavily. And Fujitsu Siemens, which last week introduced a scaleable Sparc clone server, may use the Reliance chipset for its Merced. Reliance is a subsidiary of Fujitsu, and was adopted by Compaq after it abandoned its own SMP Triflex architecture three years back. See also Seven Dramurai ride two memory standards at once Intel abandons server Rambus efforts
The lack of ability of Intel to supply .18 micron Pentium IIIs to its channel dealers and distributors in volume until next year has given AMD a rare window of opportunity to sell its products. AMD officially released its 25th October prices yesterday but the prices of its mobile processors are unchanged. That demonstrates its products are holding solid compared to the extremely volatile desktop processor market. Although AMD, so far, makes few performance claims for its mobile microprocessors, the family, ranging from a K6-III/450 at $320/1000 right down to a K6-2/350 at $74, is making considerable inroads into the marketplace. Intel's own projections estimate that notebooks will eventually displace desktops. Next year, AMD will introduce mobile Athlon processors, as revealed here in June. ®
If you take all the letters in Bill Gates III and then convert them into ASCII code and then add up all the numbers, you will get 666, which is the number of the beast. B = 66 I = 73 L = 76 L = 76 G = 71 A = 65 T = 84 E = 69 S = 83 I = 1 I = 1 I = 1 Total = 666.
Hurrah for The Sunday Times, Britain's biggest Sunday newspaper, for choosing The Register for its roll-call of "100 of the best" Web sites. Turn to page 54 in last week's ST Culture -- that's right, Culture -- section and there we are. "British technology news site with a sense of humour. Usually the best place to look to check out industry rumours." That'll do nicely. Only five "computing" sites make the ST 100 of the best sites. The others are CNet, Wired, the inevitable Slashdot and coming in a little left-field, Netlingo.com (summed up as "fast relief from confusing internet jargon"). Good company to keep. A wee bit smug? Yes, you could say so. ®