LettersLetters Our stories PGP dies of neglect: NAI shrugsHow we can save PGP - Zimmerman drew the following responses:- I'm saddened to hear NAI is trying to kill PGP (or at least failing to try to keep it alive). Are you aware that free versions of PGP for non-commercial use can be found at: http://web.mit.edu/network/pgp.html? Thank you for your article, Ben Levy Subject: PGP Personal Edition (?) Date: Fri, 8 Mar 2002 14:49:08 -0700 -----BEGIN PGP SIGNED MESSAGE----- Sir, I read your article published at this web site :(http://www.theregister.co.uk/content/54/24336.html) You made reference that a desktop version of PGP is no longer available. While this may be true for the NAI versions, MIT is still distributing a desktop GUI version that has a lot of the same functions described in the package mentioned in your article. I am only writing to seek some clarification and inform you and potentially your viewers that they can still download a use a stable and functional version of PGP for personal use. The article confused me in stating that a GUI was no longer available, except in trial form. Guess the article was not crystal clear to me (as a lot of things are) and wanted to make sure that this great product does not get washed up in NAI's foolish maneuvers. Or, I could be completely wrong in all my statements, but last I checked you can still get it from MIT (http://web.mit.edu/network/pgp.html). Lets hope that PGP lives forever! Thanks, Sean Gash IS Administrator Subject: Crypto Kong and other encryption stuff Hi Andrew Regarding your article about PGP or the lack thereof. There are some alternatives these days, none are as easy to use and not all are up to the enterprise, but good enough for a technical home user. Crypto Kong: Crypto Kong uses Elliptic Curve Cryptography and only runs on Windows. Which is a pain for those of us who avoid Windows like the plague. Elliptic Curve Cryptography is interesting in light of the latest advances in factoring technology because it provides an alternative to 'standard' discrete logarithm and factoring based crypto. I have spoken with James D via email about five years ago and he is certainly not going to sell out. He has continued to develop Crypto Kong off (IIRC) his own back. Hushmail: I'm wary of any web based system, if only because its not under direct control of the user and hosts multiple users. But that said they apparantly provide an OK OpenPGP based service and it works with most standard browsers. PGP International: Not strictly for customers, but then not strictly for the US either. Gnu Privacy Guard: *nix users should head this way; command line, fast, lots of plugins and works with most *nix sytems and comes with a slew of plugins, including GPGME which provides a simple library for applications developers. Developed with German government money and pretty quickly updated any time there has been a problem (There was a problem with split signatures that was rapidly solved, as with a lot of Free Software). I am sure there are probably more out there but I cannot recommend them as I have never personally used them, I only emailed BTW because I think using cryptography is a major issue and it is better to say something than say nothing ;-). Regards John B Everitt Bugger Network Associates! Let Zimmerman go back to 6.5.8 [which is what I still use] and start from there. Privacy is IMO more about what your application can do or prevent others doing, more than what you look like while you're doing it, although deception no doubt, has its place in a good privacy solution :-) As for Mac OSX and WinXP, that's a downside for Microsoft in selling new products, not for PGP per se AFAICS. Michael O'Neill O'Neill Quigley & Associates I was interested to see that NAI had notified their customers. Only yesterday I was looking to find if one could buy the commercial PGP Desktop. Of course, non-commercial users still have http://www.pgpi.org/. As far as I can tell, the only real commercial alternative is the free GPG, which can be used with WinPT (Windows Privacy Tray) on all Windows systems. The WinPT web site is http://www.winpt.org/ . The WinPT graphical installer includes GPG, and provides a very similar user interface to PGP. It's easy to install, but has the same naive user problems as PGP in understanding how to manage keys. But then, the TLAs would not let you advertise that, would they? Andrew Yeomans Global IT Security Architecture and Strategy, Dresdner Kleinwort Wasserstein Of course, given that Ashcroft wants his fingers in everything, and that PGP actually *works*, and given the prior treatment of Phil Zimmerman re: PGP's crypto provisions, are you really surprised it's being left out to dry? Bob Halloran Jacksonville FL
Re: Back in the Bloghouse Mr. Orlowski - I liked your article on blogging - in the best of cases they tend to be op-ed pages and nested commentaries, not new news dug up by the writers. That doesn't make them bad, it just means they aren't journalism. I find that I visit Yahoo news even more often now that I've become addicted to warblogs. The cloud of opinions makes me look for some shining facts. But the warblogs are a lot like my recollections of sitting with friends in the common room in college and reading the LA Times. We would point out interesting articles to each other and comment upon them. The bIogs and email allow that sort of interaction to go on that is a little more diffuse than just a one-to-one conversation. I know that others get that some sort of thing from chat rooms (or pubs and park benches) - but I didn't ever get into the chat room thing. Too high a density of flamers. For me at least, the blogs have the advantage of being edited/run by an individual and you can quickly get a sense of whether this is a person whose perspective is worth returning to sample. You get the advantage of an energetic person searching out things that might be of interest to you. A wetware version of a software agent. Anyway, I liked your perspective. Mike Pearson Hi Andrew, Enjoyed reading your Register comments on blogs today. I like blogs, but I see them as simple tools allowing for self-expression. I get to say things at my blog that would be inappropriate or misunderstood at my non-profit's website. It occasionally brings me an e-mail and, so far, those have been interesting. What aggravates me about many in the blog crowd is what has aggravated me about so much on the Net over the past few years: this tendency to call everything "revolutionary". Good grief. Blogs have their little role to play on the greater Internet and we should be happy they're available to us, but let's keep it all in proportion. I think critiques, such as yours, are very useful. Thanks for your contribution! Bob Adams GlobalAngst Your first two lines are clueless: "Blogs are almost as old as the web... In recent weeks - three years into the blogging phenomenon..." The web is more than ten years old. The rest of the article reads like typical blog content: meandering and self-centered. Andreas Subject: El Reg Big Media? Date: Sat, 2 Mar 2002 13:55:27 +0100 You write, in http://www.theregister.co.uk/content/6/24263.html Now as you might expect, no one wishes Big Media to fall as much as us. But aren't you now part of Big Media yourself? And isn't your heated attack on the Cluetrain Manifesto and the bloggers, more or less in defense of your position in Big Media, a proof of that? Let's face it, The Register is not exactly an underground rag anymore, it's as mainstream as tecno/IT publications can get. Gunnar Hellisen (innocent bystander) Bergen Norway Oh Gunnar, you Norwegian troll, you. Of course we're big media - just add up the value of our movie studios, affiliate radio stations, record labels, and feel the power of the Vulture. Move over, Disney. With its King James-compliant English, this next one had us completely baffled. We couldn't understand a word of it, apart from the last line, but reprint it in full for you enjoyment. Subject: Back in the BlogHouse: Now I'm *really* pissed! Andrew, I think that you have missed the point, somewhat, of the recent spat between Chris Locke (as the self appointed representative of all that blogs) and John C Dvorak (as the self appointed representative of all that journalises.) Interesting first, to consider these words, for a blog *is* a journal, albeit a public one and not usually for any fee. As I recently commented on the Gonzo Marketing - Engaged blog (http://gonzoengaged.blogspot.com/), a blog specifically dedicated by its instigator to the discussion of Chris' later work "Gonzo Marketing", the net is the threat. It is difficult for an observer/participant to not conclude that some element of the hostility of Dvorak for blogging as a phenomena in its own right (as opposed to the content of many blogs, which is admittedly puerile, self referential and ultimately irrelevant) has its source in whatever competitive threat such things might be perceived to represent by those who make their living from the more usual (historically) sort. As much as blogs, a similarly high proportion of professional journalism is self referential, self serving, puerile, and simply wrong, wrong-headed, and contemptible. Blogs do indeed, lower the average cost of journalising. and therefore, to the extent that they fulfill some innate (and apparently universal) desire to read the contents of other peoples journals, they represent a competitive threat to professional journalists. Secondarily, I was struck by the plethora of assumptions underlying your assertions regarding the valuable qualities of journalism as you (attempt to) practice it; to wit, that "new stuff" or "news" is the primary valuable quality. As a well-read Christian who acknowledges the truth in the statement "there is nothing new under the sun" I won't shock you by contrarily asserting that there is great merit in a great many things that are neither new nor news, and that ideas are at least every bit as powerful as current events and news of them will ever be. The age of these ideas is simply irrelevant to their world-shaping power, and I cannot help but think that in another context a man who makes his living in an occupation exemplified by the idea that "the pen is mightier than the sword" might be inclined to be more agreeable to this particular idea as I have stated it! I say, parenthetically, that you "attempt to" practice your self styled journalism as the presentation of new stuff simply because, contrariwise, those of us who blog knew about the ongoing contretemps between Locke and Dvorak long before you or both of your readers ever did. Which brings me to another point: the self styled representation of which Locke is assumed in your missive to be the arch-duke of blogdom is precisely mirrored by Dvoraks self referential journalism and self appointed representation thereof. To say nothing of your own pretensions to championing "real" journalism by defending Dvoraks conflicted attack on blogging, which is similarly compromised. Which "authority" was it (I note your reliance on this concept, an idea as old as sin and just as pernicious, in self-contradiction of your own stated ideals, and to go with your self referential lies* and your self appointed status) that made *you* the definer of what is real journalism and again the defender of this faith against the heretic barbarian blogging hordes? It was your self, wasn't it? *Yes, I followed the link but found only the original article. If the blog you promised exists it is concealed beyond the art of this mere mortal. There is certainly no Editors Blog link on the Home Page nor on the article itself, nor any obvious (or even subtle) clues as to where it might lie. With all these "self's" one could almost believe that every act of journalism, whether amateur or professional, is innately a selfish act, an act of self-aggrandisement, now practiced with the assistance of the net, on a global scale. The difference, it seems to me, is that bloggers explicitly recognise this, accept it, in some cases celebrate it, and often enough, hope that it doesn't get in the way of some either useful, or entertaining, or merely diverting, or sometimes all three, dialogue. We understand that real professionals prefer unidirectional broadcasting to quiet civil conversation, oh yes indeedy do we do, just as we recognise the implicit threat in the recipients of this manifestly ill-mannered shouting having a working conduit - like, oh, say, blogging - with which to communicate to your employers exactly what we think of your efforts and precisely what value we'd place on them were they subjected to contest in the open market of for ideas, news, observation, commentary, etc; in short, the sum of what may be called journalism. Hey! it's not working guys, we'd better all SHOUT LOUDER!!! Good Luck with your career, Andrew. In deepest and most sincere and utter contempt, Denver Fletcher Wellington, New Zealand. Pleasure, Denver. Can anyone translate from medieval English into something more contemporary? ®
Cray confirms that the MTA-2 multithreaded supercomputer, one of the most interesting and unusual machines ever built, has indeed been sidelined by the company. "The R&D development level has gone down," spokesman Steve Conway told The Register. Will there be an MTA-3? "That will all be determined by the market. We've shipped two of the MTA-2s, one to a Japanese customer. It all depends on the market," said Conway. The MTA represents over twenty years of pioneering work in parallel processing, and the ideas inspired today's SMT Intel chips. But very little about the CMOS-based MTA resembles any of today's high end commercial systems, let alone personal computers. Each MTA processor handles up to 128 hardware threads, and each thread has its own virtual register file and program counter. The MTA processor is attached to a system board, with up to 4GB of memory per board, and up to eight of these modules can be accommodated in a single MTA system. But that's only part of the story of this remarkable machine. It's a uniform flat shared memory system, with a full-empty bit for every word of memory providing much faster synchronization. And there's no data cache. So cache coherency - the bane of SMP shared memory systems - isn't a problem. The machine creates a large number of tasks, and ensures that each is execution stream is kept busy. A vintage slide from the Wayback Machine shows MTA machines far outscaling future Cray systems. Ironically, Cray was acquired by Tera Computing, from SGI in 2000. Dr Burton Smith, father of the MTA, co-founded Tera in 1987, and the company floated in 1995. Down in the Dell Conway defended Cray's decision to focus on services revenues from selling commodity Dell systems in clusters. "It didn't make a whole lot of sense for us to develop that kind of machine, while Dell is one the best in the world for its economics. Services are a big hole in the market." But don't the customers for PC clusters, running Beowulf, know exactly how to put such a system together, we wondered, as they devised the technology themselves? And where did leave services revenue? It's not that easy, says Conway. "One of the guys who we're talking to has tried to put big four clusters in his work lifetimes, and has broken his pick on them each time," he says. "We have a ten year history with standard machines using standard processors. The Cray T3E series is the bellweather system. That's still the one everyone is trying to emulate with partial success." So it's standard microprocessor based systems, and PC clusters. Here's what Tera had to say about this three years ago:- "In an effort to improve scaling, some vendors have abandoned shared memory and introduced distributed-memory computers. These are also euphemistically called scalable parallel, massively parallel, or cluster computers. Regardless of the name, they all suffer the same basic problem: a truly horrible programming model. "First, they require that applications be rewritten before they can even be run in parallel. Then, to achieve mediocre levels of performance, they require programs to be carefully tuned to manage communications and data placement. And since these systems are built using off-the-shelf microprocessors, they require further tuning for effective use of their data caches. Finally, these systems all suffer from inadequate communication bandwidth. Parallel applications can never be expected to run as well on these computers as on shared memory systems regardless of the programming effort invested." Shudder.® Related Stories Cray stunned by shock resignation Dell inks Linux HPC cluster deal with Cray
Letters RoundupLetters Roundup Welcome the postbag, the bag that's more irregular than it should be. Network Associates neglect of PGP has prompted many of you to look for a Plan B, and there's a good selection in PGP dies of neglect - your alternatives . You just don't GET IT do you? You just don't GET IT? Apparently I almost GET IT, according to techno-utopian Dave Winer, which I think is far worse than NOT GETTING IT... We're talking about the Cluetrain vs Dvorak row. Blog writers and blog readers have sent us comments about our perspective on this flame war. Read Blog Almighty! for more details. And Mac users fly in with their views on our coverage of the G4 SPEC benchmarks. Is Mac performance: up to snuff or up the duff?. Which brings us to a splendid miscellany. Read on for learned contributions to the IPv6 debate, Dead hippies, and readers promising to turn "red-faced" Washington Register bureau chief Tom Greene with phlegm. Geddit? From: Robert Clayton To: Tim Richardson Subject: "The Lion's Share" In an article in The Register I read this sentence: "It is no coincidence, Geist argues, that NAF and WIPO have the lion's share of the market - 34.5 and 59.2 per cent respectively." Technically, speaking, "the lion's share" is the whole thing - 100% -- not just the largest of several shares. I remember a quote concerning this, though I can't recall all of it. It went like this: "One quarter of the kill is mine owing to my right as the hunter, one quarter for my position as King of Beasts. The third quarter is due to [something I can't remember - I suppose a good web search will turn this up, but I haven't the time for that just now], and as for the remaining quarter, let him dispute it with me whomever will." If you're not the one to whom I should have sent this, can you see it gets to that person? Yes, I know it's nit-picky and pedantic, and goes against the popular understanding of the phrase, but if the public isn't told it's wrong, then how shall we ever get an informed public? When we wrote about Intel's Skamania multihoming network utility we bemoaned the lack of a decent Windows equivalent to the Mac's Location Manager. Symantec abandoned its commercial equivalent, and we never had much success with the freeware Alphaworks IBM equivalent. But a reader writes:- There is a 'network configuration switcher' available for windows, it's called Netswitcher, and has been around for ages. Works great too. www.netswitcher.com Regards Gordon Lamb Fullduplex Asia Pacific Singapore It looks just the ticket. The website promises " NetSwitcher will only require you to perform a REBOOT if you modify a parameter which needs a reboot to take effect." Which reminds of the mad scientist gag: "Our experiments are harmless: EXCEPT TO THOSE WE HARM!!". Has anyone tried this? We smugly have Location Manager, here. Subject: The reg hits the spot! I was enjoying your "Windows now friendlier than Mac" page when I got to the +/v8APAAh-DOCTYPE HTML PUBLIC +ACI--//W3C//DTD HTML 4.0 Transitional//EN+ACIAPg- +ADw-HTML+AD4APA-HEAD+AD4- +ADw-META http-equiv+AD0-Content-Type content+AD0AIg-text/html+ADs- charset+AD0-utf-7+ACIAPg- [etc etc.] My first instinct was "That's pretty cheap, to take someone to task for using an email client which doesn't gibe with yours" - being a determined non-Outlook user on my home machine, I get these things too often to get excited about. However, I have to hand it to you guys. It was a delicious selection from the genre. As you read on down the screed you find yourself quietly rocking with laughter. What is it about this piece that makes it so wonderful? Obviously humour is not meant to be analysed, but I thought the "drowning" atmosphere was nicely poignant in : size+AD0-2+AD4APA-FONT face+AD0-Georgia+AD4APA-FONT color+AD0AIw-800000+AD4-But I will not abandon my Mac...+ADw-/FONT+AD4APA-/FONT+AD4APA-/FONT+AD4APA-/DIV+AD4- +ADw-DIV dir+AD0-ltr+AD4APA- lost somewhere in the middle of it all. And yes, ASCII is deeply beautiful any day. Register hits the spot! James Minney Thank you, James! Re: Stallman issues Porte Alegre clarification Thinks RMS comment on "The very existence of GNOME is the direct result of our ideals of freedom, precisely what the open source movement was founded in 1998 to reject. " is very very harsh. GNOME was started because Qt in KDE, didn't support the licence that RMS and GNU thought was the one they should use. If they had been more relaxed, perhaps the Linux-side wouldn't be split on KDE vs GNOME; but instead have one single KDE that everyone works for. I wish RMS more open pushed GNOME and KDE to work together. (Which they in some parts already do, thought its little mentioned) Orjan Larrsson Subject: Barlow article I read your article on John Barlow's thoughts about the future of the Internet and the DMCA. I find it interesting that Barlow likely made quite a bit of money as a songwriter for the Grateful Dead in the 70s and 80s. He didn't seem to have a problem collecting royalties on the sale of and airplay of songs he wrote back then, so why does he have a problem with a songwriter collecting royalties for a song transferred over the Internet?? If he doesn't want to be such a hypocrite, he should return all the money he earned co-writing Grateful Dead songs (probably his chief source of income), something, i suspect, he would NEVER do. I would also add that the Grateful Dead was a singular unique phenomenon and cant really be used as a blueprint for general use. Just a thought. Smashmark I cannot for the life of me figure out why you guys keep referring the the HP-Compaq merger as the "Sircam Merger". I'm familiar with the email virus from a few months ago of that name, but I'm really at a loss. Please, please fill me in. It's gotten so annoying that I can't read your articles anymore, for fear that my inflexible brain will explode in frustration. Thank You, Ryan Lynch Once again, the name stems from our discovery of these extraordinary emails. ® I am curious how the press perceive IPV6. Your recent Computerwire repeat set me to wondering. Since you published the article, someone at El Reg likely finds the content at least passing of the 30 second laugh test. I have tracked various measures of Internet size since before I motivated (ahem) Kirk Lougheed and Yakhov Rechter to produce BGP. If I am the only one who finds the Computerwire IPV6 article bad for the state-of-the-world, I will be still. The specific claim I have trouble with is that IPV4 will "run out" in 2005. I appreciate various administrative problems in the address space allocation. On the other hand, one measure I like is still small and not rapidly growing. See < A HREF="http://www.telstra.net/ops/bgp/bgp-percent.html" target="_blank">this. The wide-spread use of NATs (at US$ 100 a go) seems to have both cut the demand intensity and slightly improved security at the same time. My own pet model of Internet address space use gets to 50% at about 2015. I do know how silly such a claim is; it's just my pet. My -- perhaps ungenerous -- summary of IPV6 is that it solved problems perceived in 1989 by 1993. It not quite ten years on from there and todays and tomorrows problems now look different than 1989's. If my model prediction is anywhere close, IPV6 should be put out to pasture. Come 2005 we can start two years of design for IPV7 and still have a five year rollout ending in 2012. Plenty of margin for slip, as well. Best Regards, Len Bosack Subject: Poor excuse for a "reporter." "This is the same Xybernaut that sued an online critic, Dan Whatley, in absentia and won a judgment of $450,000. Whatley made a few disparaging remarks on a BBS, and was blindsided by the company's legal beagles." My Question to you: Did you do any INDEPENDENT research regarding the above? If you had, you would have found the Mr. Whatley made much more than "a few disparaging remarks..." Your face should be red! My guess..., you will never publish a correction. Subject: Hey Smart Guy!! To: email@example.com I will keep your article on Xybernaut and will certainly shove it in your face after the company and the technology prove you to be the imbecile you are!!! Jmileolem Subject: XYBR If you do not think that wearable computers are the future for most companies,military,schools just to name a few then you did not do your DD. XYBR owns over 500 patents no company can make a wearable computer without XYBR patents. When the time comes in a few years that wearables are main stream the patents alone will make this company a must own stock.So when you write for your paper try to be fair mention Bell Canada, Fed-X the pantents the progress on the company it would make you less of a chump that only writes negitave print. Pam Danner Subject: Wearable Computers From: Guninpocket@aol.com To: firstname.lastname@example.org The day will soon approach, when you will eat those words. How stupid you made yourself look, wasn't this computer being used in Afgan.? YES, I was a reader of the Register, but no more. Not with the stupid statements that YOU made in the article. Your way of base, and time will bear this out. Maybe, the day will come, when YOU need somebody with one of these wearables, and I hope they spit in your face. [Dr Janov! They're out of control - Letters Ed.] ®
Some years back The Register heard IBM saviour Lou Gerstner, then the new boy, explain some of his big visions for the industry. It having been deemed, it appeared, that his previous claim not to have a vision at all needed just a tad of correction. As it happens these visions largely haven't come true at all yet; or they have come true, but in unfortunate ways. Lou however, now headed into a happy and prosperous semi-retirement, has become substantially richer than Croesus. One of his bright ideas, which spawned a ludicrous Register in-joke we ought to get around to killing off, was that people's shoes and/or belts would talk to items of other people's apparel. The limitations of the technology of the time prompted us to muse on people bumping their belt buckles together lasciviously, in the same way as they swing their backsides against security doors so the smartcard will work. But he was right there, Bluetooth can cope with this withotu you getting arrested, although Bluetooth itself isn't quite right here yet. Bandwidth was free, he told us, which it most certainly wasn't for all of the subsequent years. It is kind of free now, in that you could (well maybe not you, but Lou could) probably pick up one of the numerous semi-bust or bust providers for chump change. The Register confidently expects to be able to pick up something of the heft of Worldcom out of the change from a round of beers RSN. This however doesn't make it free for the users, as many of you may have noticed. Nor, oddly, does it make the busted providers drop their prices to levels cash-strapped dotcoms can afford. The other thing we remember is him saying, what goes around comes around. This was not a vision, but how true, how very true. Having saved IBM by not having a vision at all, and turning it back into a very boring company, much has been going around and coming around the man himself, and Lou is now a very, very rich man. Last year's salary and bonus netted him $12.6 million, and he cashed stock options worth $115 million. He probably has twice that many left to cash, and although we've long since lost count of how many he's received, and how much he's cashed, he was estimated as being worth $500 million in stock three years ago, and surely won't have got poorer in the interim. Although Sam Palmisano has replaced him as CEO, he remains chairman for another year, and then has a ten year consultancy contract which will include use of aircraft, cars, office space and apartment. The price of cabs and accommodation in Armonk must surely be criminal. ®
StarOffice creator Marco Boerries is to unveil his new venture, a device- and OS-agnostic system intended to allow wide ranges of appliances to communicate with each other, and across the Internet. Boerries company VerdiSoft, which he founded last year after leaving Sun, believes that for at least the remainder of this decade most people in the developed world will be using more than two non-PC appliances connected to the Internet every day. Hence CrossPoint Server. If VerdiSoft is right, then it's likely to be substantially more than two. The count will include mobile phones, PDAs, cars (a complicated and expensive Internet appliance, but one nevertheless), set-top boxes, games consoles and of course those net-enabled fridges and microwaves. As VerdiSoft says, "there will be a huge demand to keep all the applications executed on such devices and the associated preferences up to date to reflect the changing requisites and behaviours of every human being." CrossPoint is intended as an infrastructure sold to service providers, who will in turn use it to offer subscription services to their users. These service will be available on all of the devices the user has, and they'll be kept in sync, and up to date. You could maybe think of it as what mobile phones did next. The initial platform is Unix, Solaris specifically (so no hard feeling there then, Marco?) but will be portable to to other carrier-grade hardware, which usually is Unix anyway. ® Related link: VerdiSoft home
IT online news service netimperative.com faces an uncertain future after its parent company was put into receivership last week. No one from the receivers, Manchester-based Harrisons, was available for comment at the time of writing. However, it's hoped that netimperative will find a buyer. A statement from the company reads: "Heltward Ltd, operators of netimperative.com, were placed into receivership on Friday March 8th. The management of the company are currently exploring a number of options in order to continue to operate the service, however, expressions of interest would be welcomed by the receivers, Harrisons in Manchester." It went on: "The team would like to thank everyone for their support and hopes to be providing a service again soon. The site will remain live in order to allow members to continue to access the archive and updates on the situation will be given when more information becomes available." Last year netimperative asked its readers to pledge money for subscriptions in a bid to generate revenue. ® Related Story Netimperative.com extends begging bowl to readers
Damaging allegations that Compaq and HP boosted their most recent quarterly figures by stuffing the channel with inventory have surfaced in the San Jose Mercury. Analysts disagree on the amount of surplus stock in the channel, and some say they see no surplus at all, but the allegations are very serious indeed. The vote for the Sircam merger looms, and channel stuffing - where a unit shipped to a retailer is booked as revenue, even if it isn't subsequently sold - is a tactic that adds to the bottom line. HP sales shot up 40 per cent in the final quarter of 2001 over the previous quarter, triple the growth of Dell. But the high doesn't last that long, and the hangover's terrible. The most notorious recent example was Palm Inc, which flooded the retail channels with PDAs in an attempt to maintain its market share. Several hundred thousand returned Palms were finally consigned to the landfill, Palm crashed into the red and shed several hundred jobs, eventually losing its CEO. Between them, Compaq and HP supply 70 per cent of PCs in the US retail channel. UBS Warburg analyst Don Young, says that Compaq boosted its 2001 figures by at least $85 million by booking promotional extras, such as scanners and printers, as they went into the channel, a change from before when revenue was booked when the PC was really "sold" to the end user. The argument, in the context of the merger, is that HP in particular is risking future financial problems by enhancing the short-term profitability of the company. No one knows how damaging the allegation is better than Walter B Hewlett Jr.® The Mercury report is here
A flaw in the zlib/libz data compression/decompression libraries could enable an attacker to mount a denial of service attack against any Linux or AIX firewall, database server, mail or Web server. It's also possible that arbitrary code could be run on a remote machine. Because these shared libraries are used by hundreds of packages on numerous platforms, the bug is on a par with the DHCP, SNMP and Sun vulnerabilities recently reported. The problem here is a stuff-up in a decompression routine which can corrupt the internal data structures of malloc by a double call to the free() function (double-free), discovered by Matthias Clasen. This enables memory space to be cleared twice, and can crash virtually any program. At the moment there are no known exploits circulating, but of course it's only a matter of time before there will be. There is no workaround, but most if not all vendors have fixes. Unfortunately, this will apply only to the library on a given system. Apps that link to it dynamically will of course be safe to run after patching, but there are hundreds of apps and services which link statically to or contain implementations of the old code, and these will of course have to be fixed individually. The Linux kernel also uses the compression library, in the ppp layer and the freeswan IPSec kernel module. Other apps/services which contain the old code include: gcc 3.0 gpg rsync cvs rrdtool freeamp Netscape (fix in the works) ssh vnc XFree86 A number of these packages (except Netscape) will have been rebuilt either to link dynamically or with their own code updated, but this varies from vendor to vendor. Check with your vendor for the appropriate patches. Once they're installed, it will be necessary to recompile your kernel (if you have the relevant update package) or update the relevant modules. Any apps you built using the old code will have to be recompiled. Generally speaking, you should substitute apps which link dynamically to zlib/libz for ones that link statically or are based on the old code. Be sure to test all patches thoroughly before integrating them into critical systems. The repaired zlib version 1.1.4 can be downloaded from zlib.org. CERT has a bulletin with a list of currently-known vulnerable systems, but the list of unknowns is still quite large. However, CERT has proven to be fastidious about updating their advisories as new information becomes available, so do check back regularly. We'll update this item as soon as we get better information. Data is a bit sketchy at the moment, and we've found some contradictory details in the various vendors' bulletins. ®
Next generation Symbian OS phones are to feature heavily at next month's Symbian Developer ExpoSymbian Developer Expo in London, as the latest models from Sony-Ericsson and Nokia face-off. Both companies rate their own developer tracks at the show, and Symbian has just celebrated the announcement of Sony-Ericsson's new P800 smartphone by focussing one specifically on the new beast. So it'll be the Sony-Ericsson P800 versus the Nokia 7650. Neither is actually shipping yet, but at the recent GSM World Congress anybody who was anybody had a 7650 with a two digit identifier on the back. The Register, for the record, is therefore officially nobody, and remains so for the moment. Both Symbian and the mobile phone networks have a lot riding on this next generation of phone. They're not quite the devices that will catch the imaginations and wallets of the broad range of phone users, but they're a step in that direction. They're an opportunity for Symbian to steal back some of the thunder Microsoft gained with its GSM World announcements, and to demonstrate backing for Symbian phones from the major networks (Orange and Vodafone are actually co-sponsors of Symbian Devexpo, as indeed is Microsoft reference platform partner Texas Instruments, plus our old friends Nokia and Intel). There will also be keynotes from major execs from Sony-Ericsson (CEO Katsumi Ihara), Motorola, and Vodafone. ® Related Stories Nokia, Symbian score blockbuster Samsung phone win Smartphone roadmaps for 2000
Business ISP edNET has launched a new Web site so that it can map broadband demand in Scotland. If those behind broadbandscotland.net can prove there is sufficient demand in areas currently not served by high-speed Internet access, then they will either press BT to enable the exchange or unbundle it themselves. Sebastien Robin, Sales & Marketing Manager at edNET told The Register: "If BT won't do it - we will." edNET claims it can make money with just 100 customers and is happy to press ahead with unbundling. BT has said that it will enable exchanges if there is a proven demand for broadband that is commercially viable. In Cwmbran, Wales, for example, BT has told Internet users that it will enable their exchange if 275 customers register their interest, but so far only a small number of people have done so. edNET's MD, Aydin Kurt-Elli, said: "We want things to be made simpler, cheaper and quicker for Scottish business. "Whilst most ISPs, like BT, are interested only in population-rich areas, we at edNET continue to demonstrate our investment and firm belief in the Scottish marketplace with this project," he said. Depending on interest, edNET hopes to start rolling out new DSL exchanges by the summer. A spokesman for BT welcomed the initiative and said anything to generate demand was a good thing. ® Related Story Welsh clubbers get DSL
Microsoft Corp chief executive officer Steve Ballmer is expected to distance his company from past business practices at the CeBIT trade show in Germany this week by stressing that the company must be seen as "responsible" in dealings with other companies. Ballmer, who recently insisted Microsoft had never retaliated against OEMs that sell rival products, will tell CeBIT delegates that Microsoft must emphasize partnerships to prosper, he told the Financial Times. Ballmer's words are an attempt to mend the company's reputation among partners and the press, after its ongoing antitrust action with the US government and prosecuting states that has exposed Microsoft's business practices. Microsoft has always had a reputation for hard bargaining, but among its activities exposed to public scrutiny by the US legal action is licensing. A US appeals courts last year demanded that the Redmond, Washington-based company introduce uniform licensing with OEMs. Ballmer will speak just days after Sun Microsystems Inc initiated a $1bn legal action against Microsoft for allegedly attempting to control the internet by killing Java. The case, which exploits last year's ruling by a US appeal court that Microsoft practices over Java were predatory and deliberately misleading, is likely to rehash old hostilities and put partners in the firing line. The FT said that Ballmer will tell delegates that Microsoft wants to be an industry leader. "The industry wants us to be more responsible. We can't have business policies that are capricious or variable. We have to be reliable and consistent. We must redouble our emphasis on partnership," he told the newspaper. "It's fair to say that we thought that we had behaved in an appropriate way in the past. But the company has grown up since then. In the past we saw ourselves as the underdog that had to battle harder. We are now an industry leader and that implies a sense of responsibility," he said. The outspoken Ballmer recently used his deposition in the antitrust case to deny that Microsoft had entered into favorable licensing agreements with OEMs or taken retaliatory action against partners when they had supported rivals' products. However, Jim Allchin, Platforms vice president, appeared to contradict Ballmer's testimony, when he conceded that Microsoft had engaged in "unlawful" practices to maintain a monopoly in the PC operating system space. © Computerwire.com. All rights reserved.
Taiwan's notebook sector appears to be relatively healthy according to research by Merrill Lynch, although visibility remains extremely limited. A research note from Merrill Lynch said that after visiting Taiwanese vendors, it seemed that March was going to be better than January, which was in turn going to be better than January. However, the bank continued, visibility remains limited with rolling forecasts at one month with no committed production, compared to a year ago, when rolling forecasts were three months with one month of committed production. Pressures on the sector include short supply and high pricing for LCDs, with little improvement until the second half of next year. DRAM contract pricing is also high. A further blot on the horizon is increasing manufacturing capacity in China, which will enjoy an increasing cost advantage as volumes ramp up. © Computerwire.com. All rights reserved.
The global mobile handset market suffered its first real decline ever last year, and in 2002 it is likely to only return to low single-digit growth, predicts market research house Gartner Dataquest, a unit of Gartner Inc. In 2001, total worldwide mobile phone sales fell 3.2% to 399.6 million, compared to a compound annual growth rate of around 60% between 1996 and 2001. In 2002, Gartner Industry Analyst Carolina Milanesi predicts that the situation will be nearly slow, with "contained growth" for the entire year. "Not much will happen in the first or second quarter, as subsidies have come to an end in many markets," said Milanesi. The market is likely to re-ignite in the third and fourth quarters of 2002, when new mass market phones with support for new standards such as GPRS (General Packet Radio Services), color screens, and mobile multimedia messaging (MMS), pushes existing users into replacing their old handsets. But 2001 has some particular low points, which are unlikely to be repeated in 2002, suggest Gartner. The major problem in 2001 was that operators in two major global markets, Latin America and Western Europe, started to remove pre-paid handset subsidies in their markets in 2001. This meant that they were left with many unsold handsets, as consumers moved to so-called SIM-only purchases, using an old or second-hand phone for a new network connection. Handset vendors had also shipped too much inventory into the channel at the end of 2000. This saw large amounts of units moved into China and the Asia-Pacific region through gray market sales. Sales of new technologies, such as General Packet Radio Handsets were also slower than expected, taking a major source of top-end growth out of the market. The only vendor to do spectacularly in the market in the year was South Korean vendor Samsung Corp, which increased sales by 36.8% to sell 28.2 million devices, with a 7.1% global market share, which put it in fourth place, due mainly to its sales of the A300 model phone. Nokia Corp was again in first place, with market share of 35% and sales up 10.5% at 139.7 million. In second place was Motorola Inc, with market share of 14.8% and sales down 1.7% at 59.1 million, although its global sales were held up by it leading position in the Chinese market. German vendor Siemens AG managed to make the third place for the third time, with 7.4% market share and sales up 10.2% at 29.7 million. In fifth place was Swedish vendor LM Ericsson Telefon AB, with a market share of 6.7%, and sales down 35% at 26.9 million. The remaining players saw sales of 115.9 million units, and a market share of 29% with sales down 15.5%. With the entry of vendors such as Sanyo Electric Co Ltd, Toshiba Corp and NEC Corp truly moving onto the global stage in the next year, especially with new technologies such as third-generation phones, this could change the market very quickly. The second wildcard factor is the entry of operators into the markets for handsets of all types. New reference platforms from vendors such as Texas Instruments, Microsoft, Nokia, Ericsson and Motorola could substantially alter the market for named phones. © Computerwire.com. All rights reserved.
Claiming a short-term cash crunch, Linux distributor Mandrake is asking individual users to join a 'club' for $5 per month and up, and business users to join one for $2,500 per year and up. All Club members will get some manner of special treatment in exchange, depending on the level of commitment offered. The idea is to help pay the salaries of employees and avoid sackings while the company struggles to achieve profitability, which it says will happen toward the end of this year. "Since MandrakeSoft's revenue continues to grow and the company remains on target to become profitable at the end of 2002, it would be a real loss and a great disservice to the huge community of users for us to cut positions to meet short-term goals when the future looks so bright," announcement cheerfully says. Membership levels for individuals range from 'standard' to 'platinum' (with several precious metals in between), and the cost of admission ranges from $5 to $100 per month. Business users can choose from the same designations, with one above 'platinum' called 'privileged partner'. The fees range from $2500 per year to $100,000. Whether this last category includes free time-shares on the corporate jets, yachts and penthouse condos we can't say, but rather hope it would. ®
Major businesses could well be poised to embrace open source software, with cost, control over development and "an alternative to the status quo" being prime considerations, according to survey data released today by OpenForum Europe. OpenForum, which aims to accelerate the deployment of open source software in business and government, jointly funded the survey with the UK's Department of Trade and Industry. Over three months 79 CIOs and financial directors in financial services, retail and public sector were interviewed, and the results suggest both a receptiveness on their part to moving to open source, and problems ahead for Microsoft's 'upgrade escalator' sales model. Perceptions actually varied surprisingly little between users and non-users. Some 64 per cent of users felt a benefit of open source would to to decrease general costs, against a still substantial 49 per cent of non users. This was by far the biggests perceived benefit, with development control (23 per cent and 14 per cent respectively) and lower software licensing costs (23 and 24 per cent) coming next. Licensing costs are the major issue as far as total cost of ownership is concerned. Then comes "an alternative to the status quo," with 23 per cent and 14 per cent, and we think we know who they mean here. Access to source, cross-platform capabilities and customisability come fairly low down, which suggests that they really just want a cheaper, commodity alternative, rather than to be able to sing and dance as well. Another item of concern for Microsoft will surely be that reasonable numbers of them (26 per cent in the retail business) propose slowing down the upgrade cycle as an important part of their licensing cost strategy. So why don't they jump? Availability of support is seen as the major challenge, and beyond that they're all over the place. Cross platform compatibility also comes second in the challenges section (23 per cent and 16 per cent), but there's a whole raft of other concerns of a similar order. Here though the users and non-users diverge most clearly, with non-users worrying hard about no track record (24 per cent), "due diligence process unproven" (a legal thing) 19 per cent, and credibility of supplier (19 per cent). So there's still a sales job to be done on the people who haven't bought into it yet, and over at the purse strings of the banks we barely seem to have started yet. Only 6 per cent of bank financial directors admit to having heard a little about open source from their IT departments, while the balance confidently assert "none." But the good news: 86 per cent of CIOs intend to run open source at infrastructure level, 17 per cent will use it for business critical apps, and 14 per cent apiece reckon it will play on the desktop and handhelds. So we can't expect a major desktop and device rollout soon, but the server end of the business looks plausible, which is of course as it should be, given the current nature of the platforms. Unlike the more overtly geeky open source organisations we're familiar with, OpenForum Europe has set itself the tricky task of evangelising the software in business and government, which means having a few suits on board itself, and working the line between suit and geek. As we've suggested previously, blood may well be spilt on this one, but the survey is a credible first effort, and more details are available here. The next stage, spokesman Graham Taylor told The Register, is to get together some credible case studies, and a migration guide. ®
The worldwide server market declined sharply by 26 percent in Q4 last year, but the worst may be over and buying patterns are expected to return to normal this year. That's according to the latest figures from analyst firm IDC which reports that the server market was worth $12.6 billion in 4Q 2001 compared to $16.9 billion in 4Q 2000. Unit shipments declined 7 per cent to 1.143 million units. As we reported last week, IDC reckons IBM marginally edged out Sun Microsystems to take the top spot for Unix market (with a share of 26.9 percent compared to 26.8 per cent). Hewlett-Packard came in a close third with 25 per cent market share. Offset by shipments of IBM's zSeries mainframe servers, the high-end server market declined only 4 percent from $3.3 billion in 4Q 2000 to $3.17 billion in 4Q 2001. During last year as a whole the high-end market dropped 9.4 percent from $13.1 billion in 2000 to $11.9 billion in 2001. ® External links Worldwide server revenues decline 26% in Q4, IDC
More than a million Orange customers experienced difficulties using their mobile phones yesterday following problems on the company's network. Punters began experiencing intermittent problems making phone calls from around 3.30pm yesterday afternoon. According to Orange more than ten per cent of its 12.4 million customers had to redial a number of times before succeeding in making calls. A spokesman for Orange said the glitch was resolved by 9.00pm last night and apologised for any inconvenience. ®
Microsoft is attempting to settle the European antitrust matter with what it describes as concessions going beyond those offered as part of the MS-DoJ settlement in the US. The offer preempts Europe's verdict on the company's activities, which was originally due early this year but now seems to be in a holding pattern pending the verdict on MS-DoJ in the US. The offer would also seem to preempt Microsoft actually offering it - the European Commission earlier today was saying it hadn't actually received the proposals yet, although Microsoft European attorney John Frank has been spinning them up to the press. There would seem to be two elements to what Microsoft is about to offer the Commission. The company says it won't assert its IP rights over its Common Internet File System (CIFS) and Server Message Block (SMB) technologies, and that it will put CIFS forward as an Internet standard. Note that although this sounds a little like offering something as an open Internet standard, that is not precisely what Microsoft is saying. Count the teeth of this horse, friends. As regards SMB, the company says it will begin offering its proprietary SMB enhancements for license. This could conceivably help the likes of Samba, but it'll depend on the Ts & Cs, which we don't as yet know. Overall, there's probably less to this than meets the eye, and the concessions are unlikely to satisfy Microsoft's antagonists, or change the Commission's verdict, as and when it feels ready to deliver it. Frank did however tip a little more of Microsoft's strategy for dealing with Europe by telling Associated Press that it is "not workable" to have a different set of rules in Europe from "what's emerged from the four years of litigation." So, plan A, make a noise about extra concessions, plan B, try to rattle the Commission about getting out of step with and offending its US equivalent. ®
Britain and the Netherlands will be the first in Europe to roll with 802.11a high speed networking kit, as country by country European approvals for the standard proceed. By July users in the two countries will be able to buy kit that gives a theoretical maximum speed of up to 54Mbps, though the equipment needs to be set up so that only four base stations are used in a given area - instead of the eight available to US users. This concession to regulators means 802.11a kit can be used sooner than expected and applies to all vendors, though Intel is the first to detail its plans for marketing the technology this side of the pond. Existing 802.11b networking kit allows transmission rates of up to 11Mbps in the 2.4GHz spectrum. Networking kit based on 802.11a is already in sale in the US, but concerns expressed by satellite and radar operators have delayed full approval from ETSI, the European regulatory agency. Using four base stations means transmission frequencies will be restricted to the 5.15-5.25 GHz part of the 5.15-5.35GHz spectrum used by 802.11a wireless LANs. David Bradshaw, Intel's EMEA head of Wireless LAN product marketing, said there would be "no performance detriment" in using only a portion of the spectrum. UK and Dutch 802.11a wireless access points will be the same as ships in the US and user documentation will explain how users can choose four legal bands during installation. Currently prohibited bands can be opened up when full regulatory approval, expected by the end of the year, is granted, Bradshaw explained. Intel expects Germany, Sweden and Belgium will green flag the technology in a second phase of the roll out of 802.11a prior to modifications in the technology expected to satisfy ETSI's concerns. Full ETSI approval is expected by the end of the year. This involves developing a technology called DFS (dynamic frequency spectrum) which means equipment will detect and avoid transmitting in bands used by existing traffic. A technology called TPC (transmission power control) which controls the power of signals sent to wireless access point - so helping to restrict the power of stray transmissions - has already been incorporated into 802.11a wireless networking kit. ® Related Stories Cisco and Intel split on high speed wireless futures Intel poised to roll out 54MBps WLANs in Europe IEEE sets 802.11g (sort of) Synad builds dual mode WLAN chipset 802.11b market grows To be or not to 802.11b Proxim doubles 802.11a wirefree bandwidth to 108Mbps Rocky road to wireless networking nirvana
As the UK braces itself for the eagerly anticipated launch of Microsoft's Xbox games console on Thursday, some gamers are feeling cheated after their orders were rejected by etailer dabs.com. Giles Essame from Milton Keynes and Matthew Coleman, from Northampton are among 13 people who took up Dabs' offer of an Xbox Ultimate Bundle (which includes the console, memory unit, controller and three games) for £249 (ex VAT). Despite receiving confirmations last week they received an email earlier today telling them that their orders had been "processed incorrectly" and that the price of their goods was actually £379 (ex VAT). Those involved have been asked to re-submit their orders and told that it could be weeks before they receive their consoles. "I'm really annoyed at how they've handled this," Giles told The Register. "They've left it to the last minute to tell us and now they expect us to wait for eight weeks for the order," he said. Giles also claims that the price was being advertised for several days last week although a spokeswoman for Dabs claims the wrong price was only available for a couple of hours. She blamed the glitch on a software error and said that once it was spotted it was taken down immediately. Only 13 people had placed orders for the "Ultimate Bundle" during that time, she said. "As soon as we realised there was a problem we took down the site," she said. "We're very sorry for any inconvenience." ®
There are red-faces at the Motley Fool today after its UK site disappeared because it failed to re-register its UK domain. A statement by the humorous financial site reads: "Our apologies for the recent problems you may have had if you tried to access our site over the last few days. "Here's the reason why. It was a boob on our part as we allowed the registration of the name "fool.co.uk" to lapse," it said. Thankfully, the Fool managed to re-register the domain before anyone else. Phew. ®
The consumer wireless networking market is becoming commoditised much like the modem market before it, with wireless networks cards costing a fraction of the price they fetched a year ago. e-tailers like Action.com, Dabs.com and Simply.co.uk are offering 802.11b 11Mbps USB adapters from £60 (excluding VAT), compared to an estimated £250-£300 such kit fetched 12 months ago, when its use outside the enterprise was rare. A search on Dabs, reveals that wireless network cards from Belkin (£75), SMC (£69), D-Link (£65) and ActionTec (£60) are pushing price points down to a level mainstream consumers can afford. Ian Robin, sales director at ActionTec Electronics, said that by Christmas wireless networking cards might cost at little as £50. The wider availability of broadband is driving demand for wireless LANs in the small office and at home, he added. David Bradshaw, Intel's EMEA head of Wireless LAN product marketing, agreed that the wireless LAN market was been commoditised as the technology becomes widely deployed. Intel has recently announced a price drop of 30 percent across its range of wireless gateways, cards and USB adapters. Enterprise class access points still sell at a premium but increased competition at the lower end of the market is encouraging vendors to price more aggressively, Bradshaw added. He added prices for wireless networking kit were likely to level off soon. ®
Intel has extended its SMT, "HyperThreaded" Xeon processors to large servers. But the processors don't give the dramatic performance benefits that they should, given the higher frequencies, better bandwidth and virtual processor technology. The chips use ServerWorks GC-HE custom chipset and come in 1.4Ghz, 1.5Ghz and 1.6Ghz frequencies, with L3 cache up to 1MB. As with the earlier SMT Xeon for wsorkstations, the chips are built to a .18 micron process, use DDR memory and the P4's 400Mhz bus. There's a significant bandwidth improvement over the PIII, too. But does this add up to a linear scaling? Not quite. Intel's own benchmarks show that for an extra 600Mhz, or 78 per cent increase in clock frequency, the SMT Xeon adds just 36 per cent in the SAP application benchmark. (Intel compares a 4-way 900MHz PIII Xeon box against a 4-way 1.6Ghz Xeon MP). And that's with the bigger L3 cache, and much better internal bandwidth. So by Intel's own figures, HyperThreading isn't delivering. We look forward to seeing recompiled applications run through the same benchmark, and word is that these do deliver the promised speed advantages. Until we do see some comparitive figures, the jury's out. Since Windows users will need to pay a higher license fee for the virtual processors in a Xeon SMT, the price/performance advantages are beginning to look very dubious. (See SMT Xeons count double for Win2k Server licences. The new Xeons are priced at $1,177, $1,980 and $3,682 respectively. An additional Windows server license starts at little under the price of the 1.4Ghz Xeon MP. Intel has vowed to base all future IA-32 server processors on SMT technology, which was formerly known as Project Jackson. Why does Xeon MP underdeliver? Answers on a virtual postcard, please. ®