 |
 |
The
weekly newsletter for the Fed II game by ibgames EARTHDATE: April 9, 2006 OFFICIAL
NEWS |
 |
 |
 WINDING DOWN An idiosyncratic look at, and comment on, the week's net and technology news by Alan Lenton Great news for the thirsty this week - astronomers report finding a cloud of alcohol 288 billion miles long. That's an awful lot of alcohol! Mine's a Bombay Sapphire and tonic - a large one. Unfortunately, the cloud is methyl alcohol, which is poisonous, but I'm sure that by the time we reach it someone will have been inspired to invent a process to change it into ethyl alcohol! This issue is the last for a couple of weeks. Next weekend is Easter Weekend, currently a Christian festival, falling on the first Sunday following the full moon that occurs on or following the spring equinox (March 21). Try and code that into your homebrew calendar system! The week after Easter is the ACCU conference, at which I'm speaking, and so I will be away from the ibgames command module for a week, but I will be back on Sunday 30th April with another issue of Winding Down. And as for this week...
Fascinating new info on the SCO vs IBM and the rest of the world case. For the uninitiated, SCO claims to own the licence to Unix (something which is disputed, but that's another story). A couple of years ago it decided that Linux included source code from SCO's version of Unix. It therefore decided to file a case against IBM of all people. Why it picked the one organisation with both the will and the money to mount a determined defence, I have no idea. Inflated egos, maybe. To cut a long (very long) story short, IBM said (in legal jargon) OK, tell us which code, file names and line numbers and we will look into this. SCO failed to come up with the goodies, citing all sorts of spurious reasons. Eventually, IBM got the court to order SCO to produce file names and line number and details of the alleged infringement for IBM to investigate. Zilch. SCO continued to run around making vague claims and asserting that it was up to IBM to figure out where it had 'infringed' SCO's source code. And there the matter rested until recently. Now, however, IBM has moved onto the offensive with request for the judge to dismiss 198 items (the bulk of the items) listed in SCO's filing, since they haven't complied with the court judgement to provide details. IBM's filing gives some idea of what is involved, and the scale is staggering. Remember that IBM hasn't been told what these 198 items are, but it is supposed to search for them in: 11 versions of System V
code - 112,622 files - 23,802,817 lines of code That's over 5 million files and just under 2.8 billion lines of code. ( I had to work out the latter number by hand because my freebie Microsoft calculator plugging Visual C++ .NET couldn't handle it!) Interestingly enough, IBM experts reckoned that it wasn't the size of the job that was defeating them, but the fact that they had no idea what they were supposed to be looking for. So, the question is, will the court punish SCO for its failure to comply with court instructions? I'll let you know the answer when it becomes available. In the mean time, the Groklaw article the URL points to gives interesting background on this issue. http://www.groklaw.net/article.php?story=20060405002552215
There was an interesting little snippet slipped out in a Tech republic piece about businesses planning upgrading to Windows Vista when, and if, it comes out (see also this week's 'Homework' section). It seems that the analysts - in this case Gartner Research - expect that businesses will spend as long as 18 month testing the new version before they start to deploy it. That means that even if Vista makes it out of the door next January, by no means a given, businesses won't start to upgrade until the middle of 2008. I don't think this is what Microsoft want to hear, but it doesn't surprise me. The truth is that prior to Windows XP people used to queue all night and rush to upgrade their systems when a new version of Windows came out. But that didn't happen with XP. XP didn't really take off until manufacturers had sold a slew of new machines with XP installed. Even then Microsoft had to make it difficult to buy new machines with Windows 2000 installed. Apart from new machines, the bulk of upgraders were developers who just wanted to test their applications on the new operating system. So assuming a January launch, realistically you are talking about early 2009 before Vista becomes common. Still, by then there might be desktop machines powerful enough to run the system at something faster than the pace of an arthritic snail... http://ct.techrepublic.com.com/clicks?t=1968645- The Gnarls Barkley track 'Crazy' has become the first single to reach number one in the UK purely via legal download. The record reach number one last week with 31,000 downloads before it hit the shops as a CD. Partly this is because relatively few singles are selling these day, but it's also because download, legal or otherwise, is rapidly becoming the preferred method of obtaining single tracks. Maybe the media corporations will spot this trend sometime next decade, but don't hold your breath! http://www.theregister.co.uk/2006/04/03/gnarls_barkley_number_one/
'The Register' has an excellent, and short, piece by former OS/2 bug hunter Dominic Connor, discussing the sort of problems Microsoft's Vista has probably got into. In the process of which he describes some of the similar problems with the OS/2 system. It makes fascinating reading and covers such things as 'deadline fatigue', managers in denial, and shipped versions that failed to make it into the source code control system! Most programmers will recognise, from bitter experience, at least some of the things Dominic describes, but it is the inexorable piling of one on top of the other that makes it so fascinating. And right at the end of the article there is an interesting point about the destructive nature of DRM* on good programming practices. Good programs are as modular as possible, but because of its nature DRM has to spread across all the whole program, thus breaking the modularity so essential for the writing of good programs. And the future prospects? Difficult to say, but Dominic suspects that Vista may have got into a loop whereby for each fix put into the code something else breaks, meaning that you can never clear all the known bugs. A grim scenario, indeed for Microsoft. An excellent read. http://www.theregister.co.uk/2006/04/06/os_vista_coding/
PS3 priced at
Euro499-Euro599 - Sony exec Top Microsoft
programmer signs up for space mission Record ocean waves
observed Microsoft will now
support customers that run Red Hat and SuSE Linux in
Virtual Server 2005 R2. Apple, Beatles case
hinges on outmoded agreement
Alan Lenton is an on-line games designer, programmer and sociologist. His web site is at http://www.ibgames.net/alan. Past issues of Winding Down can be found at http://www.ibgames.net/alan/winding/index.html. |
|