THE YEAR 2003 ...
  IN CoMPuTeR SCienCe


For 2003, PowHertz has chosen not to give Powhertz Awards in multiple and various categories, which was the concept maintained over the last 3 years. The main reason is simple: the lack of content. This year was especially unexciting in regards to software releases, and rare were the PowHertz Awards' traditional categories that could have had a winner who fully deserves this honnor. Also, I had no idea for "the category that changes every year" and I hadn't taken any note for helping me write the current file over the last few months. Still, PowHertz cannot let 2003 end without making a little recap of it, in order to remind us what were the prominent events and phenomenons of the ending year.

First of all, let's talk about the software releases in the two categories for which I would still be interested in awarding a Powhertz Award: the best Web browser and the best Internet utility. In the first category, the competition was made between 2 players, a brand new one and an old one, Opera, already twice winner of this honnor in 2000 and 2001. The award for 2003 is given to the new one, Safari 1.0 from the Apple company. Although PowHertz' author and administrator did not have the chance to see Safari by himself because the software is only available for Mac OS X on a Mac machine, it is no wonder that Safari was the Web browser that made people talk the most in the industry in 2003. Just the fact that Apple enters this already-nearly-saturated market after all these years of retard was an event, but it's mostly Safari's excessive speed that made people talk about it so much. For the load of HTML pages, it is 3.2 times faster than Microsoft Internet Explorer for the Mac and, according to Apple, it would even be faster than the already extremely fast Camino!! However, the little Camino, known as the fastest browser for HTML processing on the Mac, was not performing as well with the execution of Java or JavaScript code. That's where Safari impresses even more, being faster than any other Mac Web browser for both Java and JavaScript! We would tend to think that it's too good to be true and that Safari probably does these outstanding figures by loading much more information in memory at the launch of the application. Well, PowHertz does not have any data on the use of RAM by Safari, but what is sure is that it is crushing the competition even for the application launch time! It lanuches approximately 35% faster than Internet Explorer and nearly 72% faster than Netscape! Safari's performance is so amazing that Microsoft deceided to stop the development of Internet Explorer for the Mac, leaving the market off to Apple! Microsoft did not openly and directly admit its defeat, but it's definitely what it shown by its actions. Congrats to Safari! As for Opera 7.x, it was still an excellent runner-up anyway. It continues to be the excellent, fast and full-featured browser that it ever been, but it also found ways to innovate once again and add to its list of exclusive and useful functionalities. However, it's at the internal level that the changes were the most major, with a cleaner rewrite of the engine's code and an internal restructuring that will allow to release the browser's new versions simultaneously for Windows and Linux (instead of releasing the Windows version first and the Linux version only a few weeks/months later) in the future. Congrats to Opera as well. Lesser congratulations to Mozilla, last year's winner, that continues to be an excellent browser that beats Internet Explorer at many levels, but which didn't really have major improvements in 2003 and even sometimes brought new problems to its Linux version (accents were no longer working in Web-page forms, the e-mail client and the Web-page editor in Mozilla 1.3 Beta/RC and others, the Java module that could stop working with the version 1.4 and that, even if you're using the online installer while using Mozilla 1.4, etc.). I repeat it though, PowHertz is still supporting Mozilla and would not hesitate to recommend it to any Internet Explorer user.

As for the second category, the Internet utility, the honnor goes to KaZaA Lite K++, an unauthorized version of KaZaA that allowed to use the extremely vast directory of files available via KaZaA, but without having to accept the installation of spyware or adware on your computer. For any  Windows user (because KaZaA Lite was available only for Windows; it claimed to be supported on Linux as well via the WINE emulator, but PowHertz did not succeed in making this configuration work in his internal tests), it was awesome: KaZaA's impressive directory without all this "sh*t" that was making KaZaA Media Desktop a product that was so much unrecommended to install on your computer! Victim of its success, KaZaA Lite K++ is no longer available on the Web since not long ago, following threats of legal actions. It deserves the title of Internet utility of the year anyway, actually being almost alone at the finish line.

Except for the release of the above mentionned software releases, let's see what were the prominent events of 2003 in computer science. To make it short & sweet, they group into 5 categories, 2 that come back every year, 2 more or less new ones and 1 really new of this year. Let's go from the most recent one to the more classical one.

#1 - The SCO Group
Right in January, we were learning that the company, formerly Caldera International which had purchased Santz Cruz Operation (SCO) and some rights on the UNIX operating system at the same time, had hired the famous lawyer David Boies to check if other operating systems were violating their acquired rights. Less than 2 months later, SCO was suing the giant IBM for US$1 milliard, mainly for having revealed UNIX secrets and contributed to their integration into the Linux OS. SCO's refusal to provide any proof of their claims is causing a lot of controversy, but we also got to remember that at the time of launching its lawsuit, SCO was still selling its own distributions of Linux, SCO OpenLinux and SCO Linux Server, which also was pretty bad for the lawsuit's credibility. In the months that followed, SCO pulled off its Linux products from the market, sent a letter to about 1500 companies telling them that Linux contains some of their code and that continuing to use this code without paying them could result in legal actions, put on sale "licenses for commercial use of Linux", increased to US$3 milliards the damage they seek from IBM and revoked Big Blue its UNIX license, used for its AIX operating system. The SCO Group, a company that was until now largely ignored, has quickly became in 2003 the most controversial company in computer science.

SCO has made a few friends in this controversy, mainly Microsoft and BayStar Capital who both invested a few million dollars in SCO Group following these news. However, if they added some allies, their enemies have not just added, they multiplied! It is useless to mention IBM, so let's start with Novell and The Open Group, two companies who opposed not only to SCO's legal actions against IBM, but also to the definition of SCO's intellectual property, both of them claiming to own the rights of which SCO believes to be the owner! There is also the German group LinuxTag, that has managed to shut down SCO's site in Germany, after the court recognized that SCO was using unfair techniques to damage the reputation of Linux. Next on list is Red Hat, the largest Linux distributor in the world, which sued SCO for having attacked integrity of Linux and its development process. Then came the Open Source Development Labs, Free Software Foundation and Transmeta of which employees (including Linus Torvalds) received subpoenas from SCO in November, and then SGI (Silicon Graphics) to which SCO has revoked the UNIX license for contributing to the development of Linux, with code that was yet clearly written by SGI themselves (!?!). Let's add to that of course almost all other companies of the Linux world (including SuSE, MandrakeSoft, etc.) who did not like that someone dirties up the reputation of their products without providing relevant proofs of their claims, as well as the millions of Linux enthousiasts, a few tens of which having made a few denial of service attacks against SCO since the month of March.

In short, a huge amount of 2003's computer-related news have turned around The SCO Group and let's bet that it is not finished yet, since SCO will have to face a few things in 2004, including making proof of their claims against IBM as early as in January.


#2 - Online music stores
Apple did not invent the online music store, but at least it managed to make it popular. In fact, the pioneers in this domain were not meeting expectations, facing too strong competition from free services such as KaZaA, Morpheus, Grokster and iMesh, which legality varies from one judgement to the other in court. With its iTunes Music Store, Apple has proven for the first time that it was possible to make interesting profit out of this kind of business, despite the KaZaAs of this world. Just one week after the launch of the service in late April, Apple was already announcing the sale of the millionth song downloaded perfectly legally through its pay-per-song service, while it was only available for Mac OS X! On December 16th, a month after releasing a Windows version of the iTunes Music Store, Apple was hitting the incredible mark of 25 million songs sold online.

Obviously, everybody in the industry is jealous of this outstanding success. Starting in May, many companies announced that they were releasing or about to release their own virtual store "à la iTunes": PureTunes, Microsoft (MSN division), Puretracks, MTV, Wal-Mart, Hewlett-Packard (HPShopping.com), Destra Music and even Coca-Cola. That is without mentionning the influence that iTunes on the other non-free services Pressplay, Listen.com and MusicNow, as well as on Bell Canada, which also deceided to launch an online music service (different from iTunes, though) following the immediate success of the iTunes Music Store.

In all this, we shall not forget to mention the much-anticipated relaunch of Napster by its new owner Roxio, in a non-free service similar to Apple's iTunes. Many think that online music downloads will replace the CD as the #1 format of distribution of music in the world within a few years. Even if services of this kind already existed in 2002, it's 2003 that we will recall as the catalyst of this revolution.


#3 - The 64 bits
The transition into the 64-bit era was already started by Intel with the release of the Itanium (2001) and Itanium 2 (2002) processors, but the commercial success of these server architectures was still remaining relatively marginal. However, AMD has made it much more mainstream in 2003 with the release of its Opteron (for servers) and Athlon 64 (for all) processors in April and September, respectively, making more accessible the biggest revolution in the PC architecture since the 80386, which was announcing the beginning of the 32-bit era that lasted for over 15 years. Let mention that the Athlon 64, cheaper than the other 64-bit processors, is the only one to support the 32-bit as well, becoming (I think we can say that) the first viable 64-bit option if we don't necessarly want to replace all software equipment as well. Various software manufacturers have also released a 64-bit optimized version of their software, in order to take advantage of this technological advancement. Among them, we find SuSE Linux Enterprise Server and Red Hat Enterprise Linux, the first two operating systems to officially support AMD's 64-bit architecture. Microsoft is taking some time to make it happen on his end, but a beta version of Windows XP for 64-bit architectures is already out and its final version should release somewhere in 2004.

Beginning of a new era in the world of computers.


#4 - Linux
Linux continues its progressive rise to the top. Relatively not much affected by the legal actions of The SCO Group against him, the operating system continued to gather new fans in 2003 with, among others, the version 2.6 of its kernel as well as support for AMD's 64-bit architecture by two of its commercial distributions  (from SuSE and Red Hat) before Microsoft or any other one. Many more institutions announced plans of migration to Linux in 2003, as much companies as governmental organizations.

However, several events have changed the face of the Linux market during the year, all being perceptible as positive, negative or neutral. The first case that comes to mind is obviously the announcement of the discontinuation of Red Hat Linux, the popular product that currently holds over 50% of the market of Linux distributions. Red Hat has deceided to concentrate on its product that pays off better, Red Hat Enterprise Linux, which sells for between US$179 and US$18,000 per year per machine. The legendary Red Hat Linux will then stop at version 9, which end-of-life is set for April 30th, 2004, while versions 7.1 to 8.0 are already no longer supported since December 31st in the evening. There is some kind of relief in Fedora, a young project "for new technology enthousiasts" (it's them who are saying that) that comes with absolutely no guarantee of stability, with which the  Red Hat company merged in September. Companies and individuals need, in this end of the year, to evaluate alternatives and turn over to other distributions of Linux. We should then expect a decline of Red Hat's market share in 2004, but it is hard to predict in what proportions. Future will tell us. One thing's for sure, it is a factor that might very well help Linux Mandrake reach to the top of the market of the Linux distributions for the desktop.

Speaking of Mandrake, the French company MandrakeSoft lived a lot of different emotions this year. First of all, it was forced to put itself under the law on bankruptcy protection in January, causing a lot of uncertainty in regards to the company's future. It seemed pretty sure that we would see a Mandrake Linux 9.1 despite the enterprise's cash-flow problems, but not necessarly a version 9.2 or 10.0 . Well, we got a version 9.1 and we got a version 9.2 as well! The most positive thing though remains the news of December 15th, which were announcing that they reduced their losses "by a factor of 7" as well as much better forecasts for the beginning of 2004 than for the beginning of 2003 on the financial front. Mandrake Linux has not given up the race, then.

For the title of the Linux-oriented company that made the news the most in 2003, Red Hat and MandrakeSoft are facing solid competition from Novell though, better known for its NetWare network operating system, which reorientation towards Linux was greatly reaffirmed this year with the acquisition of Ximian (best known for its Ximian Desktop and Evolution software, both for Linux) and the #2 of the Linux-distribution market, the German company SuSE Linux AG. Impressive offense.


#5 - Microsoft
If the Redmond giant has made everyone talk about them over and over again because of security holes, new worms & viruses and other negative points this year, we cannot neglect the positive impact that the new Windows Server 2003 has had on the company. In fact, the new product allowed Microsoft to strengthen its position on the server market, especially at the Web-server level, where the company's share of the market has risen as soon as the new baby was born in April. Office System, also known as Office 2003 or Office 11, was also released in October, but did not have as much success. It is becoming more obvious than ever that no version of Microsoft Office will ever imitate the success that Office 97 had.

For Microsoft, it also was the year of the settlement of many lawsuits against him.They wanted to get rid of these chains as much as possible and made big efforts to take all this to an end. Among the settled case, we find class-action suits and the Netscape/AOL lawsuit. However, RealNetworks has played the troublemakers at the end of the year with a huge lawsuit against the giant, with damages that could easily reach US$1 billion. The gaming company Mythic Entertainment is also among those who filed a new lawsuit against Microsoft rather than settling one in 2003.


Finally, that wouldn't belong in any of the above categories:

In short, that was the year 2003 in computer science. We can now look forward to 2004.


Powhertz
2004/01/03


Back to Powhertz' main page