VOICE Home Page: http://www.os2voice.org

June 2001 [Newsletter Index] [Previous Page] [Next Page] [Features Index] editor@os2voice.org

Quo vadis OS/2

An operating system stands or falls on the software available for it, or a bit more specifically, on the availability of so-called killer applications. According to this definition OS/2 is more than dead. It's simply non-existent, since none of the big software manufacturers offers one of their killer applications also for OS/2, except for Lotus and its SmartSuite. The recent problems with development tools I leave out on purpose; trying to focus on software programs that are important for the end user.

So what keeps today's OS/2 users toeing the line? It's surely not its uniformly, highly praised, enormous system stability. Neither NT/2000 nor Linux, when configured properly, have to fear comparison regarding the overall system stability. Furthermore, even in OS/2, a badly written device driver can undermine the system's foundation. And to people who are used to frequently experiment in their system internals, or programmers, a crashing system is part of day's work. So OS/2 must have some unique qualities, otherwise the total number of clients would be drastically smaller. And I believe I'm not alone with my suspicion that out there a large number of silent users exists, who sporadically turn up in Usenet, or online forums, just to disappear again.

Let's count the reasons:

• OS/2 still has a very stable kernel and superior multitasking
  
• OS/2 can be configured highly individual, if someone needs it and knows how to do
  
• OS/2 today is quite a lean system, compared to the operating systems of the consumer market leader. Only for comparison: Windows XP in its current beta 2 release fills roughly one GB of your harddisk's space! The reason is obvious: even Warp 4 was just a somewhat enhanced version of Warp 3, plus a lot of at that time innovative technologies (OpenDoc, VoiceType), one can simply leave out today (OpenDoc is dead, VoiceType allows discrete speech only). Furthermore, the system has not been significantly enhanced since 1996 - just fixes have been applied. In my opinion a certain market leader did the very same thing, but he knew how to sell his products as new to the masses, just by clever marketing more and more doubtful features.
  

So, are OS/2 users very thrifty people, who very rarely invest in new soft- and hardware (the latter one is even an ecologically positive thing)? Maybe, but that still doesn't answer my initial question: what are people working with in OS/2? The answer is as simple as it is correct: with OS/2 software. Lacking the infamous killer applications doesn't mean that there is no way to achieve your goals. Without the dominant males there's space for other firms, I immediately recall Blueprint Software, R.O.M. Logicware, Sundial Systems, and Matrica. I could list many, many others as well.

And even these little vendors, all of which offer their products for other platforms as well, leave some gaps, into which most of the users sooner or later will fall.

Well, I'm surely not the first one to realize this, and there have been many efforts in the past to fill the worst gaps at least. So I wish to name two of the most popular projects: XFree86 for OS/2, that was started by Holger Veit and is still being maintained by him; and the probably better known OS/2 Netlabs (Project Odin, port of GIMP).

Unfortunately I can't help thinking that many - too many - projects at the Netlabs have reached a certain level of stagnation. Others show only very little progress. One reason for this was indirectly uttered by Adrian Gschwend (Netlabs founder and maintainer) himself, when the question came up about whether or not to do an OpenOffice (formerly StarOffice) port to OS/2: for projects like these the OS/2 community is lacking resources. Well, obviously not only for that.

But all in all, OS/2 users can think positively about the year 2000, because in no prior year there has been so much movement especially as far as free software is concerned: Mozilla, SANE, XWorkplace, Odin, GIMP, and others, surely have made quantum leaps in the past year. PMMail, PMView, Papyrus, even ProNews, and maybe ImpOS/2 (see current poll at OS2.org!) will be continued. Thanks to the generosity of HOB, it's also possible to run X11 applications seamlessly (at a moderate price), Scitech Software shows par excellence what capable programmers can do with OS/2. The past few weeks brought us MPEG and DivX ;-) players. Christopher Hodges (MIDI Station Sequencer) had pity on all of us and started development of a decent WAVE-editor for OS/2. It's somehow amazing that a program like this one is not already available. Additionally, some decent pieces of software still slumber on their developer's PCs: there do exist a DVD player, an Outlook clone, and I believe that eComStation will bring some very positive news (thinking of SmartSuite 1.6, and a re-newed UI). So let's wait for the GA - according to Bob St. John we'll all know more by the second half of June. And to all vacillators: the upgrade offer is still valid!

The only real sad news came from Stardock Systems: support and product service have ended with IBM's Warp 4 support. In fact it ended long before. While IBM still offers product support in forms of the Convenience Packs, such famous names like Object Desktop, Process Commander, PMINews, etc. will soon be dead products lying somewhere in Brad's office. After the mud-slinging he had to endure, among others at OS2.org, I doubt he will ever release even a single line of his programs' sources. Moreover, I can even understand it. To my astonishment Stardock increasingly gives business reasons for not opening the sources. Those who have the time, may rummage through the support newsgroup at Stardock, which will remain publicly accessible, to get an idea of what was going on...

What is wasted on ammunition on the one hand, is saved on the other - wrong - hand. Frequently people ask for new software and new projects - only to pull faces when it becomes available: the software doesn't look like one had wished, it doesn't do (yet) all of the fancy things one had hoped, and is difficult to install and awkward to use. And once you are disappointed, you give vent to your feelings, and leave the software, which often bears great potential, on its own.

Yet, the only thing most projects need is simple feedback. If you download free software from the internet, figure out that it doesn't work for you, then please take time to write a short note to the developer(s). Especially for hardware-sensitive software, or bigger projects, any bug report is highly appreciated - the same with praise, of course. Taking a look at some mailing lists, as for example SANE, Win-OS/2 Sound Driver, or even Odin, lack of feedback seems to be a common problem. Paul Floyd, who did some very good ports of SANE to OS/2 (up to 1.0.4) has announced that he would stop porting it unless he gets some feedback, since the software works for him. However, he's not entirely out of business - he'll have a look at the coming 1.0.5 as well...

At this point, I also have a suggestion to the developers: not every potential user of your software is capable of thinking in the same arbitrary ways as you do, and quickly gets frustrated when he or she gets some binaries with poor, or even no, documentation. What did I hear Adrian Gschwend saying at Warpstock Europe? "If you'd like to help OS/2 Netlabs, use WarpIN". I can just say: if you want to make your software attractive for the end user, use WarpIN! I find it quite astonishing that dozens of helpful people adopt software packages for eComStation, while there's only a very few developers wrapping their programs into WarpIN, which, as an Open Source project and by its conception, should be first choice over Wise Client/Wise Manager.

We have talked several times about the problem of developers not announcing their pieces of work, but let me mention a thought provoking example. A Japanese group of programmers has written a whole bunch of applications to significantly enhance the multimedia capacities of OS/2. Alas - no one knew about it, until recently. When such a large project can't make it over here, how many smaller ones still flourish in secrecy or are already wilted? What kind of potential is there in the former Soviet Union, where OS/2 has an excellent reputation and where the community is a very active - unfortunately they mostly speak Russian only.

What makes it clear for me: the global OS/2 community lacks communication. There seems to be no vision about a common goal and how to reach it. When developers in three or more regions of the world re-invent the wheel (whether they have to or want to I don't want to discuss) then that's a waste of valuable resources. I can not, nor is it my intention to tell someone what program to write, how to write it and why, but I have come to the conclusion that OS/2 users should all pull the same rope.

In order to achieve this, there is not that much to do. At least two established international forums do exist entirely in English: OS2World.com and OS2.org - maybe eComStation.com will become a good one in the future. With the OS/2 Netlabs we have an online resource with great potential, offering some important basics to developers who are interested in working in a coordinated fashion on an open source project. Moreover, OS/2 Netlabs has a good reputation, even at IBM.

However, this will only work if everyone is helping. This is the core idea behind it. It is actually very simple: publicize news about, and upload programs and sources to Hobbes, which is the only remaining, well maintained OS/2 software archive on the internet.

So cooperation and engagement are the issue. But is this really supposed to work on the lowest level, whilst it doesn't on a higher level? Most people might already have guessed that there won't be a Warpstock Europe this year, but no one knows the real reasons for it. It has absolutely nothing to do with financial problems, or a lack of professionalism, as you might have read on some online forums. Rather it fails because of a lack of manpower. Last year we gave a good performance and yet it seems as if within the German TeamOS/2 not even a single person is feeling brave enough to commit to the project this year. Please don't get this wrong - last year's planners would have been with it again, but they demanded more support... and didn't get it.

Warp Doctor offers a similar example: set up with lots of energy by Dan Casey in order to replace the very outdated Warp Pharmacy, news and maintenance have become a rare thing. That's a real pity in so far as this site was set up for the community - but it can only survive if the community actively participates in it.

Not to paint everything black, I saved the good news for the end of this article. VOICE and the German TeamOS/2 Trier have agreed to share articles in the future. So each will get new articles: VOICE for its newsletter, and TeamOS/2 Trier for its website. Not to mention that the articles will all be translated into either English or German.

The other very positive news is about three editions old, but I'd like to mention it nevertheless: OS/2 eZine finally found a new editor and was able to publish its issues living up to its high standards of quality without any problems.

Also Walter Metcalf, editor of the recently closed down OS/2 part of About.com, finally found a new place to publish and will write from now on for POSSI's Extended Attributes and the VOICE Newsletter.

Let's all hope that the positive news won't be less in the future - everyone can help with it!