From ns.itep.ru!Radio-MSU.net!news.phys.msu.su!news.rssi.ru!demos!news1.relcom.ru!EU.net!news-res.gsl.net!news.gsl.net!usenet.eel.ufl.edu!spool.mu.edu!munnari.OZ.AU!comp.vuw.ac.nz!canterbury.ac.nz!newton!physmsa Tue Jul 30 16:38:39 1996 Path: ns.itep.ru!Radio-MSU.net!news.phys.msu.su!news.rssi.ru!demos!news1.relcom.ru!EU.net!news-res.gsl.net!news.gsl.net!usenet.eel.ufl.edu!spool.mu.edu!munnari.OZ.AU!comp.vuw.ac.nz!canterbury.ac.nz!newton!physmsa From: physmsa@phys.canterbury.ac.nz (Mr M S Aitchison) Newsgroups: comp.os.os2.beta,comp.os.os2.advocacy Subject: Re: A better beta (Long..) Date: 30 Jul 1996 03:25:36 GMT Organization: Physics and Astronomy Department, University of Canterbury. Lines: 107 Distribution: inet Message-ID: <4tjvbg$5dk@cantuc.canterbury.ac.nz> References: <4s6415$4rn@panix2.panix.com> <4sb4re$7r1@rome.ntr.net> <4sb766$4k9@cnn.Princeton.EDU> <1996Jul19.120746@cantva> <4eG8xowaVriH091yn@os2bbs.com> <4t733v$5hv@cantuc.canterbury.ac.nz> NNTP-Posting-Host: newton.phys.canterbury.ac.nz Xref: ns.itep.ru comp.os.os2.beta:40698 comp.os.os2.advocacy:244433 >>But *lots* of others take the view "what is the person next to me >>using?" or "what does all the shelves at the local computer shop >>contain?". > > Sure, this is a big concern for a big slice of the market. But I don't >believe IBM has thus far tried to serve that part of the market with OS/2. My whole contention all this time has been that, given the fact that IBM aren't trying to ensure OS/2 customers will be able to run the "mainstream" software wanted by the unwashed masses, support by 3rd parties is going to suffer more and more, as OS/2 is seen to be a "dead" operating system. My operating system more useless to me without the ability to run mainstream software as it would be without a driver for my favourite video cards. We have gone through arguments about the size of IBM and the slim chance they will drop OS/2, but I see corporate software decisions based on long-term prospects for products that include more than "will the supplier still sell it?" We have discussed various ways of determining whether OS/2 is software-rich or software-impoverished. More controversial is the question: "Is what IBM is doing now, with the direction of new versions and with OS/2 marketting, going to ensure it captures mindshare in the future?" By far the greatest reaction to my original comments has been: "I don't care if OS/2 runs Win32 apps - in fact I hope it never does - since I prefer the apps that I want now, thanks." I maintain that letting the 3rd party applications support slide downhill is a tailspin that can destroy OS/2 no matter how fervntly you cling to the idea that IBM will stil be there to sell it next century. We keep coming back to the question: "What is OS/2 meant to be?" or "What is the reason for its existance?". Most people don't buy OS/2 to do IBM a favour. Most developers don't produce OS/2 versions of programs out of the kindness of their hearts. What is in it for them? Well, if OS/2 is trying to be another DEC VMS for instance, or Amiga DOS or Data General AOS, then there will certainly be a number of people who like it just because it is the best, and there will be some suppliers that write for it because nobody else is covering that market. The amount of development that can be justified decreases with the size of that market. (There is another, odd, factor that means that many developers judge the market to be smaller than it really is. I have my theories as to why they are acting so stupidly, but the simplest answer is the rule of thumb that you need to be an order of magnitude better than the opposition to crack a monopoly mindset). If a company has ten times the market share for its software product than its nearest competitor then it has a great advantage - it can distribute development costs over more sales, and therefore have bigger development teams and bigger profits, as well as the psychological benefits of customers having to be brave to go to the competition. But if it has 100% of the market (or even 99%) then it has even more power. No news to anybody there? In Monopoly parlance, MS is about to build hotels; in Risk terms Bill is not just going for a continent but the whole world. These are *not* unreasonable fears. (anybody disagree? e-mail me). I know the problems of running software on a proprietry operating system which, although not cancelled by the supplier, is a constant pain to find modern decent software for - stuff that is plentiful in Unix or DOS or Windows. OS/2 was very appealing at the time it ran DOS and Windows apps as well as the latest MS system would. Being able to run CP/M or TRSDOS programs at that time instead simply wouldn't have been as good. Running Win 3 now instead of last year's version is just as bad. My knowledge of what was going through the minds of IBM when they beefed up OS/2 support of DOS and Windows enhanced mode isn't all that extensive. Possibly IBM didn't realise this would be a such a good strategic idea to give OS/2 leverage. Now IBM have to decide where OS/2 goes from here. There are 4 possibilities, as I see them: (1) Say that OS/2 applications choice is good enough (or somebody else's problem) and simply try to make OS/2 a better operating system. (2) Compete with MS by either producing apps that are as good and cheap as the majority of software for W95/NT or encouraging/forcing/whatever others to do so. (3) Say OS/2 isn't trying to compete on the desktop - it is really a (say) file or Web server, that MS is what you want if you want a desktop O/S, but you will want to run an OS/2 server for your MS apps on MS operating systems, won't you? (4) put the time and money into AIX instead. My crystal ball tells be loud and strong that anything other than (2) will be a downward, unstoppable spiral for OS/2, and MS will be put in an incredibly powerful monopoly position. I have argued that OS/2 needs to woo back support (by incorporating Win32 support; I honestly cannot see it happening otherwise) then do what they should have done before Win95 came out, which is to ensure a flood of good software comes out that is better than anybody writing for W95 could produce (including MS themselves) so that developers see they can beat the MS monopoly in office products, etc by way of OS/2 apps. In this way MS would have to play catch-up with OS/2 instead of the other way around. Thanks to all those who have read so far, or even posted. I had hoped we could get onto the details of what OS/2 should have in it but we kinda got sidetracked by arguements like "Of course OS/2 has better apps than MS right now! What are you worrying about?" and the perrenial problem of "if IBM puts support for Win32 in then nobody will produce native OS/2 apps". I hope it is obvious that current thinking of IBM encoraging apps that run on OS/2 and Unix and Macs and Win95 will not give OS/2 any lead - it must have more apps that cannot run under Win95 because that O/S isn't good enough. ------------------------------------------------------------------------------- Mark Aitchison, Physics & Astronomy \_ Phone : +64 3 3642-947 a.h. 3371-225 University of Canterbury, (/' "Computologist to the stars." -------------------------------------------------------------------------------