From: Tim Bird Message-Id: <199703121623.JAA12667@caldera.com> Subject: [opendos] Re: your mail To: gtc AT ozemail DOT com DOT au (Graeme Cruise) Date: Wed, 12 Mar 1997 09:23:27 -0700 (MST) Cc: opendos AT mail DOT tacoma DOT net In-Reply-To: <2.2.16.19970312224451.0a6feea8@ozemail.com.au> from "Graeme Cruise" at Mar 12, 97 10:42:52 pm Content-Type: text Sender: owner-opendos AT mail DOT tacoma DOT net Precedence: bulk I'll try to answer this one, since I have been involved in some of the discussions at Caldera regarding these very issues. I am not an official spokesperson for Caldera, Inc., nor do I play one on TV. > > Dear DOSbodies, > > I invite your considered responses, please. My apologies in advance if I > have chosen the wrong forum. > > I have only just joined this list, but I have read through the available > archives on Caldera's OD pages, and as many other links as I can find, > trying to get a feel for what I can expect of OpenDOS in the foreseeable future. > > I was simultaneously heartened and intrigued to learn via another > (unrelated) mailing list that Caldera was going to not only revive the lost > DR/Novell DOS, but put it into the public domain too (and sue MicroBloat as > well for good measure!). > > My own interest in things DOS revolves principally around the need for a > simple, reliable, supported multi-tasking DOS for business apps which do > *NOT* need the gigantic overhead of Windoze, both in terms of machine > resources and perpetual software maintenance. God, I'll even settle for a > measly *two* swappable tasks -- the enhanced batch language and other > goodies are simply icing on the cake to me. > > And I want to actually *pay* for it, as I need it to be supported by a > recognizable organization, and it will be used in everyday commercial > operations anyway, so it needs the credibility of a recognizable "brand" > behind it to be able to sell it to clients and their managements. > > So, to my main concern: release of the source code into the public domain. > Forgive my apparent ignorance, but will that not invite a zillion variations > to flower? Will it not put DOS in a sort of "feral" category way beyond the > problems that beset Unix because of its (comparatively few) recognized dialects? Caldera hopes that DOS will develop roughly the same way that Linux has. If you are not familiar with how Linux advances, it is something akin to controlled chaos. Everyone pretty much does whatever they want, but Linus Torvalds (or a duely appointed representative) approves and coordinates all changes that go into the kernel. Although there are several Linux "distributions" from different vendors, the kernel is the same (barring version differences) for all of them. Where the distributions differ is in the selection of tools and applications to include, how things are installed, etc. Caldera has carefully worded the binary license this first time around to restrict redistribution of OpenDOS. Right now, OpenDOS is only available from Caldera. We are still working on the source license, which will have to be looser than this (in my opinion) if it is to encourage ongoing development by the community at large. The rationale for a tight license (tight by Linux, or GPL standards - and incredibly loose by normal software business standards) at first was that we want to open the gate slowly to measure the effect. We would like to find the right balance between offering the software freely and being able to control it. Once the gate is opened a certain degree, it is really quite difficult to try to close it more. There is every possibility that we will use less restrictive licenses in the future, but we'll have to wait and see (I have argued in internal meetings for using the GNU Public License for some components). Feedback on this issue is greatly appreciated. I will try to forward feedback I see on this mailing list to Caldera, but it is more likely to be seen if you send it to Caldera's mailing list at: caldera-opendos AT caldera DOT com You should subscribe to that list also if you want to follow OpenDOS developments directly from Caldera. Although I must admit this list has recently had better technical content. The two main issues for restricting distribution this first release were: 1) We wanted to be able to gauge the market demand. We have our form in place on the web site which helps us gather information about who is interested in OpenDOS. If we had thrown OpenDOS to the wind (so to speak), we would not be able to make early market assessments that we feel are important. (Linux suffers from the problem that no one really knows who and how many are using it) 2) At this stage, we were unsure whether we want to do a CDROM release ourselves. We would like to reserve that right for ourselves if we decide to do it, so for now we have made it impossible for the CDROM vendors to come in and make their own products from OpenDOS. If we allowed any redistribution (even only for free), it would be difficult to prevent the CDROM vendors from skirting our intentions and making their own products. > > Or, is it really the plan to let it go feral on the basis that once "it's > out there" it will survive by constant mutation and natural selection? We _would_ like to see mutation and natural selection. But we'd like to control it for our customers and our own products. In fact we have to so we can support it. > > OK, fine, but what are serious commercial app developers to do who really > want to just stick with a good, solid, supported multi-tasking > non-MicroBloat DOS? Even if we eventually allow others to make DOS distributions based on OpenDOS, Caldera will always have its own product, which we will advance and support for our commercial customers. Sorry for the long-winded response. I hope this helps people understand what we're trying to do with OpenDOS. Tim Bird