Date: Wed, 3 Jul 1996 14:03:03 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: ALHACK@am.pnu.com Cc: questions@FreeBSD.org Subject: Re: FreeBSD vs. Caldera Linux Message-ID: <199607032103.OAA11252@phaeton.artisoft.com> In-Reply-To: <1da8cba0@am.pnu.com> from "ALHACK@am.pnu.com" at Jul 3, 96 11:01:43 am
next in thread | previous in thread | raw e-mail | index | archive | help
> Regarding applications, are you aware of growing ISV support for FreeBSD? Yes and no. FreeBSD has binary compatability with BSDI 1.x a.out, SCO COFF, and Linux a.out binaries. FreeBSD 2.1-stable (the release candidate for the soon to be released 2.1.5) has binary compatability with Linux ELF and BSD 2.x binaries. Support for SVR4 ELF binaries is currently undergoing testing. So the number of binaries which FreeBSD is increasing dramatically. One of the problems with supporting binary compatability modes is that the vendors believe they only have to do one port. Because FreeBSD is compatible with other systems binaries, but other systems are not compatible with FreeBSD's (NetBSD and OpenBSD are, to an extent, and BSDI will run statically linked binaries), the vendor is incented to port to systems other than FreeBSD, and get their "FreeBSD binaries" that way. The problem with this approach is one of supported platforms; generally, validation testing (if a vendor tests their products at all) is only done on the porting platform. Because of this, there is sometimes a lack of support for running products in an emulation environment. If FreeBSD sales of products to run in an emulation environment are sufficient to fund a FreeBSD native port, it is generally a good idea for a vendor to do the port, since it will allow them to pick up people on the "buy" margins for support reasons. Whether or not this is economical depends on the cost to port, which is dictated, really, by how portable the code is in the first place. If your software runs on SunOS, then a port is probably quite trivial, and if you are getting FreeBSD sales at all, will probably pay for itself: most of the people on the FreeBSD lists are computing professionals, with the ability to dictate purchase decisions among "equivalent" products on other commercial UNIX platforms. It's worth the brand loyalty alone, to many of these companies, since UNIX software has traditionally been sold directly rather than through store fronts. Because of this, we have seen slow, but steadily increasing commercial product support of FreeBSD as a platform, especially for ISP and NSP products (for instance, as far as I know, FreeBSD is the only free OS software being used for NSP peering at the MAE-East national network access point in the US). > Are you aware of any plans by anyone for a SunSoft Wabi port to > run Windows apps? Not WABI itself, because of the cost and because SunSoft must protect its Solaris market, since they did not reduce their per unit cost to Microsoft OS levels after the royalty buyout from USL. They are treating the royalty buyout as an amortizable cost instead of a fixed overhead, and so they will not be benefitting from the buyout for 2-3 more years (unless they wise up and kick their accountants into fixing the problem). Because of this, they must take any competitive advantage they can. On the other hand, Sun's is not the only WABI product out there. Willows software, another startup (like Caldera) with funding from the Noorda Family Trust (Ray Noorda), has already been ported to FreeBSD. Willows took a slightly different (and highly successful) approach to porting their commercial product: they asked for help from the people who wrote the OS, and from computer professionals who have done work on the application on the OS. As a result, because the FreeBSD community wants to promote commercial products running natively on the FreeBSD platform, Willows was given a large amount of *volunteer* effort towards their porting problem: they did not have to eat porting costs on an unknown, other than nominal effort for non-disclosure and licensing agreements. I'm quite certain that other developers could get similar porting help, if they requested it. This would drastically reduce the time to return on any porting investment by significantly dropping the up front costs (if you are a developer interested in doing something like this, you should contact jkh@freebsd.org to set up a porting effort). > Or something similar to Cadera's Office Suite? The Caldera Office Suite is largely runnable on FreeBSD as of 2.1.5, since it directly supports Linux ELF binaries. Intermediate -stable and -current (active developement trees) have supported Linux ELF binaries for almost six months now. In addition, there is a German company producing an Office suite for UNIX systems, which is expected to be nearly identical function to the Microsoft Office suite. This should be (like Microsoft's own) significantly better integrated than the Caldera Office Suite. The company has shown a strong interest in porting to the freeware OS's, including Linux and FreeBSD. Hope this answers your questions; if anyone else has any additions or corrections, feel free to jump in. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199607032103.OAA11252>