Date: Thu, 15 Sep 2005 22:12:56 -0700 From: "Ted Mittelstaedt" <tedm@toybox.placo.com> To: <jahnke@fmjassoc.com> Cc: youshi10@u.washington.edu, freebsd-questions@freebsd.org Subject: RE: IE in FreeBSD? Message-ID: <LOBBIFDAGNMAMLGJJCKNEEGBFCAA.tedm@toybox.placo.com> In-Reply-To: <1126794190.9885.45.camel@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
>-----Original Message----- >From: Frank Jahnke [mailto:jahnke@fmjassoc.com] >Sent: Thursday, September 15, 2005 7:23 AM >To: tedm@toybox.placo.com >Cc: youshi10@u.washington.edu; freebsd-questions@freebsd.org >Subject: RE: IE in FreeBSD? > > >> My opinion on WINE is that it merely harms people who are writing >> software >> for FreeBSD. If I write a wordprocessor for Linux or FreeBSD and try >to >> sell it, why would a customer buy it when he can just use his >Microsoft >> Word under Wine? > >> As a result the existence of these programs discourages interest in >> native >> FreeBSD programs, and encourages people not to wholeheartedly switch >> over to FreeBSD. It also gives an excuse to software developers not >to >> bother >> writing software for open source development since "they can always >run >> it on wine" > >> Ted > >I often hear this said, but I don't think it is true. I saw this kill OS/2. I ran OS/2 exclusively as a desktop OS for a number of years, it had excellent networking integration with UNIX, better than Windows. But IBM spent way too much effort in keeping Windows emulation going in the OS and as a result didn't put the development effort where it would have helped - primariarly strengthing the OS on different hardware. This is what put companies like DeScribe out of business. >As far as I can >tell, there is essentially no commercial software written for FreeBSD >(and very little for Linux) as it stands, and while the FOSS software >has improved a great deal, much of that targeted for the desktop is >either not good enough or simply does not exist at all. > Not desktop but there's a lot of commercial back-end software that uses FreeBSD. >Wine will always be a compromise: some (but with hope, an increasing >number of) important programs will work very well, some will perform >with limited functionality which may be OK for a few selected tasks, and >many or most will not work well enough if they work at all. They will >also continue to be difficult to integrate with other desktop programs, >even more so than Linux programs which are bad enough already. They >simply are not a replacement for native programs unless no alternative >exists. > An alternative always exists. >Your early proposed solution of running a remote desktop to run the >"real" windows program also does not encourage writers to introduce a >FreeBSD program version. That was a joke. Anyway, in a way it does because it forces the user to go through a lot more trouble than an emulator, and the only way to get users to invest the time to learn how to use FreeBSD is to make the alternative more difficult. Look at Macintosh software sometime, the UI for most apps is little different than what it was under System 7 except more colorful and glitzy. Most Mac users don't even know UNIX is involved with their OS. The Mac isn't a gateway to UNIX by any means. Apple made it easy for Mac users to continue to be stone stupid, and the Mac users by and large chose to stay stone stupid. Apple knows it's customer base that's for sure. >Instead of saying "run it on Wine," one could >always say "run it on a remote desktop." Old computers that may well be >good enough for such occasional use are very inexpensive. Why then >would anyone run a native version? > >I think that the best way to increase the number of native programs >written for or ported to FreeBSD is to increase its market share, >particularly on the desktop. The rapid acceptance of desktop-oriented >versions of FreeBSD, such as PC-BSD and DesktopBSD, I find very >heartening. But as long as the OS has such a small market share, we >will have to rely on such "non-optimal" solutions such as qemu, Wine, >CrossOver Office and the like. Sadly, I think this will be the case for >the near-term future of a few years at least. It will likely be longer. > Simply increasing the market share numbers won't do jack. Look at MacOS, Apple has far less of a market share than FreeBSD yet has tons of software for it and more every day. You must increase the market share among the people that pay money for software in order to interest ISV's in porting. This is one of the famous catch-22 of FreeBSD. Skilled and smart techs can make free applications that run under free OS's like FreeBSD work for them, or fix them if they don't work. Garden variety users don't want to learn much and are willing to pay money to not have to do so. If you dumb-down the OS like Windows and MacOS is, you attract the garden variety users and you get a lot of money which atttracts all the ISV's who want to port to you, but the skilled users get sick of the shit and they are out of there. For commercial OS's that's not a problem they just pay people to continue building them, but it will break the back of an Open Source volunteer effort. RedHat understood this and that's why most RedHat Linux users today are pretty basic, and the skilled Linux people have fled to Suse and Debian, and even some to Fedora, while the RedHat owners are smiling all the way to the bank, and you have ISV's like Oracle who are porting to it. >In the short term, I have work to do that requires windows programs, or >at least the function of certain windows programs. Not IE, as the >original poster of this thread, but others that are common in the >Windows world. I'd like to use a single computer and its tools for this >purpose -- the workflow is so much more convenient. As it stands, I >cannot turn "wholeheartedly" to FreeBSD until I can perform the sort of >tasks I need to -- I will always need a Windows box for too many things >otherwise. And I certainly can't subject my employees to this >situation, unless they are "Unix heads" like me. > The problem is your thinking like Microsoft wants you to think, that your workflow must be desktop-driven with desktop programs. I use Windows every day and will probably continue to use it for years - but I don't use it for much more than what you would use a Wyse Winterm for. In short, most of my real work is done on the server and the Windows box serves to run xterms and terminal emulators and such. The last real app that I have left on Windows is Outlook and that's only because I have about a decade of mail stored in it. But that isn't going to last much longer because the IMP/Horde framework has finally matured to the point that it can replace it. I've been waiting for that to happen for years. >That's why I started the petition to CodeWeavers to port CrossOver >Office to BSD. That product may not be the "perfect" solution, but it >would sure help me a lot with most of the needs I have now. > >That petition is located at http://www.bsdnexus.com/petition.asp >and to date we have nearly 900 signatories. If you have not signed, I >would encourage you to do so. > I think it's a better thing to continue the work on OpenOffice under FreeBSD, frankly. I would rather run an xterm on my Windows box and run OpenOffice on the FreeBSD server. The era of the desktop being the main computational engine is really drawing to a close. It is simply too expensive to continue dumping a thousand bucks or so into a new desktop every 2 years and probably an equivalent amount into new desktop software. Do you realize that if you had a full blown set of Microsoft software, and you did a forklift upgrade every 2 years with a nice Dell business system that you would be spending about as much on your desktop computer as you do if you buy a new car every 12-13 years? That's why piracy is rampant and Microsoft is going to have the casual home user pirates shut down in another 5 years. We are going to have to move away from this new-machine-every-2-years model that the hardware industry is trying it's best to shove down our throats and start treating the personal computer like an automobile. Desktops need to last at least a decade or two. And that will only happen if the desktop becomes totally insignificant and simply a means to get to the important stuff on the server. Ted
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?LOBBIFDAGNMAMLGJJCKNEEGBFCAA.tedm>