Date: Mon, 9 Mar 1998 11:38:01 -0500 (EST) From: Christopher Sedore <cmsedore@mailbox.syr.edu> To: freebsd-hackers@FreeBSD.ORG Subject: Re: WINE (was: Uncle Sam, got a million bucks?) Message-ID: <Pine.SOL.3.95.980309111250.24074D-100000@rodan.syr.edu> In-Reply-To: <350403F3.7AC7@opengroup.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 9 Mar 1998, Kaleb S. KEITHLEY wrote: > Christopher Sedore wrote: > > > > I'd like to see an lightweight X11 replacement (eg photon), > > Where is there information on photon? www.qnx.com > Lightweight in terms of what? Bandwidth? Memory footprint of the > application? Something else? Simply stated: yes. I want better bandwidth utilization and lower memory footprint, and simpler app development. > What strategy does photon offer for preserving the investment that > various enterprises have made in their existing X11 applications? Or are > you going to pull a Java on me and tell me that the apps have to be > rewritten? Well, if we had apps to be rewritten, I'd be more concerned :). There hasn't exactly been a stampede to the X11 "Office" type apps. This is the problem--the existing investment in X apps is (in comparison with the desktop market size) negligable. There are probably already more business machines running Windows NT workstation for desktop applications than there are running X in that arena. If you want the desktop market, forget about the existing investment. Its inconsequential in the context of the market you're after. Start by designing what you need to capture a non-trivial portion (ie one larger than the current X userbase) and go from there. Key to the strategy will be coming up with an architecture coupled with development tools that make it much less painless to develop apps. (Now if you want to argue that X can do all these things with some work, maybe it bears investigation, but so far I haven't heard anyone claim that.) > > and apps to > > run with it. A FreeBSD that ran (with GUI and some apps) on an 8MB 486-66 > > would be a killer. > > Five years ago I was running FreeBSD on a 486-25 with 8MB of RAM -- and > I was using X too. Three years before that I was running 386/BSD 0.1 on > a 386SX-16 with 4MB, and Thomas Roell's X server -- it was no speed > demon, but it did work. Me too! I still have my 486-25SX in a corner somewhere. Never did mess with 386/BSD stuff, though. > I'd still be using a 486DX2/66 with 8MB if Frye's hadn't had a special > last year on a Pentium motherboard w/ 100Mhz CPU for $150. I run > Netscape 3.0x, xfig, a dozen xterms, and more. Yes, the Pentium makes it > quite a bit faster, but it was a very servicable machine before I > switched to the Pentium. > > All it takes to have that killer system you want already exists -- and > has existed for quite some time now. :-} > > I suspect that there's some other set of criteria, like performance, > that you aren't mentioning. Performance is important, and X11 is a bit heavy. Properly done, this scheme would have wide application across computing applications. Microsoft had to do WinCE for palmtops, Windows 98 for home, and Windows NT for business. Why? Because their system is not component based and was not designed to scale across small to large environments. FreeBSD has the ability to cover all these areas, and isn't that far from it on the kernel end. Again IMHO, FreeBSD needs to have dedicated (full-time) effort to making it a viable desktop platform. This is the same place that Linux suffers and where BSDI has missed the boat. BSDI went after a market that FreeBSD effectively serves better. On the freeware front, there are lots of folks out there dedicated to enhancing what already exists in the direction of what we need, but I don't know of anyone starting with a careful analysis of the whole situation and then building from there. I know it takes time and money to do this, but in today's US economy money is around if you're willing to go through the work to get it, and I think you could find developers who would be interested in working on such a project (particularly if your funding was adequate). The other side of this is adopting some of the functionality without adopting the cruft that goes along with it. We need apps that meet our needs--the bells and whistles are a stretch. If Microsoft apps were dinner utensils, you'd have 40 different forks, knives and spoons for each course of your meal--and some of them would be there depending on your mood! All IMHO, -Chris To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.SOL.3.95.980309111250.24074D-100000>