Date: Sat, 9 Oct 1999 01:26:24 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: tcrimi+@andrew.cmu.edu (Thomas Valentino Crimi) Cc: brett@lariat.org, chat@FreeBSD.ORG Subject: Re: Targeting the server: Not such a good idea? Message-ID: <199910090126.SAA02965@usr09.primenet.com> In-Reply-To: <4rzaRDa00Uw_1CoKk0@andrew.cmu.edu> from "Thomas Valentino Crimi" at Oct 8, 99 05:51:11 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> The question in my mind, though, is: > > "What can one add to the kernel to make FreeBSD a better desktop > environment?" > > And the answer, is... nothing (almost). GGI would be a good start. So would either a CORBA enabled VNC viewer, or some other OMG style component, that could use the GGI to do its diplays. > The OS's job is to get mouse and keyboard clicks to the appropriate > programs, manage their resources, and otherwise let the number crunchers > keep crunching, and the netscapes keeps scaping. Not a single new > kernel feature or utility is really going to change the fact that 99% of > the Desktop "experience" is up to the Gnome/KDE guys. I think the OS's job is actually to abstract the hardware from the software, so that the software doesn't have to worry about where it is running. You can achieve this in two ways: (1) run the software on a server where it is a native binary, or (2) run it in a virtual machine, either dynamically bound (e.g. Java) or installation bound (e.g. ANDF). > In terms of code, FreeBSD is fairly well latched onto Linux's train of > development as far as X and window managers go.. if anything, FreeBSD > can at least state that they are commited to keeping KDE and Gnome > well-ported and up to date, and have stated support for the programs > (which I feel already exists) which Desktop users may wish to use. > Marketing and visibility, Brett, are your cup of tea, and maybe you > should go about and state the following things: X is really a dead end. There should be no difference between the desktop display software, and any other software which exports a canvas into which another application can be embedded. The historical and artificial division of X as a frame buffer acccess protocol has been as damaging as anything else to UNIX, and is also a serious limitation on Windows, even in the face of the Citrix code. Even the Citrix WinFrame stuff, which allows multiple sessions under NT, has a serious limitation in that each session -- like an X server -- consumes inordinate (and unjustifiable, for the function) resources that could be better applied elsewhere. Where X is an "outie", the Citrix WinFrame code is an "innie". > The job of the FreeBSD group is to make the best underlying OS > possible. There aren't nearly enough commiters to make the > WordProcessors, spreadsheets, cute window managers and graphical mp3 > players the world wants. It only requires committers when the code goes into FreeBSD's source tree. I believe that the division between the kernel and the rest of the OS allows Linux to take advantage of aggregation far better than FreeBSD, at the obvious cost in incompatabilities between distributions. Yet such incompatabilities are in user space, and could, with effort, be ameliorated. > The REAL problem, frankly, is in Xserverland.. FreeBSD can never > install and boot you into X because it still takes a somewhat > knowledgable human to configure. This seems to be changing as of > release 4. GGI would resolve that, in allowing FreeBSD video drivers (written under NDA and distributed in binary format) to come on vendors adaptor driver disks. Such drivers would replace the default VGA (or MDA, etc.) driver only if installed, so minimum functionality could be obtained. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199910090126.SAA02965>