Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!quiche!peterd From: peterd@cs.mcgill.ca (Peter Deutsch) Newsgroups: comp.sys.next Subject: Re: X11 for the NeXTstation Summary: Long, and a bit of a flame.... Keywords: X Server, X11, XNeXT, Interpersonal Computing Message-ID: <1991Jan10.020437.14465@cs.mcgill.ca> Date: 10 Jan 91 02:04:37 GMT References: <34009@athertn.Atherton.COM> Sender: news@cs.mcgill.ca (Netnews Administrator) Organization: SOCS, McGill University, Montreal, Canada Lines: 119 In article <34009@athertn.Atherton.COM> dlw@Atherton.COM (David Williams) writes: >But, getting back to the real question at hand....What has been changed >in 2.0 that breaks XNeXT? The distribution of mouse and keyboard events seems to be handled differently, so the X code isn't get these events. At least that's what has broken McGill's X11R4 port (known as "mouse-X" to distinguish it from the earlier, X11R3 port known as "XNeXT"). >How hard would it be to fix it, given the availability of source to it? This of course depends on how or why it is broken. :-) Seriously, if NeXT would document the procedure for obtaining mouse and keyboard events from the kernel, the guy who did our port could probably get a working X11R4 for NeXTStep 2.0 out the door in a couple of days, max. They just don't seem able or willing to release this info. God knows, we've asked. >Is it something that would make one conside just starting over with an R4 >implementation? Actually, as I mentioned above, the McGill port of X is an port of the 8 bit MIT server from the X11R4 release. [ For those who known all about mouse-X, skip ahead a few lines ] mouse-X was ported by Mike Parker of the McGill Research Centre for Intelligent Machines (aka as "mouse@mcrcim.mcgill.ca") and he did it by converting the MIT 8 bit colour server into a 2 bit greyscale server. It works under NeXTStep 1.0a (which means it gets its mouse and keyboard events from NeXTStep, but when active it paints a DPS window in front of the NeXTSTep windows and scribbles directly into video. I use it, but rue the day we convert to NeXTStep 2.0, because (sadly), although mouse-X works fine (with a few known, documented bugs) under NeXTStep 1.0a, it fails to receive any mouse or keyboard events under NeXTStep 2.0. We have asked NeXT for help many times, but so far, none has been forthcoming, except for one suggestion to email questions to ask_next@next.com. Mike has shelved the project for now, althought has promised to come back to it later, if he can get infomation on event handling under 2.0. He would like to build a stand-alone version, by getting events directly form the kernel, but NeXT seems less than enamoured with the idea of supporting our efforts. Anyone who could provide the needed info would obviously get the first copy we release..... ;-) . . . >You'll notice that at the hardware level NeXT's floppy drive will accomodate >PC floppies (and eventually Mac formatted floppys one would hope), in the >interest of data sharing...support for the X Window Server would be an >equivalent in the software world. As evidenced by the development of XNeXT, >technically this is not something that would overly tax the R&D of NeXT. Actually, I don't believe NeXT R&D developed XNeXT. I understand it was a student project at MIT with some NeXT background support. McGill has had a working X11R4 2 bit server for months, but can't get anyone at NeXT technical support to even return my calls. I spoke to several people, and Mike Parker apparently spoke to one or two people, but we've have zero support. A NeXT person told me mouse offended some by casting aspersions on NeXTStep, he denies it. To ensure it was not a personality clash, I spoke to a couple of people myself, who promised to call back, but never did. :-( I can only conclude that, if they do want X (and I have a letter from the company promising to "support but not endorse X") they do _NOT_ want a university developed port. Fair enough, but I _do_ wish they'd announce where we can buy it from. I've been waiting over six months since I received that letter and I now have all my 63 new machines. I can't use them for one of their primary applications because we need to have X for the network-based debugger on our BBN parallel processor. Unless BBN plans to release a NeXTStep version of their debugger and no one told me? Somehow, I doubt it. Sorry to seem snarky, but I've been put on the spot over this (I was one of the people here who pushed for NeXT, on the understanding that they would solve the X problem) and I've been waiting for X since July 2. We offered to do it ourselves, we gave Pencomm the first copy of our beta release (and if that is what they want to copy protect, I might even call the McGill lawyers ;-). Now I'm waiting as apparently another firm is working out a deal with NeXT for an X port. Sorry to be crud guys, but as me 'ole dad used to say "It's time to S**T or get off the pot". >NeXT has been rather quiet on the internet of late, one would hope that in >any case they have been following discussions like this. Well, last time I spoke to my sales rep, she seemed to indicate that she'd received word of my various postings, so _somebody_ is seeing this stuff. I only wish they'd act on this one. It really is a no-brainer. You need X, and you need it in a big way. It's costing you guys sales, and it will only get worse. I can't say it any more plainly. Do it, get someone to do it, or help us to do it. And if you don't, I'm calling Sun for the next order.... - peterd ------------------------------------------------------------------------------ " Although botanically speaking a fruit, in 1893 the U.S. Supreme Court unanimously ruled that tomatoes are a vegetable (and thus taxable under the Tariff Act of 1883) because of the way they are usually served. " ref: Smithsonian, August, 1990. ------------------------------------------------------------------------------