chouw@galaxy.cps.msu.edu (Wen Hwa Chou) (03/06/91)
Have anyone try to run the XNeXT on NextStep 2.0? I have some problem in starting it up. (Are the two versions of NeXT OS really binary compatible? I have several programs having problems running under the new systems. All of them requires services from the window system.) Wen
charlie@wam.umd.edu (Charles William Fletcher) (03/07/91)
In article <1991Mar5.232508.20719@msuinfo.cl.msu.edu> chouw@galaxy.cps.msu.edu (Wen Hwa Chou) writes: >Have anyone try to run the XNeXT on NextStep 2.0? I have some problem in >starting it up. > >(Are the two versions of NeXT OS really binary compatible? I have several ^^^^^^^^^^^^^^^^^ I also question binary compatiblity-I have had several problems. Is there really suppose to be binary compatibility between the old and new chips/systems, or have I misunderstood this. Thanks, Charlie > programs having problems running under the new systems. All of them requires > services from the window system.) > >Wen
howie@cunixf.cc.columbia.edu (Howie Kaye) (03/08/91)
I had some problems using 1.0 X binaries on the next. Anything run under xdm (on an X terminal served off of a next) did not use the resolver or netinfo correctly. Something to do with the way shared libraries work no doubt. I rebuilt the whole X tree, and everything works correctly now. ------------------------------------------------------------ Howie Kaye howie@columbia.edu Columbia University hlkcu@cuvma.bitnet UNIX Systems Group ...!rutgers!columbia!howie
shanega@athena.mit.edu (Shane G. Artis) (03/08/91)
I heard that somebody had fixed mouse-X (from McGill) to work under 2.0, and that the diffs were posted somewhere. Could somebody build the thing and post the server binaries? I have a modem and there is no way I can modem to an internet site, download 30+Meg of code and build it on my machine, so having server and clients all built and at an FTP site makes everything a billion times nicer. Can someone do this for those of us without direct internet access? Shane -- ------------------------------------------------------- Why waste bandwidth on signatures?
scott@erick.gac.edu (Scott Hess) (03/08/91)
In article <1991Mar7.140806.13975@wam.umd.edu> charlie@wam.umd.edu (Charles William Fletcher) writes:
I also question binary compatiblity-I have had several
problems. Is there really suppose to be binary compatibility
between the old and new chips/systems, or have I misunderstood
this.
Well, binary compatibility only means so much. What NeXT really means
(and I think it's in there, small print, you know) is that programs
which follow the API should be binarily compatible. XNeXT, for one,
did not follow the API. For speed (or conveinience?) MIT chose to
go under it and use more raw commands to interface with the windowserver.
Thus, XNeXT broke under 2.0.
mouse-X has similar problems, I believe, in that the author, while
a good programmer, is not a good _NextStep_ programmer (from our
discussions, he's an X-person, not so much a NeXT-person). If you
look at the sources, you realize rather quickly that these were
not generated by someone out of developer's camp! I guess I was
sort of surprised that they ran under 1.0 :-).
There are some other apps with problems, too. I'm not sure what I
can say about them. What I can say is that almost all of the apps
that I've seen which had binary compatibility problems did not
conform to the NeXT UI guidelines, and thus I suspect did not
fully conform to the API.
Later,
--
scott hess scott@gac.edu
Independent NeXT Developer GAC Undergrad
<I still speak for nobody>
"Tried anarchy, once. Found it had too many constraints . . ."
"I smoke the nose Lucifer . . . Banana, banana."howie@cunixf.cc.columbia.edu (Howie Kaye) (03/10/91)
In article <1991Mar7.195211.29685@athena.mit.edu> shanega@athena.mit.edu (Shane G. Artis) writes: >I heard that somebody had fixed mouse-X (from McGill) to work under 2.0, and >that the diffs were posted somewhere. Could somebody build the thing >and post the server binaries? I have a modem and there is no way I >can modem to an internet site, download 30+Meg of code and build it on my >machine, so having server and clients all built and at an FTP site makes >everything a billion times nicer. Can someone do this for those of us >without direct internet access? > >Shane > Ask and ye shall recieve. I fixed mouse-X a while ago and sent the diffs back to McGill, but since they don't seem to be available for anonymous ftp yet there, i've put them up. They are on cunixf.cc.columbia.edu (128.59.40.130) as X11R4-Next-2.0.tar.Z I don't know if they are completely in synch with McGill yet, and have had a couple of minor reports about it (like the shifted top row of the new keyboard's keypad aren't supported). Please report other problems as you find them. ------------------------------------------------------------ Howie Kaye howie@columbia.edu Columbia University hlkcu@cuvma.bitnet UNIX Systems Group ...!rutgers!columbia!howie
mouse@thunder.mcrcim.mcgill.edu (der Mouse) (03/10/91)
In article <SCOTT.91Mar7204438@erick.gac.edu>, scott@erick.gac.edu (Scott Hess) writes: > In article <1991Mar7.140806.13975@wam.umd.edu> charlie@wam.umd.edu (Charles William Fletcher) writes: >> I also question binary compatiblity As do I. I had a perfectly good program that didn't even *touch* the NeXTisms and it broke subtly on 2.0. I'm still not sure just what broke. (To make things worse, it wouldn't recompile on 2.0; I had to hack on it. Apparently NeXT decided they wanted to get even farther from UNIX compatibility than they were before.) > mouse-X has similar problems, I believe, in that the author, while a > good programmer, (Thank you!) > is not a good _NextStep_ programmer Not a NextStep programmer at all, if I can help it. Why do you think I ported X? (For the rhetoric-impaired: because I can't stand NextStep.) > If you look at the sources, you realize rather quickly that these > were not generated by someone out of developer's camp! I guess I was > sort of surprised that they ran under 1.0 :-). :-) Well, I really had little choice. The raison d'etre of the thing violates the GUI spec - or so I assume; it seems unlikely that the spec permits programs to take over the whole screen - and NeXT didn't document their lower-level interfaces. So I had to dig through .h files and what documentation there was, until I managed to make it work. Of *course* it looks kludged up; it is! der Mouse old: mcgill-vision!mouse new: mouse@larry.mcrcim.mcgill.edu
mouse@thunder.mcrcim.mcgill.edu (der Mouse) (03/10/91)
In article <1991Mar10.000354.15364@cunixf.cc.columbia.edu>, howie@cunixf.cc.columbia.edu (Howie Kaye) writes: > In article <1991Mar7.195211.29685@athena.mit.edu> shanega@athena.mit.edu (Shane G. Artis) writes: >> I heard that somebody had fixed mouse-X (from McGill) to work under >> 2.0, > Ask and ye shall recieve. I fixed mouse-X a while ago and sent the > diffs back to McGill, Quite right. Thanks much, Howie. > but since they don't seem to be available for anonymous ftp yet > there, i've put them up. > They are on cunixf.cc.columbia.edu (128.59.40.130) as > X11R4-Next-2.0.tar.Z Good. > I don't know if they are completely in synch with McGill yet, Hint: they aren't. The McGill distribution has some 2.0 support in it, but it is probably not as good as what Howie has on cunixf, at present, so get his, not mine, for now. Once this slab I have in my office gets access to some disk space, I'll clean things up and wrap it all up into a neat package with a nice bow on top :-) > and have had a couple of minor reports about it (like the shifted top > row of the new keyboard's keypad aren't supported). Any idea how to find out what sort of keyboard is in use? It's just a matter of xmodmap to fix this, but it *should* be dealt with automatically - and if I can sense the keyboard type, it will be. der Mouse old: mcgill-vision!mouse new: mouse@larry.mcrcim.mcgill.edu
scott@texnext.gac.edu (Scott Hess) (03/11/91)
In article <1991Mar10.121059.18635@thunder.mcrcim.mcgill.edu> mouse@thunder.mcrcim.mcgill.edu (der Mouse) writes: In article <1991Mar10.000354.15364@cunixf.cc.columbia.edu>, howie@cunixf.cc.columbia.edu (Howie Kaye) writes: > and have had a couple of minor reports about it (like the shifted top > row of the new keyboard's keypad aren't supported). Any idea how to find out what sort of keyboard is in use? It's just a matter of xmodmap to fix this, but it *should* be dealt with automatically - and if I can sense the keyboard type, it will be. I don't think that the keyboards really make much difference. On an '030 system with the old keyboard, running 2.0, try out Shift-(keypad)= and shift-(keypad)/. Right! They yield | and \, resp. So, my guess would be to support the shifted-(keypad)= and /, _and_ the main keyboard |\. On the new machines, this won't hurt a thing, as there's not a keyboard |\. Unless I miss the point? Later, -- scott hess scott@gac.edu Independent NeXT Developer GAC Undergrad <I still speak for nobody> "Tried anarchy, once. Found it had too many constraints . . ." "I smoke the nose Lucifer . . . Banana, banana."
howie@cunixf.cc.columbia.edu (Howie Kaye) (03/11/91)
In article <1991Mar10.121059.18635@thunder.mcrcim.mcgill.edu> mouse@thunder.mcrcim.mcgill.edu (der Mouse) writes: >> and have had a couple of minor reports about it (like the shifted top >> row of the new keyboard's keypad aren't supported). > >Any idea how to find out what sort of keyboard is in use? It's just a >matter of xmodmap to fix this, but it *should* be dealt with >automatically - and if I can sense the keyboard type, it will be. > > der Mouse I just put up a new (split) version on cunixf. Someone complained that a 12M file transfer wasn't working out. I also made the code always put in the shifted versions of the top row keypad on new keyboards. /h
howie@cunixf.cc.columbia.edu (Howie Kaye) (03/11/91)
Since a bunch of people have asked about this, to get from mouseX back into NextStep you use a hotkey. The version I have up on cunixf has two key bindings for this key. 1) Command-Command-Keypad-* (hold down both Command keys, and type the * on the keypad). This is the same hotkey which mouseX has always had 2) Command-Command-Delete I added this one, because some people with smaller hands couldn't reach the other easily. ------------------------------------------------------------ Howie Kaye howie@columbia.edu Columbia University hlkcu@cuvma.bitnet UNIX Systems Group ...!rutgers!columbia!howie
jm7e+@andrew.cmu.edu (Jeremy G. Mereness) (03/12/91)
Excerpts from netnews.comp.sys.next: 10-Mar-91 Re: XNeXT on NextStep 2.0????? der Mouse@thunder.mcrcim (1506) > - and NeXT didn't > document their lower-level interfaces. So I had to dig through .h > files and what documentation there was, until I managed to make it > work. Of *course* it looks kludged up; it is! Well, I would sure like to see an X11R4 server for the NeXT, even one that switched between interfaces full-screen instead of having to deal with a window (it seems like it would be easier that way... maybe I'm wrong). I think NeXT would sell a lot more machines if people had the option of running X if they wanted to. I'd sure wanna buy one! As I understand it, NeXT is being very unfriendly as far as telling people what they need to know about writing alternative window servers... except for the folks who are writing CoXist which has to be paid $$$ for. I thought one idea behind X was that it was to be public domain (read: free). ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ |Jeremy Mereness | Support | Ye Olde Disclaimer: | |jm7e+@andrew.cmu.edu (internet) | Free | The above represent my| |a700jm7e@cmccvb (Vax... bitnet) | Software| opinions, alone. | |staff/student@Carnegie Mellon U.| | Ya Gotta Love It. | ------------------------------------------------------------------------
waltrip@capd.jhuapl.edu (03/12/91)
In article <kbqufqq00WCv075WBT@andrew.cmu.edu>, jm7e+@andrew.cmu.edu (Jeremy G. Mereness) writes: [...] > As I understand it, NeXT is being very unfriendly as far as telling > people what they need to know about writing alternative window > servers... except for the folks who are writing CoXist which has to be > paid $$$ for. I believe this has been rebutted by a number of people in previous postings...including the good folks at McGill who have given you a free X11R4 so now you can go ahead and buy your NeXT:^) > I thought one idea behind X was that it was to be public > domain (read: free). ^^^^^^^^^^ No, MIT's X development was sponsored by commercial interests (e.g., DEC, HP, IBM) who were interested in promoting a standard that would help them sell workstations. As it happens, the standard was embodied in sample servers (not reference servers) and toolkits which helped the rapid widespread acceptance of X as a de facto standard. Thus, X was often "free" on a number of platforms but the vendors were left with plenty of room to offer value-added (read: more efficient, supported and integrated) products. Products such as CoXist are quite consistent with the aims of the X project and with the aims of a company such as NeXT which is interested in having people offer software that helps them sell workstations. In particular, note that CoXist (this is my understanding from other posters--I have no first- hand knowledge) is integrated into the NeXT environment rather than existing as a separate environment into which toggles back and forth. A NeXT/CoXist user is thus able to co-exist naturally in both environments...a feature of significant value to many of us although for many others, the free implementation originating at McGill is a tremendous service which provides needed functionality. The MIT strategy has been tremendously successful in gaining acceptance of X. A comparable strategy for NeXTstep would probably require something like Adobe developing a number of Display PostScript controllers for popular architectures and distributing a generic NeXTstep source AND a generic X server that used DPS. Sample implementations for popular platforms would have to be included. I'm not suggesting that this strategy would be successful...only that it would be comparable in scope to what was done by MIT. c.f.waltrip Internet: <waltrip@capsrv.jhuapl.edu> Opinions expressed are my own.
scott@texnext.gac.edu (Scott Hess) (03/12/91)
In article <kbqufqq00WCv075WBT@andrew.cmu.edu> jm7e+@andrew.cmu.edu (Jeremy G. Mereness) writes:
Well, I would sure like to see an X11R4 server for the NeXT, even one
that switched between interfaces full-screen instead of having to deal
with a window (it seems like it would be easier that way... maybe I'm
wrong). I think NeXT would sell a lot more machines if people had the
option of running X if they wanted to. I'd sure wanna buy one!
Well, the way it looks, there are now two of them - one which is a port
of the MIT code, and one from Pencom (which will cost). So, what's the
problem?
As I understand it, NeXT is being very unfriendly as far as telling
people what they need to know about writing alternative window
servers... except for the folks who are writing CoXist which has to be
paid $$$ for. I thought one idea behind X was that it was to be public
domain (read: free).
The idea behind MIT's code is that it is free. At no point did anyone
state that every new machine has to have an X port, though! NeXT
hasn't been all in favor of X because NeXT's do not run X as their
native windowing system. That's a business decision, no matter the
ramifications. There is no requirement that NeXT to require any
information to anyone on anything - that they've chosen to do so
with many apps is because they'd like to have those apps on their
system.
On the other hand, they are basically in the position of being forced
to have X. It's no wonder that they aren't all that happy about it.
[Note to those of you waiting in the wings to restart the Great X
Debate: please! I'm not making any comment on X windows as a
windowing system - just pointing out that there's not anything
forcing NeXT to have it. -scott]
Later,
--
scott hess scott@gac.edu
Independent NeXT Developer GAC Undergrad
<I still speak for nobody>
"Tried anarchy, once. Found it had too many constraints . . ."
"I smoke the nose Lucifer . . . Banana, banana."rca@cs.brown.edu (Ronald C.F. Antony) (03/12/91)
In article <kbqufqq00WCv075WBT@andrew.cmu.edu> jm7e+@andrew.cmu.edu (Jeremy G. Mereness) writes: >As I understand it, NeXT is being very unfriendly as far as telling >people what they need to know about writing alternative window >servers... except for the folks who are writing CoXist which has to be >paid $$$ for. I thought one idea behind X was that it was to be public >domain (read: free). NeXT, I think, welcomes ANY compatibility product. They do however not welcome products that try to turn a NeXT computer into a "SUN?HP?APOLLO?IBM?DEC?You name it"-type workstation. i.e. as long as NeXT stays next, that's ok, if you want to have another computer, they probably think you should buy another computer. coXist coexists with NextStep, MouseX replaces it. If you run SoftPC, that's welcome, if you want to port MS-DOS to the NeXT and turn the MegaPixell display into a Hercules or CGA cardin 80*25 character mode, then that's not welcome. Easy? If you WANT X or MS-DOS then buy an X-workstation or an MS-DOS PC (does this stand for Pocket Calculator??), if you want a NeXT and NEED to run X or MS-DOS once in a while, then buy SoftPC and coXist plus a NeXT. Ronald ------------------------------------------------------------------------------ "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rca@cs.brown.edu or antony@browncog.bitnet
howie@cunixf.cc.columbia.edu (Howie Kaye) (03/13/91)
In article <68196@brunix.UUCP> rca@cs.brown.edu (Ronald C.F. Antony) writes: >coXist coexists with NextStep, MouseX replaces it. > Is this true? Does CoXist run within NextStep? Is the X root a single NextStep window (with a separate X environment in it), or is this more like the XNeWS merge which supports both windowing systems in the same environment? If it is a separate root window, i don't really see much of a difference between having to iconify/deiconify a window to switch between NextStep and X, and having to use a hotkey. ------------------------------------------------------------ Howie Kaye howie@columbia.edu Columbia University hlkcu@cuvma.bitnet UNIX Systems Group ...!rutgers!columbia!howie
rca@cs.brown.edu (Ronald C.F. Antony) (03/13/91)
In article <1991Mar12.204258.23440@cunixf.cc.columbia.edu> howie@cunixf.cc.columbia.edu (Howie Kaye) writes: >In article <68196@brunix.UUCP> rca@cs.brown.edu (Ronald C.F. Antony) writes: >>coXist coexists with NextStep, MouseX replaces it. > >Is this true? Does CoXist run within NextStep? Is the X root a >single NextStep window (with a separate X environment in it), or is >this more like the XNeWS merge which supports both windowing systems >in the same environment? > >If it is a separate root window, i don't really see much of a >difference between having to iconify/deiconify a window to switch >between NextStep and X, and having to use a hotkey. The difference is that even if it is a single rootwindow within NextStep, you can resize this to the size of the X-window you are interested in. i.e. you can simlate a merged envrionment. Of higher importance is however that you might want to be able to monitor several windows at once, some of them X windows, some of them NextStep. Also, I hope that coXist supports cut&paste between the two environments. Ronald ------------------------------------------------------------------------------ "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." G.B. Shaw | rca@cs.brown.edu or antony@browncog.bitnet