steve@fnord.umiacs.umd.edu (Steven D. Miller) (12/21/88)
I hope that this is new information, and that I'm not uselessly repeating
something that someone else has said; I haven't been following this group.
However, I thought that this was worth repeating, if that's what it comes
down to...
I just spoke to Tony Bonidy of NeXT's East Coast sales office. I told
him that there's a rumor (actually, more than that) going around that NeXT
doesn't want to make its sources (to anything!) available. He confirmed
that, saying that while the decision is not yet final, that is the way the
wind is blowing. I stressed to him the importance of sources, both for
system administration and for research, and mentioned that I know of
machines that have been turned off here in large part because sources for
them haven't been reasonably available. He's going to carry that message
back to California, to see if that will influence someone's final decision.
I'd suggest to those with interest in NeXT boxes and sources that they
get in touch with their NeXT sales reps and point out to them the error of
their ways. NeXT's phone number is (415) 424-0200; the folks at that number
should be able to point you at your regional sales rep.
An educational-market machine that won't sell in the educational market
because no one can get sources doesn't have much of a market left, does it?
-Steve
Spoken: Steve Miller Domain: steve@mimsy.umd.edu UUCP: uunet!mimsy!steve
Phone: +1-301-454-1808 USPS: UMIACS, Univ. of Maryland, College Park, MD 20742)) (12/22/88)
> Posted: 20 Dec 88 18:01:32 GMT > Organization: UMIACS, Univ. of Maryland, College Park, MD 20742 Steve Miller says: > I stressed to him the importance of sources, both for system administration > and for research ... Gee I got alot of work accomplished w/o MVS, RT-11, or VMS sources (pre-DEC of course :-) > An educational-market machine that won't sell in the educational market > because no one can get sources doesn't have much of a market left, does it? The focus of the NeXT cube being educational doesn't [necessarily] warrent the availability of sources. Do you have sources for MAC OS or MS/DOS? I cannot see NeXT easily licensing their sources since it's their implementation that sets them apart from the competition. (Apple's fanatic control of the Mac's ROM has kept cloning at bay for 4 *very important* years) Kevin
verber@cheops.cis.ohio-state.edu (Mark A. Verber) (12/23/88)
In article kevin@hiatus.dec.com (Kevin Baranski-Walker) writes: >> I stressed to him the importance of sources, both for system administration >> and for research ... > >Gee I got alot of work accomplished w/o MVS, RT-11, or VMS sources (pre-DEC >of course :-) There can be a lot of arguments about source/no-source. Almost all sites *think* that they do need source. Most sites have at least one hacker, and hackers by definition want source, if only to look at what other people have done. I agree that many sites don't need source, but saying that sources to the OS aren't needed is just as extreme as saying that everyone needs source. There are certain types of sites have a real need for source to the OS. Two examples if sites that don't need source: Application development. A friend of mine has a few machines (Macs currently, NeXT in the future?) and is doing some software development. He is writing a very slick tool to do dance notation. He has no need for source. Normal System Admin. Another friend is a sys-admin for a network with a few Suns. The Sun provided servers (such a YP) handle all his problems just fine. There are other sites that this just doesn't work. These sites would be characterized by one of three things: size, security concerns, or computer science research. If a site has one of these attributes it is somewhere between diffucult and impossible for them to function effectively without source. #1. Large and Complex Integration Problems There are many sites out there that have a large collection of hosts of varying types. These sites often want the systems to have access to identical services and resources. Sometimes these sites have need to drop things into the kernel. One site might have home grown tools that support distributed systems. You could say they should use the services of Mach, but they don't have Mach on most of their machines. If you can't integrate these services researchers will have trouble getting done what they want to do. Other sites might have a locally developped networking protocol, file system, etc which they need to permit integration with the rest of their machines. #2 High Security Sites. Lets face it, all software vendors are slow when it comes to distrubuting fixes. This is a fact of life right now. What happens if a major hole is found in the NeXT OS at a secure site? If they don't have source they have two choices: disconnect the machines from their network, or stop using the machines all together until the hole is fixed. Neither options is very good since it could be months until they get fixes from the vendor. #3 Research Sites I work for a computer science department. My users are researchers. One research project is investigating system preformance. Another project is looking into different ways to do distributed systems. Without sources these tasks can get very diffucult or impossible. One of the reasons that Mach has been successful (and useful) is because there was a large body of code that didn't have to be written because sources *were* available! Thinks like Mach, Project Athena, etc. can't happen without source. This is a good way to kill the future of a product. Mark A. Verber Ohio State Univ.
raymond@pioneer.arc.nasa.gov.arpa (Eric A. Raymond) (12/24/88)
In article <29954@tut.cis.ohio-state.edu> verber@cheops.cis.ohio-state.edu > >Two examples if sites that don't need source: > >Application development. A friend of mine has a few machines (Macs >currently, NeXT in the future?) and is doing some software >development. He is writing a very slick tool to do dance notation. >He has no need for source. Don't underestimate the utility of source. It is an extreme pleasure to work on a Lisp Machine with full sources. You don't like the way the editor works, change it. Find a bug with the file system, change it. Don't quite understand how XXX works, look in the source for examples. Of course, the real benefits come from the synergy with the truly integrated environment (in which your never a keystroke away from the source/documentation of a function in question AND there is no concept of functions being associated with a single program). I could go on and on, but .... RELEASE THE SOURCE. P.S. When Symbolics restricted source code a few years back, everyone screamed bloody murder. Fortunately they have rereleased the source. --------- Name: Eric A. Raymond ARPA: raymond@pioneer.arc.nasa.gov SLOW: NASA Ames Research Center, MS 244-17, Moffett Field, CA 94035 Nothing left to do but :-) :-) :-)
allanh@sco.COM (Allan J. Heim) (12/28/88)
steve@fnord.umiacs.umd.edu (Steven D. Miller) wrote: >An educational-market machine that won't sell in the educational market >because no one can get sources doesn't have much of a market left, does it? Absolutely! Just look at Apple (with the Apple II series and the Macintosh) and IBM (with their line of PCs).... Seriously, since when does the lack of sources spell doom for a computer? I've never heard of companies giving away source with their computers. Am I wrong? -- al allanh@sco.COM ...{decvax!microsoft,ucbvax!ucscc,uunet}!sco!allanh ...Lacrosse...Keyhole...KH-12...Milstar...NSA bait....
debra@alice.UUCP (Paul De Bra) (12/28/88)
In article <1981@scolex> allanh@sco.COM (Allan J. Heim) writes: ]steve@fnord.umiacs.umd.edu (Steven D. Miller) wrote: ] ]>An educational-market machine that won't sell in the educational market ]>because no one can get sources doesn't have much of a market left, does it? ] ]Absolutely! Just look at Apple (with the Apple II series and the Macintosh) ]and IBM (with their line of PCs).... ] ]Seriously, since when does the lack of sources spell doom for a computer? ]I've never heard of companies giving away source with their computers. ]Am I wrong? Yes you are. When my university was looking for a "big" (for those days) Unix (or Unix-like) machine for our computer science dept, getting the source (basically free of charge) was one of the top requirements. Just a tape with the standard System V or BSD source was not good enough. It had to be *the* source for their particular machine. For several companies this was no problem at all, and the source for UTX/32 we got with our Gould machine has proven to be extremely useful. I can assure you that if 2 comparable computer products are available, one with and one without source, the one without source won't go far in the educational market. Paul. -- ------------------------------------------------------ |debra@research.att.com | uunet!research!debra | ------------------------------------------------------
annala@neuro.usc.edu (A J Annala) (12/31/88)
Enhanced SECURITY is a widely geographically distributed network is another very serious reason to release sources. If it were not for the day to day familiarity with source code in the system administration groups of SUN and BSD UNIX sites around the USA during the recent virus attack, we probably wouldn't have recognized or been able to effectively disassemble, understand, and develop good immunization approaches to inhibit the spread of the virus. Neither SUN or BSD manufacturing sites were able to produce a solution to the virus ... the work was performed in a distributed fashion by academic sites working together throughout the country. The perpetrator of this crime actually worked with sources obtained while employed at a manufacturing site (AT&T) ... so restricting sources to manufacturer sites is ineffective and dangerous for the INTERNET at large. AJ Annala, USC Neuroscience Program
rang@cpsin3.cps.msu.edu (Anton Rang) (01/06/89)
In article <29954@tut.cis.ohio-state.edu>, Mark Verber (verber@cheops.cis.ohio-state.edu) writes: >In article kevin@hiatus.dec.com (Kevin Baranski-Walker) writes: >>> I stressed to him the importance of sources, both for system administration >>> and for research ... >> >>Gee I got alot of work accomplished w/o MVS, RT-11, or VMS sources (pre-DEC >>of course :-) >There can be a lot of arguments about source/no-source. Almost all >sites *think* that they do need source. I agree with this--most people think they want source for the OS. They probably don't need it, though. >I agree that many sites don't need source, >but saying that sources to the OS aren't needed is just as extreme as >saying that everyone needs source. So make the source optional (as DEC has done, I believe). That way those of us who DON'T need source don't wind up paying for it. >There are other sites that this just doesn't work. These sites would ^^^^ [not having source] >be characterized by one of three things: size, security concerns, or >computer science research. > [ ... ] >There are many sites out there that have a large collection of hosts >of varying types. These sites often want the systems to have access >to identical services and resources. Sometimes these sites have need >to drop things into the kernel. A decent OS should let you write kernel-level code *WITHOUT* needing to modify the OS itself. For instance, VMS lets you write device drivers (where I/O is involved) and user system services (for general types of things). The interfaces are documented. No source needed (though it can help by using parts as an example). Having sites modify their OS is a vendor's (support) nightmare. If a large number of changes are being made (say, implementing NFS under VMS :-) the chances that the customer will be willing to understand all of the relevant code (some 250 microfiche sheets for VMS/RMS) are pretty low. And what about upgrades? >Lets face it, all software vendors are slow when it comes to >distributing fixes. This is a fact of life right now. What happens >if a major hole is found in the NeXT OS at a secure site? This can be a problem, but having source isn't a "magic" solution. The larger, business-style vendors (DEC and IBM, say) generally are quite good about security patches (we reported one major VMS problem and got a patch tape back within a week). If we had the source, maybe we could hack around and fix the problem. Maybe we could have broken the three or four modules which depended (in non-obvious ways) on various parts of its behavior, too. >I work for a computer science department. My users are researchers. >One research project is investigating system preformance. Another >project is looking into different ways to do distributed systems. For this kind of thing, you really do generally need source. Why? Because you're writing an operating system. It's not a question of "modifying" the existing OS as much as making it do things it wasn't intended to do--adding paradigms. Of course, you then wind up with a myriad of differing versions, but that's OK for research. Just my opinions. >Mark A. Verber >Ohio State Univ. Anton +---------------------------+------------------------+----------------------+ | Anton Rang (grad student) | "VMS Forever!" | "Do worry...be SAD!" | | Michigan State University | rang@cpswh.cps.msu.edu | | +---------------------------+------------------------+----------------------+
gnu@hoptoad.uucp (John Gilmore) (01/08/89)
I think it's great that NeXT will not provide source (except to IBM,
and except for source to GNU stuff like gcc and gdb, which presumably
goes to everybody).
After all, I have a lot of Sun stock...
--
John Gilmore {sun,pacbell,uunet,pyramid,amdahl}!hoptoad!gnu gnu@toad.com
Love your country but never trust its government.
-- from a hand-painted road sign in central Pennsylvaniawebber@aramis.rutgers.edu (Bob Webber) (01/08/89)
In article <6193@hoptoad.uucp>, gnu@hoptoad.uucp (John Gilmore) writes: > I think it's great that NeXT will not provide source (except to IBM, > and except for source to GNU stuff like gcc and gdb, which presumably > goes to everybody). > > After all, I have a lot of Sun stock... Sun? Aren't they the company that doesn't provide online copies of their manuals for things like SunView? While I will grant that sources are more useful than documentation, it does seem that we have a duel here with both players firmly aiming at their own feet. ------ BOB (webber@athos.rutgers.edu ; rutgers!athos.rutgers.edu!webber)
tim@hoptoad.uucp (Tim Maroney) (01/08/89)
In article <6193@hoptoad.uucp>, gnu@hoptoad.uucp (John Gilmore) writes: > I think it's great that NeXT will not provide source (except to IBM, > and except for source to GNU stuff like gcc and gdb, which presumably > goes to everybody). > > After all, I have a lot of Sun stock... In article <Jan.7.21.59.35.1989.28140@aramis.rutgers.edu> webber@aramis.rutgers.edu (Bob Webber) writes: >Sun? Aren't they the company that doesn't provide online copies of their >manuals for things like SunView? Everything Sun does is perfectly well documented on-line. Just look in all those .c files. They're in plain text format for convenience. You could hardly do an NFS port without them, after all. -- Tim Maroney, Consultant, Eclectic Software, sun!hoptoad!tim "But don't you see, the color of wine in a crystal glass can be spiritual. The look in a face, the music of a violin. A Paris theater can be infused with the spiritual for all its solidity." -- Lestat, THE VAPIRE LESTAT, Anne Rice
mrc@Tomobiki-Cho.ACS.Washington.EDU (Mark Crispin) (01/28/89)
Although I work at an organization which uses NeXT computers, I
am speaking for myself here.
I am a serious software developer. I own two DEC-20 mainframe
computers (including source license). I have a NeXT at my regular job
(that's what Tomobiki-Cho is), and I'm seriously considering using the
NeXT as a software development platform off-work. I'm attracted to
the NeXT as it has the basic functions needed, they seem to have the
right ideas for software development tools, and (in spite of a
personal distaste for Unix) I need a Unix platform in my environment
as part of my support to my DEC-20 users (for whom continued effective
communication with Unix systems is a high-priority item).
I have an order form on my desk to purchase a NeXT for my
personal use at home. It's filled out, and I have the funds to cover
the check. There's no way in hell I'm gonna send that order in unless
sources are available.
Steve Jobs' arguments against distributed sources are utterly
unconvincing. It seems that he has failed to learn an important
lesson from the Apple experience. The Macintosh operating system is
often lampooned as the "Fisher-Price operating system". It is a
Mickey-Mouse OS, and the non-availability of sources guarantees it
will always be a Mickey-Mouse OS. This was a strong factor which
influenced me not to purchase a MacII.
In my opinion, ALL sources to critical components should be
available. This is not just limited to the operating system; it also
includes the window system, directory browser, etc. Maybe it'd be OK
if the InterfaceBuilder sources were kept secret since you could run
your machine without it -- although it seems a bit silly. Source code
gives the power to fix critical problems, is the ultimate
documentation, and, yes, even allows customization by customers.
This is not an argument to bundle sources with every NeXT. It's
perfectly alright to sell sources as a separate product. Not everyone
cares about sources. Not everyone will buy sources. But those who do
care, care a lot. [And I have yet to hear people who don't care
chorusing that "sources should not be released."]
Presently, because I decline to use YP in my development
environment, my NeXT is not doing host name recognition in virtually
all network utilities. I am now going through the agonizing task of
getting sources for individual components from the net (many of which
are publicly available via FTP from Berkeley!) and rebuilding all
these components to use the resolver version of gethostbyname().
It is sheer arrogance not to make sources available to customers.
My reaction as a potential customer is to answer arrogance with not
buying the product and not committing to the platform. Sure, I'm
using a NeXT now, but a better box will come along in the future, and
I'll ask my boss to take the NeXT away and give me the new box.
It is not too late for NeXT to undo the damage. But every day
that NeXT waffles on the issue, more damage gets done. If I didn't
care, if I didn't wish NeXT success, I wouldn't be sending this
message.
Is NeXT listening?
-- Mark --bob@tinman.cis.ohio-state.edu (Bob Sutterfield) (01/28/89)
In article <651@blake.acs.washington.edu> mrc@Tomobiki-Cho.ACS.Washington.EDU (Mark Crispin) writes:
It is not too late for NeXT to undo the damage. But every day that
NeXT waffles on the issue, more damage gets done. If I didn't
care, if I didn't wish NeXT success, I wouldn't be sending this
message.
Lest <BOB.89Jan27183103@allosaur.cis.ohio-state.edu> ("NeXT vs
sources") seem too harsh, let me chime in again with Mr Crispin in
wishing NeXT well. The NeXT cube is a nifty machine, and I look
forward to developing applications for it. It is inappropriate for
our department's general instructional and research environment
because of the lack of several features, one of them being source
availability. But that can be changed, and the machine could be a
major force in moving educational workstation technology forward.rick@Jessica.stanford.edu (Rick Wong) (02/01/89)
I just wanted to throw in my two cents worth to the outcry for source
availability.
Most of the folks arguing for source availability seem to be coming from
a UNIX or networking standpoint. I would like to make a pitch for source
availability for the AppKit.
I develop applications on the Mac, using MacApp, Apple's object-oriented
applications framework. It is analogous to NeXT's AppKit. Apple distributes
MacApp in source-code form, and I find having the source absolutely essential
to getting anything done with it.
I have four points:
1. Object-oriented programming requires the programmer to have fairly
precise knowledge of the protocols that have to be followed when
"plugging into" an existing set of classes (i.e., the AppKit).
Although _extremely good_ documentation may be able to describe
the protocols sufficiently, it can never provide as precise a
description as the source of what the source is expecting.
2. Software (even NeXT's!) often has bugs. When I run into a problem
because some method call to an AppKit object isn't behaving as I
expect, I would like to be able to find out if it's because of
something I'm doing wrong, or because of a bug in the AppKit. It's
much easier to determine this from the source than from documenta-
tion.
3. Having the source available ought to greatly improve the quality
of bug reports to NeXT. When I find a bug in MacApp, for instance,
I can usually find out the exact portion of code that's in error,
fix it myself, and send an _exact_ description of the bug _and_
its fix to Apple.
4. Being able to fix bugs also saves me from having to kludge around
them until they're "officially" fixed.
I suppose many of these points apply to having source available for all
levels. I think they especially apply for object-oriented programming,
considering the often complex interactions between different objects.
(Smalltalk is distributed with source, isn't it?)
Rick Wong
Stanford University
rick@jessica.stanford.edubob@allosaur.cis.ohio-state.edu (Bob Sutterfield) (02/02/89)
In article <169@Portia.Stanford.EDU> rick@Jessica.stanford.edu (Rick Wong) writes:
Most of the folks arguing for source availability seem to be coming
from a UNIX or networking standpoint. I would like to make a pitch
for source availability for the AppKit.
I agree with your reasons, and will add another (which I already
mentioned long ago) for distribution of user-level sources.
When we get another computer on our network, I can almost always give
it a set of X11 clients with very little trouble, so that the new
machine can be used "seamlessly" from our Suns, HPs, RTs, etc. I can
do this because I have sources to the X11 protocol and toolkit
libraries as well as to the clients themselves. It becomes a
(relatively) simple matter of copy-and-compile, with some occasional
porting woes, to get onto a new architecture. Most NeWS clients are
almost as easy, for the same reasons.
The NeXT machine's only interface to the compute power other machines
on the network is via a (buggy) terminal emulator running a remote
login application (rlogin, telnet, etc.) It's embarassing to have
such a primitive asynch-tty worldview in a window next to the spiffy
applications that can run on NeXT machines.
In The Old Days, we were happy to have systems connected by standards
at the level of RS-232. Then if it didn't have TCP/IP it didn't come
in the door. Now our users are surprised at the existence of a
machine (our Encore Multimax) that still doesn't integrate into our
NFS network (two years after they sold us NFS, BTW). Just as YP makes
every workstation look just like another, the users simply expect
remote X and NeWS clients to run on everything on our network.
NeXT should distribute sources to their clients and protocol and
toolkit libraries so that users can integrate the NeXT machines into
their existing environments.gillies@p.cs.uiuc.edu (02/03/89)
I think that by not distributing the sources, NeXT is making an extremely heavy commitment to fix all bugs almost immediately and generally support the software in their machines incredibly well -- much better than UNIX. Furthermore, they are promising to document their software systems superbly, a la "Inside Macintosh" plus 230 (currently published) Apple technical notes. Finally, they are taking on the burden of making their system a lot more general, so that certain users can build on existing packages, rather than having to reimplement features that have been overlooked by NeXT, Inc. NeXT does not have the resources to accomplish this. Also, the computer architecture (UNIX) is not as conducive to this type of support as the Macintosh, where anything can be patched/hacked by a knowledgeable software expert. Can anyone from NeXT respond to these issues? P.S. I can understand why the sources to Objective C, Display Postscript, and perhaps WriteNow cannot be released, since these are commercial products owned by other companies. But what about NeXT-developed sources?