ANNOUNCEMENT: Alpha release Linux/GNU/X ...

64 views
Skip to first unread message

Jamie Mazer

unread,
Nov 26, 1992, 5:18:42 AM11/26/92
to
> ...w...@cs.brown.edu writes:
> ...wes...@smurf.sti.com (Will Estes) writes:
> > This is a bit embarrassing to have to point out, but did you realize
> > that Microsoft is selling Windows/NT on CD-ROM for $69? I mean, is
> > Microsoft's proprietary 32-bit operating system even more free than
> > free GNU UNIX? :)
...
> As for your assertion that MS's NT is "more free" than the Linux stuff..

Perhaps I'm loosing track of all this, but I think you're all missing
the meaning of the term "free software". I do believe that several
people (including RMS, if I remember correctly) have frequently pointed
out that the "free" in "free software" w.r.t. to GNU doesn't really mean
that the software is available for no-money-down, but rather, it refers
to the fact that the software is unencumbered (tho, I guess some may
argue over the details of the GPL).

Please correct me if I'm wrong, but it seems that you guys are all
arguing about something off the track -- some people will want W/NT
for their boxes, while others will want GNU/Linux. Anyway, the
difference between 70 and 100 bucks is pretty small compared to the
cost of the hardware. The real issue, I think, is that NT and Linux most
likely appeal to different crowds. I'd run Lunix (if I had a 486 box
lying around), but most of the non-computer people in my lab certainly
wouldn't.

Hmm, perhaps I shouldn't have gotten involved.. It's been a LONG
time since I posted to an impending flame war..
/Jamie
--
EMAIL: ma...@cns.caltech.edu PHONE: (818) 356-6816 FAX: (818) 449-0679
USMAIL: Div. of Biology 216-76, Caltech, Pasadena, CA, USA 91125
"don't shoot someone tomorrow that you can shoot today", housemartins

Chris G. Demetriou

unread,
Nov 26, 1992, 4:43:27 PM11/26/92
to
In article <1992Nov27.0...@u.washington.edu> t...@carson.u.washington.edu (Tim Smith) writes:
>The GPL is a significant encumberance. Is there anyone working on anything
>significant (compilers, operating systems, editors, etc.) that will be
>public domain? I've tried a couple of times on the OS front, but these
>projects have always died due to lack of time. Is anyone else either
>doing this, or planning to do this?

amen, on the fact that the GPL isn't unencumbering...

386bsd is being done under a UCB-style license. That is,
it's not strictly PD, because the it's still copyrighted, etc.,
but it's basically of the form:

You can use this s/w for any purpose whatsoever so long as
(1) you include this copyright,
(2) don't advertise w/our names,
(3) give us some credit,
and (4) don't try to blame us for any problems you encounter w/it...

I don't think you're going to ever find a *large* truly-PD package,
because *usually* authors of large things want to keep at least
some form of minimal copyright notice on their works...


Chris
--
Chris G. Demetriou c...@cs.berkeley.edu

"Sometimes it is better to have twenty million instructions by
Friday than twenty million instructions per second." -- Wes Clark

Sakari Jalovaara

unread,
Nov 27, 1992, 10:03:57 AM11/27/92
to

>> The GPL is a significant encumberance. Is there anyone working on
>> anything significant (compilers, operating systems, editors, etc.)
>> that will be public domain?

Well, I've been writing an ANSI C compiler as a hobby. It'll be PD
if it ever gets good enough to be published. The front end is mostly
done (very standard conforming, fairly portable, very fast!) I'm just
starting work on a code generator (and CG generator.) Ask again in
a year or so...

cgd says:
> I don't think you're going to ever find a *large* truly-PD package,
> because *usually* authors of large things want to keep at least
> some form of minimal copyright notice on their works...

Define "large"... Is 50 KLOC large? (15 KLOC estimated for a simple
back end; add 20..infinity KLOC for optimization.)

As to copyright, I have yet to find a _reason_ to restrict my work
(except when I do do something to keep the fridge filled with
lasagne&beer.) I am a great believer in code re-use; everyone's
"don't change this copyright" is inconvenient in that respect
(though better than the GNU copyright which is actively hostile
against any other form of copying terms, including PD.)
++sja

bryan ford

unread,
Nov 28, 1992, 1:08:55 PM11/28/92
to
>>the meaning of the term "free software". I do believe that several
>>people (including RMS, if I remember correctly) have frequently pointed
>>out that the "free" in "free software" w.r.t. to GNU doesn't really mean
>>that the software is available for no-money-down, but rather, it refers
>>to the fact that the software is unencumbered (tho, I guess some may
>>argue over the details of the GPL).
>
>The GPL is a significant encumberance. Is there anyone working on anything
>significant (compilers, operating systems, editors, etc.) that will be
>public domain? I've tried a couple of times on the OS front, but these
>projects have always died due to lack of time. Is anyone else either
>doing this, or planning to do this?

Did it ever occur to you that there might be a connection between the
abundance of excellent GPL-based software and the lack of
"significant" public domain software? Public domain software gets
snatched up and proprietarized by greedy people and companies, and
thereafter gets shattered into a thousand different competing threads
of development which can never be reconciled. The "encumberances" of
the GPL hold the software together and make it possible for the
software grow into what we have today.

If the GPL is a "significant encumbrance" to what you want to do with
the software, then what you want to do with the software is probably
not in the interests of the original authors of that software or of
the "general public". If you think this statement is wrong, then I
would be happy to see a counterexample.

Bryan Ford
baf...@cs.utah.edu

Tim Smith

unread,
Nov 28, 1992, 8:31:14 PM11/28/92
to
baf...@labhp23.cs.utah.edu (bryan ford) writes:
>
>"significant" public domain software? Public domain software gets
>snatched up and proprietarized by greedy people and companies, and
>thereafter gets shattered into a thousand different competing threads
>of development which can never be reconciled. The "encumberances" of

So, having *more* software available is bad? You can use a given
piece of public domain software regardless of what other, more greedy,
people have done with it.

>If the GPL is a "significant encumbrance" to what you want to do with
>the software, then what you want to do with the software is probably
>not in the interests of the original authors of that software or of
>the "general public". If you think this statement is wrong, then I
>would be happy to see a counterexample.

OK, how about a small company trying to compete in an area dominated by
one or two large companies. The large companies have a few great
programmers, and a lot of good programmers. The small company has a
few great programmers. To make this a clean example, assume that
the area these companies work in is one that most programmers could
not do at home (e.g., assume the software goes into some embedded
system and to develop this stuff requires in-circuit emulators and
special target systems and stuff like that).

The large companies are *NOT* going to make their source code available.
If the small company makes its source code available, the large companies
will be able to drive the small one out of business, because their good
programmers are good enough to take the work of the small company's great
programmers and put it in their software, and that will negate any
competitive advantage of the small company.

If the small company could use various routines from GPL'ed source,
they would not have to spend as much time rewriting wheels, and could
more effectively compete with the large company, leading to more choice
for the user. The large companies would not benefit as much if they
could use GPL'ed stuff, because they have lots of good programmers that
they can use to reinvent wheels.

Thus, if the small company could use various GPL'ed software without having
to put their entire product under GPL, there would be *MORE* software
available, and more choice (indirectly) for the general public. Isn't
this good?

Another way to look at this is that there *is* a cost in using GPL. If
a company will lose by GPL'ing their software, that has to be evaluated
against their costs in developing their own proprietary versions of
whatever GPL'ed thing they want to use, or licensing something like that
from someone else. In this sense, GPL is as much as an encumberance as
an AT&T license: I have to pay to use the software -- the only difference
to my pocketbook is that with the AT&T licence, I can calculate what I
have to pay, rather than having to try to estimate how much letting my
competitors use my source will cost me.

This is why if I write something useful, other than at work, I will either
make it public domain or only put on a vanity copyright (e.g., my name has
to stay on it). The public benefits both if my software is used by others
in freely distributable software *AND* they benefit if a commercial software
company is able to use my software to make a better product. The benefit
the most of both of these things can happen.

--Tim Smith

Peter da Silva

unread,
Nov 28, 1992, 3:34:28 PM11/28/92
to
In article <1992Nov28.1...@hellgate.utah.edu> baf...@labhp23.cs.utah.edu (bryan ford) writes:
> If the GPL is a "significant encumbrance" to what you want to do with
> the software, then what you want to do with the software is probably
> not in the interests of the original authors of that software or of
> the "general public". If you think this statement is wrong, then I
> would be happy to see a counterexample.

Well, I have written software components (subroutines, libraries) that
have been included in commercial packages by third parties. If I had used
the GPL I couldn't have done this... even if I were willing to make a
separate deal with the third parties they would probably have been scared
off in the first place.

What I want people to do with my software is be the most productive people
they can be. If I could support that goal with the GPL, I would, but for
the sort of things I do I can't. Yet I believe that what I'm doing is in the
best interest of the original authors (myself) and the general public.

The problem is, there are applications that are insufficiently fun to attract
GPL type support, and are of insufficiently high value on a person-by-person
basis to allow customer #1 to pay all the development costs. Commercial
software publishing as it now exists is the best way to pay for the work
required to develop such software, and if you want to help those people (and
I do) you need to give them a deal they can afford to make.

The library GPL seems like it would be a good solution to this dilemma, but
I haven't seen any software actually use it yet.
--
%Peter da Silva/77487-5012 USA/+1 713 274 5180/Have you hugged your wolf today?
/D{def}def/I{72 mul}D/L{lineto}D/C{curveto}D/F{0 562 moveto 180 576 324 648 396
736 C 432 736 L 482 670 518 634 612 612 C}D/G{setgray}D .75 G F 612 792 L 0 792
L fill 1 G 324 720 24 0 360 arc fill 0 G 3 setlinewidth F stroke showpage % 100

Marc Unangst

unread,
Nov 29, 1992, 3:58:22 AM11/29/92
to
In article <1992Nov29.0...@u.washington.edu> t...@carson.u.washington.edu (Tim Smith) writes:
>If the small company makes its source code available, the large companies
>will be able to drive the small one out of business, because their good
>programmers are good enough to take the work of the small company's great
>programmers and put it in their software, and that will negate any
>competitive advantage of the small company.

Of course, you have conveniently neglected to mention that if the
small company releases its source under the GPL, and the large company
picks this up and incorporates it into its product, the large company
will then have to release all source to *its* product under the GPL.
(This is why some people call the GPL the General Public Virus -- it
infects all the software it touches...) And the small company will
then be free to take the source code from the large company's program
and incorporate it into its offering.

The real place the GPL falls down is from the perspective of the large
company -- it will have to re-invent its own version of the small
company's code to avoid being forced to release all its code under the
GPL. This effect is intentional, of course; the idea is that
companies will either have to release their code under the GPL, or be
handicapped by not being able to use other GPL'd inventions in their
works.

>for the user. The large companies would not benefit as much if they
>could use GPL'ed stuff, because they have lots of good programmers that
>they can use to reinvent wheels.

And this is why the virus-like nature of the GPL hasn't caused the
entire software industry to be infected by now. Companies
distributing proprietary software without source code have decided
that the benefits of using GPL code in their programs is not worth the
"cost" of having to adhere to the GPL.

>This is why if I write something useful, other than at work, I will either
>make it public domain or only put on a vanity copyright (e.g., my name has
>to stay on it). The public benefits both if my software is used by others
>in freely distributable software *AND* they benefit if a commercial software
>company is able to use my software to make a better product. The benefit
>the most of both of these things can happen.

Except if you leave off a copyright, or worse yet say "This is
copyright 1992 by Tim Smith; you can use this for whatever you want as
long as my name stays on it", it allows the code to be abused. A
commercial software company can take your code, put it into their
application, and only give you a footnote on page 1538 of their
manual. You don't get to see whatever modifications they made to your
code and whatever improvements they made, and neither does anybody
else. The GPL, while not perfect, at least ensures that the code and
its derivations will always be available under the same conditions as
--
Marc Unangst, N8VRH | "There are two ways to solve this problem:
m...@mudos.ann-arbor.mi.us | the hard way, and the easy way. Let's start
| with the hard way."
| - W. Scheider, from a Physics lecture

Tim Smith

unread,
Nov 29, 1992, 7:41:11 AM11/29/92
to
[Note the Foloowup-To line. I accidently started this in too many groups,
so I'm trying to remedy that]

m...@mudos.ann-arbor.mi.us (Marc Unangst) writes:
>>If the small company makes its source code available, the large companies
>>will be able to drive the small one out of business, because their good
>>programmers are good enough to take the work of the small company's great
>>programmers and put it in their software, and that will negate any
>>competitive advantage of the small company.
>
>Of course, you have conveniently neglected to mention that if the
>small company releases its source under the GPL, and the large company
>picks this up and incorporates it into its product, the large company
>will then have to release all source to *its* product under the GPL.
>(This is why some people call the GPL the General Public Virus -- it
>infects all the software it touches...) And the small company will
>then be free to take the source code from the large company's program
>and incorporate it into its offering.

My assumption is that the large company would be looking at the small
company code to figure out how certain things are done, and would use
that knowledge to modify the large company code accordingly, rather than
trying to actually take the code itself from the small company.

At least in the area that I work in, device drivers for SCSI host
adaptors (and sometimes for network cards or other assorted cards),
the big thing that someone would get reading my source code is how
to work around bugs in various operating systems, how to work around
bugs in various chips, and techniques for doing certain things more
efficiently. They would probably not want to actually use my code,
because my stuff and their stuff is likely to differ quite a bit
in internal organization. They would just want to grab the ideas
from it.

If my code is GPL'ed, it is not going to hinder them from doing this.
I think what I'd have to do if I wanted to release my source code
so that individual programmers could use it in their free products,
yet prevent myself from being seriously harmed by the big company,
is get a *patent* on my stuff, and then make a GPL-like patent
license.

>Except if you leave off a copyright, or worse yet say "This is
>copyright 1992 by Tim Smith; you can use this for whatever you want as
>long as my name stays on it", it allows the code to be abused. A
>commercial software company can take your code, put it into their
>application, and only give you a footnote on page 1538 of their
>manual. You don't get to see whatever modifications they made to your
>code and whatever improvements they made, and neither does anybody
>else. The GPL, while not perfect, at least ensures that the code and
>its derivations will always be available under the same conditions as

But I don't consider this to be abuse of my code. If they write their
own, rather than use mine, I still don't get to see what they do, and
neither does anybody else. On the other hand, if they do use mine,
then perhaps the time they would have spent writing their own will be
spent writing something else, and I, if I am a user of their product,
will then benefit by getting more features in the next version of
their product.

Has there been abuse of the kind you mention with X? From a quick
look at the X source (a very quick look, so I may have missed things),
it appears that MIT is distributing under the terms that I would
use for my own software. Have they run into problems with this?

--Tim Smith

mathew

unread,
Dec 1, 1992, 9:47:29 AM12/1/92
to
baf...@labhp23.cs.utah.edu (bryan ford) writes:
> Did it ever occur to you that there might be a connection between the
> abundance of excellent GPL-based software and the lack of
> "significant" public domain software? Public domain software gets
> snatched up and proprietarized by greedy people and companies, and
> thereafter gets shattered into a thousand different competing threads
> of development which can never be reconciled.

Can you give an example?

> If the GPL is a "significant encumbrance" to what you want to do with
> the software, then what you want to do with the software is probably
> not in the interests of the original authors of that software or of
> the "general public". If you think this statement is wrong, then I
> would be happy to see a counterexample.

Can you start giving us examples? Then we'll start thinking of
counterexamples...


mathew

mathew

unread,
Dec 1, 1992, 10:04:11 AM12/1/92
to
m...@mudos.ann-arbor.mi.us (Marc Unangst) writes:
> Except if you leave off a copyright, or worse yet say "This is
> copyright 1992 by Tim Smith; you can use this for whatever you want as
> long as my name stays on it", it allows the code to be abused. A
> commercial software company can take your code, put it into their
> application, and only give you a footnote on page 1538 of their
> manual. You don't get to see whatever modifications they made to your
> code and whatever improvements they made, and neither does anybody
> else.

And why should you? *They* made those changes and improvement, it's *their*
work, not yours. The fact that they started off with your free code is
irrelevant.

Sure, you can say "I only want you to use my code if you're going to let me
use your code", which is basically what the GPL tries to do; but I see no
reason why this is fundamentally the One True way of doing things. Oxford
Publications don't demand the right to my articles just because I use words
and grammatical structures I've copied from their books.


mathew
--
Lung cancer and arteriosclerosis.
For people who like to smoke.

der Mouse

unread,
Dec 1, 1992, 5:07:48 PM12/1/92
to
In article <1992Nov28.1...@hellgate.utah.edu>, baf...@labhp23.cs.utah.edu (bryan ford) writes:
[intermediate attribution line seems to have been lost]

>> The GPL is a significant encumberance.
> If the GPL is a "significant encumbrance" to what you want to do with
> the software, [...]

"Encumbrance" and related terms (eg, "encumbered") are legal terms with
specific meanings. The GPL is indeed a significant encumbrance.
Whether or not it gets in the way of what you want to with the software
is irrelevant.

> Did it ever occur to you that there might be a connection between the
> abundance of excellent GPL-based software and the lack of
> "significant" public domain software?

No.

der Mouse

mo...@larry.mcrcim.mcgill.edu

Doug DeJulio

unread,
Dec 1, 1992, 5:01:55 PM12/1/92
to
In article <uXy4uB...@mantis.co.uk> mathew <mat...@mantis.co.uk> writes:
>baf...@labhp23.cs.utah.edu (bryan ford) writes:
>> Did it ever occur to you that there might be a connection between the
>> abundance of excellent GPL-based software and the lack of
>> "significant" public domain software? Public domain software gets
>> snatched up and proprietarized by greedy people and companies, and
>> thereafter gets shattered into a thousand different competing threads
>> of development which can never be reconciled.
>
>Can you give an example?

X-windows on the NeXT. There is one free (money-wise) version for the
machine I've got, but it doesn't come with source code for the server
so I can't recompile it and move up to X11R5. There are something
like three or four commercial X servers, all sold for upwards of $250
or so, and all claiming to be better than each other. The basic
feature is to have one big NeXTstep window in which an X server runs.
Then some offer the "enhancement" of putting each X window in a
separate NeXTstep window, and some offer the "enhancement" of having
the X server use video memory directly to speed things up. I'm sure
each of these enhancements was done in virtually the same way
conceptuially by each company, but without having any code in common.

Now, how about BSD Unix? SunOS and Ultrix derived from it. Would you
say they compete? Would you say they could ever be reconciled?

>> If the GPL is a "significant encumbrance" to what you want to do with
>> the software, then what you want to do with the software is probably
>> not in the interests of the original authors of that software or of
>> the "general public". If you think this statement is wrong, then I
>> would be happy to see a counterexample.
>
>Can you start giving us examples? Then we'll start thinking of
>counterexamples...

Once again, X-windows. Let's say a vendor makes a graphics
accelerator card, and makes an X server for it. Let's say X11R2, it
was done a while ago. Now the users can't upgrade. They can't grab
the normal publicly available X sources and recompile them on their
machine, because the support for their graphics system isn't free!
They've got to wait for the vendor to release a new version, and often
have to pay for the upgrade (even if all the vendor needed to do is
relink against new libraries).

THIS SUCKS BIG TIME!
--
Doug DeJulio
dd...@cmu.edu

Tim Smith

unread,
Dec 2, 1992, 12:00:41 AM12/2/92
to
[Note the Followup-To line! This shouldn't be taking place on a zillion
newsgroups. If you respond to this, check the newsgroups line. If it
has anything other than gnu.misc.discuss, change that, please.]]

In article <ByLp79...@cs.cmu.edu> dd...@cs.cmu.edu (Doug DeJulio) writes:
>Once again, X-windows. Let's say a vendor makes a graphics
>accelerator card, and makes an X server for it. Let's say X11R2, it
>was done a while ago. Now the users can't upgrade. They can't grab
>the normal publicly available X sources and recompile them on their
>machine, because the support for their graphics system isn't free!
>They've got to wait for the vendor to release a new version, and often
>have to pay for the upgrade (even if all the vendor needed to do is
>relink against new libraries).

You are still free to grab the normal X sources and port them to your
machine. The vendor having used X for their proprietary port in no
way prevents you from doing your own port and giving it away (or
selling it).

By not giving you their port, the vendor is not harming you. They
are failing to help you. You are basically in the same situation that
you would be in if they had decided to write an X clone from scratch.
The fact that they may or may not have used actual X source is
irrelevant, except that if they did, they can presumably update their
port easier as new version of X become available.

--Tim Smith

der Mouse

unread,
Dec 5, 1992, 5:07:56 PM12/5/92
to
In article <ByLp79...@cs.cmu.edu>, dd...@cs.cmu.edu (Doug DeJulio) writes:

> Once again, X-windows.

("It's a window system called X, not a system called XWindows.")

> Let's say a vendor makes a graphics accelerator card, and makes an X
> server for it. Let's say X11R2, it was done a while ago. Now the
> users can't upgrade.

What users? Users that were stupid enough to buy the card and thereby
lock themselves into a single-vendor solution?

Stupidity always brings its own reward.

der Mouse

mo...@larry.mcrcim.mcgill.edu

Reply all
Reply to author
Forward
0 new messages