TEAM-ADA Archives

Team Ada: Ada Programming Language Advocacy


Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
"Crispen, Bob" <[log in to unmask]>
Reply To:
Crispen, Bob
Mon, 5 Mar 2001 10:47:44 -0600
text/plain (249 lines)
Tom Moran [mailto:[log in to unmask]] sez:

>  We agree.  It's quite unlike land, sheep, etc.  But given digital
>distribution, we need to come up with a some way of encouraging authors
>to produce works useful to the public.

As someone who's involved in some open source projects, I
assure you, authors don't need to be encouraged to produce
works useful to the public.  The GNU tools, majordomo, Linux,
FreeBSD, apache, hypermail, and hundreds of other projects
have proven that.  In fact, the problem is just the opposite --
willing programmers far outnumber able ones.

I don't mind someone saying that we need other incentives
besides the ones associated with Free Software.  Making money
is a very good thing.

I've even said it myself: I've often said that I think it
would be a very good thing if people working on 3D technologies
for the Web got rich.  I haven't said this recently, but I'll
say it again now: if the people who are working on technologies
like Ada that promote good software engineering practices get
rich, that would be a very good thing, too.  Perhaps an even
better thing, since it would affect so many other endeavors.

What I do object to is someone else trying to pass regulations
and legislation saying that the *only* incentive for making
useful inventions should be money, and that Free Software with
certain kinds of licenses should be outlawed in certain large

>Simply saying computer
>programmers should not ask for money for their work is a bit

No, no, no!  The question isn't whether programmers should or
shouldn't ask for money.  The question is whether programmers
will be allowed *not* to ask for money.

If Allchin and his masters get their way, programmers (and,
more relevantly, project managers) on government projects will
not be permitted to issue certain kinds of free software.

Agree or disagree whether programmers should get money all
you want.  If you do absolutely nothing but fight resolutely
against this latest campaign of Microsoft's and win, you'll
be able to continue that argument and persuade people to your
point of view.  If Microsoft wins, the argument itself

>It is downright annoying when the people telling me I
>should not ask for money (work on a farm, programming in the
>are themselves being whelped by large companies or universities (i.e.,
>government grants).  It is doubly annoying when those people are being
>paid by, e.g., IBM, to preach ideas which make it yet more
>difficult for
>small competitors to arise and challenge IBM.  It is triply annoying
>when those people claim the flags of  virtue, "The Public Interest" and

I'm with you.  As Samuel Johnson said, "No man but a blockhead
ever wrote except for money."  I expect money if I'm to put my
fingers on the keys and write prose for publication.

However, just because I want money for my writing, I understand
that good citizenship obligates me to do certain kinds of writing
for free.  While I haven't written many papers, I've refereed a
large number, often without even getting a free copy of the journal
(journal editors who are reading this, please take note that this
is a grumble).

And that's OK.  I have the choice, just as I have the choice to
either participate in some Free Software projects or not.

But, just like navels, everybody's got an interest.  I think
it's reasonable to say that Microsoft's interest is in removing
competitors from the playing field, and that prohibiting GPL
licensing for government software will have that result -- that
is, to take away that freedom of choice.

And that is emphatically not OK.

>> protected by the GPL
>  The GPL is most assuredly a copyright, a claim that the owner has
>rights over distribution.

Just a technicality -- the GPL is not a claim of copyright.  The
author (absent a work for hire condition or assignment of copyright
or a few other things -- IANAL) owns the copyright from the
moment the work is created.

The GPL (etc.) is merely a license describing the terms and
conditions under which copying the software is permissible.
That license is freely granted by the owner of the copyright,
who is free to license her own software under any terms she

>I object to the particularly draconian GPL
>rights, just as I would object to the oil companies buying the
>proverbial patent for an engine that runs on water, and then  highly
>restricting use of that invention. I'd argue that such limits on use
>violate the intent "To promote the Progress of Science and useful Arts"
>of the Constitution.  And most certainly if government money helped pay
>for a development, the GPL would be a highly unreasonable set of
>restrictions on its use.

Well, if you'd ever dealt with the problem of getting software
from a company that your company has replaced on a program,
perhaps you'd welcome a broader use of open source licenses.
And there are practices going on in programs today -- I won't
do more than allude to them, but we all know there's too many
of them like this -- where things I'm profoundly ashamed go on
in the aerospace and defense segments of government would be
completely wiped out by a vigorously enforced open source
licensing policy.

Further, your argument that GPLed software is licensed in such
a draconian fashion that the world is made a poorer place thereby
is perfectly tenable -- so long as you never, ever open your own
eyes and take a look around.  GPLed products have taken the world
by storm, and Linux, the poster child for the GPL, is springing
up everywhere.

But let's go from the big to the small.  I've written a number of
utilities for VRML and GPLed them.  On at least three separate
occasions I've been given credit for some snippets of code that
others have taken to give themselves a leg up on utilities of
their own -- which they've also GPLed.  I used Steve Chenney's
GPLed code for converting viewpoint, look-at point, and up vector
to VRML x,y,z,theta notation a number of times.  You want to talk
massive code reuse?  That's it right there.

Nor is the GPL the all-contaminating virus it's claimed to be.  I
wrote a bad little newsreader for Windows called Foxynews, and as
the documentation makes very clear, it contains GPLed code, BSD
code, and code where I obtained permission from the authors for
this particular use.

Do people who claim that the GPL contaminates everything it touches
seriously believe that, by using the code for uudecode in my little
newsreader, I've changed the licensing on uudecode from BSD to GPL?

Or do they claim that somebody who uses the uudecode source from
Foxynews is bound by one license while somebody who gets the same
code elsewhere is bound by another license?

If not -- and it ought to be clear that those claims are just
absurd -- just what the heck *do* they claim?  This is FUD, plain
and simple.

And let's talk about the issue of lasting usefulness and software
longevity.  If you're into practical 2D computer graphics at all,
you're aware of PhotoShop plugin filters.  These useful little
utilities let you perform image manipulations inside PhotoShop
and compatible programs like Paint Shop Pro.  You've no doubt
heard of Eye Candy and Kai's Power Tools, and you may even have
heard of Blade Pro.

But there's a filter you probably haven't heard of unless you've
been around PhotoShop for quite a while: Almathera.  It turns
out that the Almathera filters are a terrific little package,
very reasonably priced (or they were), and they do things that
simply can't be done in Eye Candy, KPT, or the Filter Factory

So is this a blatant ad for Almathera?  Not at all.  You can't
buy it.  The company has gone out of business, and there is
literally no way you can get a legal copy of those filters at
any price.  If you need that filter in your work and you weren't
one of the lucky few who bought it originally, you're out of

Had Almathera been published under a different sort of license,
I think we both agree that the Constitutional goal of promoting
science and the useful arts would have been achieved much better.

And had Almathera received a software patent for the technology
used in their filters and were the patent now in the hands of
receivers who had no idea how to sell it, but knew how to sue
someone who infringed, we'd be farther still from the goal of
the framers.

So I think, in comparison with the traditional model -- whose
faults I think we can live with, btw -- the GPL comes off quite

Now let's suppose that you go to my Free Software page on the
Web and say to me, "Bob, that little utility you wrote that puts
the HTML RGB values on the clipboard when you select a color in
a Windows color selection tool is brilliant.  I'm going to sell
that program for $50 a copy and we'll make a fortune: 50/50.  But
you've got to take the GPL off the code."

Suppose for a minute that I lost my marbles and didn't bother to
tell you that 12-year-olds can turn out that code in 10 minutes,
and I went along with your visions of wealth, changed the license
on the code, and took it off my website.

Would I have done anything I wasn't allowed to do?  Absolutely
not!  The copyright to that code is mine, and I can decide
tomorrow to start charging for it, to turn it from GPL to BSD,
or to require you to paint yourself blue before you use it.  I
can even decide not to let *anybody* have it.

But if one of the thousands who've downloaded that code with the
GPL (I presume they're all 11-year-olds) wanted to distribute a
copy they obtained under the terms of the GPL, could they do
that?  You bet.

So while the GPL doesn't contaminate the code forever, it does
set up a situation where my product, if I take the GPL off, is
in competition with a freely distributable version of the same
product.  But as the customers for just gobs and gobs of
software prove every day, that's not an insuperable barrier to

All this said, I think an LGPL license is more appropriate for
most software produced for the government.  It lets huge companies
and mom-and-pops have an equal shot at getting the software out
there to the public, and even lets people make a buck or two.

However, to say that 60 or 70 or even 90 percent of software that
we don't have actually sitting in front of us to evaluate, but
are only hypothesizing about, *should* be LGPLed is a far cry from
saying that 100% of this software *must* be LGPLed or BSDed.

That's what I'm arguing against, and I think, as a serious person,
you should be arguing against it too.

>> ad hominem
>  I don't claim your argument is bad because of who you are, but I do
>find it odd that someone who seemingly is willing to produce programs
>for hire wants to be paid and *also* to retain all rights to the

That would indeed be absurd if I'd said it.  But so far as I
know, I didn't.

But, y'know, you've already said that you don't like it when
somebody who's supported by a university or a competitor of
Microsoft argues against your points.  And you've cast doubts
on the validity of comments from employees of defense contractors
(who, I remind you again, don't speak for their companies).
Just who *would* you take arguments from?
Bob Crispen
[log in to unmask]