Any java lib. to parse .cab file?

[Andreas Leitgeb]

The whole point of GPL is to prevent anyone from closing off
(making proprietary) the GPLed code, ...

Now that's entirely agreeable. I do like the GPL, and I've created
GPL'ed software in my spare time even without a legal need.

But it's about the *code*!

What we're disagreeing on is the interpretation of what is
a derivative work, and what is not. Since my last question
there have been posted urls about clean-room conditions, but
so far no excerpt of the GPL itself, that would be explicit
about this. (no, I didn't look at the wiki-articles; they
*aren't* the GPL, and the GPL doesn't contain any of the words
"room", "clean", "taint" according to a grep I did. The word
prefix "deriv" only appeared in version 2 but not in version 3)

 

[Mayeul]

But since he wants everyone I sell software to to be able to
redistribute it freely, he doesn't want me to be able to sell it for
enough to make a living.

If the people you sell software to made the software themselves, they
would benefit the 4 freedoms. But somehow when they pay you to make
their software they don't. One might argue there's little point,
therefore little point in supporting this business model.

If you don't want to consider a software model the GPL works fine with,
then one might argue it is legitimate GPL software doesn't work so well
for you.

GPL or other license, if one does not want to deliver working software
solutions without being paid for it, one can afford not to.

 

[Arne Vajhøj]

Apache and BSD licenses are founded in a radically different point of
view and intent than GPL.

Yep.

But the term "taint" in this thread has a different meaning than in
the GPL vs Apache/BSD license discussion.

The point here war whether seeing code tainted a person.

Not that some license taint code that is linked with it.

Arne

 

[Arne Vajhøj]

For one thing, you would never do any business with many large
companies, the sort ISVs can't afford to write off. Not very
idealistic, I know, but it's a fact. So at times I've had to
re-implement stuff I couldn't find an Apache or BSD-licensed version
of, rather than use or adapt the perfectly good GPL'd version, all
because Richard Stallman thinks that what I do for a living is
immoral.

This is getting a bit off topic.

I have never seen any evidence that RMS is against people
making money from software. He is just insisting (very insisting !)
that you should give the end user the source code and allow them
to modify it and distribute it.

Arne

 

[Arne Vajhøj]

Stallman came up with the idiotic "It's immoral to charge money for
software you put time and effort into cretaing."

Not at all.

RMS and FSF is fine with you being paid for your time or
you selling your software. As long as you distribute the
source.

> Without him, I doubt
that many people who want to contribute their software to the world
would bother to distinguish between commercial and non-commercial uses
of it. I wouldn't.

What?

There are not a single open source software that makes that
distinction. Because it is explicit forbidden in the open
source definition.

The "not for commercial usage" is not coming from RMS/FSF or
any other part of the open source movement.

Arne

 

[Arne Vajhøj]

My code downloads wsusscncab2.cab file (http://go.microsoft.com/
fwlink/?LinkID=74689) and extracts it. Currently I am identifying the
OS and either using "cabextract" on *NIX OR "EXPAND" on Windows and
now we are removing the dependency on 'cabextract'(as user has
explicitely install it). To unifiy the approach I feel to use some
java lib. to extract and which works on any OS.

I still don't see the point in extracting windows specific
files on *nix.

Arne

 

[Arne Vajhøj]

What we're disagreeing on is the interpretation of what is
a derivative work, and what is not. Since my last question
there have been posted urls about clean-room conditions, but
so far no excerpt of the GPL itself, that would be explicit
about this. (no, I didn't look at the wiki-articles; they
*aren't* the GPL, and the GPL doesn't contain any of the words
"room", "clean", "taint" according to a grep I did. The word
prefix "deriv" only appeared in version 2 but not in version 3)

GPL and a commercial license is very different in the terms
between the supplier and the user, but they use the exact
same legal mechanism as foundation: copyright.

There are no reason to believe that whether a reimplementation
after seeing code is a copyright viloation should depend on
what license the code is under.

Arne

 

[Lew]

I still don't see the point in extracting windows specific
files on *nix.

I, too, am curious about this matter. We have seen a circular argument - we
need to extract .cab files because they're on the *IX system because they
download a .cab file to *IX, but no word on the value of .cab files themselves
in a *IX environment once downloaded. I am left with the impression that the
..cab file is downloaded without regard for whether it's a good idea to do so,
and that it probably isn't a good idea in the *IX environment.

 

[Lew]

Of course it is. If anyone I sell a piece of software to has the
right to distribute it freely, the first copy I sell terminates my
rights as its owner.

"What are you doing in my house?"

"Remember the Krauses? You leased them the place in July, when you
were back East? Well, they leased it to me for the rest of the year."

"They had no right to do that!"

"What's wrong, don't you believe in Open Housing"?

Flame war! Flame war!

 

[Arne Vajhøj]

Of course it is.

No.

Check item 6 in:
http://www.opensource.org/docs/osd

If anyone I sell a piece of software to has the
right to distribute it freely, the first copy I sell terminates my
rights as its owner.

Not at all.

You put the source under an open source license and that code
is forever under that license.

In fact open source could not exist if that was not the case.
It is your rights that makes the open source valid.

Arne

 

[Arne Vajhøj]

But since he wants everyone I sell software to to be able to
redistribute it freely, he doesn't want me to be able to sell it for
enough to make a living.

Redhat is making 650 M$ a year on open source.

I would expect you to be able to make a living.

Software get sold with hardware, software support contracts are sold,
custom software is developed.

The only business model that does not work well with GPL is the
write the software, burn a million CD's, wrap them in plastic,
sell them and you are done with the customers until next product.

Among the big IT companies the only company where that is the primary
business model is Microsoft.

Not surprisingly they are not too keen on GPL.

If you happen to be using that business model as well then you
will not want to use GPL.

That is your privilege as copyright holder.

RMS may think that is a bad model for the end users. But he can
really not do anything about it. The same copyright that you use
to not pick GPL is the same copyright he uses to enforce GPL
on his software.

Arne

 

[Nigel Wade]

Mike Schilling wrote:

Stallman came up with the idiotic "It's immoral to charge money for
software you put time and effort into cretaing."

No. He came up with the perfectly reasonable "It's immoral for you to charge
money for software which *someone else* put time and effort into creating, and
allows you to use for free".

If you want to charge for your software which makes use of code someone else
owns then you need to obtain the correct license from them. It may well be that
you can pay the owner of the code to use it in your application. That's up to
the owner to decide. Why should you be allowed to make money, but deny others
that right?

 

[RedGrittyBrick]

Mike Schilling wrote:



No. He came up with the perfectly reasonable "It's immoral for you to charge
money for software which *someone else* put time and effort into creating, and
allows you to use for free".

If you want to charge for your software which makes use of code someone else
owns then you need to obtain the correct license from them. It may well be that
you can pay the owner of the code to use it in your application. That's up to
the owner to decide. Why should you be allowed to make money, but deny others
that right?

Perfectly put.

 

[Mike Schilling]

Mike Schilling wrote:



No. He came up with the perfectly reasonable "It's immoral for you
to
charge money for software which *someone else* put time and effort
into creating, and allows you to use for free".

If I read a journal article describing an algorithm, it's perfectly
moral for me to implement that algorithm and sell the result. It's
also perfectly moral for me to sell a Java program, which is useless
without the JVM and rt.jar that Sun puts time and effort into creating
and then gives away for free. It's also perfectly moral for me to
repackage and modify Xerces and Xalan, and sell a product that
includes the modified versions of those. I've done all of those
things, and don't feel at all dirty.

I agree that the GPL is based on the proposition you've stated, but it
reflects a feelings that selling software is itself immoral, and
that's crap.

 

[Mike Schilling]

Redhat is making 650 M$ a year on open source.

That is, by packaging and supporting software written by others.
That's a fine business and I wish them well, but it's not the business
I'm in.

 

[Andreas Leitgeb]

import i.am.not.a.lawyer.Disclaimer;

No, not even that!
You can even legally earn money with someone else's GPL'd software,
as long as you pass along the source (on their request).

However, customers are more likely to actually pay *you* for someone
else's work (rather than directly using that work for free) if you
add some value that they'll be willing to pay *you* for.

Once you sell GPL'ed software, the customer has the *freedom* of
either continuing to pay you e.g. for further maintenance, or
finding someone else for further maintenance based on the sources
that the customer has a right to get from you (after you decided
to include GPL'd code into your work).

That's the freedom (of the customer!), Stallman and others have in mind.

 

[Seamus MacRae]

Mike Schilling wrote:



No. He came up with the perfectly reasonable "It's immoral for you to charge
money for software which *someone else* put time and effort into creating, and
allows you to use for free".

No. He came up with the perfectly reasonable "It's immoral for you to
lock up and not share your improvements to code which someone else
provided to you for free on the condition that you share such improvements".

Red Hat mainly sells support services, but they do make some money
selling actual plastic disks with Linux on them. According to your
interpretation, this would be immoral according to Stallman. I'm not
aware of him complaining, however, and the GPL doesn't forbid it. It
just forbids Red Hat making it closed source or having a distribution
monopoly. Interestingly, they still manage to make some money selling
the disks WITHOUT the monopoly. (Recording and movie industries, take note.)

 

[Andreas Leitgeb]

Not even that. The _goal_ of RMS is to avoid a lock-up situation where a
user of a (software-based) system has no recourse in case of
malfunction, due to a monopoly on knowledge related to that software,
and unwillingness from the monopoly holder to share that knowledge. [...]

Very well written (the whole of your post, not just the quoted paragraph).
Thanks! I see it the same way.

PS: if it isn't already, it should also be preserved at a public place
more stable than newsgroups. (I save a copy of it locally, anyway)

 

[Lew]

That is, by packaging and supporting software written by others.
That's a fine business and I wish them well, but it's not the business
I'm in.

Your two posts above seem to contradict each other. First you complain that
you cannot redistribute GPLed software for money, then you claim you aren't in
that business.

 

[Lew]

import i.am.not.a.lawyer.Disclaimer;

A completely unnecessary and paradoxical disclaimer, as only a lawyer would
need it. Unless you actually claim that you *are* a lawyer, no one can rely
actionably on your advice as if you were one. This is particularly true for
someone posting in a computer-programmers' forum, where the presumption is
that the participants are not lawyers. This is a free-speech forum expressly
for the purpose of people offering opinions for which others do not pay and
for which the grain-of-salt rule is understood.