GPL defenders frequently attack any other options as unholy. In
particular, they love attacking the BSD license as "not really
sharing", because it can be incorporated in proprietary products,
therefore being "stollen" or "not free."
It goes something like this:
"GPL is (implies) sharing. So if you have the option of using the GPL
together with non-GPLed software, you're not *really* sharing.(*))"
(( P -> Q ) /\ ~P ) -> ~Q
GPL "is really sharing" non-GPL(**) "not sharing"
It's called the *fallacy of denying the antecedent*.
(*)( The name of the fallacy refers to the negation of P, which
appears as a premise - the whole premise that, e.g., proprietary
software can't conjoined with the "pure" GPL. This very
fallacy is probably the origin of the viral enforcement of the license.)
(**)("bad","pernicious","proprietary")
Best regards.
HL
HL wrote:
> GPL defenders frequently attack any other options as unholy. In
> particular, they love attacking the BSD license as "not really
> sharing", because it can be incorporated in proprietary products,
> therefore being "stollen" or "not free."
Here is my favorite 'flaw' in the GPL attack on the BSD license:
You cannot STEAL what has been given to you.
There it is. Take my BSD licensed code, do what you want with it.
Make a new product, leverage my work, whatever. Good luck to you. You
can't steal what I gave you to begin with. You didn't steal it from
me, you didn't steal it from the community.
Of course, anything you ADD to that code can be your own. The concept
and design for making a lightbulb is free for anyone to use. But
Company X can add their own filament materials and design, and make
their lightbulb better than the one Edison designed. Of course, they
also assume the risk in doing so.
Once you take the code I 'gave' you, add proprietary stuff,
improvements, whatever... you now have to assume the risk involved in
convincing the world that your additions are worth using and/or buying,
rather than the original base code, and thereby recoup your investment
and make a profit. Or just gain market share with a closed product.
And you will no longer benefit as easily from fixes/update/improvements
to the original code. And someone else can always take the original
code, 'extend' it with top-secret stuff, and compete against you. They
might even clean-room copy your features. But that's the price of
freedom. In my opinion, TRUE freedom, as far as it concerns open
source software.
"jmckitrick" <···········@yahoo.com> writes:
> HL wrote:
> > GPL defenders frequently attack any other options as unholy. In
> > particular, they love attacking the BSD license as "not really
> > sharing", because it can be incorporated in proprietary products,
> > therefore being "stollen" or "not free."
>
> Here is my favorite 'flaw' in the GPL attack on the BSD license:
>
> You cannot STEAL what has been given to you.
>
(...)
Exactly.
>
> Of course, anything you ADD to that code can be your own. The concept
> and design for making a lightbulb is free for anyone to use. But
> Company X can add their own filament materials and design, and make
> their lightbulb better than the one Edison designed. Of course, they
> also assume the risk in doing so.
Sometimes these additions are only relevant to a small subset of
situations. For example, there was some complaining regarding the
Linux kernel as being to bloated because it had to be installed
everywhere, on every weird device possible it supported. This is the
kind of enforcement of the software license that does not make a
whole lot of sense.
>
> Once you take the code I 'gave' you, add proprietary stuff,
> improvements, whatever... you now have to assume the risk involved in
> convincing the world that your additions are worth using and/or buying,
> rather than the original base code,
(...)
Right. So, in order to do that, you really must come up with a
productive that's really competitive.
> And you will no longer benefit as easily from fixes/update/improvements
> to the original code.
This aspect actually is the one aspect that makes one want to
contribute back to the original codebase. If you fork too much, it
becomes harder to keep up with the general improvements made by the
community. It's probably why Apple contributed back to FreeBSD.
> And someone else can always take the original
> code, 'extend' it with top-secret stuff, and compete against you. They
> might even clean-room copy your features. But that's the price of
> freedom. In my opinion, TRUE freedom, as far as it concerns open
> source software.
I believe there's sort of like a tipping point, beyond which the
proprietary fork becomes irrational (the difficulty in maitaning the
patchset in sync with original codebase). The GPL, on the other hand, has
a huge flaw, in that in permits dual-licensing. Which means you
contribute to someone else who then dual-licenses and makes money
with your work.
HL
> I believe there's sort of like a tipping point, beyond which the
> proprietary fork becomes irrational (the difficulty in maitaning the
> patchset in sync with original codebase). The GPL, on the other hand, has
> a huge flaw, in that in permits dual-licensing. Which means you
> contribute to someone else who then dual-licenses and makes money
> with your work.
>
1. You submit your work as GPL only : it can only be used in GPL or GPL
compatible licensed works, not in the above dual-license.
2. You submit your work as equally dual-licensed, it can be used in the
above situation.
Besides (to be pedantic), what has this discussion to do with Lisp ?
Regards,
Jurgen
On Fri, 15 Sep 2006 20:08:26 -0300, HL wrote:
>
> GPL defenders frequently attack any other options as unholy. In
> particular, they love attacking the BSD license as "not really sharing",
> because it can be incorporated in proprietary products, therefore being
> "stollen" or "not free."
>
> It goes something like this:
> "GPL is (implies) sharing. So if you have the option of using the GPL
> together with non-GPLed software, you're not *really* sharing.(*))"
>
> (( P -> Q ) /\ ~P ) -> ~Q
> GPL "is really sharing" non-GPL(**) "not sharing"
> It's called the *fallacy of denying the antecedent*.
That misrepresents a straw man's argument. (a fallacy you
really *should* look up). A translation of the real argument
from natural language requires the predicate calculus, which you will not
be taking next semester, since you obviously failed even the sentential
calculus. It is simply a particular instance of the statement
that -- a quality of a thing can be compromised by a
quality of one of its constituents.
For-all-p (if ((p is program)
and
(x in p)
and
(x is ~free)
)
then (p is ~free).
mbstevens <·············@XmbstevensX.com> writes:
> On Fri, 15 Sep 2006 20:08:26 -0300, HL wrote:
>
>>
>> GPL defenders frequently attack any other options as unholy. In
>> particular, they love attacking the BSD license as "not really sharing",
>> because it can be incorporated in proprietary products, therefore being
>> "stollen" or "not free."
>>
>> It goes something like this:
>> "GPL is (implies) sharing. So if you have the option of using the GPL
>> together with non-GPLed software, you're not *really* sharing.(*))"
>>
>> (( P -> Q ) /\ ~P ) -> ~Q
>> GPL "is really sharing" non-GPL(**) "not sharing"
>> It's called the *fallacy of denying the antecedent*.
>
> That misrepresents a straw man's argument. (a fallacy you
> really *should* look up). A translation of the real argument
> from natural language requires the predicate calculus, which you will not
> be taking next semester, since you obviously failed even the sentential
> calculus. It is simply a particular instance of the statement
> that -- a quality of a thing can be compromised by a
> quality of one of its constituents.
>
> For-all-p (if ((p is program)
> and
> (x in p)
> and
> (x is ~free)
> )
> then (p is ~free).
Outch!
(for-all (p)
(if (and (is p 'program)
(in x p)
(is x (not free)))
(is p (not free))))
--
__Pascal Bourguignon__ http://www.informatimago.com/
"I have challenged the entire quality assurance team to a Bat-Leth
contest. They will not concern us again."
On Sat, 16 Sep 2006 03:36:24 +0200, Pascal Bourguignon wrote:
> Outch!
>
> (for-all (p)
> (if (and (is p 'program)
> (in x p)
> (is x (not free)))
> (is p (not free))))
Yes, that would be a more lispy version.
mbstevens <·············@XmbstevensX.com> writes:
> That misrepresents a straw man's argument. (a fallacy you
> really *should* look up). A translation of the real argument
> from natural language requires the predicate calculus, which you will not
> be taking next semester, since you obviously failed even the sentential
> calculus. It is simply a particular instance of the statement
> that -- a quality of a thing can be compromised by a
> quality of one of its constituents.
>
> For-all-p (if ((p is program)
> and
> (x in p)
> and
> (x is ~free)
> )
> then (p is ~free).
Dear mbstevens --
An odd character you are. First, you come all arrogant and highbrow,
making inferences about me and addressing me in a rude manner (why, I
don't know - if this is how you get your kicks, you have bigger
issues to deal with...)
Then, you claim quantifiers are needed - which is about the most
clich� reply involved that you could possibly give("you need
quantifiers to translate natural language"). I don't think
I need to show you examples of when quantifying is unnecessary. Do I?
Furthermore, you seem to think there is only one interpretation
possible (you should really worry about this aspect).
Let's try this:
Assume, as GPL zealots do, the following equivalence (Note: I know
"zealots" is derogatory - but the term refers to the small parcel of
irrational advocates of the license as well as followers of cults of
personalities).
BSDL = non-GPL = bad
Then, the following inference rule:
1) If GPLed software is used then you are "sharing"
2) BSD is not GPLed
-----------------------------
BSD software is not "sharing"
This is (( P -> Q ) /\ ~Q ) => ~Q
I'm sorry you've never heard of it.
This argument crops up everywhere when people claim the "only" way to
share is to use the GPL. All other ways are disqualified as not in
fact promoting true sharing values. In the case of a liberal license,
they additionally claim you let "them" _steal_ from "you" (how can you
steal what is infinitely replicated?)
See, one criterion of the sentential theory of inference is that, if a
proposed rule results in a false conclusion form true premises, then
you should reject it. Like this:
I say the rule is:
From sentence Q and sentence P->Q, we infer P.
Then (the example goes):
P = "Lincoln was born in Illinois"
Q = "Lincoln was born in the United States"
Q and P->Q are true, but the fact is that Lincoln was born in
Kentucky.
If anyone misrepresents anything ("the straw man argument"), it's the
GPL zealots. There are many fallacies committed by this crowd. This is
only one.
HL
HL wrote:
> GPL defenders frequently attack any other options as unholy. In
> particular, they love attacking the BSD license as "not really
> sharing", because it can be incorporated in proprietary products,
> therefore being "stollen" or "not free."
>
> It goes something like this:
> "GPL is (implies) sharing. So if you have the option of using the GPL
> together with non-GPLed software, you're not *really* sharing.(*))"
>
> (( P -> Q ) /\ ~P ) -> ~Q
> GPL "is really sharing" non-GPL(**) "not sharing"
>
>
> It's called the *fallacy of denying the antecedent*.
>
> (*)( The name of the fallacy refers to the negation of P, which
> appears as a premise - the whole premise that, e.g., proprietary
> software can't conjoined with the "pure" GPL. This very
> fallacy is probably the origin of the viral enforcement of the license.)
> (**)("bad","pernicious","proprietary")
>
> Best regards.
> HL
Should...not...feed...the...trolls...
HL wrote:
> GPL defenders frequently attack any other options as unholy. In
> particular, they love attacking the BSD license as "not really
> sharing", because it can be incorporated in proprietary products,
> therefore being "stollen" or "not free."
Anyone making that argument doesn't understand the GPL. The BSD
license can be attacked from the point of view that using it can
improve proprietery software. But it is in no way "not sharing".
Note the FSF does not make this arguement, they call the BSD license a
"free license".