Re: GPL vs. sweatshops (was Re: Bias)

On Tue, 11 Jun 2002, Steve Langasek wrote:

> So you believe that the number of hobbyist Open Source programmers in the 
> world far outstrips the market demand for writing software?

Certainly not.

> In any case, my question concerned the compensation of those *employed*
> to write Open Source software, versus the compensation of those employed
> to write proprietary software.

I imagine the salaries would have to be comparable to get decent employees.  
However, those salaries could be coming from venture capital based on the 
promise that the open-source business model will be successful.  More than 
a few such businesses have failed already, taking those jobs with them.

> Open Source or not, the fact remains that if you want a business-critical 
> piece of code written, the only way to make sure it gets done on your 
> timetable is to pay someone to write it -- that's true whether you're an 
> operating system vendor or a manufacturer of fine industrial equipment.

But this is exactly my point about "free beer" problems with Open Source.  
You still need to pay people to write software if you're not willing to 
wait indefinitely for completion.  Unfortunately, "free beer" provisions in 
the license diminish the ability to derive revenue from the software to 
actually pay for the development.

Yes, sometimes it's important enough to other business needs that paying 
for it completely is worthwhile.  However, in such cases, it may well be a 
competitive advantage, and being forced to release the code might turn it 
into a relative competitive _disadvantage_, because your competitors can 
achieve the same crucial business benefit from the code, while you alone 
bear the cost for having written it.  In this case, you're better off with 
the code kept as a trade secret, but the GPL doesn't really permit that.

The Open Source benefits of "many eyes" for debugging, better security from 
code transparency, and collaboration among all users of the software are 
all real, pragmatic benefits.  None of these depend on "free beer", only on 
receiving the source and having rights to share patches with other users of 
the same code.  Allowing any user to use all the code gratis is part of 
Stallman's moral code, but it's not much a part of the pragmatic arguments 
for the quality of open source code.

> For what it's worth, I *do* believe that the supply of Open Source
> programmers far outstrips the demands for new and different *operating
> systems* and *desktop productivity software*, and I believe Open Source
> will eventually (though it will take a while) eliminate markets for
> proprietary versions of some of these types of software.

I'm not sure that's true.  I think those markets have been artificially 
suppressed by Microsoft's actions, so it's really hard to gauge just how 
vital those markets would be without monopolistic influences.

> But I'm not going to shed any tears when it happens, because while more 
> money flows from the end user to the pockets of Microsoft than to any 
> other software company, I believe of all software firms they 
> proportionally return the least value to the economy in exchange.

Of this I'm certain.  Microsoft exists to enrich its shareholders and make 
Bill Gates into the (effective) king of the world.  That man's ego knows no 
bounds, even moreso than Stallman.

> > Are you saying that the GPL does, or does not eliminate market imbalances?
> 
> That it would.

It creates an imbalance between the empowered (those willing and able to 
use and benefit from GPL code) and the disenfranchised (those unable to 
participate and forced to compete against this ever-growing codebase that 
the empowered need not pay for).  Maybe BSD eliminates imbalances by not 
playing favorites, but the GPL emphatically plays favorites and it shows.

> It takes hard numbers before it should be considered a serious danger.
> Theoretical discussions of what might happen given the right market
> conditions, that may or may not ever exist, don't hold my attention.

Very well then.

> Then you are far more misguided than I thought, because Transgaming never
> *had* an "open-source business model", except to the extent that it
> benefitted them more to return some of their non-core enhancements to the
> public tree than it did to maintain a complete fork on their own.

Transgaming publicly claimed that they were attempting to implement the 
"Street Performer Protocol" as their business model.  This model was 
proposed at least three years ago and recommended as a supposedly viable 
model for funding development of public works.

Now, rather than setting a particular price for each piece of code, they 
set a threshold of ongoing subscribers -- once that particular amount of 
revenue was coming in, all their finished code would be released under the 
Wine license, and all new code would continue to be released on an ongoing 
basis as long as that subscriber base remains above the threshold.

I see nothing inherently bad about this business model -- it's reasonable 
to establish a bargain like this with the community -- we'll devote our 
time, effort and skills to enhancing this public code, but only if those 
people hoping to benefit can band together and give us enough money to make 
it worth our while -- and by the way, here's how much that is.

The flaw, if any, in this model is that it trusts too much in users to pay 
when they could benefit more by being one of the freeloaders after someone 
ELSE meets the subscriber threshold.

Transgaming could have refused to write anything until the threshold was 
met, and nobody would be complaining that they hadn't released any code, 
but because they took the risk of making the investment in coding BEFORE 
receiving the requested payment for it, they've been maligned.  Is that 
really a fair response?

> While I for one don't entirely approve of Transgaming's licensing practices,
> and while I did favor the Wine conversion to the LGPL, I don't harbor any
> ill-will towards Transgaming, and I don't believe that most participants
> in the Wine community (which I'll be the first to admit that I am /not/ of
> late) do either.  I would like to think that most of the condemning
> comments Transgaming has had to weather have come from outsiders, not from
> those who have actually invested time and code into the Wine project and
> who are therefore in a position to appreciate the positive contributions
> that Transgaming *has* made.  I suspect Transgaming also feels this is the
> case, since they have continued to participate in the community in spite
> of any negative reactions they might have endured.

Well, I don't know about the composition of Transgaming's critics as a 
whole, but Alexandre seems to have been quite upset with them, based on 
statements he's made and his disparagement of their business model -- and 
as the leader of the project, he can hardly be written off as an "outsider" 
unaware of their contributions.  I'm sure he has his reasons, and maybe 
he's in the minority among Wine developers; I don't know.  However, he is 
certainly not the only outspoken Transgaming critic out there.

> I also suggest that you provide evidence to support your otherwise
> slanderous suggestion that the switch to LGPL was motivated in any large
> part by a desire to harm Transgaming.  While some of those voicing support 
> of the LGPL may have had this on their minds, I certainly did not:  I 
> favor copyleft, but I willingly made my (meager) contributions to Wine
> under the standing license and I do not begrudge anyone the benefits
> they might have reaped as a result.

Excuse me?  I didn't say anything about what the motivation was for the 
switch to LGPL.  I said that Transgaming evidently made an honest attempt 
to follow a recommended open-source business model, it failed, and the 
community has treated them rather poorly for not releasing their code 
anyhow.  That's just adding insult to injury.  If Transgaming actually 
broke any promises, nobody has been able to point them out -- they've only 
pointed out statements where others jumped to unwarranted conclusions which 
then proved false.  That's not really Transgaming's fault, even if they 
could have been a bit more clear about their intentions.

If Transgaming did anything clearly wrong, please, point to it.  Everyone 
who's tried pointed at ambiguous statements that were misinterpreted by 
those who later complained.

Deven