Re: Transgaming's promises mean little (was: Re: bias...)

On Wed, 5 Jun 2002, Francois Gouget wrote:

> I am actually quite happy with that. According to the above, now they
> are working on a proprietary implementation of the DirectX API based on
> the X11 ReWind tree. At least it does not create false expectation that
> their code will be returned to Wine/ReWind in the near future.

You know, I think this is the crux of all the animosity toward Transgaming.
The assumption that their promised code release would be in the NEAR future 
as opposed to "eventually, if all goes well".  The promise I read didn't 
really imply "near future" -- it said "if and when" the subscription goals 
were met.  That could be near future, or a very long time away, or never.

Why did everyone assume (relatively) near future, then?

> When things don't work out the way they want, companies sometimes have
> to change their business model. Sure this is tough for their subscribers
> and they should definitely send them an email of some form to announce
> the change (which maybe they did). But you have to give them some
> leeway.

Often the alternative is to go out of business, which probably isn't what 
the subscribers would want...

> It may be better to not promise anything than to make a promise which
> you know you are very unlikely to ever be able to hold.

Their promise was conditional in nature.  I've seen no evidence that 
they've actually broken it, only people assuming they have because their 
expectations of "near future" were never met.  If they had more realistic 
expectations, maybe the situation wouldn't have become more acrimonious.

Mind you, it's possible that Transgaming had unrealistic expectations as 
well, and maybe that led them to imply "near future", even if it hadn't 
been promised as such.  Presumably they expected their original business 
model to have reached their goals by now (or be getting close), otherwise 
they probably wouldn't be trying a new business model now...

> I disagree. Call me naive if you want, but I believe that Gavriel State
> and Transgaming were sincere (though I obviously cannot prove it) and
> that they have attenuating circumstances.

I tend to agree.  Everything I've written sounds sincere, but "tainted" by 
the mentality of a businessman.  Nothing wrong with that; it's good to have 
a businessman at the helm of a business, of course.  But I can see how that 
mentality could clash easily with the altruistic mentality of a volunteer.

> The source of their problems is that things did not work out the way
> they expected or are taking much longer than expected.  The issue wrt.
> Wine is the way they handled this change in circumstances. They should
> have kept the Wine community informed of their difficulty to reach the
> set number of subscribers.  Publishing their subscriber numbers as you
> suggested would have been the best way to do so. But that's a difficult
> decision to take for a company.

It's scary for a company to make it easy to see if failure seems impending.  
And yet, hiding information often causes worse speculations to take hold, 
as seems to have happened here.  I'm not suggesting they open their books 
completely, but rather just on their subscription numbers, since they have 
direct bearing on the promise they made to the community.

> Yes, I find this quite upsetting too. And I have stronger doubts about
> them ever intending to let others ditribute binaries of their AFPL tree.

If it happens on a small scale (say, among buddies), they might never 
object.  I'm sure they'd object if it happened on a large scale, as Debian 
would have been.  And why shouldn't they?  It directly threatens their 
revenue stream from subscriptions, which is evidently already disappointing 
them in the difficulty of getting people to subscribe.

Really, how could you advise a businessman to acquiesce to such binary 
distributions, given the circumstances?  Transgaming is probably very deep 
in the hole from paying developers (or perhaps developers working for the 
promise of future compensation) and running their business, compared to 
the limited subscription revenue they've evidently received.  Is it any 
surprise that they'd want to discourage easy binary distributions?

> I think it would be good to clarify the situation. And if the only way
> to clarify it is to actually package their AFPL tree for Debian then I
> believe it would be worth it.

If that happens, they'll probably change the license on future versions.

> Transgaming cornered themselves. If they contribute patches to the
> ReWind tree, then then cannot trade them against Wine patches.

It's a moot point if people refuse to trade patches on principle. :-)

Deven