Re: os discuss (http://www.osopinion.com/perl/story/7263.html)

[Robert G Brown <bob@xxxxxxx>]

>< I have significantly snipped down Bob's epic post >
>< of a few days ago.  Most of what I snipped either >
< now resonates with my own thinking or appeared to >
><....evasive tactics on Bob's part.
I so prefer the term "evocative" to "evasive". I don't
evade issues, preferring to EVOKE a clearer response
before going on. Sometimes I do that by direct questioning,
at other times I use various forms of verbal theater to 
clarify something...

>>>Open Source may be a nice tool for felling a giant, but I'm not so sure
>>>it's a viable model in the long run.
>>Well, David went on to become quite an effective King, and even though
>>he made some mistakes,... business models that support the free UNIX 
>>systems have had quite a run so  far,...

>Having quite a run doesn't imply success.
Quite true... of ALL the "Business Models" we've seen.

>The Soviet Union had quite a run, but we eventually saw that some
>underlying assumptions failed and the entire model eventually
>collapsed.
This is a somewhat extreme case, but a good one. The Soviet system
underwent extensive and almost constant change from the moment the
Bolsheviks took power in 1917. The extent to which this "Business
Model" changed (and why) makes it almost completely silly to com-
pare this with the evolution of Open Source or Free Software models.
Our system has ALSO undergone extensive changes since its inception, 
but has on the whole changed a lot less than the Soviet system ever did,
been around longer (changed more slowly) and has proven more resilient to
tough times and change. Perhaps this is why we like to think that it is
"better". I do remind everyone that many still refer to this system as
"The American Experiment", which is still evolving. Open Source Software
is a part of that experiment.

>I'm concerned that this is the feeling of many; that the current success
>of the OS/FS movement implies its long-term viability.
It is NATURAL to assume this, and I agree that it is not nessesarily
correct. Advocates of OS/FS believe that this is true, and no prudent
opponent can assume it is untrue.

When I was 13 years old, my Uncle gave me a book, called "Toward the
Year 2018", which was a collection of essays by leading authorities
in their fields, all having been asked to write about progress over
the next 50 years. Most all of them predicted that they would make 
mistakes, and now, 33 years later, some of them are easy to spot!
The most glaring one was that EVERYONE assumed that the Soviet Bloc
would remain in place... NO ONE discussed the consequences of the
evaporation of the Warsaw Pact and the breakup of the Soviet Union.

When something like this (IBM, Microsoft, the USSR, Free Software,
the Beatles) is having a good run, it's hard to project the world
without it....

>It is important to look at the underlying motivations behind
>the movement to see if they will hold under most conceivable 
>conditions.  
Agreed. I think they do. We don't have to wait around for proof, 
it's already all around us.

>Saying that a model is flawless says that you haven't dug deep 
>enough to find that case where the model doesn't hold.  
Yes, there are always some exceptions.

>Bob, if I may be so bold as to say, believes that there are cases 
>where the OS/FS model doesn't work.  
I'll be bolder.. I can cite cases where I KNOW it will not work.
There is software that embodies trade secrets which CANNOT opened
without revealing things a company (or a country, in which case we 
tend to call it "National Security") cannot make public. I'm not 
referring to software companies protecting how they coded a copy-and-
paste word processor, but rather things like pricing plans for 
products, recipes (like Coca-Cola), and compensation agreements
that give an organization it's position in a market. A lot of 
companies in financial services have comission plans that, if made 
public, would in effect be blueprints for having their competitors
recruit all their worthwhile employees! 

These are all cases where opening the software causes harm to the 
organization that originated it, not due to the opening per se, but 
rather because of what the software contains.

HOWEVER, I feel that he above is very much an exception to the rule. 
Software can be constructed so that the really secret stuff is confined 
to a module, isolating it from everything else. The REST of the software
may be released under the GPL, and a stub module can be supplied. 
Anyone who wants to do so may then write their own module, as needed
(presumably incorporating THEIR secrets).


>The article that started this thread implies that there are no such
>cases.
I'd have to reread the article to see, but in any case I feel no 
sense of obligation about defending the article in general, or that 
position in particular. If I met the author, I'm sure I could show
him cases (like the above) that he probably assumes, or simply does
not account for.


>It is important to close ones eyes to blind idealism
>and step back to see what works and what doesn't.  
Quite true. As a business owner/operator, I can't afford a lotta 
idealistic claptrap. For a long time, I felt that this FS stuff
was largely for Academia, but in 1993 I woke up and saw that it
had potential elsewhere. 

>My feeling, at first, was that the OS/FS model derives
its motivation from a strong anti-corporation bias.  
Yes, some are. Most successful players in the FS movement are
so in a largely corporate environment. 

>Also, one of the biggest sticking points with me is
>that a bigger ego doesn't necessarily put food on the
>table.  Ultimately, we must stop living off our parent's
>income and provide a living for ourselves.  
That's true, so much so that no one claims it happens.
Eric Raymond, it ought be pointed out, has been a successful
consultant and contract programmer for years BEFORE he was
working on any FS projects, and he has incorporated that
into his business. Eric never "lived off" his parents
income.

>...there is no place for a community of hackers who do nothing
>but work on free software, for free or for ego alone.
If there aren't, they will wither away. I think there ARE
reasons to work on at least some of this software, and some
of those people are really benefitting from what they have
written. 

For example, who is going to write BIND? This is the very
backbone of DNS, yet it was supplied FREE to everyone. Is
the author better for it? He thinks so, and I agree. 

The corporate worlds is full of people who were in essence
ripped off by corporations who never compensated inventors
for their work. If these inventors had used their inventions
to build infrastructure and reputation, they might have been
a lot better off. The inventor of FM radio, for example, 
fought corporations tooth and nail for years in the courts, 
and was ultimately driven to suicide. Would it have been better 
for him if he had worked WITH corporations, build a career on his
reputation, and created win-win-win situations? Probably.

It is very hard to judge where the next innovation (Allchin
notwithstanding) is going to come from. If someone can afford
to sit around and write free software, who are we to decide that
they're not doing the right thing? I'm glad you weren't in charge
of things at CERN in 90 and 91... you might have thought that
some physicist on the staff and diddling with some ideas for
electronic publication was fooling around with something without
need....

Ask yourself if you believe there are cases where the
OS/FS model doesn't work.  If you do, name one, for
your own benefit.  Ask yourself why you think the OS/FS
model lasts for the long term.  

>Can you envision a set of circumstances in the future where 
>the model doesn't hold?
Outside the commercial shrink-wrap-and-warranty market (which
is under 5% of all software creation), and other than my cases
above....not really.

>...my gut feeling is that there IS something wrong with it.  
>But, this may be a product of my money-oriented thinking.  
I think it's a product of your "industrial-style" thinking. We
live and work in a service economy, where most software activity
does not exist in a vacuum, but in a customer-professional
service relationship. Money is exchanged for the service, not the
software, which is a vehicle for delivering the service. You need
to realize this, since it's the basis of your own position.

                                                 Cheers,
                                                 ---> RGB <---