[Opensource] [OT] RE: Opensource Discussion [Couldn't find original thread name]

[Michael Rimov]

I want to throw my .02 in here. [It seems I always do! ;) ]

I think there's a few aspects that are getting missed here in the discussion:

1 - RedHat is NOT discontinuing Red Hat Linux, rather it is rebranding the 
freebie as Fedora and throwing a few of it's engineers to manage that project.

2 - Although RedHat is now moving into a new level trying to build 
Enterprise Linux with a heavily modified threading capability, etc, I don't 
think ANYBODY would argue that RedHat got its reputation as a good Linux 
distribution from the "freebies".   Their Enterprise Linux would never have 
gotten off of the ground without a good reputation to start with.  What I 
suspect that they're trying to do is now increate the quality of the 
enterprise Linux distribution in such a way as that it's no longer worth 
their reputation to keep the free implementation around.  They feel that if 
people evaluated their enterprise Linux based off of the performance of the 
freebie Linux, it would hurt sales.  Although, on one hand, I am not an 
advocate of what they did, I do understand that it's true that people would 
evaluate the enterprise version based off of the free kernel.

So what comes out of the opensource then?  RedHat still has engineers paid 
to work on Fedora.  And since the last Software Development survey said 
Silicon Valley programmers are in the $100K/year range, that's quite a bit 
of money they're pumping into it.

-R&D:  RedHat itself says that Fedora is going to be sort of their R&D 
version of Linux.  It's very easy to experiment with ideas in opensource, 
kill it if it fails to take and not pollute the enterprise product.

-Feedback:  Tons and Tons of feedback.  Feedback and testing is 
critical.  Expresso wouldn't be nearly where it is without the community 
feedback driving it's continued improvement.

-Patches:  There ARE a few linux red hat hackers out there, just like there 
are a few Expresso hackers out there.  Their patches, while requiring 
testing and inspection, are absolutely invaluable.  For example, Larry has 
a machine that has some features that I don't have access to.  Because of 
this, he's found some bugs that would have been absolutely impossible for 
me to find.  So again, quality goes up.

-Ease of other projects and innovation:  My predecessor once said to me 
that he was amazed at how much easier it was to make commercial projects 
since Expresso had become opensource.  The organic nature of Expresso had 
created abilities he had never thought of!

Now, as to the revenue model, consider that  MySQL makes a healthy living, 
JBoss is profitable, and RedHat is profitable [even before splitting off 
the freebie Liunx!]. [As are others, I'm sure].  Then add to the equation 
that SuSE and Ximian were acquired by Novell partly because of the fact 
that they were generating decent revenue and Novell believes it sees how to 
make money off of those company's products! [Not to mention that Novell 
still is supporting the Mono project which is so specialized that very few 
people outside of Ximian can work on it!]

	If you consider $ values, despite the fact that money often has to go into 
opensource projects, the value on return is extraordinary.  If Larry, 
Peter, Kris, Malc, Tino, Geeta, and all the others were paid an hourly wage 
for the amount of work they've done on Expresso, the resulting bill would 
be extraordinary.  [Consider much greater than $10,000/month minimum, even 
when they're busy! :)].  Or, to put it another way, if you considered two 
full time programmers to sort of make up the difference, then you're 
dealing with $200,000/year + benefits! [And the feedback and patches still 
wouldn't make up the value of the opensource project!]

I think this is where you will find the true value of these opensource 
projects.

							-Mike