A case of FreeBSD users shooting all FreeBSD users in the foot

[Terry Lambert]

Roger 'Rocky' Vetterberg wrote:
> Lets say that my friend would start a bussiness, which is not totally
> hard to believe since he is a really smart guy with sometimes brilliant
> ideas. What do you think the odds are that he would base his bussines on
> Open- or FreeBSD?
> My guess is that it would be yet another bussiness turning to Linux
> instead of BSD, due to reasons that have nothing to do with the
> capability of the os.

That really depends on whether he puts his personal or business
interests first, doesn't it?

If your friend really is smart, he'll pick the OS that best deals
with the intellectual property issues surrounding his core value
invested in his "brilliant idea".

For example, the current legal harrassment of Linksys over the
Linux core for their router software in an attempt to get code
unrelated to the core OS released under the GPL would be enough
to put me right off the idea of Linux.  That, and the SCO attempt
to revoke the UNIX licenses from IBM and SGI, and the implied
threat against anyone else that uses Linux in their business.

You can ignore this type of crap when you are an SGI or an IBM,
but if your pockets aren't so deep, because maybe you are in the
first round of funding from a V.C., and your entire budget is
$3M, or worse, you've done an angel round for $250,000 and have
yet to get a first round, or even worse than worse, you are self
funding because you and one or two other people have quit your
jobs and are living off of savings while you try to build your
company, then the FUD alone is enough to eat your entire wallet
in legal fees.

Personally, I'd only use Linux if nothing I was doing was going
into the kernel, and the system was only being used as a platform
-- in other words, when no part of the system which I considered
a strategic asset could be used to throw code that I intended to
maintain in house and proprietary, and the only code involved
with the  GPL was tactical.  For example, like TiVO (and definitely
*not* like the Linksys product)... or SAMBA, which is *all*
tactical code, with nothing strategic there at all.

Other than that, I think there's just too incredibly much in terms
of value add that could be done at the kernel level for pretty much
any serious product.  For example, if I spent time optimizing the
networking stack to get the performance up, then I could use much
less CPU, which might in turn allow me to get rid of the need for
a fan in the box, which in turn drastically lowers my overall COGS,
my MTBF, and my overall risk.  At the same time, it raises my profit
margins.

Frankly, IRC is a sewer.  Your friend would  be very foolish, indeed,
if he weighed all of that as being equal to something some 14 year
old typed on IRC, and ignored all of the business considerations for
what, after all, is supposed to be a business venture.

-- Terry