the real problem?
there's too much open source. not not enough, all agenda driven by the people making it, with not a hint of an invest-able opportunity.
That's the headline, now here's what I mean.
How many variants of Linux are there? how many are actually needed?
how many forks do exactly the same as a different fork, but the developers didn't get along so they fork a project and both follow the same road map competing for the same users in the same limited space throwing confusion into the market?
How exactly is your average middle manager mean to pick the wheat from the chaff? how do they know which one to go with? what support arrangements are they getting? so many open source projects are brilliant but are written by only a few people.
What happens when project developers have disagreements and a project just dies? (taking a useful tool with it).
damn small Linux is a perfect example of this, a really nice fast tiny OS, that stopped development after the guys that were developing it had a spat and went their own ways.
yes the project was forked and went somewhere else, but that's hardly the point, if I'd invested my time specifically developing for this fork, and the new fork IS different then where does that leave me? wasted work, uncompleted project, no more time being spent helping me?
lets say I've embedded this software into products half way round the world, it's not like I can just recall and redeploy lots of devices...
sure I can follow the new fork, but what if the developers fall out again? where does that leave me now?
what if my CLI driven service suddenly gets a new developer who want's to move all the configuration to a GUI driven interface but I don't want that on my server, do I have to stop using the service, never update again, expose myself to all the risks that a new exploit will be found and I can't patch...
I say that open source is un-invest-able not from the point of view of investors funding the software house but from the point of view of people picking up the technology...
with no clear road maps given by the developers, lots of OSS being developed by people in their back rooms when they have the time, how do I know what the future will hold?
(to be fair this perhaps doesn't apply to Redhat, or Suse, etc) if the future of a project is uncertain then how can I ever justify to a CTO that we should go this route? what happens when the project folds for whatever reason? then your whole business is left out in the cold whilst you scrabble to find another OSS project that will complete your needs.
-note I'm talking about business using the software, not development houses.
how is your average accountant going to decide what OSS package to use?
I'm not saying that OSS is impossible, just without clear heavyweight winners in place it's difficult for the small businesses to decide.
I suppose the simplest way to describe what I mean is this.
hypothetically, I'm an IT manager in a school, what version of linux should I roll out as our desktop OS. plus what office software, what presentation software what art software? in the technology shops, what CAD software, etc...
that's before you've even got to the office side of things and looking at accounts packages attendance registers, server stuff to hold it all together.
The answer to the question is not important, what is important is that I could ask here (a magazine full of technical people who would know what they are talking about).the point is that if there were 100 answers they would all be different.
there is a phrase that nobody ever got fired for buying IBM.
nobody ever got fired for buying Microsoft either, because whilst it costs money, you can at least depend on them to be there to answer the phone and fix your problems, even if it does cost a lot to log a ticket.