what canonical did for linux is what apple did for bsd.
one day we will wake up and it will hit us "ubuntu IS Linux" all the other distro's will be the hobby linux's.
1. to fast of release cycles, six months is too fast. and the rolling release method doesn't work either IMHO, i had to reinstall arch the other night cause of a outdate repo, and installig a bunch of out of date software (about 8 months old from what i could tell), and then fixing the repo problem, and doing the pacman -Syu to get the updated software, it didn't make my system unusable, it was the little bitty annoying problems that kept happening that annoyed me.
2. bleeding edge, honestly what is the point? whats the point in having on the latest and greatest, if it is filled with bugs and issues? one of the biggest selling points to me for linux has and always will be stability, this cutting/bleeding edge mentality sorta started with ubuntu, and caught fire in other distro's.
3. hardware - there is an incredible amount of hardware that simple does not work with linux, and this isn't really linux's fault, well it sorta is. the biggest problem is userbase, nvidia has been good to the linux community in the last few years, and for one reason, whatever our user base truly is, they know the ones commited to running linux and using it daily, know that nvidia prop drivers work fairly well with linux, and most people i know who run linux, purchase nvidia cards, i don't think they give two cents about the linux users or open source, but they like a healthy profit margin, and this gives them a slight edge over ati.
so i believe it will come down to that, like mushroomboy was saying, whichever distro has the largest userbase will be the one who gets the goodies linux users have wanted, like photoshop.
4. standards - i believe this is another thing that holds the linux back, there is no standardization between the distros, and i know some will make claim that this is what sets the distro's apart, that one has a bsd style init scripts, the other uses upstarts, this one uses bizzaro and that one use tiddles the cat fur, but that drives me mad, i guess to each his own.
5. not posix - linux is not "posix", its not certified that is, and that is cause the lack of "standards" that go along with the name, the only thing that each distro have in common usually that they use the linux kernel and mostly use open source software, not much of a standard there...and posix simply isn't enough of a standard, there should be slightly tighter reigns, so a "linux OS consortium" should be created, have ubuntu, debian, redhat, novell, all the big distro's get together to set standards, i guess you can have some of the niche distro's there as well, but i honestly would only let the slackware folks in
6. better funding - that is the big one, open source is free, but the developers are not usually, for small things they typically are, but for large projects to be good or even successful they need quality developers who know what they are doing, not you 13 year old script kiddy.
7. limit amount of developers - as a programmer myself, i know how devestating having 30 ppl working on one project that only honestly needs 3 or 4 working on it, you get a lot of overlapping in code, a lot of changes that get pushed that over write a fix or others code, you actually get more delays or mess ups with the larger amount of people.
time and energy that i would spend creating such things would distract me and take away from the things i do that make money
all the software that we use for those are all done professionally, periodic updates and good priced (some of the ones we use are free, others that we use do cost but the ones that do cost are more than reasonable.)
the projects i talk about needing to merge are the more critical, specialized parts of a OS, like a sound system, its great that oss, alsa, phonom, pulse, etc, all compete with each other, but honestly, they fix each others deficiencies, but also at the same time, create new weak points that the other didn't have, so its a circle. i might be a programmer, but i don't have any exp programming a sound system, i can't imagine a whole lot do, so when several of these projects get started, with limited resources that they would have, i would imagine it would only be logical to merge them together to benefit the whole.
now for the desktops, Gnome and KDE are pretty solid (yes they both have problems and i would love to see more unity from them.) but some of these others that pop up randomly, they set a fire for a few years, then they burn out and become outdate and obsolete cause no one is working on them anymore, the time and energy spent on them could have went to one of the DE's or another project. thats all i'm saying about this subject, instead of redoing redundant work, why don't some of these guys figure out how they can help the exisitng projects or even distro's.
9. unified packages - one of the greatest selling points for me when it came to linux was the concept of the packages and repos. i agree with what they had to say in the video on this subject, and it goes with point 8. i'm not talking about package managers, i'm talking about the packages themselves, would arch run any different if it used deb packages? or would slackware be horrible if it is used rpm's?
they all do the same thing, but now we have split resources again, we have debian repo's, we have fedora repo's, we have arch repo's..etc.
don't get me wrong, i love linux, but it does have a long way to go before it is used by everyone, these are things i would like to see happen.