i will swithch to Ubuntu.
i dont know when but im quite sure the day is not far away.
Currently the only things holding me back are:
- my system is perfectly configured
- i have no time at all for a reconfiguration (even under ubuntu)
- i still have some nerves left (or occassionally even the joy to tweak / try stuff out)
- from old suse days im very sceptical about version based distros. (they used to break on distro-upgrades).
- im a fan of the gentoo program versioning system (complete system is (should) be up to date.
For me gentoo could not decide what they want. The audit is splitted between
- elitists which
enjoy it differentiating semselves over the masses
with their OS choice and configuration.
- power users
- occasional users with interest for os insights
and so on
Gentoo shows no clear direction in which they want to go:
- getting k1k4zZ 3litist even more manually fixing ebuilds, fixing code, setting bugs through users
- beeing a source-driven user-friendly distribution ?
- beeing faster (until 2004 gentoo defined itself also over performance on the
main page)
- beeing more stable (with the current manpower no one talks anymore about gentoo on servers.. and there is a good reason why)
When i switch to Ubuntu it will be because:
- i have no time at all for this hassle. I have a life and working 12 hours the day at least in the IT business, and coming home tierd i really like to update my system without having to read the forums for possible emerge errors, blocks. Sometimes it just has to work without any tweaking. (Im still a huge "fan" of baselayout changes (yes even running dispatch-conf afterwards) )
- overall the ebuild quality got worser since 2004. maybe its only my feeling, but it seems also less developers are working on gentoo, and those who are have to care now of at least two very popular archs (~x86/~amd64 if you didnt guessed)
- i will still be able to tweak / compile the parts which are possibly important for me to optimize
- installing Ubuntu
for my wife on her notebook took 15 minutes, and afterwards everything worked out of the box. you know what i said: WOW !! i could not belive it. i was prepared for some hours chipset and wireless research and configuration, but hey, the only thing we had to change was using the binary graphic driver and the resolution.
And she already could start using the system productive.
Do you think she would have enjoyed "learning" all customize and tweaking stuff gentoo would have to offer ?
- Would i still enjoy that ?
- Actually even if i still want to stay in touch with some of the lower level parts of linux
i want to be in control when i want to tweak a system-part for fun and what im interested in. Having emerge blockers, unresolved dependencies and so on are not that kind of tweaking stuff im still interested in. Freedom of configuration and such should also include functionality. It should mean that
you have the freedom to search your own problem if you like, not vice versa.
but this directly also lead to
- general gentoo product Quality lowerd. Leading to these "unforseen" errors.
So if you want to make gentoo better
1) determine your target audience
2) determine your target [dekstop/server/source/stable/instable/supertweak/elitist/whatever]
3) make that target clear
so no misunderstandings will arise and less people will be frustrated.
frustration only raises where expectations differ.
this guy
here brought it also very clear (read all his posts, because his first is a little p*sed off) to a point. But reading the reactions on it shows clearly why this distribution lacks a clear product target definition.
to give a short background about me:
i grew up using computers, was one of the first Fidonet BBS in germany (running DOS than OS/2), started linux with suse 5.x. While im not a linux guru im not considering me a newbie. If i run into problems which anoy me i know a lot of other people will run into much heavier problems. Although im a project manager now and no longer on the first line of fire, i used to use several gentoo servers prodcutively as Test Engineer for inhouse and customers test-setups.
(this turned out to have been a bad choice btw. Sys-Admin team which took the systems over later, gave up trying to have the updated, after they ran into several emerge problems and such. The servers run now on debian)