Fedora.next in 2014 -- Big Picture and Themes

[Adam Williamson]

On Tue, 2014-01-28 at 20:34 +0100, Robert M. Albrecht wrote:
> Hi,
> 
> > * Although it's certainly not the only reason, Fedora as _solely_ a hobbyist
> >    desktop is not ideal for an upstream for RHEL server and cloud products.
> 
> No other system can be reinstalled / upgraded every six months. That 
> single fact IMHO kills all other use cases.

Please stop repeating this argument. Fedora's lifecycle is ~13 months,
not six.

I have been running four production servers on Fedora for five years,
and find this to work perfectly well. Upgrading them takes me a weekend
every year.

> If I need a stable Fedora-like server, I get CentOS. It's kind of a 
> Fedora-LTS.
> 
> > * General trend in Linux towards the base distribution being "boring" and
> >    not mattering. I asked several dozen different people at a gigantic Amazon
> >    conference why everyone was using the distribution they chose instead of
> >    Fedora, and the answer was almost universally "oh, I don't care; that's
> >    not really an interesting question because there's nothing important at
> >    that level".
> 
> All (of the big) distros are mature today. At the early days one chose 
> his distro by drivers, by installation-tool, by packages, ...
> 
> Nowady every distro has a working installer, has a bazillion packages, 
> ... they basically all work.

There's a huge range of stuff to fix at the distro level. Lately I want
to clone myself ten times just to work on all the stuff I see that needs
fixing.

Just to wax philosophical for a minute: I think there's a lot of value
in building boring stuff that works well, and I might be weird, but I
wake up every day enthused by the prospect of doing it. The world needs
its Lennarts, but it also needs its pocket-protector dorks who spend all
day tightening the nuts and bolts on a platform that was "cool" ten
years earlier and is now so boring, out of date, stable...and essential.

It's not cool to work on a language that's been around and used in
production for decades (Python, perl, GCC...), or an operating system
installer, or a terminal app, or a text editor people have been using
for thirty years. You won't get on Hacker News or get a breathless
write-up in Wired. You won't get a massive VC funding round. But it is
*important*.

If you look at things that have been around forever and ever but are
still actively and carefully tended - python, GCC, vim, emacs, apache,
mariadb, etc etc etc etc etc etc etc - all the boring, 'stable' bits
that you're talking about - and compare them to projects that are
marginally maintained or not maintained at all, the difference is huge.

And yes, it's much harder to sell the story "we work all day on
carefully and painstakingly improving something you've been using for
decades that already works pretty well" than the story "we're building
something SHINY and NEW!", even if the first thing is vital to the world
economy while the second is a lolcat generator. People do not always
have their priorities correctly wired, and neither do journalists
(rimshot, please!)

But perceptions are not reality, and I do think that people *do*
perceive the work done on 'boring', 'stable' projects, and differentiate
between them. They don't do so as consciously or immediately or clearly
as they look at SHINY NEW things, but they do. You see it happening
gradually, over time, and usually not as explicit statements, but rather
implied in other things.

Yeah, we spend a lot of time building the 'boring' nuts and bolts of an
operating system, and yeah, operating systems in general have got to the
point where they're 'boring', 'stable' things. You can probably pick one
blindfolded and be doing productive work on it in a half an hour (a
browser and a text editor and you're good to go!), and maybe if you're
asked about it straight up, you'll say the same kind of things as I'm
replying to here: it's a boring space, there's no real difference
between the choices any more, etc etc etc.

But I don't think we should rely on that conscious perception. The nuts
and bolts of this 'boring' operating system *are* important, and we
*are* providing substantial value by working on them, and people *do*
perceive that over time, even if they don't necessarily know it. Maybe
we don't get our Wired write-up (except when we do something really
crazy, or there's a giant political flamewar, or whatever), but that
doesn't mean we're not doing good and important work.

This doesn't mean I'm against doing Big Exciting New Things in general
or Fedora.next in particular, but I do want to stand up for the value of
just keeping your head down (hah, I know, Adam, practice what you
preach) and doing good, dull engineering work. With your pocket
protector firmly in place.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net