[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: ::scr open sores
On Mon, 14 Jan 2002, simon wistow wrote:
> Hark at my witty pun!
"It's been done before."
> I keep being disapointed by Open Source.
It's a development philosophy. One of many. A lot of the time I'm
surprised that it works at all -- the whole herding cats thing -- but it
has had some remarkable, if imperfect, successes. I mean, the mere fact
that you can run a modern computer with more or less all the applications
and utilities that one would expect, and most all of it based on people's
unpaid volunteer work, is amazing. Yes, the duplication of effort and the
spotty quality of a good deal of it is frustrating, but those sports on
the record aren't enough to dismiss it outright.
I think it's safe to say that OS development has earned a permanent place
along side all the older, more traditional development philosophies. It
will never replace them, no, but it will continue to provide valuable
contributions. It may never be a true "innovator" -- Linux, Mozilla,
Apache, MySQL, and the rest are all basically reimplementations of earlier
projects, most of which were commercial or academic/research originated.
Even if every big OS project is a clone of something else, if they can do
it better than the original, or try out new functionality that the
original lacks, or at the very least get the developers of the original to
incorporate long needed improvements (such as the new found emphasis on
stability in Windows now that Linux is making it look so bad), then the OS
projects are playing an important role.
Even if they never catch on, just the fact that they prod on improvements
in the closed-ware -- compare & contrast Apache & IIS here -- then it's
worth it for the OS project to persist.
> I'd get a Mac but they crash even more, they only have 1 fricking mouse
*troll gong!*
There's a nice argument that one button is all a mouse should have -- a
pointer to select, and a button to chooose, hey presto it's easy as pie.
If that's a bit too philosophical for you, then what would be the optimal
number of buttons? Two? Why just two? Why not one for each finger? Hell,
why not have a glove-type mouse top that allows you to insert your
fingers, and execute different functions depending on which way you flick
which fingers? Confusing? No more so than the two button paradigm -- why
left click some things & right click others? I see no intrinsic meaning in
"left" or "right", and more than I would in "index finger up", "thumb
down", etc. If you want extended click-functionality, then the mac gives
it in a much cleaner way: cmd click, shift click, option click, etc. At
least something like this can be logically organized a bit.
But then you were just trolling, and I was just baited :)
> * Apache.
> It's actually pretty shit. Really. You can write your own HTTP server
> in Perl which is faster. But it does have a huge extensible API.
> So Apache 2 is better but if a commercial company to $long_time to
> release its next version which wans't as shit everybody'd scream blue
> murder. It just happens to be better than everything else out there.
But speed isn't a design goal, extensibility & throughput is! If you want
fast, try thttpd or whatever -- it'll fly but it can't be adapted at all.
And as for the long release time, I'd rather wait for a version to be
complete than on time. Granted there's a lot of balance there, but e.g.
OSX went out because it was on time, not because it was complete, and
though I understand Apple's reasons for not wanting to wait, I also wish
that it had been a lot more polished before it went out the door. So it is
with Apache 2.0: it'll be ready when it's ready, and 1.3.2x should keep
everyone busy in the meantime.
> * Mozilla
> /me raises an arched eyebrow
I'm with you here: it reduces my expensive computer to a simple stone.
> * Perl
> Tricky one here. It's incredibly useful. But hang out on Perl 5 Porters
> for a bit and you're amazed that it works. It's riddled with bugs, the
> internal design is best described as, err, baroque.
Just like ...most big, complex software projects? Rare is the application
that was fleshed out in detail right from the outset...
--
Chris Devers chdevers@xxxxxxxxxxxx
Apache / mod_perl / http://homepage.mac.com/chdevers/resume/