85% languages (was Re:Linux fat/bloated)

[Sy Ali]

On 4/6/06, Paul Nash <paul-fQIO8zZcxYtFkWKT+BUv2w at public.gmane.org> wrote:
>
> >Ruby is a lot older than you realise.  =)
>
> You're right, it is.  Ruby dates back to Feb 24, 1993.  I thought that it
> was a child of the late '90's.

I thought it was a bit older.. I'm reading some more into it and I'd
say Ruby is 1995 (first release appearance).


> In most cases, the documentation stinks.  That's why you should RTFS :-).
> We don't call it "open source" for nothing, you know ...

I'll start getting into that when my Ruby skills improve.  Then I'll
also start hacking away to put inline documentation (RDoc is a nice
concept, although I disagree with the markup)



> >As a user, I think that I would end up learning the tool and bending
> >it to my uses instead of working on the remaining 15.
>
> Can you say "Microsoft"?  There, it's not that hard.  Try again a few
> times, and when you can do it without barfing, pop into Best Buy and get an
> application that you can just "learn and use".

Here's the problem:

Most mainstream software (read: Microsoft / Windows) has defaults
which aren't user-definable and which are geared towards people who
have trouble walking and blinking.  I'll admit that it's gotten a lot
better.. but I wasn't enough of a drooling idiot to want to think in
the ways that were defined for me.  Plus I was unsettled at needing to
steal all the software I used.

Most mainstream linux software is built from another perspective. 
Much of the mainstream stuff (think the popular KDE apps) have sane
defaults and are very configurable and nice.  This is about where I
sit these days.

Other popular but less "complete" software can have widely varying
sanity levels for defaults but can also be very customizable.  With
effort these can be shaped to a user's desire.  This is where I was,
although I spent more time learning and configuring than I'd like to
admit.  I'm not stupid, but a lot of software sure makes me seem that
way.  =)

Fringe linux software has poor documentation and is only barely
functional, or used to be barely functional but has aged and is no
longer contemporary enough to be dropped in and used in a common
environment.  A lot of this stuff is "holy crap that's cool!" but
won't ever work out of the box.

There's lots of other stuff out there that's even crazier and which
could promise incredible things but they're just not ready for actual
use.   This stuff is pre-pre-alpha, barely done, specialized or
narrow-sighted use, etc.


Now when I say that I'd rather spend my time bending the software to
my needs than developing to get it to 100%, this is only because I'm
so desperately lacking in the qualities needed.  I end up doing
documentation, bug reports and the like rather than actual "useful"
coding.

And so as a regular user, I end up spending my time shopping for
better software or learning how to configure software better rather
than cutting it open and fiddling about with its code.


> It depends on the developer and/or development team.  *YOU* think that they
> are driven to develop a spectacular product that will install easily, work
> right out the box, and do everything that you want.  Oh, and have decent
> documentation as well.

I'd say it's a safe assumption when it's a more public project which
has a user-facing website -- i.e. not a man page turned into HTML, but
good docs, tutorials, a mailing list or forum and the like.

These are more my concern.  These are the applications which are 85% to me.

The applications which a hobbyiest would create wouldn't generally get
its own URL, attract multiple developers, have user resources like a
public mailing list etc.  These I wouldn't put so much of a
perfectionist's eye on.


For the 85s, it does seem to me that their goals include making "best
of breed" software .. things like features, configurability,
usability, documentation etc.

Stepping back and looking at all the choices.. there are cases when
I'll see bits and pieces of the "right way to do it" in one
applciation which are missing in another.. and vice versa.  This is
why I end up feeling like each application is capable of more.

However, I guess I can't fault there being so many choices..
programmers all have their tool preferences.
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml