Re: some thoughts..
- From: "Kenneth R. Kinder" <Ken KenAndTed com>
- To: robert havoc pennington <rhpennin midway uchicago edu>
- Cc: "Kenneth R. Kinder" <Ken KenAndTed com>, gnome-list gnome org
- Subject: Re: some thoughts..
- Date: Wed, 06 May 1998 21:37:19 -0500
-----BEGIN PGP SIGNED MESSAGE-----
> On Wed, 6 May 1998, Kenneth R. Kinder wrote:
> > I disagree; resource usage IS major concern. No one wants to "pull a
> > microsoft" (bloat software) here. Personally, I'd put memory over CPU or
> > hard disk usage, so I'd be in favor of a large, but very modular file
> > manager.
> It ain't a major concern when the file manager already has near-100%
> functionality, and is only 800K; when some hyper-modular do-everything
> wonder program would be vaporware forever; when there are already a
> zillion Linux file managers, some of them quite tiny. I also don't see how
> it makes any sense to make the file manager a web browser when we already
> have a help browser and Gnomified Mozilla.
I see your point, and I don't want a web browser built into my file
manager - in fact, it's what I'm arguing against. Modularism allows you
to decide what you want where.
And I also understand the time constraint; I myself am deciding on whether
to go with a super-duper modular system on a project of mine, that would
require about 100 hours, or a simpler, more monolithic program.
BTW, we have a gnomified mozilla??? Where???
> This is a project to have a working, easy-to-use system in a reasonable
> amount of time. hyper-size-optimization when size isn't even a problem is
> not going to help. There are many more important things on the TODO list
> than re-writing Miguel's perfectly good file manager.
I agree; memory usage of gnome, at present isn't bad. And it is a fine,
fine file manager (he has a nack for great file managers). At any rate,
with folks taking about integrating the file manager with everything else,
and integrating everything else with the file manager, we need to look at
a modular approach. Do you understand my position? No need for
super-duper modules on a NORMAL file manager, but with developers talking
about a file manager with everything from FTP to SMB to ice cream :), I
have to think a modular approach would be best for this kind of thing.
> Don't worry, there is no danger of Microsoft-style bloat. There probably
> is a danger of not getting a working system in time to compete with other
> platforms, or not making the system easy enough to use.
I know, and appreciate your concern of time. Us "software idealists" are
often too quick to think about what a project should be like, and fail to
ever get it done.
> I don't mean to snap at you in particular, it just seems that every other
> day someone posts saying "I don't want to run this program" "Isn't the
> Help bloat? I don't need help, I'm too manly" etc.
If you're too manly for ghelp, use the man pages! <g>
And BTW, documentation is NOT bloat!! :) If you don't like docs, don't
d/l them!
> If you want
> functionality, you have to run software. Software uses resources. Everyone
> is doing their best to minimize resource use, while preserving
> functionality; there's no need to remind them to do it all the time.
I know.
> My $.03. Input is appreciated, by the way, and I'm also guilty of giving
> it all out of proportion to my code contributions (my contributions are
> kind of lame even when they exist) - I just don't think this particular
> topic is productive, FWIW. Let's talk about something else.
Yeah. In the interest of documentation, I'm doing a DocStructure project
myself -- to allow developers to quickly and easily write documentation!
:) And I'm taking resource usage as a lower priority than modularity and
speed of development; I'm writing it Python! :)
The advantages to writing it in Python: it gets done VERY quickly; it is
VERY modular (at least, the way I'm building it is); I get to program in
C/C++ for the parts that need it....
I couldn't help but notice you talked about CORBA a little. Which CORBA
implementation do you suggest, and what's a good way to learn its C/C++
bindings?
At any rate, thanks for the comments!
- --
H3C-N--C=N | Kenneth R. Kinder, aka Bouncing
"code=caffiene*hours"
\==/ | Ken@KenAndTed.com http://www.KenAndTed.com/KensBookmark
/
O=<__>N-C3H |
H3C-N \\O | Ken & Ted's Software
_________________| http://www.KenAndTed.com/
Software@KenAndTed.com
-----BEGIN PGP SIGNATURE-----
Version: 2.6.3i
Charset: noconv
iQCVAwUBNVEsayUbN6MeoG1RAQFn5QP/b3I7QtYZu4nCKHII344pyESwCpqJ2zhI
Pi53/uyjMRPOwGPVQi8C4ZZ4lwLMi0uWoMtH7EV+je96ghc6HOxR8fmDp0lsTABU
tFazVViaWfsj852fx4gUezg2HGWbBNzIwAFpELcSfKZEGAZZ4hHz7rJA0c4HpWCE
ocP6AEhJyP0=
=Kd3f
-----END PGP SIGNATURE-----
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]