Re: Looking for an editor

Dances With Crows <danSPANceswitTRAPhcrows@xxxxxxxxx> wrote:
? What are you *doing*?

Possibly, using Linux for everything and not having a single Windows
install both at home and at the lab. Last weekend I had one kernel oops
with my firewire PCMCIA card and several problems with kino (Ubuntu
Dapper). That would have been 3-4 bug reports, if I cared and if I had the
time to make the problems repeatable. Of course, the fact that the PCMCIA
+ firewire + kino worked OOTB was in itself astonishing. Yesterday I had a
problem with a scanner installation (Debian unstable). Thats just three
days. I do not even want to start describing problems specific to our lab,
which is fully Linux based, with one or two serious servers and storage
units, dozens of workstations, lots of services, and integration in the
campus Windows network. OK, so that isn't necessarily impressive, but I am
not a computer scientist and this is not a computer science lab.

And no, I am not complaining. I am just tired of hearing "you should have
sent a bug report" from Linux converts who probably still have Windows
floating around somewhere to do most of the real work and gaming (nothing
personal). Filing a decent bug report is some work, and while I do that
occasionally, I do not have the resources to do it for every bug I find
out there. Especially especially not for packages that

Running things that would be marked with
"unstable" keywords in Gentoo? I've been using various distros since
the middle of 1999, and have faced tons of problems. "Failure to RTFM",

I have been using various distros since Slackware 2.0, and using Unix much
earlier than that. Just as a side note.

"Redhat packages", and "Oracle" have been far more troublesome for me
than "badly written code". Obviously, YMMV.

Whatever this acronym means.

"now or never principle"? If I have any kind of alternative then I
just send [a bug report] to /dev/null. It works now or else.

I sometimes wonder if this principle written large is why everything in
software engineering is such a damn mess--but that's probably too meta
for this here discussion.

Why not? It is way more interesting than the job that I should be doing
right now :-) Maybe it is another way round? Maybe, if the programmers
took this principle seriously, everything would be as polished and
thoughtful as a MacOS X interface.

The chaos of pen+paper, various colors, notes, and pictures, and all
that is something that is *difficult* to replicate with the ways
document-writing software works. The software has to impose a structure
of some type on the data that gets put in. Then there has to be a UI

Well, many problems are difficult in computer science. And creativeness is
very often chaotic.

for entering data and categorizing data. Actually, there are usually
different UIs for different types of data, since it's hard to draw
pictures with a keyboard and hard to enter text with a graphics tablet.
Then the UIs make the users react differently, and that affects the way
the documents get produced.

I have now a very precise idea how the perfect interface should look like
in my program. Actually, I made a working model yesterday using
perl/tk/zinc.

If you really need something that can handle lots of text in different
fonts, with pictures interspersed, with a pretty simple UI, your best
bet is OOO-Writer. It's a resource hog. So what, enough CPU+RAM is
cheaper than wasted time.

The more I think of that, the more I come to the conclusion that actually
Impress would be more the thing that I am looking for, if only it had not
been so cluttered with features and had a braindead interface, copied from
PowerPoint.

If you need to use a tool without thinking about it, you train with the
tool until you can perform the required actions without thinking much
(if at all). This applies to programs, printing presses, paintbrushes,

Some tools require less training than the others. It is always a matter of
optimal ressource allocation. There is no time to learn everything. You
need first to decide what to learn.

pellet guns, Perl, poodles, and a plethora of other things. So once
you've found a tool that works, learn it good, and use it. (Or get a
newish Epson scanner, do work on paper, convert paper to TIFFs....)

This is supposed to work the other way round, you know. From the
computer to the paper documentation.

Cheers,

j.

--
Uskadara is a little town in Turkey
And in the old days, many women had male secretaries
Oh, well, that's Turkey!
.

Relevant Pages

  • Re: Latetes Slew of patches.
    ... I've and others I visit often who use Linux, ... Why haven't you mentioned these things about windows? ... There may be an alert out for 'pppd', but it will tell which distro it's for. ... Where did I say anything about your post other than the computer science part? ...
    (microsoft.public.windowsxp.general)
  • Re: developing next windows versions
    ... programming, a computer science degree might be your best choice. ... of people developing windows. ... Being a Linux fanatic myself (the news server I am posting this on is run in my ...
    (microsoft.public.cert.exam.mcse)
  • Linux software equivalents | a list...
    ... I'm slowly moving the computers in the lab I run from Windows to ... GNU/Linus. ... particular) we do with software that has native Linux equivalents ...
    (comp.os.linux.misc)
  • Re: what happened to openness?
    ... just and equally as bad as people who release Windows only roguelikes! ... So Windows is more POSIX than Linux. ... "Computer Science is no more about computers than ...
    (rec.games.roguelike.development)
  • Re: Future of IT in Lebanon
    ... working knowledge of Indian programmers DNA, nor of their intuitive Java ... > So Longhorn is not an experiment and Linux is an experiment? ... another chapter in the Windows story, and the Microsoft marketing machine is ... > application opens, Check the about, it says Microsoft Visual Basic 6.3. ...
    (soc.culture.lebanon)