Re: [seul-edu] School Networking Guide (was: need detailed network advice)

[Manuel Gutierrez Algaba <irmina@ctv.es>]

On Sun, 3 Sep 2000 bill@mail.tasis.ch wrote:
> > getting the stuff done in any format !!!
> 
> I couldn't agree more.  Help me redefine the current project, take over the
> current project, or create a parellel project.  Feel free to help.  Please. 
> There is room enough for more people in the effort.  

Very sensible words. One of the best ideas of Sourceforge ( the
bad one would be use SSL ) is to have a page of projects asking
for help , does SEUL have one? Well, If you clearly state which
is the current status and what do you need,... that would help
me to help you.

> 
> I realize progress has been very slow.  I am probably not the best suited for
> this project because I just don't have much time to dedicate to it. 

About time, don't worry. Nobody has. One of the matters I've always
thought is to minimize the effort for getting progs (or doc) done.

Currently, information, like Aristotle said, shows us two faces:
- content
- shape

To put it short, I've come to a "design" which could minimize the
effort of writing large ammounts doc. This is done in two steps:
- Writing attributted chunks of info. 
- Combine those chunks into greater units. 

(what's the hell this man is talking about, some of you would be saying
right now)

Look at this thing :
\jiji
where can I find the format of .dvi files and orther (fonts,...) used
or
generated by tex ?
\jaja
These are described in the tex.web and mf.web, from ctan:systems/knuth
use `weave tex.web' to generate a .tex file
A better source might be dvitype.web.
And you can probably easily read it as text without formatting it.
If you do format it, you need "weave" (of course) and probably also
webmac.tex.

\indexformatodvi
\indexweave

\jiji

As you see it's a very small ammount of info about a very specific
issue. If you notice it's a question-answer piece of info.
And besides, you have these things : "\indexformatodvi".

- What's this ? It's an attributed chunk of info.
- What can It do with it? It can combine with other chunks.
- How? Relating all the chunks labelled "indexformatodvi", or 
stablishing groups of labels, like translators = { indexformatodvi,
indexlatextodvi,...} We can relate that using a python script, 
for example.
- But, can you do that? In fact, that system exist:
http://www.ctv.es/USERS/irmina/TeEncontreX.html
and it's fully downloadable, GPL,...

- Anyway, how can that speed up the creation of doc ?
Well, it's far easier to think and focus in a very limited issue
each time. In the example of "Schooling Network Guide" we could
have a chunk like:
\chunk
NIS is useful for allowing different logins from different machines
to the same account
\indexintromatter  \indexdef \indexNIS
\chunk

When you write that chunk you don't need to worry where to place it,
before or after that or this section , chapter. It's a kind
of "fact" or "ethernal truth". 

You ***___REALLY___*** don't worry about structure, you just
define a chunk!

- Ermm, But how can I join those chunks into something usable ?
Which tools I have for that?
Well, apart of what you see at TeEncontreX, which is the lowest
form of organizing chunks, plain index.  It's easy to imagine
that in a Object oriented database of chunks ( this is dead easier
to do than to say , in python, for example), you can say:
"let's go chunks about NIS, let's order yourself in the following
order: intromatter, otherissue, anotherissue...

Those tools don't exist yet, but any programmer here may see they'd
be easy to do.
 <as these is one of the weak points of the email,
attack here, I might counterattack with python-bullets>

- Then, editor has a lot of work !
Well, my attempt is to minimize the authors work . Is minimized
if he's required to write a very small chunk each time, without
trying to relate each chunk with a "whole"? The answer is "Yes". 
The editor's work would be "to choose" the indexstuff's are
 related each other.

- Well, how many of those \indexstuff exist ?
Any author can write as much as they want to, we can do them
compatible, afterwards. \indexintromatter == \indexintroduction ==
\indexfirststeps...
We can use search&replace for that.

So basically, everybody has _absolute_ freedom to put whatever he
likes in the chunk. Obviously, small chunks are preferred.

- But, this would be a very weird way of writing a book's chapter.

Yes, it'd require that ,while writing, you define a chunk and attribute
it and then you continue writing, if you really want to go on
writing the old way: section or section after section. 

But, you could fully embrace the "chunk-method" and abandon the idea
of "sequential" writing, so you could write chunk after chunk and 
organize them ( putting some \index... if you like), afterwards.

Absolute freedom. 

The question is to give author no excuses to help:
- Any chunk can be written in , let's say, three minutes. Any
chunk is reusable and adaptable to any doc. 

If people don't help is because:
- we haven't explain it correctly
- GPL doc would go on growing only under the extreme effort of 
some individuals, and slowly.

Well, I won't waste more of  your time explaining this system.
I'd be more than happy if you understood what I meant.

> I really won't be offended if someone else wants to do this project or set up a
> parellel effort -- I promise I won't be offended.

:) I believe you.

> Another alternative is I can keep all the authors from doing overlapping work in
> the newly redefined standard and the multitude of authors can write away and the
> project documents every Linux/Internet project related to schools.  
> 
> When I started I got very little feedback, I assumed it was because either no
> one else was interested or that I had astonishingly good ideas ;-)  Now that I
> know there are some flaws, I welcome your joint effort.  I don't want to be
> stuffy or stiff and as long as I have docs to help my staff and students I have
> won too.  So please help get more momentum to this project. 

I've put all the material I have. I don't mind spend a bit of 
my time in it, whenever I feel I'm not doing something repetitive
or boring.

> Cool documents wasn't one of my design criteria.  (I could be one of those
> boring geeks for all I know).  I just needed easy to organize documents which
> could easily become xml and easily be searched for future indexing, etc. 

Well, " my proposal is a  very, very lax form of XML", let's
say it's a keyword-oriented-XML. I did TeEncontreX in 30 hours,
so it's easy to see that It's a system that grows very fast.

> I also
> need graphics for screen shots.  I don't want to spend time making cool
> documents -- probably a fault, but nice design takes a lot of time.

Again, we could have attributed chunks to be uuencode/decoded or
makefiles... And we could decide afterwards how to use that chunks.
Again, absolute freedom. The less ropes I can imagine.

> I look forward to this project moving forward at lighning speed, now.
> 

As, you've been so kind and sensible, I'd like to ask you a favor. 
Please, try to understand what I mean with my "chunk-system". 

I really can't think of anything so flexible, reusable, cool
 and easy. 

The bottleneck here is : getting people to collaborate.

A small chunk is the smallest ammount of help anyone can contribute.

( I know it's hard to change such a basic view of such a basic
thing.)

Regards/Saludos
Manolo
www.ctv.es/USERS/irmina    /TeEncontreX.html   /texpython.htm
/pyttex.htm /cruo/cruolinux.htm ICQ:77697936 (sirve el ICQ para algo?)

  senility, n.: The state of mind of elderly persons with whom one happens to disagree.