[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Next #d-women forum; topics anyone?


I just noticed that the old mail I sent reached the mailinglist, 
but since I tried to spellcheck my post and correct a few things, 
here is the post again. (With spelling errors fixed, I hope).

tuesday, 2004-12-28 kl. 01:46 skrev Erinn Clark:
> * Helen Faulkner <helen_ml_faulkner@yahoo.co.uk> [2004:12:28 00:37 +0000]: 
> > I had a few ideas, but I reckon we can come up with better ones if we 
> > discuss it together.  My ideas included stuff like
> > 
> > - making a list of the locations of important Debian documentation and 
> > posting it on the Debian Women wiki.  So people can benefit from the fact 
> > that someone else already spent an hour searching the website to find a 
> > particular thing out.
> This might be useful, but I'm concerned that giving a long list of links,
> while more helpful, is not *all* that helpful.
> I'm also not inclined to believe that Google should be the one indexing the
> docs -- there has to be a better way to centralize the information and make
> it easier to search. I suppose one thing we could start with is asking people
> here whether they know of any documentation that is really comprehensive,
> well-written, and centralized.

One piece of documentation collection that I personally like  is the 
"info" thing.  I use GNU/Emacs a lot, and its very handy for instance 
when I write a LaTeX document that I can look up the "info" on either 
LaTeX or GNU/Emacs itself.  When I write a "make-file", i can take a 
look at the info that one, if I cannot remember a automatic dependency 
or whatever.

Some argue that info has a bad interface, but I like the structure of 
info-nodes quite a bit, and I am used to the info-interface in 
GNU/Emacs.  I know there are several other interfaces, and I'm sure we 
can create a new interface, if someone has an idea and a need for one, 
however if I understand it correctly, most agree that the centralised
structure of info is nice.  Personally I wonder if a "lynx-like" 
command-line info program was made, or simply a web-interface, would help?

Info nodes are organised in circles or discs. Remember the game "towers 
of Hanoi"?  Well, imagine you have a stick, and a set of round disks 
with a whole in the centre.  They're of different diameters, and you put
the one with the largest diameter in the centre.  Now, with info, the 
documentation start in the centre, on the top most disk, and the 
more specific it gets, the further "down" from the "centre" you have to 
go. I explain how in the PS[1], in case you're interested and don't know
already, however, if you find it boring, just skip it.

At any rate, I find that a great way of organising documentation.

I know others have different opinions, but I can only represent my own, 
however, I will note that a gopher site was organised a bit like this, 
and many homepages try to simulate this in some way (or maybe not 
simulate, but do a similar thing).  The difference is, that in HTML (and
much documentation) you don't really have a centralised structure, 
unlike gopher and info.  You might have an index, and possibly a 
subindex, but levels below that often seem tedious.

Now, not all people like "info" (the command-line program), but I am sure
its not the structure of the info-nodes themselves they dislike, but 
rather they lack tools that fit their habits, to navigate the 
info-nodes.  As a GNU/Emacs user, its quite natural for me to use info 
inside GNU/Emacs, but I'm aware that others use other tools.  Whether 
info is really what we're looking for, I don't know, but I guess we're 
looking for a structure of some kind, and I happen to like the structure
of info a lot.  Info was developed by the FSF to make better
documentation than man pages for the GNU project, and while I am not 
saying that they found the ultimate solution, I do think that at least 
the structure is a step in the right direction.  I also like the
structure of man pages, but I think man pages do lack means of
navigation and centralisation, that info offers.

	In the hope that we can make Debian better,
	Dabian .


In the following, I assume that we're in GNU/Emacs in X11, and have
typed "CTRL + h" then release them, and then "i" - in GNU/Emacs
terminology thats called "C-h i".

Info has a central document, like a front page in a wiki, the "(dir)"
node.  From this node you can navigate to other top-nodes, for instance,
if you want to go to the LaTeX documentation, you type "m" to access the
menu, and either type "latex2e" or use the tabulator and/or mouse.

Now you're in a new top info-node, and you can choose from a menu of 
LaTeX topics. 

If you want to read about, say "Debian menu", you now type "u" (for up)
until you're at the top level again, type "m" again to access the menu
and type (on my system) "Deb <tabulator> <enter>" and you're at the top
node of the documentation for "Debian menu".  Now, at this level, you
can either decide to enter one of the subtopics, say "Chapter 2", or you
can decide to browse all the documentation in this level in a sequential
way.  If you choose the latter, you type "n" to go to "next" node -
these nodes might have sub-nodes again, which work in the same way. 
Now, if you at any point went to the wrong node, and want to back to the
last node you were watching, you can type "l".  Note that you have to
choose from the menu, in order to get to the next level of the circle,
unless you follow "a note" (a kind of hyper-link).

Reply to: