Re: Word Processors

[Olof Oberg <mill pedgr571 sn umu se>]

[since the first try obviosly didn't get through (?)]

John R Sheets <dusk@smsi-roman.com> wrote:
[snip]
> For example.  Let's say you're using a Resume template.  The
> template would probably only imbed a single entry for each
> employment or education history.  To add 2 or more, you have to
> cut & paste the original entry.  Wouldn't it be nice (my favorite
> phrase) to let the template spit out the text blocks for you? 
> (This could also be automated if you had the data stored in a
> database.)  If you change to a different Resume template, the
> format for these blocks will most likely change, so it seems
> natural for the template to keep track of this.
> 
> How would all this fit into the XML/XSL/scheme model?

I envision a word processor (or any other application that 
interactively produces information) where you have a generic 
template and file format - namely XML. Scheme (Guile) will 
take care of binding resources to documents and building 
macros (dialogs and wizards).

An example.

Lets say Jim Allchin wants to send out a memo to his employees 
that they have to leverage Windows to gain control of the 
Office market. Since he has already done something similar 
before he already has a template for it. The new instance of 
the template makes dialogs pop up and asks him about title, 
partners in crime, etc. The input is processed by Guile and 
looks up relevant resources through LDAP via a CORBA exported 
interface. He could even import a spread sheet to show the 
monetary gains of the move, which of course would be a Baboon 
object connected to Gnumeric.

I was once actually paid to type down templates for 100 or so 
documents and to code automatation in Word Basic was cumbersome 
to say the least. I think with a good tutorial on doing it the 
users will benefit immensely from having a real programming 
language. With good I mean something easier than the Elisp 
tutorials *looking at a non-working CSS mode*.

The style part I am divided on. I don't like to mix things 
with different meaning and same syntax (i.e. XML and XSL) so 
I am leaning towards DSSSL. I do see the benefit of needing 
just one parser (gnome-xml), but since we already have Guile
 present...

What I don't see any benefits of is to put the physical markup 
of information as the first thing the users see. It will 
inevitably result in people thinking "14pt bold Arial" instead 
of "Header of a sub section". 'Creativity' that means waste of 
money crippling of assumed benefits of computers in a corporate 
environment (hard to index and search such documents for example).

 /mill

- -- 
#############################################################
# S-mail: Olof Oberg     #  mill@pedgr571.sn.umu.se         #
#         Pedagoggr. 7A  #  mill@ludd.luth.se               #
#         S-907 30 Umea  #  tdv94oog@cs.umu.se              #
# Phone:  090-197395     #  http://pedgr571.sn.umu.se/~mill #
#############################################################