[Edu-sig] Things to come

[Dinu C. Gherman]

Steve Morris wrote:
> 
> I guess I am not making myself clear. It is not a matter of blaming or
> criticising pdf. I just don't think it is right for collaborative
> projects. You don't seem to disagree in spite of your strenuous
> rebuttal.

Ok, as it seems we have more in common than expected we should
be able to settle this rather sooner than later... ;-)

I guess what irritated me most was your comment about illegi-
ble PDF files that were almost as bad as graphic bitmaps. I
then tried to point out that you can actually do more to PDFs
then your posting would let people believe.

Let's not forget Jeff's original question: "What format is best 
suited to creating content that can be both web ready and gene-
rate good looking printed copy from the same source?"

Up to here the answer is, in all likelihood, PDF. If you add
the requirement of collaborative editing and/or directly edit-
able content, you're stepping, most likely, into the ?-ML 
world.

My point, that I did not make yet, is that (human-)editable 
content that is web-ready and printer-friendly does not exist, 
unless you install a tag and escape-sequence filtering plugin 
into your brain ;-) (as you see I'm not putting enough trust 
in tools, here). If you do you will find out that LaTeX is, 
perhaps, as good a choice as other "tag sets".

If we can agree on this, too, it all comes to some ASCII or 
UTF-8 format, minimally enriched with some of the tags that
are currently en vogue, plus a set of tools to start conver-
ting this "real content" into whatever you like or need or 
think you need, be it PDF, ?-ML, JPEG, paper, flying rugs...

> I'm not sure where you see a religious war happening. Do you disagree
> that pdf is less accessable than say html or DocBook? Do you disagree
> that that makes it less desirable as a collaboration language? Do you
> disagree that these are significant concerns when deciding how to
> publish in a collaborative project?  

As I hope I clearified by now, PDF is great to distribute
and "publish" documentation. It is almost as accessible as 
paper; certainly less than any pure ASCII/UTF-8-based format
(but more than flying rugs). It is certainly hard to "colla-
boratively" edit a PDF directly and it does not make sense.

I *do* have trouble seeing is why ?-ML is *fundamentally*
more appropriate (and not just more convenient) for that pur-
pose than the good old vanilla ASCII/CVS combination (yes,
I know it can better enforce structure, anything else?)? My 
guess is that any sophisticated ?-ML solution is going to 
sacrifice "easy collaboration" and "content editing" for
rigid structure enforcing mechanisms. In any case, we'll 
need the tools anyway, but as Kirby said, it's the content
that matters.

If you want collaboration at the extreme go looking for web
tools like Zope combined with products like ZWiki[1] (still 
premature, but already quite nice) and its siblings[2] in or-
der to work on the "pure" content. While the emphasis is much
less on structure (by definition) it will get you going much
faster than any set of a few zillion tags. 

Having-it-all-at-once-is-never-easy'ly,

Dinu

[1] http://joyful.com/zwiki
[2] http://c2.com/cgi-bin/wiki?WikiWikiClones
    (seems down right now...)

-- 
Dinu C. Gherman
................................................................
"The only possible values [for quality] are 'excellent' and 'in-
sanely excellent', depending on whether lives are at stake or 
not. Otherwise you don't enjoy your work, you don't work well, 
and the project goes down the drain." 
                    (Kent Beck, "Extreme Programming Explained")