Run, Forrest, run!

["Piroumian, Konstantin" <KP...@flagship.ru>]

Drill Sergeant: 
	Gump! What's your sole purpose in this army?
Forrest Gump: 
	To do whatever you tell me, drill sergeant!
Drill Sergeant: 
	God damn it, Gump! You're a god damn genius! This is the most
outstanding answer I have ever heard. You must have a goddamn I.Q. of 160.
You are goddamn gifted, Private Gump.
(http://us.imdb.com/Title?0109830)

Hey, Nicola, I'm back. So, what do we do now?
As I can see, you've used the same old book format for content menu. Ok,
it's not bad for the start. But for the documentation part we need something
more than that, I mean something more dynamic or pseudo-dynamic. 

The main idea behind the book.xml is to provide links to files (documents),
but we need also comfortable navigation between sections of the document
(topics). This can be solved by using the XPath-enabled directory generator,
but in that case we don't need book.xml anymore, otherwise we will have to
create some XPath enabled transformer, that will iterate through the book
menu items and generate book topics retrieving them from the according
document.

So, what is my proposal? It's simple. Let us create the document topic map
(chapter or section list) using only the directory generator. And we can use
the old book.xml to provide labels for the documents in the current
directory. If book.xml does not contain a label for a document then
document's label will be used.

What do you think?

Best regards,

Konstantin Piroumian
Leading Developer

Protek Flagship LLC
Phone: + 7 095 795 0520 * 1288
Fax: + 7 095 795 0525
E-mail: kpiroumian@flagship.ru
http://www.protek.com 

RE: Run, Forrest, run!

[Steven Noels <st...@outerthought.org>]

Hi G.I. Gump ;-)

> The main idea behind the book.xml is to provide links to
> files (documents),
> but we need also comfortable navigation between sections of
> the document
> (topics). This can be solved by using the XPath-enabled
> directory generator,
> but in that case we don't need book.xml anymore, otherwise we
> will have to
> create some XPath enabled transformer, that will iterate
> through the book
> menu items and generate book topics retrieving them from the according
> document.

Although I favor dynamic generation, we are still left with the
'sequence problem' - unless we leave some sort key inside each doc that
can be picked up by the XPathDirectoryGenerator ... not sure if this is
a feasible/good idea, either.

Apart from my proposition of having some less book-oriented
sitemap-describing thing dispersed throughout the directory structure,
please re-read my ideas in
http://marc.theaimsgroup.com/?l=forrest-dev&m=101605443113305&w=2 and
http://marc.theaimsgroup.com/?t=101637949200001&r=1&w=2 concerning this
ordering issue.

Thanks,

</Steven> - still trying to find some time within a week or two to fully
participate again :-(