[Typo3-doc] question about the glossary

[Mark Ravitz]

> Sorry, if I am rude. But I think this importing definitions
> script is just a small part of the Glossary extension. So I made
> a Glossary extension page and *copied* your information there:
>   http://wiki.typo3.org/index.php/EXT/glossary
>
> Your page is still there, so if you don´t like my suggestion, we
> could just put your link back again (but the one to the glossary
> extension should remain). Ok?
>

Go for it.  I really don't mind if people edit my work here.  I'm
not sure how to delete the 'Glossary:extract_definitions' page.

> About extracting definitions:
> I don´t think it matters much if we do the extracting from the
> database or the wiki page (saved as a text file).

Yes.  I'm not sure at this point which would be the best way.

> The main problems - as I see it - is how to:

>   1. handle "syntax errors" in the text

Perhaps it would be possible to write a syntax error
correction script.  This is a good reason to know how
to query the wiki database.

>
>   2. check the quality of the explanation
>
> When I do the "extracting" by hand, I get this auotomatically. I
> correct typos, syntax errors and see if I understand the
> explanation. But how do we do it automatically? I suggest:
>
>   1. Someone has to do it manually before extracting. And the
> most important is to keep the syntax (on the wiki page) very simple!

I do wish there was one other field which says what areas of knowledge
the definition belongs to: typoscript, backend, php classes and files,
or
general typo3, or general web knowledge.

>
>
>   2. The extraction tool has to ask for confirmation before a
> definition is added/updated into the database. It should show
> both the current (if existing) and the new definition at the
> same time - for you to decide what to do. Maybe even give a
> possibility to edit one or both of them before
> adding/updating/rejecting.
>
> What do you think?
>

Sounds like a lot of work.

It might be simpler not to try and edit or
approve Glossary entries in the Glossary extension or
the in the script used to import the defs from the wiki.

Wouldn't it be easier to get things right in the wiki
(which could include protecting finished defs) and then
do the import in one step, which would always update
the entire Glossary Ext. database to reflect the wiki?

If we classified the defs with a keyword and protected
the finished ones we might be able to, for example,
have the script extract only  the protected ones concerning
general typo3 knowledge.

Since the wiki is up an running, and the Glossary Ext.
isn't, I would say it would be best to do as much as
possible from within the wiki.  There may be all kinds
of interesting open source tools for messing with the
wiki.  I'm going to try and learn more about it's
capabilities.

Mark Ravitz