Le 15 juin 05, à 11:43, Bruno Dumon a écrit :
>> ...It would be good to differentiate what's imported, for example to
>> generate warnings like on
>> http://forrest.zones.apache.org/c2/build/cocoon-2-2/c21-index.html
>>
>
> While this wouldn't be hard to do, I don't like it very much because:
>
> * it's additional work to remove it again afterwards..
But can't we generate such a warning from a document attribute which
says "this document has been imported automatically from a corpus which
is known to contain a lot of cruft and needs to be cleaned up before
we're happy about it" ?
In a more concise way of course ;-)
Then, it's simply a matter of removing the attribute, or changing the
document state, once it is reviewed.
> * the purpose of this import would not be to remain a static legacy
> copy, but rather something based on which we start working for the new
> docs, which also means asap removing all completely outdated pages, and
> putting a disclaimer on pages of dubious quality...
Sure, but with 250 docs, ASAP might mean quite a long time...
> * since documentation will never be as correct, complete and up to
> date
> as we want, we could potentially place a disclaimer on each
> documentation page.
Sure, see above - what I suggest is just to differentiate imported docs
in a way that allows the unreviewed ones to be identified in the
published pages (if we want it), and to be retrieved with a Daisy
query.
If there's a Daisy query that tells us which imported docs have not
been reviewed yet, I'm happy.
-Bertrand
|