More on p:directory-list

Hello.
First of all, I'd like to say "thank you" to the WG for accepting my
p:(directory-)list suggestion. Now, I have some comments about it ^_^.

I liked (almost) all of the changes that were made prior to this step's
addition in the draft. In particular, the "depth" option seems like a much
better idea than "recursive" one to me (I wish I thought about it :D).

I think the whole paragraph
"
When a directory entry is a subdirectory, that directory's entries are not
output as part of that entry's c:directory. A user must apply this step
again to the subdirectory to get that directory's entries.
"
Should be edited to respect the recursive option (or the "depth" if that is
going to make it). Something like:
"
When a directory entry is a subdirectory, that directory's entries are not
output as part of that entry's c:directory, unless the recursive option is
set to "yes" or "true". A user must apply this step again to the
subdirectory to get that directory's entries.
"
OK. I guess that's not good either (I sure didn't understood a thing), but
you see the point.

About symbolic links. I don't think they should be treated. This step should
map the file system, not the server system. In fact, I'm not sure if the
later could be indexed to begin with. I mean, web servers just return an
HTML page (in their own format) with links to the files, not metadata about
folders in an RDF like fashion, right? Still, I guess c:other is a good
thing.

BTW, should extension information be added in other namespaces to avoid
conflicts, or are implementations free to place such information without a
prefix as well as long as it's not in conflict with a standardized
attribute? I think they should only be allowed to add such in their own
namespaces, as who knows what later versions might standardize for
c:directory and c:file.

Regards,
Vasil Rangelov

Received on Wednesday, 8 August 2007 22:08:44 UTC