Andreas Sewe wrote:
Compare this to the following statement:
If I'm uploading a file, the server can automatically generate that
header and the users will probably get what they expect.
If your client can automatically generate a slug, the server can
probably do the same thing, unless your client simply uses the filename
for a slug. But since servers are supposed to do things even more clever
automatically, like populating atom:summary, I don't see what would make
it impossible for a server to generate a Slug which "probably [will] be
what [the users] expect."
If I select c:\documents\me-at-the-beach.jpg from the disk, I can
generate the slug "me-at-the-beach.jpg" or "me-at-the-beach".
If I just post the image/jpeg content to the collection URL, there
is nothing but the title to extract a name from. Now, if the title
and the filename aren't related, then you can't do so.
If the user really wants the resource to end with the same filename,
they won't be able to do so without something like the Slug header.
Lots of ifs, but I think this is an important feature. It should also
be optional, so ignore it if you don't like it.
--Alex Milowski
