On Oct 31, 2005, at 1:32 AM, James M Snell wrote:
While I can definitely understand your argument from a purely
architectural point of view, existing blog software implementors may
(and are likely to) disagree.
Then let them disagree. It isn't the first time I've had to deal
with bad browser implementations.
They're already doing this kind of stuff [1] regardless of whether it
is the right thing to do architecturally and they're not likely going
to change just because APP doesn't support it.
They will change if the server doesn't support it, which is where
standards tend to get enforced. APP only needs to provide one
correct way to configure blogs.
Most likely, they'll do what blogger has done and find some way to
hack it into the API... most likely using extension elements directly
on the atom:entry [2].
[1] http://www.sixapart.com/movabletype/docs/mtmanual_programmatic.html
[2] http://code.blogger.com/archives/atom-docs.html#extensions
So the question is: do we just ignore the fact that folks are already
doing this stuff (and potentially doing it badly)?
No, we stamp them out like bugs. We expose them and watch them
scurry away. We do not standardize crap just because a lot of
people do weird things with older APIs. APP is supposed to be
cleaning that stuff up.
If you think those features are necessary, then place the
configuration information where it belongs and edit it there.
APP clients will do the right thing when given a chance.
....Roy
