Re: general question : why keeping it so hard ?

[Prune <prune@lecentre.net>]

Oliver Egginger wrote:

> I think the problem is the debug-output from the server.
> If the Server rejects a query as a result of schema-violation,
> he simply reports "object-violation" or that he don't know an
> token as attribut, but which token ?!
>
> It would be very nice if the server report these tokens,
> also he should report which attribute cause a 
> "object-violatation". 
>
> I think this expansion would be pretty simple
> and help a lot.
>
>
> personal note:
> In my opinion schema rules are very useful.
> If you don't have them you have to deal with
> stray entries (entries without a predecessor), multiple entries, 
> undefined attributes (which are actual tokens and no attributes).

Yes, Schema is usefull. The main problem is how we deal with it. Even if 
it's easy to use an already built schema, it's hard to know what to do 
when you need new stuffs.
Drowbacks are that everybody add it's own schema to existing one. 
Finaly, the protocol is not a protocol anymore.
It would not be the same thing if it was simpler. Then you'll ask : and 
then, who will be authoritative to validate a schema ??

Think...

Prune

--
   - le Centre - 
a Mad Cow Tribe product

(Very uncommon, but we should please everybody anyway, even disturbed minds)