The reliance on Content-type

Ernst Johannes at
Mon Nov 14 12:28:28 PST 2005

Comments below.

On Nov 11, 2005, at 0:10, Martin Atkins wrote:

> Ernst Johannes wrote:
>> (Note that these concerns where the very reason why originally  
>> had  the
>> meta=lid / meta=capabilities arguments -- this problem is far   
>> smaller
>> when doing so.)
> I see that as not being much different to using a "magic" filename,
> transforming (for example) into
> Using the URL that the user entered seems like the only safe way to
> ensure that it'll work everywhere. Making up URLs requires that all
> server software be able to support your view of how URLs work,  
> which in
> a world that's moving away from a directory full of static files and
> towards crazy content management systems is not always a valid  
> assuption.
> If any "magic" URL transformations are going to occur, I'd rather  
> it be
> to another "filename" than adding a query string argument. I don't
> really like either, though.

I think we have a number of non-perfect choices here. (Real  
engineering ;-))

I have attempted to summarize them here:

Are they correct? What am I missing? (I apologize if I do)

If so, which alternative do people prefer?

On balance, I lean towards alternative "Draft-002-c", specifically  

Johannes Ernst
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lid.gif
Type: image/gif
Size: 973 bytes
Desc: not available
Url :
-------------- next part --------------

More information about the yadis mailing list