[jdev] GEOLOC Support

[Norman Rasmussen]

>
> I misunderstood the purpose of the hash. I thought the client was expected
> to unhash it and use it to determine the capabilities. I didnt realize it
> was used to compare caps efficiently.
>

Hashes are one-way, so it's very hard (i.e. you often need to brute force),
to reverse the calculation.  Hashes are a quick and easy way though to
compute a 'fingerprint' for a set of data, so that you can easily determine
if you've seen it before (by comparing the hashes), or not (and then you can
do the 'expensive' disco query to get the data that the hash was computed
from).  Clients need to store the disco result along with it's hash, so that
when they see the same hash again, they know what the result was, and they
don't need to re-query it.

-- 
- Norman Rasmussen
- Email: norman at rasmussen.co.za
- Home page: http://norman.rasmussen.co.za/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.jabber.org/jdev/attachments/20081031/33864c86/attachment-0003.htm>