Re: [Bug-gnupedia] Architecture Questions

[Mike Warren]

"Imran Ghory" <address@hidden> writes:

> So we don't have to re-invent the wheel. Usenet has tradionally gone
> down the direction of storing files in human readable form and it
> has had in the past problems with the OS not being able to cope with
> xxxx thousand articles in a directory.

They could be in directories based on their unique-IDs (i.e. all IDs
starting with XXXX could be in one directory). Apache's mod_rewrite could
change:

http://www.gne.org/article/some-unique-id.xml

into:

http://www.gne.org/article/some/some-unique-id.xml

> Also searching and querying are integeral to a database and are far
> easier and faster to use then trying to grep a massive directory.

Searching should never take place on the back-end servers.

-- 
address@hidden
<URL:http://www.mike-warren.com>
GPG: 0x579911BD :: 87F2 4D98 BDB0 0E90 EE2A  0CF9 1087 0884 5799 11BD