I just wanted to mention that I made an interwiki gadget for the Dutch wiktionary long time ago, that just crawls through the toolserver databases to list interwikilinks (pages with equal names as well as all interwikilinks found on these pages). Only in case the toolserver database servers are down, http requests are send out. If there is interest I can share the php code so it can be generalised it in an interwikibot for wiktionary (or with some more work for the other project families as well).

Regards
Annabel

From: Paul Selitskas <wizardist@ya.ru>
To: toolserver-l@lists.wikimedia.org
Sent: Sat, December 4, 2010 11:25:54 PM
Subject: Re: [Toolserver-l] Interwiki handling on the toolserver. (Was: FYI: Changing of interwiki-bots-account-approval)

By the way, this approach lied into my account approval request ACCAPP-97. Though that tool is just a statistical one, the equal approach can speed up interwiki business in times. This is how appropriate interwikis in be: and be-x-old: (their absence actually) are listed in tables for future processing.
 
>Using the replicated data instead of making http (api) requests should speed
>up the data collection phase of large inteerwiki groups from several minutes
>to a seconds or so.
 
---------
С уважением,
Павел Селіцкас/Paul Selitskas, skype: p.selitskas
Wizardist @ Wikimedia projects