>> On Fri, 3 Jan 2003 10:58:24 -0500, >> Bennett Todd <betat_private> said: B> If we're closing on an implementable design, would it be worth giving B> some thought to the registration management for the tags? Definitely. B> I think we're in agreement that we can't hope to cover every possible B> need a priori, there needs to be some open-ended flexibility for people B> to add tags where the existing lexicon doesn't cover their needs. Could we do something similar to the mail-header custom of prepending "X-" to non-blessed headers? Let the namespace do the work? B> But we really ought to try and find a way to encourage people who run B> into that situation to submit their new tags to a registry that feeds B> them back as updates expanding the lexicon... If we use a namespace convention like the above, it might be easier to write a tool which grubs around in existing logs to find added tags. Any lines found by this tool could be sanitized (IP addresses removed, etc) and emailed to this list, or a similar list for classification suggestions. -- Karl Vogel I don't speak for the USAF or my company vogelkeat_private http://www.dnaco.net/~vogelke We live contentedly with things our ancestors would have crossed oceans to escape. --Joe Sobran _______________________________________________ LogAnalysis mailing list LogAnalysisat_private http://lists.shmoo.com/mailman/listinfo/loganalysis
This archive was generated by hypermail 2b30 : Fri Jan 03 2003 - 18:44:37 PST