Remembering 9/11

This is the day. I started Archivist going at approximately noon. I checked to see which versions of 9/11 was getting the most 'use,' and found three: 911, #911, and 9/11. Twitter also indicated that Sept 11 was trending. So I put Archivist to work collecting all four. Unfortunately, it will not do a combined search. So there are four searches. Archivist does not distinguish the '-' and the '/'. So the search for 9-11 captures both.

All four searches found 1500 messages when I initiated the search. So I have missed some number of messages since I did not start earlier. Clearly there is a very big burst of messages and then it tails off quickly.

The trendistic figure shows that it started increasing volume very early on 9/11. This is the 24 hour figure through up to and through noon. It is not dynamic so I should probably get another 24 hour figure later today.

In this chart you can begin to see small differences in the volume of messages by 'designation,' especially if you shift the show and hide for the three streams.

I checked Google at 12:30 and it listed only one story about 9/11 and twitter http://features.csmonitor.com/innovation/2009/09/11/on-911-anniversary-twitter-users-remember-whereiwas/

and it is about #whereiwas rather than 9/11, etc.

The figure for the week suggests that it is time to stop collecting messages.

1) the minor streams, #911, 911, and Sept. 11, stopped being about '9/11' a few days ago. They will require considerable cleaning before they are useful. stopped 21:45 September 17, 2009

2) the larger files, 9-11, seems to have become other references by today, but I cannot stop it until September 18.

The final totals are: 6932 for #911.xml, 92980 for 9-11.xml, 18,949 911.xml, and 15,486 for Sept 11.xml

The .txt files that can be read by Excel to acquire this data are: #911.txt, 9-11.txt, 911.txt, and Sept 11.txt