-
Website
http://www.louisgray.com/live/ -
Original page
http://blog.louisgray.com/2008/07/tweetstats-shows-impact-of-instability.html -
Subscribe
All Comments -
Community
-
Top Commenters
-
charlieanzman
60 comments · 11 points
-
Jesse Stay
221 comments · 70 points
-
Ari Herzog
43 comments · 21 points
-
ChangeForge | Ken Stewart
133 comments · 18 points
-
drewolanoff
64 comments · 53 points
-
-
Popular Threads
-
FTC Disclosures Made Simple For Bloggers With Conflicts
6 days ago · 46 comments
-
Still Waiting for An Evil Google? It's Not Going to Happen.
6 days ago · 30 comments
-
Fighting Bots With Bots on Twitter, Leveraging SocialToo
1 day ago · 5 comments
-
Simler Adds Likes, Favorite Tags, Revamps Homepage
1 day ago · 4 comments
-
Gowalla Raises $8.4 Million for Location Check-in Service
1 day ago · 2 comments
-
FTC Disclosures Made Simple For Bloggers With Conflicts
I am still somewhat on Twitter but focusing my efforts on Identi.ca (which is where I saw the link to this article). I'd love to see you and anyone else reading this chat more over there, instead of just using it to alert to new blog posts. The community (and feature set) is growing quickly and I'm excited that we may have a viable alternative.
@dacort tweeted about it here: http://twitter.com/TweetStats/statuses/868447439
The conversation between @dacort and @al3x is available in a conversation in the Twitter Development Talk Google Group: http://groups.google.com/group/twitter-developm...
Examples:
Duncan Riley: http://tweetstats.com/graphs/duncanriley
Jesse Stay: http://tweetstats.com/graphs/JesseStay
Steve Rubel: http://tweetstats.com/graphs/steverubel
In contrast, I am pretty confident about the data shown in the graphs used in the blog post.
As it stands, if a bot or user hit the graphs page for a user and it hadn't been updated in 3 days, TweetStats would automatically update. For top bloggers/tweeters like those posted above, this probably helped keep the data somewhat current. I've changed this to 1 day at this point to help with the recent pagination changes.
I may also try to utilize Gnip to keep stats up to date. Over 20,000 people have used TweetStats at this point, and I struggled for a while as to how to keep the data up-to-date, especially once Jabber went away. Gnip can make this easier, but I hope that Twitter can provide the type of data access they used to someday soon. It is now impossible to retrieve Tweets retroactively, for the time being.
That being said, I can understand if their new round of funding has placed limits on their primarily value - the huge database of user data. We have already seen evidence of this in that it was stated PubSub will only be made available on the terms that no resyndication of content will be allowed.
Twitter is growing up and we are experiencing their growing pains right along with them, unfortunately.
I can't really trust these graphs. When I pulled my stats from TweetStats, it showed zero tweets for all of April, May and June 2008. Zero tweets for three straight months? I think not.
Maybe people are also finding it difficult to be the conversation all the time. Hard to manage several thousand people on a daily basis.
I do agree though, instability has impacted stats, and ultimately the loyalty of some users.