Recently I wrote a guest blog post for Michael Q. Todd’s about using ifttt to backup your own tweets. I actually recorded the video back in January out of frustration from not being able to easily find an old tweet. Well, writing the post to accompany the video stirred up evil and vile feelings towards Twitter’s ridiculous 3200 API limit all over again. And that led to the battle cry I am hoping you will join in today. Tweet #FreeMyTweets
Hey @twitter, #FreeMyTweets! We the tweople demand access to our old tweets. << HELP me spread this message!
— Kevin Mullett (@kmullett) May 25, 2012
Now to make sure everyone understands that this isn’t about me, by all means craft your own message, augment it, write your own posts, whatever, just make sure to include #FreeMyTweets so the message has a chance to trend and is recognizable as the same unified voice. Whatever it takes to liberate our data!
Now for a little back story on the issue. Feel free to tweet from above and skip this part if all you care about is getting Twitter to Free Our Tweets.
Backups can only capture what I do moving forward, so how do I get back tweets extending to that fateful September 3rd, 2008 day when I joined Twitter? This is roughly the same question Robert Scoble asked in this very lengthy G+ thread, back in November 2011. A question I had previously been researching myself. You see, Twitter’s search won’t help because it only goes back roughly 2 weeks (10 days or so, but less if you tweet a lot). A Google operator search of “site:twitter.com/kmullett” only yeilds around 6,890 of my 42,291 total tweets. Google shelved replay-it, which used to work like a timeline of tweets. I didn’t connect my Twitter account to Memolane until February 5th, 2010 and connected Tumblr back in October 7th of 2009, but both are horrible for archive or searching. FriendFeed is worthless to search and also would only be a backup from when you connected it. Other tools mentioned in my post listed above, and a lot that are not mentioned, also have to have been setup when you started twitter.
The only site that didn’t require a previous connection, was free, and came close to having all my tweets, was social search site Topsy. Topsy was able to take me back to September 19th, 2008, which was the easiest and oldest, but it still only exist on Topsy’s servers, is not guaranteed to have all my tweets, and they are not all in order.
Why would you want old tweets anyway? A fair question, so here we go:
- As record of exactly what you said, in case someone recalls it differently.
- To see responses, and who responded or shared the tweet.
- To recall an event or happening and the context of it.
- To recall a link you posted but forgot to bookmark, or can’t find.
- To recall when you something historical occurred and your response to it.
- As a type of journal or timeline.
- To avoid rethinking or retyping a well constructed and thought out response to a common question.
- For possible use or inclusion in a presentation or a book/ebook.
- Because the Library of Congress Twitter Archive back to 2006 and twitter shouldn’t be the only ones who have your data (thoughts).
- And lastly, because if you don’t back them up, you currently don’t have the option to get more than 3200 of them later.
I could go on with more examples, and could include business reasons, but those were just top of mind. And yes, I am sure there are people who, based on how and why they use twitter in the first place, simply won’t care, and that is fine too.
So Twitter, what is the end game here? Why isn’t there a way to download all of our tweets right from within our account settings area? Why are developers limited to just 3200? OK, to be fair I know that part of the issue is resources based on how they currently allow it, but that isn’t an answer. Do I have THE answer? No, but I also don’t work there or have an intimate knowledge of their system, but no solution isn’t acceptable. Here are a few hair brained ideas.
- Put a backup option within each accounts settings area.
- Allow developers, with proper authentication, to incrementally download successive batches of data via the API to avoid load issues. No seriously, I will wait weeks if that is what it takes.
- Share the ID algo so people could query past infomation based on a specific user.
- And finally, though I believe it would be a bad move overall and I would resent having to do so, why not allow me to pay for a backup of all my tweets?
As it stands, right now, there is NO KNOWN way for me to go retrieve every tweet I have posted. Now if you know of a way to retrieve MORE THAN 3200, AND have done due diligence to make sure it also isn’t limited to 3200, by all means…comment below. Feel free to comment on your favorite backup solutions too, but I already know of a truckload of those.
Don’t forget, if you want your tweets liberated » Tweet #FreeMyTweets