Modern Software Experience

2009-03-13

Retweet Thief

simple

Retweeting is simple. Just copy a tweet and add RT @username in front of it. You can not just retweet a tweet, but even retweet a retweet - until you run out of characters of course. At that point you need to consider what to keep.

retweet thief

Less than a week, twitter was buzzing about JustinSMV’s Are You A Twitter ReTweet Thief ? article. It is a great article that explains all you need to know about retweet thievery. It does not just explain what it is and why it is wrong, but gives concrete examples that demonstrate the combination retweet, the shorten URL switch and finally the retweet thief.

I recommend you read JustinSMV’s article if you have not done so already. This article discusses some investigative techniques.

tools & techniques

Twitter Search

text search

Your basic tool is Twitter Search. Finding retweets is very easy. Simply enter the text of your tweet in the twitter search engine.

When your tweet is so long that a retweet is unlikely to fit in 140 character, people are like to shorten your text in some way. They may decide to abbreviate some words or leave out others, particularly adjectives that do not really change the meaning of the tweet.
It is still very easy to find those: just search for your own twitter handle.

link search

Things get slightly more complex when your tweet includes an link.

Now, because someone retweeting your tweet may prefer a different URL shortener (the shorten URL switch), you should search for just the text, without the URL.

However, because yet others may decide to shorten your text to make it fit, you should also search for just the shortened URL.

Again, the simplest way to find all retweets is to search for your twitter handle.

third-party tools

retweetist

Actually, the easiest way to find retweets is to use a web site like retweetist. The one major limitation of retweetist today is that only lists honest retweets, and makes no effort to show potential retweet thievery. In fact, it hardly tries to distinguish retweets from messages that merely contain RT and your twitter id.

BackTweets

There are many URL shortening services. You could try using them all and then search for the links you get in return. It is probably easier to simply search for the short links you encounter. But the easiest method of all to find all tweets that link to a particular page or domain is to use a service like BackTweets.
BackTweets lets you search tweets using the original, fully expanded URL.

researching

finding retweet thieves

no single way

I am sorry to say that there is no single sure-fire way to find retweet thieves. They generally do not advertise themselves with handles such as @RetweetThief or bios that unabashedly state I steal the best tweets for you.

thief?

Moreover, even if you discover what appears to be stolen tweet that does not credit you, you still have not proven that its author is a retweet thief. However brilliant, witty and stunningly original your tweet may seem to you, someone else could have the same thought. Someone else could have stumbled across the same interesting link as you, or heard about it in a roundabout way, say through digg.com.

Even the use of an identical shortened link does not prove the tweet has been stolen. As JustinSMV points out, URL shorteners may return the same short URL.

Thus, proving retweet thievery seems almost impossible. There are two problems, finding the tweet and proving it has been stolen, and there is no sure-fire method for either.
However, research is not entirely impossible, and can certainly unearth some interesting facts. Here’s how I researched what appears to be thievery of one of my better tweets.

case study

tweet, retweet

A few days ago I wrote the article Twitter is for microblogging and then tweeted about this to let all followers know:

2009-03-10 00:07 TamuraJones Article: Twitter is for microblogging, not for link-spamming your macroblog. http://tinyurl.com/twitisblog.

quotable quote

When I hit upon that particular formulation while writing the article, as its central thesis and summary in a single sentence, I rather liked it. I thought it sounded good. It seemed a chuckleworthy quote. It seemed retweet-worthy. Sure enough, retweets followed soon:

2009-03-10 00:14 cmic RT @TamuraJones: Article: Twitter is for microblogging, not for link-spamming your macroblog. http://tinyurl.com/twitisblog.

2009-03-10 00:54 antains RT @TamuraJones: Article: Twitter is for microblogging, not for link-spamming your macroblog. http://tinyurl.com/twitisblog.

2009-03-10 07:31 JoachimNiemeier Reading: Twitter is for microblogging, not for link-spamming your macroblog - http://ow.ly/LHG.

2009-03-10 08:04 arimue Damn right: RT @JoachimNiemeier: Twitter is for microblogging, not for link-spamming your macroblog - http://ow.ly/LHG.

2009-03-10 08:17 RicardaBock Put that in your pipe and smoke it: RT @JoachimNiemeier: Twitter is for microblogging, not for link-spamming your macroblog - http://ow.ly/LHG.

I initially thought the retweeting had stopped after the first two. I did not the discover the other three tweets until a few days later. That is when I started to consider that I might have somehow have missed a few, that I needed to do something smarter than a straightforward search for my name or the URL.

search technique

I thought a good search technique would be search for something relatively rare. The shortened URL is a unique, but someone may use another URL shortener, and there are quite a few of those.
A technique I often use with Google to narrow down search results is to search for a small phrase or relatively rare word. I decided to do the same. I searched for macroblog, a word that does not get much usage, and thus found the additional three tweets shown above.

surprised

Twitter Search shows tweets in antichronological order, so I saw the last ones first, and was rather surprised to see my tweet retweeted but attributed to someone else, and with another URL.

Seeing someone else’s name in front of your tweet makes you appreciate just why it is called retweet thievery; it is not so much about their tweet, but about all others who retweet the stolen tweet correctly, attributing it to the thief instead of you; they steal not just your tweet, but also your retweet attribution.

case-sensitive

It was only when I started to think more about how to find retweets that I took the trouble to search for all occurrences of the additional short URL. That resulted in more hits than those three tweets, and immediately exposed a major weakness of this technique; The ow.ly URL shorten service is case-sensitive, and the twitter search is not. Putting the URL within quotes does not help, the twitter search is not case-sensitive.
I did not do so when I first tried to find retweets, but using BackTweets is much easier, and finds them all in one fell swoop.

credit

link

The last two retweeters were crediting someone else for my tweet. Now, that short URL still leads to the article on my web site. The site is still getting hits, and some readers may become repeat visitors. Some visitors may even follow the twitter link to my profile. Arguably, I have lost nothing, and gained a few readers.

credit

As JustinSMV already explained, it is not that simple. It is not just about the link and the visitors. In fact, I do not even particularly care about the number of visitors. I am not selling anything and there are no adverts making me any money either. I do not care about reaching many readers, but about reaching the right readers. Yet, something about the retweets bothered me.

I was not bothered much by one user leaving my name of. Mistakes happen. No, what bothered me was that subsequent retweeters crediting him instead of me. They were not making any mistake, but it just wasn’t right. This guy was getting props for my tweet, and I was getting none. I felt like these retweeters were complimenting him on my work, and that made it feel like he was showing off my tweet as his find. It sure felt like my tweet was stolen.

plagiarism

What makes it worse with this particular tweet, is that this tweet is not just about just the link. That one sentence is the essence of the article, and I feel pretty sure that it is that single sentence, not the link to the entire article, that caught the readers attention and begot the retweets. This is not just about a link I found, which someone else liked too, as in JustinSMV’s example. This is about right-out plagiarism of the tweet itself, a tweet that seems quite retweet-worthy even there were no link. That it does include a link does not excuse the plagiarism, nor does the inclusion of a completely obfuscated link feel like proper attribution. This was plagiarism.

what to do

I am still fairly new to twitter, and I do not doubt that I have made mistakes. What I did next may be one of them, I am still not sure. I decided to contact all three tweeters.

The result now, a few days later, is that the last two retweeters are now following me. They are thus not entirely unlikely to retweet my tweet about this article.

I still have not heard from the guy who tweeted without mentioning my name. Perhaps he is just on short absence, but the fact that he did not respond at all does not contribute to my confidence that his tweet wasn’t deliberate right-out plagiarism and retweet theft.

retweet thief or not?

Having just read the article on retweet thieves, I sure wondered whether I had just caught one.

To the casual observer, there appear to be just two possibilities; either he just happened across the article and decided to tweet about it or he is a retweet thief.

not indexed

For a short while I thought that there was no method to determine the truth of either, until I realised that it was hardly possible for him to have happened across my article just yet. I had just posted and barely tweeted about it myself. It was very unlikely that any search engine had indexed the article between my posting of the article and his tweeting about it.

another possibility

That no search engine had indexed my article yet may seem to clinch the case, but it does not. A third possibility is that he was a regular visitor of my web site already, noticed I had posted a new article, and then decided to tweet about it.

server logs

Well, even though I have never shown much interest in them, my host keeps server logs as a matter of course. Let’s assume for a moment that analysis of these log would show that he is not a regular visitor - that still would not prove he is retweet thief.

See, yet another possibility is that he just happened across my website, perhaps because of a link to another article, and decided to read my latest article. It may seem unlikely that it would happen in just those few hours, but it is not impossible. In fact, the more popular a site is, the likelier it is that several people will post a link to the latest article on various sites and link sharing services. I myself had posted a link on the Diigo link sharing service.

quote

What about the quote? Doesn’t that prove the tweet was stolen? No, it certainly does not. The article opens with that sentence, it ends with that sentence, and highlights that sentence as a pull quote. It also captures the essence of the article in one short sentence. If you happened across the article and decided to tweet about it, you’d probably take that quote.
However, when you do, you should make it clear that it is a quote, not your own text.

That’s it?

So, summarising the finds so far; he could have happened across the article in various ways, and the only valid criticism that remains is that he did neither mentioned my name, nor used quote marks? No, there one more argument to consider; the article’s subject.

twitter twist

Generally, the subject of the article would not matter in a discussion like this, but the subject of this particular article happens to be twitter itself, and that gives it the twitter twist.

Because the article is about twitter, anyone who did not find the article through of my tweet, but somehow happened across it, would still assume or expect me to be on twitter - and the Twitter button that links to my Twitter profile confirms that thought. Being a curious visitor, you would probably follow the link to my profile and thus have noticed my recent tweet about that article.

The article is not just about twitter, but specifically about posting links on twitter, and if you have read it, you know that I post links to my articles on twitter. The article actually includes its own TinyURL as an example.

So, if you are a twitter user yourself, and you just read that article, then you know that I am on twitter, and you know that I tweet my latest articles. Being an experienced twitter user, you know how you should retweet that - yet you tweet essentially the same message, and do not bother to use either quote marks or even mention my name? Wow, that is just plain dumb.

Perhaps it does not matter much that it is somewhat unlikely that you found my article independent of my tweet, but it does matter that the tweet neither quotes nor attributes my text.
As JustinSMV remarked, karma is a b*tch. Expanding that succinct wisdom into more polite English: almost everyone who reads this article is going to assume you stole my tweet, despite all my protestations that they should not do so.

curious

I was curious about the apparent retweet thief, who seemed to be making good show by plagiarising my work. I looked through the recent tweets on his profile and immediately noticed several tweets with links that started with Reading as the tweet to my article did.

The thought occurred to me that he might be in the habit of tweeting interesting links he found about that way. I decided took a look at his recent tweet about URL shorting.

Reading: URL shortening

Twitter search: URL shortening Yet another security risk

2009-03-09 11:59 TRBlogs URL shortening: Yet another security risk - http://tinyurl.com/9y7aov

2009-03-09 12:02 Aprazeth RT @TRBlogs URL shortening: Yet another security risk - http://tinyurl.com/9y7aov

2009-03-09 12:11 robertpohl RT @TRBlogs URL shortening: Yet another security risk - http://tinyurl.com/9y7aov

2009-03-09 12:21 robertpohl RT @TRBlogs URL shortening: Yet another security risk - http://tinyurl.com/9y7aov

2009-03-10 12:24 JoachimNiemeier Reading: URL shortening: Yet another security risk - http://bit.ly/eOM

2009-03-09 12:29 PCdidIT RT @TRBlogs URL shortening: Yet another security risk - http://tinyurl.com/9y7aov

2009-03-10 12:31 rainerhelmes RT @JoachimNiemeier: Reading: URL shortening: Yet another security risk - http://bit.ly/eOM

I just checked, and found that JoachimNiemeier does follow TRBlogs. He has seen the original tweet, yet less than half an hour later, posts the link himself without crediting TRBlogs.

That sure seems to clinch it, does it not? This guy is a freaking retweet thief! Let’s not make any snap judgement, but continue our research. There are more issues to consider. Perhaps he simply does not know about retweeting?

twittiquete newbie?

twitter oldie

Perhaps this particular twitter user is a relative newbie, who has yet to learn twittiquete in general and about retweets in particular?

No, this particular user is a twitter oldie. He posted his first tweet about two years ago, on 2007 Mar 24. That particular post is

2007-03-24 14:29 JoachimNiemeier First post about twitter in my weblog: http://www.7daysandmore.blogspot.com/.

identity

That post links to his blog and his blog archive contains the blog entry the tweet refers to. We thus learn several things. Most importantly, we confirmed identity. An issue I did not address yet is that anyone could have used some name to create a twitter account, but with the blog and the twitter account repeatedly linking back to each other, it is reasonable to assume that both are by the same single person.

twitter thinker

Another interesting thing we learn is that he has not just been using twitter for two years already, but has also been thinking and writing about twitter for two years.

His blog does not use tags, but a quick google (google twitter inurl:7daysandmore) finds multiple articles about twitter, from that earliest one posted two years ago, to the latest blog entry less than a week ago. There are many blog entries with web 2.0 in the title, so this is not someone who merely uses twitter. This is someone who actively thinks and writes about twitter, and that makes it highly unlikely that he does not how to retweet.

retweet knowledge

Browsing through his tweets, acknowledgements of other tweeters seem rare, but does turns up positive evidence that he knows how to retweet:

2008-07-01 16:21 JoachimNiemeier Retweeting @centrestage: Wiki und vertrauliche Informationen? Interview zum Einsatz eines Wikis bei der CIA: http://tinyurl.com/6furjy.

Another recent tweet shows that he has not forgotten how to retweet.

2009-03-09 15:18 JoachimNiemeier RT @aponcier Enterprise Web vs Consumer Web [2.0]: Top Six Differences - http://bit.ly/U47SP /// Good points!

The obvious conclusion is that he knows how to retweet, yet rarely does so. It sure seems that, although he is probably far from a twitter newbie, and knows about retweeting, still rarely credits his source. It seems that, more often than not, he just doesn’t bother to do so.

caught in the act?

Did I catch a retweet thief, a recidivist even? No, sorry, I have to disappoint you. Although the circumstantial evidence may seem overwhelming, damning even, it is just that; circumstantial evidence. There is no proof.

All the above is merely circumstantial happenstance. Call it a statistical fluke worthy of the Guinness Books of Records. Do not let appearances fool you, do not let coincidences lead you to premature conclusions of retweet thievery.

real-life professor

How can I be so sure, after presenting all this evidence, that this guy is not a retweet thief? Well, I browsed through his tweets and blog posts, I followed the links. I know who he is.

Joachim Niemeier is a real-life professor at the University of Stuttgart. An academic who would never dream of taking credit for someone else’s hard work. Someone for whom it is not second nature, but first nature to always and diligently acknowledge all his sources as a matter of course.

Sure, he did slip up when he forgot to at least put quote marks around that sentence, but we all slip up from time to time. That is only human. I am sure that I have made my share of twitter mistakes, and I would not be happy if anyone tried to blow some honest mistakes out of proportion. Besides, there is nothing unusual about somewhat absent-minded professors.

This particular professor kindly shares his interesting finds with his followers. I want it to be perfectly clear that the mere fact that is he is rarely the first one to do so does not mean anything. Rarely is anyone the first to tweet an interesting link.

conclusion

I only used the tweets and retweets for my previous article - and thus his tweet and subsequent retweets - as an example because it is the example that got me thinking about researching retweets, and it turned out to be an excellent case to demonstrate various research techniques with.

I am confident that he accepts this modest case study in retweet research in the same academic spirit that I offer it, and I do thank him for creating such an excellent research example. I offer links to some of his blog posts about twitter in return.

I am honoured to have this respected professor among my readers. I am especially honoured that he found my article interesting enough to share it with his followers.

The one conclusion I draw from all the research is the one thing I knew as soon as I saw that tweet of his: he has excellent taste in articles. I therefore encourage you to follow him, but promise not to object if you decide to follow me instead.

Tweet!

I value you as a reader, and would not want anyone to think you stole a tweet about retweet thieves, so here is a tweet for your retweeting convenience.

2009-03-17 02:30 TamuraJones "Catching a ReTweet Thief?" article updated with "thief" response http://tinyurl.com/rthief

The original tweet about this article is

2009-03-13 01:59 TamuraJones ReTweet research techniques article: Catching a ReTweet Thief? http://tinyurl.com/retwhief

but the more recent tweet is better because its TinyURL is for a permalink.

updates

2009-03-16 response

Joachim Niemeier has replied to my email and tweeted a response to start a discussion. He notes that when he says reading, he really means recommended reading. I expected as much and why anyone leaves out recommended when restricted to 140 characters needs no explanation. To him, reading means more than just RT. I had to think for a moment how reading could be better than retweet, but I think I get his point; a retweet calls attention to anything interesting someone else tweeted, while reading is a direct, personal endorsement.

He notes that the author remains recognised, because he merely used a different shortener. Sure, if you follow the link in that particular tweet to my website, it will be clear that I am the author of the article. But the tweet does not make it clear that he is quoting the key sentence of the article nor that this sentence is really a retweet that came to his attention by an earlier retweet; neither I, nor the first retweeter in the retweet chain is getting any credit for putting it into the twitterverse. The intention may be to give more props to the article than a retweet does, but done this way, the twitter props are lost.

2009-03-16 12:28 JoachimNiemeier @TamuraJones started a discussion about #ReTweetThief: http://bit.ly/RjGT using me as an example. For me "RT" is different to "Reading".

2009-03-16 12:44 JoachimNiemeier @TamuraJones (1) When I write "Reading" it means something like "Recommended Reading" and explicitly not only RT. #ReTweetThief

2009-03-16 12:47 JoachimNiemeier @TamuraJones (2) You as author are fully recognized, I used a different shortener: http://tinyurl.com/twitisblog vs http://ow.ly/LHG

2009-03-16 12:48 JoachimNiemeier @TamuraJones (3) I will follow you from now on to be able to Re-Tweet your Tweets :-) #ReTweetThief

2009-03-16 12:48 JoachimNiemeier @TamuraJones (4) But I still think "Reading" has a different value proposal #ReTweetThief

2009-03-16 12:50 JoachimNiemeier @TamuraJones (5) This is NOT a viral campaign for @TamuraJones and @JoachimNiemeier :-) #ReTweetThief

2009-03-16 12:51 JoachimNiemeier @TamuraJones I'm very interested what the community means about #ReTweetThief

2009-03-16 13:04 JoachimNiemeier Here is the right link to "Catching a ReTweet Thief?": http://bit.ly/6IJA7 #ReTweetThief

2009-03-16 13:27 RicardaBock @JoachimNiemeier: "Retwittering" as a science in its own rights. Understand your meaning of "reading" but think, RT is better. #ReTweetThief

2009-03-16 13:40 JoachimNiemeier @RicardaBock For me it's not "better", it's "different". Eg. with RT I give credit to someone who shared content she/he found. #ReTweetThief

2009-03-16 13:55 RicardaBock @JoachimNiemeier For me, the difference is not big enough to avoid misunderstandings (sic!), therefore: rt ist better.

updates

2009-03-17 linkage

Original short URL did not use permalink. In this article, the titles of all short URLs remain as they are, but the destinations have all been replaced with the permalink to avoid future dead links.

2011-04-23: antains

Twitter user antains is gone. The link now points to twitter instead of his tweet.

2011-04-24: Zudfunck

The 2009 Mar 9 Zudfunck blog post Retweet Theft, A Cancer Upon Twitter is gone. The blog archive starts with September of 2009. The broken link has been removed.

links

articles

services

professor's blog