Anyone looking to actually verify the source (this is at least partially why they're cited, right?) is going to have difficulty locating the original tweet without the permalink. The original tweet, along with the link, can be deleted - but that's true for any web resource.
Ultimately, I wonder why tweets are cited any differently than any other web resource. I'd find the following more useful:
news.yc popular (newsycombinator). "How do I cite a tweet? http://j.mp/ygeHyP" 19 Jan 2013, 5:05 p.m. Tweet.
I guess it's handy to have the full body of the tweet right in the citation, but that would be true for any short quote from any resource - both on and offline.
I think the assumption is that the tweets will go away. Look at the twitter search; I mean, they can't seem to search tweets more than a few days/weeks old. Clearly twitter does not place a high priority on keeping historical data.
They don't prioritize finding historical data, which is why the permalink URL is all the more important. There are (or at least were) third-party services dedicated solely to finding older Twitter data.
Again, this is no different than any web resource. We even have an HTTP header specifically to indicate content has been deleted.
I am not, but if the purpose of a reference is to allow others to verify and find the original source, using anything but the original link is ridiculous. I added the JSON link to prove the point that you can get exact dates/times for a tweet (including info about the currently active timezone for the author's account)
A link is still valuable because it shows where the writer of the publication accessed the information. Yes, it might be deleted, but it might be possible to retrieve a copy of what was at that URL, or it might be discovered someone keeps a record of an alternate location of the content that was at that URL, or at least the revieer can verify that a valid-looking URL as cited if no other evidence can be found. A citation is supposed to supply a lot of information to make research easier.
For this reason (and others) I vastly prefer the APA's standard which would certainly require the URL, and would put the tweet text in as the title unambiguously as Twitter's site is structured that way. Thankfully the APA is still the gold standard so hopefully not many will be citing tweets in the MLA's useless way.
Submitted w/o opinion, I found this comment in a discussion [1] of the general MLA guidelines for websites:
MLA no longer requires the use of URLs in MLA citations. Because Web addresses are not static (i.e., they change often) and because documents sometimes appear in multiple places on the Web (e.g., on multiple databases), MLA explains that most readers can find electronic sources via title or author searches in Internet Search Engines.
If you want to help the reader as little as possible, imitate Krantz's hypothetical example of a minimalistic bibliography entry: ``Machedon, 1988.''
Readers may complain. In response, point out that the reader can easily find the cited work given a tiny amount of information, in this age of Google and CiteSeer and MathSciNet and the Collection of Computer Science Bibliographies. For example, those of us with MathSciNet access can do a search for machedon and 1988 and find details of two papers by Machedon in 1988; presumably the reader of Krantz's hypothetical citation can figure out which paper is the right one.
A URL is the only way to go and apart from sheer convenience, it conveys that it's coming from Twitter, instead of the solitary "Tweet" at the end of the MLA's citation. Twitter isn't a timeless utility, it's a company running a service that happens to be popular at this time and in certain countries. You can't just say "Tweet" and expect every reader to always get the meaning of it.
I think it's okay to include the actual text too, as the short length makes it possible to do so and given that tweets don't have any title and are probably less likely to be preserved if Twitter went down than most other online references a scholarly article might cite.
Citing a tweet without including its globally unique ID seems terrible. Usernames an be easily changed. Copying the entire contents of the tweet into the citing seems odd as well.
> Copying the entire contents of the tweet into the citing seems odd as well.
On the contrary, it builds in redundancy. One of the virtues of Twitter is that this approach is feasible because of their character limit, and it means that should anything happen to Twitter (or the internet) in the future, that anything of cultural importance* is preserved. It also allows the tweet to be read offline or in countries where Twitter might be censored or banned, but where the paper is available. Proper citation is just to enable verification if necessary.
(*though admittedly this is probably a very small subset of all tweets)
It's terribly inconsistent. I'm in the middle of developing a library to generate citations, and I've had to write out the formatting for each separate type of citation manually (for example, a whole book's title is italicized: in a series, it is underlined).
It's intended for schools and will be a pay-for service, but the core citation-creating tool is going to be available as a RubyGem. http://github.com/noted/scholar
As a student, I shouldn't have to know about complex citations rules while writing a paper. There are a few tools that take care of this, but they are either non-free (EasyBib), non-comprehensive (other online generators) or require that you use certain software (EndNote). To my knowledge there is no free, universal citation generator.
Great, thanks for the reply. I'm also working on something to compete with Easybib. But at the moment it definitely falls into the 'non-comprehensive' category that you mention.
Tweets have a known level of volatility and a strict maximum length. Websites are 100% unpredictable. Why not take advantage of the known tweet properties to optimize this case?
I think the most troubling feature is the time convention "time zone of the reader". How should anyone make a valid assumption about that? I think its a better practice to include a common time zone abbreviation.
This bothered me as well, particularly for the example that they provided. The convention may work for tracking conversations, but knowing the exact time that the helicopter was spotted relative to other events seems more important. Yet, by their convention you can only accurately compare it with the time of other cited tweets.
I think this is the only MLA style that is tied to the offerings of a single commercial entity. That seems weird to me. Does Twitter own the idea of a tweet or has it escaped into the general culture?
General culture. It seems standard practise now in print and video journalism to quote tweets about whatever current affairs you have: "user FooBar on twitter said 'I don't like this thing'"... regardless of whether FooBar has anything to do with it at all. A lazy way to get bystander comments from people who have even less of a relationship to the happening than bystanders do.
It's strange that when you cite a tweet, MLA wants you to include the tweet contents, whereas for every other citation, they want you to embed just enough info to form a pointer to the actual content. Tweet citations are closer to footnotes than to citations.
What about other specific online media? How do I cite a Reddit comment? An insightful meme in QuickMeme (the text could be helpful and part of an overall commentary, regardless of how you feel about memes in general)?
I feel the best thing is like an online source - aka like taking a 140 character quote from a blog post. MLA feel differently, obviously.
To obtain a stable link I've seen some publishers use a service called WebCite (http://www.webcitation.org). It archives the referenced page as the author would have seen it at the time of publication.
"Begin the entry in the works-cited list with the author’s real name and, in parentheses, user name, if both are known and they differ. If only the user name is known, give it alone.
Next provide the entire text of the tweet in quotation marks, without changing the capitalization. Conclude the entry with the date and time of the message and the medium of publication (Tweet). For example:
Athar, Sohaib (ReallyVirtual). “Helicopter hovering above Abbottabad at 1AM (is a rare event).” 1 May 2011, 3:58 p.m. Tweet."
Why would one ever cite a tweet? You can't write a useful secondary source in 140 characters; surely a tweet is only ever going to be a primary source, like an interview or experiment that you've conducted yourself? While we might include transcripts or raw data in an appendix, we don't really cite such things per se.
Wasn't the example the MLA gave a pretty good one? It was a tweet by a resident of Abbottabad noticing a helicopter, which later turned out to be US Special Forces coming to kill Osama Bin Laden.
Exactly; that's not a secondary source, it's a primary one. I don't think you'd want to cite it any more than you'd want to cite an interview with that resident.
Or the JSON API, which gives _exact_ times including timezones: http://api.twitter.com/1/statuses/show/64780730286358528.jso...