I've noticed for a while now that the value of n seems incorrect. Idly, I asked on Twitter if people have noticed this.
@IgorCarron said that "Jim plus n others" means there should be n+1 people in total. This is what I would expect, given my fragile grasp of the English language, but I suspect this is not what is happening. I think n is actually the number of retweeters including Jim, not excluding him as the wording would suggest. Then the correct form should be "Retweeted by Jim and (n-1) others" or "Retweeted by n people including Jim".
My cynical brain says people would object to and complain about a lower number being displayed so Twitter would put the higher number, but then why not use the "n including" language, which has the virtue of being correct?
@christianp has seen "and others" when only one person has retweeted a message! @AlexHymers reported having seen inconsistencies when only two people have retweeted a message. If the problem is at the n=1 end of the scale only, that is interesting. If this were a poorly coded student project I might suggest the error is in the part of the code that decides whether to pluralise the "other(s)". (Of course, this is a major international company and not some poorly tested bit of code...)
With an sensible practical mindset, @relinde suggested an experiment: "Ask two people to retweet your message, and see if it says 'by Jim & 1 other' or 'by Jim & 2 others'. I think the first."
I sent a message (Tweet 1) to @relinde and @christianp asking them to ReTweet (RT) my message. @christianp did so. Impatient, I tweeted out generally for one person to RT the message (Tweet 2) giving a link to Tweet 1. @AlexHymers retweeted Tweet 1. Then @relinde did as well. My impatience had got me three RTs. The result? Tweet 1 was "Retweeted by relinde and 3 others". The wording suggests four, when only three had retweeted it.
However, at the same time two users, @OnThisDayinMath and @englishblonde had, misunderstanding the instruction, retweeted Tweet 2. This now displayed "Retweeted by englishblonde and 1 other". The wording here was correct: 2 retweets.
Remembering hyenas and monkeys can count to 3, is it the case that Twitter can count reliably only as far as 2?
One final note: @timtfj says "n in 'retweeted by A and n others' seems to vary between meaning n and n-1, but always to be an integer". Does anyone have a counterexample?
3 comments:
I was looking at my followers list earlier and, for some people, Twitter couldn't even be sure if I was following them or not.
Maybe the authors of the paper "Does Quantum Interference Exist in Twitter?" - http://arxiv.org/abs/arXiv:1107.0681v1 - were right in more ways than they thought!
I've seen this before. I thought that perhaps the tweet was retweeted by people with locked accounts. Then (perhaps) their picture wouldn't show up, but they would be counted. So in your example, maybe your tweet was retweeted by @relinde and 3 others, one of whom has a locked account.
I don't think that would explain all of it - it's definitely not the case with my tweets which had "and others" immediately after posting.
Post a Comment