> You can say 3-4x as much in a single tweet in Japanese or Chinese as you can in English (example: 象 = elephant, yet is only one character instead of eight.)
Can you really? I thought the "140 character" restriction was due to the "160 bytes in an SMS" restriction, and therefore using multibyte unicode characters consumed multiple "characters."
Twitter's web UI suggests it's allowed, and I can post 140 象 in one tweet.
In practice, tweets get fragmented into multiple SMS anyway, so while this might have been the aim, it's not the implementation they've ended up with.
Also, while you may wish SMS was Unicode, that's not actually likely.
The number of characters you can fit in one SMS will vary depending on whether you use more than one charset, and what your operator and device support. Most SMS in the US/UK/Aus etc will be sent as 7 bit GSM, allowing for 160 characters. 8 bit allows for 140 and UCS-2 allows for 70. (This is assuming no UDH, i.e. no fragmentation)
I can send fully unicode SMS here in Germany, and while my old phone doesn’t support all the characters, it definitely displays even multibyte characters properly.
They got rid of the SMS-related limitations ages ago, but kept the 140 limit because it was "part of what makes Twitter - Twitter".
It's also why Twitter is really popular in Japan. 140 characters is more than enough to hold actual conversations (even if most statements/responses are typically 12-20 characters and the rest of the space is used for emoji)
I think Twitter's popularity in Japan, despite the fact that 140 characters there can convey a message that's 3-4x larger than in English, is indicative that increasing the English character count to somewhere between 280-560 (or much better, moving to word counts instead) would not be the death sentence for the service that many make it out to be.
Which one might call a "natural experiment" for how increasing the limit would affect English speakers' use of the service. (But I don't think Twitter will do it - the number 140 is so well known, it's basically part of its identity.)
You can try it for yourself. Go to twitter.com and put 象 into the box. The number of remaining characters only goes down by one.
I don't know why that is, because I'm pretty sure you're right that the tweet length limit comes from SMS limits. But these days the 140 limit seems to be in terms of Unicode code points, not bytes.
Can you really? I thought the "140 character" restriction was due to the "160 bytes in an SMS" restriction, and therefore using multibyte unicode characters consumed multiple "characters."