5 ms·

It would seem to me that the solution to the problem described in the blog would be to reduce the number of characters allowed for Japanese, Korean, and Chinese

• kijin 9y ago
  There is precedent of treating CJK characters as "double-width", but it's anachronistic in a world of proportional fonts and UTF-8. CJK users also don't use CJK characters exclusively. There's almost always single-byte characters mixed in, such as URLs and numbers, and in the case of Korean, even punctuation. If some characters count as 1 and others count as 2, it will result in fluctuating limits that are hard to grasp intuitively.
  • kccqzy 9y ago
    If you used CJK before UTF-8 was popular, you would know that people had absolutely no issue counting some characters as 2. It’s entirely natural. In typical proportional fonts CJK characters still occupy about twice the width of a Latin character.
• saghm 9y ago
  From the standpoint of not wanting to alienate users, I can see why they might be hesitant to go this route; it's much more likely to piss people off if you take away some of their characters rather than giving them more.
• 482794793792894 9y ago
  Yeah, this reasoning especially also struck me as weird, because Twitter's 140 character limit was chosen from an English-speaking viewpoint. The limit for Asian languages should have been adjusted downwards as early as Twitter launched (though I suppose that's actually somewhat hard to do from a technological viewpoint, as people might mix Western letters and Asian symbols). And then they point out that in Asian countries people tweet more, so alright, who cares about history, do what works best. But you shouldn't just be looking at how many tweets are written. People are going to post more, just because it's easier to post something. That doesn't mean that people actually read these longer tweets just as much. Like, there is a feedback loop here, people would also start to tweet less, if they didn't bother reading tweets, because many tweets are also responses to other tweets, but still, this reasoning seems quite a bit oversimplified.
  • hkmurakami 9y ago
    In Japan at least, the anonymity factor was at least as strong as the information density factor (TWTR > FB for a very long time there because people there generally don't want to tie their irl identities to online accounts).
  • bogomipz 9y ago
    >"Yeah, this reasoning especially also struck me as weird, because Twitter's 140 character limit was chosen from an English-speaking viewpoint." I don't believe this is correct but rather 140 characters limit was chosen because SMS will break a message greater than 160 characters into two messages. 140 characters leaves 20 characters for a username.
    • e12e 9y ago
      Well, it's correct in that 160 characters was chosen from a German (I presume, possibly English/"European") viewpoint, for sms...: http://latimesblogs.latimes.com/technology/2009/05/invented-text-messaging.html http://latimesblogs.latimes.com/technology/2009/05/invented-...
      • bogomipz 9y ago
        I've heard this story before and I think it might be apocryphal. Usually that same LA Times piece is cited as well. More likely is that 160 characters was what the GSM signaling channel comfortably allowed, essentially making it free. Its still this way today - SMS uses an out of band signaling channel and cell phone companies charge you as data.
    • mads 9y ago
      140 characters is an 8 bit SMS message. 160 characters is a 7 bit SMS message. You can also send a UCS2 encoded SMS, which can contain 70 characters.
  • improbable22 9y ago
    When I type an actual SMS, the 160 limit seems to be reduced if I add non-English characters, like emoji or possibly even just an em-dash. So they could simply have followed SMS more closely, 140 bytes or something.
• rtpg 9y ago
  A lot of Japanese people use Twitter quite effectively as a micro-blog. This is harder in English because of the character counts. So it could be that "Japanese-style twitter" is the true product vision, and it's not working in English The original 140-character constraint was an external restriction from formatting, it could be that 280-characters is, in fact, the "right" number of characters for the service they want. So many people complain about the overly-terse nature of tweets. This can help tackle this and get to a better product.
• jpalomaki 9y ago
  Doing this distillation seemed to be too much trouble for many users. You see people splitting things to multiple tweets, using images or simply posting links. Trying to for example summarise interesting stuff you seen online to 140 chars is hard. Every increment in the character count makes it a bit easier. 140 to 280 has much larger effect that say 280 to 420, so I don't see slippery sloap here.
• freyir 9y ago
  They found out that Japanese Twitter users don’t run up against the character limit, and are less frustrated as a result. The point is to make everybody else less frustrated, not to make the Japanese as frustrated as everybody else.
  • benguild 9y ago
    Yeah, you can type a lot more in Japanese. Most characters stand for at least two US characters, which is most likely how they came up with the 2x decision.
    • Mithaldu 9y ago
      Sadly that is not going to be sufficient. Japanese uses very few context indicators and thus as a whole is a lot shorter.
  • odbol_ 9y ago
    Found the UX designer in this sea of clueless engineers
• vacri 9y ago
  > the system forced you to reconsider how you said it until you got down to the bare essence of what you were trying to convey ... and stripping away context and nuance, and ultimately providing such a paucity that people these days have trouble recognising context and nuance. Twitter's played a major hand in the damage to people's attention spans and ability to interpret communications effectively. Recognising subtext no longer seems to be a thing anymore.
  • zimpenfish 9y ago
    > Twitter's played a major hand in the damage to people's attention spans and ability to interpret communications effectively. That, right there, is going to need a hefty wallop of [citation needed].
    • vacri 9y ago
      > [citation needed] - vacri, 6 hours ago.
• Grue3 9y ago
  You're missing the fact that Twitter's engagement is way higher in East Asian countries and that's because they can pretty much use Twitter as a normal blog. On the other hand in the countries with very long words (like Hungary and Finland) Twitter is not very popular. It's been clear for a while and I'm surprised they haven't implemented this sooner.
  • numerum-bestiae 9y ago
    Must be true, as a native speaker of Croatian, I find most Tweets in English understandable, not in Croatian, however. People tend to send 4 or 5 tweets instead.
• developer2 9y ago
  >> I'd be in favor of a maximum number of tweets per day, too. A while back, I spent a few days hashing out a concept for a social network that would impose posting limits. The initial obvious approach is a single post per 24 hours, to really make each person think about whether hitting send is worth it. However, I cast this aside as being too restrictive. Introducing the concept of categories: you can post once per 24 hours, but per user-defined category, up to some maximum category limit per account. You could post once about your cooking hobby, once about your programming hobby/career, once about the news, etc. Your followers could subscribe or unsubscribe to your individual categories; someone can opt to see your cooking and news posts, but ignore your posts about programming. The concept starts to collapse at the "policing" stage. People would cheat by using all of their categories to spam a single topic, or to bypass followers' unsubscribing from a topic they just must be heard on. This necessitates a moderation system whereby followers could vote on posts, or report those which are miscategorized. With such a system in place, the possibility exists to increase the limit to 2-3 posts per day in categories for which someone receives overwhelmingly positive (or nonnegative) feedback. Dealing with users' and bots' vote manipulation is of course a never-ending battle. The concept completely disintegrates when you consider who, as a publisher of content, would choose such a platform. A lot of people love sharing their every thought, meal, personality test, quiz, BuzzFeed article, etc. Such people would never opt in to a platform that limits their ability to do so - and why should they? Companies, celebrities, and other popular figures rightfully depend on being able to communicate whenever it suits them, so this concept is utterly useless to them. Thus the market is tiny, to the point of being non-existent, for limiting the publishing side of the equation. This pushes the entire problem to the side of the readers/followers. We as consumers really just want to be able to intricately tailor what we see. I'm hoping we will soon have AI that can appropriately categorize posts, so I can filter out every single post having to do with personality tests and quizzes, BuzzFeed-like links, anything and everything having to do with US politics (I'm not American), animal abuse (I'm really tired of "save the animals"), and the constant stream of photos from new parents parading their children. I stopped using FaceBook entirely because their filtering options are just not adequate. Nor do I expect FaceBook to ever offer such filtering, as it would nullify all the mindless drivel that keeps their users engaged with their platform. Lastly, I considered a monetization strategy whereby a user could pay (heavily) to post above their allocated limit. It sounds useful if you imagine people using it very rarely to push out that one extra gem that surely all followers would love to see. Of course, "pay to play" has no place in a social network designed to reduce the number of posts, and I couldn't come up with a solution to have my cake and eat it too. A "pay per post" strategy would be the foundation of a standalone concept - one for which no market exists. tldr; A social network that limits publishers' ability to maximize engagement will never gain traction. What we need, as consumers of social networking, is the ability to perform advanced categorization and filtering of existing content. Don't get your hopes up - FaceBook and Twitter will never allow us to skip straight to the "good stuff", as viewing only a dozen posts a day would decimate their engagement stats, including ad views.
• VectorLock 9y ago
  Most of the original design intent that made Twitter great has been washed away by employee turnover.