It appears our "tweet your answer" button does not truncate the text part of the answer if the resulting tweet will be over 140 characters. This was from an answer on Sci Fi, I haven't edited the tweet at all:

The tweet displayed inside the Twitter share window. The tweet consists of the site name, question title, answer link and some accompanying text. Twitter indicates the tweet is one character too long.

For those who don't know, that means you can't send this message without editing it. You should just be able to tap the Tweet button and it works.


This seems OK in my opinion. Twitter gives a useful error message to a situation which ought to be handled by the user. You wouldn't want to just truncate the title by an arbitrary number of characters; you'd end up with gibberish, misinformation, or obscenities (depending on your luck).

The user can trim this as appropriate, by:

  • Converting "Stack Exchange" to "SE"
  • Converting "Science Fiction and Fanasy" to "scifi"
  • Removing "In production" from the title
  • Removing your tracking reference /3255 (though this might get stripped out by t.co; I don't use Twitter enough to know whether this is important.

Or, best yet, writing your own tweet! A stream of "My answer to Stack Exchange [sitename] Q: [title] [url] is more than a little inane. How about something like

Turns out the Ferengi "Divine Treasury" doesn't exist in this(?) life. Bummer! https://scifi.stackexchange.com/a/8365/3255

Real-world uses of Latinum in Star Trek: https://scifi.stackexchange.com/a/8365/3255

It seems that Latinum, unlike gold, isn't really useful: https://scifi.stackexchange.com/a/8365/3255

or something else a little more...human than the stock phrase.

Granted, "I quoted a list from MemoryAlpha" isn't an answer I would tweet. Preferably, you'd have answers with creative insight or that Used Science that seemed interesting. Summarize the answer in a hundred characters or give a teaser, and provide that link. Tweets like that seem far more interesting (again, I don't use Twitter much, so I wouldn't know) than bot-generated ones.

  • Expecting the user to trim the tweet is a pretty big burden vs hitting "Tweet" and it goes. An advanced user can skim the tweet and check for a malformed tweet but it should be as easy as possible for anyone else. – Ben Brocka Dec 30 '11 at 22:50
  • 2
    An advanced user would open a new tab and enter a clever, fascinating tweet of precisely 140 characters without their fingers ever leaving the keyboard. A busy person who didn't care about the quality of their tweet would click the button and let it go. – Kevin Vermeer Dec 30 '11 at 23:44
  • 1
    @BenBrocka It's probably a better option than trying to programmatically produce a meaningful trim of the message. – Adam Lear Dec 30 '11 at 23:50
  • @KevinVermeer Assuming "advanced users" don't care about usability is an extreme pitfall. Share buttons are supposed to make sharing easy, not be ignored because "oh, the user will do X anyway". – Ben Brocka Dec 31 '11 at 19:51
  • @AdamLear but if I have to make my own message, why is there a share button? Using twitter for any period of time you'll notice very very few people actually edit the messages from "tweet this article" buttons, you're not supposed to. If you wanted to make up a completely new title you can do so, but it shouldn't be required. Low friction (but not frictionless) sharing is a huge part of what makes the social web powerful; people actually use it. There's a reason twitter and Facebook are more popular than sharing links on IRC. – Ben Brocka Dec 31 '11 at 19:57
  • @BenBrocka - Blech. Count me out of your social web with unedited tweets! I'd rather read one person sharing a few dozen truly interesting links with keen insights over a hundred people sharing thousands of pieces of banal, machine-generated fluff. The last thing we need is another source of links; there's more than anyone can or should digest available already. Stack Exchange shouldn't represent itself with crap like that. – Kevin Vermeer Dec 31 '11 at 20:52
  • 2
    @BenBrocka Popping up a dialog instead of asking the user to go to Twitter and manually copy the post URL, etc. already removes a lot of friction, IMHO. It's not perfect, but there's only so much you can do when the target network limits the message length. How would you programmatically modify a sentence to shorten it without losing semantic meaning? – Adam Lear Dec 31 '11 at 22:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .