views:

506

answers:

2

These people are selling a proprietary exclamation mark for sarcasm, and without going through the Unicode standardisation process, are proposing that it's possible to download this mark (in what format it's not completely clear), and use it in your written communications (to what extent it's not clear). My question is what exactly is being downloaded and how exactly are they transmitting it across a network.

I'm not sure if it's some sort of multilayered sarcastic joke and I don't think it's a nice idea on any level, but I'm curious if they're doing anything clever to implement it and get it to interoperate (curious enough to override my reluctance to add to their word of mouth).

  • They provide an image that you can paste into a document - so far it's a $2 piece of clip art, nothing mysterious about it.
  • They provide a font with the new glyph (or is it a character?) - presumably if you were going to send that successfully to someone they'd have to have the same font installed and the message would have to have enough information for the recipient to choose that font for viewing it?
  • They claim to have it working for SMS on Blackberries - how do they get that to work? How is the glyph transmitted through the network? The only way I can think of is that they're transmitting Unicode private use codepoints, but those would be tricky to get through an SMS network, wouldn't they?

EDIT: There's a vote to close with a "non programming-related" justification. I'm a programmer, I spend a lot of time tracking down problems where ancient and well established characters fail to get transmitted across the network correctly. So when someone claims to be able to invent a character unilaterally and get it to transmit between devices correctly and without degradation, I'm curious, I wonder how one would program it. It's programming-related - it's related to my programming and to the programming of the company that implemented the symbol.

EDIT 2: A basic question, raised by Jeremiah's answer below, that someone who knows about fonts should be able to clarify: does it suffice to provide a font with just the one new glyph in it? What I mean is, if you're viewing a document in say Arial, the system comes across the U+E001 codepoint in the document, and it's not part of the active Arial font, will the system check all of its installed fonts to find one that has a glyph for the codepoint? Or does the new glyph have to be somehow added to every font you intend to use? I'm being vague about the word "system" - I don't know if that functionality is implemented by a word processor or browser application for example, or by some underlying font service in the operating system.

+1  A: 

Probably the first place to start is to see what applications this new character works in. I think the most useful place for this character would be forums because that seems to be where sarcasm is mostly seen. So if it is forums than the only way to add sarcasm using your guesses to how it's implemented is 1. use [ img ] (only works for certain forum software) tags 2. modify a font family though I'd imagine you'd need to modify every font family for the character to show well (i.e., if one website uses Times New Roman and another uses Arial) My only other guess is that there's some sort of tag being uses and that it's a addon to IE (and other webbrowsers) that catches the tag and changes it to a SarcMark. I know that addons such as Greasemonkey for Firefox (and other webbrowsers) are capable of reading the text and html on a webpage and making changes to that html as it processes the data and sees certain fields of data. I know that one Greasemonkey extension is used to shorten long sentences that people write up on forums so you can quickly jump to the next persons post. This sarcmark could work like this but of course we wont know unless someone tries it and is able to discover if and how it works.

Jeremiah
The proprietary markup + app-specific plugins idea is interesting - it's like tunneling the new character from application to application in a way that's opaque to the infrastructure. It would also explain how two devices with the decoder ring installed could send the special character to each other in an SMS. You're right that it's all speculation until someone installs and investigates - I really should do it myself seeing as I asked the question, but I'm reluctant to support the enterprise.
d__
+4  A: 

Found a good answer in this blog post.

First off, the mark is mapped to Unicode code point U+E001, which is in a Private Use area, and is not part of the Unicode standard. Next, in order to see the SarcMark glyph, you need a special font which contains the mark at U+E001.

Below is a sarcmark. What it looks like or whether it appears at all is dependent on what fonts you have installed, and on it not being filtered out by StackOverflow. In order for the SarcMark itself to appear, you'll obviously need the proprietary font installed.

THE SARCMARK IS BETWEEN THE BRACKETS: []

I did a Google search for the character and found nothing. This Twitter search will find some uses of it out in the wild, and also some uses of the codepoint that don't seem be intended as sarcmarks. In fact, U+E001 is used in the SoftBank variant of Emoji (which, based on its usage on Twitter, seems to be used beyond the Japanese SoftBank network) for the "BOY" or "BOY'S HEAD" character (see it here). There isn't very much to be done about that when using Private Use codepoints - if someone Twitters, "[U+E001] that was a terrific movie", they either mean "Boy, that was a terrific movie", or "My next statement will be sarcastic: That was a terrific movie", so it sometimes works ok with either interpretation.

To send it through SMS, something like UCS2 would have to be used rather than a 7- or 8-bit SMS encoding.

For comparison, both in terms of the idea and the implementation, have a look at the project for making Emoji part of Unicode. If you have a GMail account, you can try sending yourself the message below. If all of the various parts work (cut and paste, databases, email infrastructure, browsers), Gmail will display an Emoji smirking face. U+FE343 is the Private Use Area codepoint where the glyph is being parked until an official codepoint is assigned to it, and GMail substitutes an image for the character when it displays it.

EMOJI SMIRKING FACE: U+FE343, utf8 f3be8d83 []

The Private Use Area codepoint + special font technique is sufficient to make the symbol transmittable through networks and storable in files and databases (most of the time), and viewable by users. The only technical thing to add to get the whole system to work is to make the symbol easily inputtable, which presumably involves installing keyboard shortcuts on whatever platforms people want to use. On top of that technical infrastructure, getting the system to work is a question of marketing, giving the symbol currency, and making the Sarcasm Club one that people want to join. Given the constraints, it's a nice solution that could be used for other things, for example to allow Prince fans to communicate with each other.

d__