Spanish language and SMS messages

At the end of the last decade everybody in Spain decided to buy a cell phone. As all foreigners know, Spaniards party and dance flamenco on the streets all day long so cells were very useful to call your friends and locate the next place to dance flamenco. However the voice service was very expensive so the people started using the SMS messaging system. As we don’t have a lot of time to write messages between one glass of sangria and the next one, we created our Spanish SMS language.

The Spanish SMS language makes use of compressed words taking off vowels and shortening expressions. In consequence it is faster to write a message in SMS language and you can pack more information in each SMS.

This SMS language is seen as an aberration for Academics, which are seriously concerned about those simplifications finding their way onto regular Spanish. One of the most widespread problems is that nobody writes SMS messages with accents.

What not everybody knows is that if you want to write proper Spanish on your SMS messages, you will need to pay more for each message.

The technical reason

All the cells in Spain are configured to send messages of 160 characters. However, the real payload limit for an SMS message is 140 bytes. We can encode 160 characters on only 140 bytes using a 7 bit character set.

A bit after the dinosaurs ruled the world, the computers spoke to each other on complete 8 bit words. American people were not entirely happy with this solution. You have plenty of space to encode all English characters on 7 bits so using 8 bits per word was seen as a waste of space. One bit does not seem a lot, but it was actually a big deal when computers could just manage a few kilobytes of memory.

So they took off the first bit and the 7 bit characters sets were born. The most widespread is named ASCII. On 7 bit you can refer to 127 characters but instead of including foreign symbols, American people decided to use the free space for drawing symbols and to control printers. This was a decision what made easier to develop software but it also made sharing information encoded in different character sets a big headache.

On the other hand GSM was an European effort, a lot more concerned about interoperability on the entire region. Instead of using standard ASCII they decided to create a new 7 bit character set removing the useless symbols and control characters and including the special symbols of the European languages. They included German, French, Danish and Finish symbols, but they just included the Spanish symbol “ñ“, essential to write wonderful words as coño,maricastaña,ñoño, guiñapo or moño.

The GSM character set is spoken by all the GSM cellphones on earth. This is how it looks:

On the GSM character set there are all the lowercased French accents but only the lowercased Spanish é. Why they included only one Spanish accent is a mystery to me. It’s useless.

You have probably noticed that this character set has 255 spaces but I said that GSM is a 7-bit character set. It actually was at the beginning, but eventually they decided to extend the character set with a beautiful hack on the system. They allow referring the upper 127 symbols using 14 bits (one control symbol plus the upper symbol). That’s why when you write a ‘[',']‘ or € symbol on your cell it counts as 2 characters instead of one. As you see there is still lot of empty space. Why they didn’t seize the opportunity to include more symbols? No idea.

GSM is also used out of Europe. Can Chinese, Japanese, Korean, Russian and Arabic people send messages in their languages? Yes, but they don’t use standard GSM character set. They use UTF-16/UCS2 (Unicode) to encode their symbols. Unicode solve all the problems because it includes on one character set all the symbols of all the present and future real languages  you can think of. Unicode is the character set equivalent to a nuclear power plant.

Unicode requires 2 bytes per character. As we have a maximum payload of 140 bytes, all those countries using Unicode have a limitation of 70 characters per message. This is not a big deal for Korean, Chinese and Japanese people because they can transmit very dense information with their symbols. I guess Russian people have the same problem as we have with Spanish.

What is the problem? Spanish is not a dense language. Using Unicode to write Spanish means we can only use 70 character messages. 70 characters is simply too short. Remember that we already have a very compressed SMS language to deal with the 160 character limitation.

You can set up your phone to send messages in proper Spanish in Unicode. It’s very easy. Why the Spanish cells are not configured by default in Unicode? I guess it would be a nightmare for PR Operator’s department because thousands of people would start complaining about extra cost. Spanish operators charge concatenated messages as independent messages.

There was a nice alternative which was also the ethically correct solution. The operators could have changed their billing processes to allow sending sequences of two or three Unicode-written concatenated messages at the cost of one message. That would have meant that sending 160 character messages of proper Spanish in Unicode would have cost the same as one message. That nice solution would have forced the Operators to assume extra transmission  costs. Let’s say they could lose 0,00001% less. I got that number out of thin air but I don’t think it’s far from reality.

The operators chose between giving the possibility of write proper Spanish in concatenated short messages or ignore the problem and force the people to write bad Spanish.

Everybody knew what happened. In fact using SMS language was used by the operators on their marketing campaigns to attract young people. It was just cool and modern to use it. The operators forced the people to write bad cool Spanish.

So if Spaniards start using SMS-based language while singing flamenco,  remember that part of the problem was that one executive working on a Spanish operator thought that it was a good idea to sacrifice our language in exchange of a bit of extra revenue.

No related posts.