Open webOS indeed [LWN.net]

Open webOS indeed

Posted Jun 29, 2012 13:47 UTC (Fri) by mattdm (guest, #18)
In reply to: Open webOS indeed by man_ls
Parent article: Open webOS "Community Edition" released

I think the inherent problem is that English is a more information-dense language than Spanish. Not in a bad way, but we fit more in fewer words -- whereas you use more words but speak faster so the information communicated per time is the same. This means that when you import English words, they'll inherently sound funny -- when we say "web oh ess", there's no real risk of it sounding like "webohess".

(Log in to post comments)

Open webOS indeed

Posted Jun 29, 2012 22:09 UTC (Fri) by dgm (subscriber, #49227) [Link]

That's the most absurd comment I have read in much, much time.

Just take any random phrase in both languages and compare. Let's see the first headline in google news right now:

"Rajoy logra que la UE dé oxígeno a España"

in English

"Rajoy manages to get the EU to give oxygen to Spain"

it's two words longer. It's a bit odd though, because it's a literal translation. Let's see if we can get it a bit better:

"Rajoy wins an oxygen line for Spain from the EU"

which is still one world longer.

But, anyway, all this is absolutely UNRELATED to foreign words sounding funny.

If it's impossible that you can say "web oh ess" and it sounding like a word, it's a mere accident. In Spanish, like in English, some acronyms are read letter-by-letter, but others are not. It's a question of convention, there's no rule. This particular case is read as a word in Spanish, and letter-by-letter in English.

Open webOS indeed

Posted Jun 29, 2012 22:24 UTC (Fri) by man_ls (subscriber, #15091) [Link]

I think you are right: anyway the informational content of each language is not a good measure of anything -- specially if you take written language as a rule. Some languages like English contort the written form so much that it seems arbitrary. Others like French contort the spoken form so much that the amount of information per phoneme is completely different than information per letter. In Spanish or (even better) German pronunciation is mostly literal, but spoken dialects often depart dramatically from it in colloquial dialogue.

I once computed the Zipf exponents for a number of written works taken from the Gutenberg project; it seemed to me that Russian was the most compact language, but it depended greatly on the author. This particular computation depends on the choice of words, not on meaning, but it should correlate somehow with the information content.

In this case what really matters is rules of pronunciation and of syllabic division. And a stupid similarity between words of different languages. This kind of paronymia (similar unrelated words) has always fascinated me: how "island" and "isle", with similar meanings, come from completely different roots.