[Ffmpeg-devel] Re: Semantic Error?

Måns Rullgård mru
Tue Jan 31 21:05:29 CET 2006

Erik Slagter <erik at slagter.name> writes:

> On Tue, 2006-01-31 at 12:36 +0000, M?ns Rullg?rd wrote:
>> Erik Slagter said:
>> > On Mon, 2006-01-30 at 23:48 +0000, M?ns Rullg?rd wrote:
>> >> > 	No, it's 'ou' vs. 'o': compute_mb_neighbours(). Now you see
>> >> > 	why British English is so suboptimal.
>> >>
>> >> It's still much more efficient than French.
>> >
>> > Interesting, because many words in English are either exactly copied
>> > from French words or with slight alterations. Also at least French has a
>> > very strong relation between written and spoken language and a whole lot
>> > less exceptions.
>> In an average piece of French text, how many words have four or so silent
>> letters at the end?  How many English words look like this?
> This fact I do not debate ;-) I guess the obtainable compression rate of
> a French text would be somewhat higher than English text. But that's
> also due to somehow French needing more _words_ for valid sentences.

That last bit is probably because English has more words, some of them
with very specific meanings.

M?ns Rullg?rd
mru at inprovide.com

More information about the ffmpeg-devel mailing list