> > Unless there's a large-scale disaster that takes our civilization to the brink of extinction (and perhaps not even then), I don't think we'll ever lose significant knowledge of how to read languages, from this point on. Sure, paper/discs/other storage media eventually deteriorate, but faster than they do we are making more and more copies of information in updated media formats, globally.
>
> We may still have all the information, but no one will be able to read it except maybe ancient history professors. At the rate language is changing, the English we use today will be exceedingly obfuscated in 50,000 years. It'll be like watching today's 16 year olds trying to read Chaucer or something even older, but magnified. It took a long time for the English language to invent spaces between words. I can't even imagine what new reading processes humanity will have in 50,000 years. That is, if we're even able to read at all. I see it working two ways. The first, letters forming words are obsolete, and there is a single symbol for almost any word you may want to say, even a single symbol for entire sentences. We have no letters, just a myriad of symbols that are impossible to interpret from an outside perspective. (Darmok and Jilad at Tanagra, anyone?) OR, our alphabet will expand. Words will be made of fewer letters, but we'll have more to choose from. Instead of 26 letters, with various accents and whatnot, we'll have 100 or more.
>
> -relishgargler

Ah, but you're not taking the progression of technology into account. If there's even one ancient history professor who can understand it, the knowledge is not lost. There will then be systems which can translate anything from modern-day languages into whatever is in use then (in practically no time at all, given how fast computation will be by then), plus more systems for humans, post-humans, or AI entities to easily access these translation systems from anywhere at any time. It will be virtually the same as if every intelligent entity spoke all languages ever known.

We're just about at that level of translation capability already - within a decade or two, not to mention fifty millennia, I expect automated translation services to be easily superior to human translators.