Good news.
The threatened extinction of 3,000 out of the 6-7,000 languages worldwide is a conservative one. Some experts suggest a 90% extinction rate by the end of the century. (At this rate one language disappears every week). In any case, the 3,000 (50%) rate (one language gone every 2 weeks) is seen as just about unavoidable, given that many of the threatened languages have fewer than 1,000 speakers and are not being learned by children (See http://www.ethnologue.com/ethno_docs/distribution.asp?by=size#2).

Re: "The threatened extinction of 3,000 out of the 6-7,000 languages worldwide is a conservative one. Some experts suggest a 90% extinction rate by the end of the century."

It must be kept in mind that these kinds of calculations are made by statisticians based on an "initial" rate. At some point in time, the inital rate will become a "fixed" rate and the number of languages dying on a daily basis will begin to drop.

It is still possible, however, that in another 1,000 years or so, there may be fewer than a dozen languages spoken worldwide.

These estimates don't take into consideration the future of technology. It's conceivable, at least, that the human language faculty will be augmented in the near future; the rate of language learning could increase dramatically.

The assumption that just a few languages will be spoken in the future seems to me to disregard the association between language and identity: all Welsh speakers speak English, for instance, and yet they choose to maintain Welsh. There are already dialects of "World English," and why should we expect that the diversification of major languages is impossible?

I'm not arguing that language death is not a serious problem, it certainly is. I'm just pointing out that we should also try to factor in possible antidotes.

That said, efforts like this one by the NSF have an obvious urgency; the short term is critical.

I seem to remember reading in Jared Diamond's The Third Chimpanzee that Papua-New Guinea hosts most of the world's langauges...

"a new, multi-year effort to preserve records of key languages before they become extinct"

What a joke!

1) Languages don't become "extinct", only a people, or a kind of animal (what High Priests of Academia Science call a species) can become extinct.

2) If these languages were "KEY", an ever increasing number of persons would choose to learn them and to insist upon using them.

Having fewer languages is better for humans as it enables more persons to interact, share, comprehend and perhaps, through love, understand each other.

A plethora of languages.. Good RIDDENCE!