My reaction as well. Additionally, I'm not in agreement that the other point the OP posted in discrediting the article -- that it would fail A/B testing -- is a cause for great concern. If you look at the method of statistical sampling used in the article, and not to mention this research was published in Nature, such an expedient reason to discredit their research seems inappropriate.

My comment is only about one word ("recent") in the title of this "news" article on gizmodo.com, not about the title or content of the actual research. Sorry it wasn't clear.

The Nature commentary has the title: "Greenland once lost nearly all its ice — and could again"

http://www.nature.com/news/greenland-once-lost-nearly-all-it...

The actual paper title is: "Greenland was nearly ice-free for extended periods during the Pleistocene"

http://www.nature.com/nature/journal/v540/n7632/full/nature2...

I'd say the actual paper title wins by a mile.

I agree the scholarly article's title is better.

Your comment did mention another point which appeared to discredit the article/research:

>probably failed A/B testing

That is also a reference to the title of the "news" article on gizmodo.com. Sorry that this wasn't clear.

In particular, I was imagining that gizmodo A/B tested "Most of Greenland Melted about a Million Years Ago" vs "Most of Greenland Melted in the Recent Past".

The latter (to a non-geologist) is quite a bit more sensational.

Oh I gotchya, sorry about that.

And this entire thread could be detached and marked OT.