Archive link, as the original is paywalled.
OP, it would be great if you could update your post to add this link.
The article itself sounds a bit fishy, but the study associated with it is a tiny bit more reasonable IMO.
It's important to highlight that "Proto-Indo-European" refers to at least two languages, that I'm going to call here "Early PIE" and "Late PIE". One is the descendant of the other, but they were spoken in different times, and have a different set of descendants - all Late PIE descendants (like Sanskrit, Latin etc) are also descendants of Early PIE, but the opposite is not true because of the Anatolian languages (Hittite, Luwian etc.)
That's relevant here because odds are that both had different homelands - and the study hints that; acc. to the study Early PIE would be spoken in Anatolia, while Late PIE would be still spoken in the Pontic Steppe.
Another relevant factor is that there are three mainstream hypotheses for the original homeland of [early] PIE, once you include the Caucasian hypothesis. It used to be a fringe hypothesis but recent genetic studies added some weight to it, specially within hybrid/mixed hypotheses.
Once you take that into account, and contrast it with what the article proposes, they aren't too different in spirit - you have one original homeland for PIE in the East, then a group migrates to the West and irradiates from there.
[Personally I place my bets on Early PIE being spoken in the Caucasus. Not just because of the genetic data, but because the reconstruction itself looks a lot like Northwestern Caucasian languages like Ubyx and Adyghe - the "odd" two vowels system, lots of consonants, heavy usage of agglutination. Almost like they were part of the same Sprachbund.]
For example, the Proto-Indo-European language had a word for axle, two words for wheel, a word for harness-pole and a verb that meant “to transport by vehicle.”
The words and roots in question are
- *h₂eḱs- "axle"
- *kʷékʷlos "wheel, circle"
- *Hróth₂-eh₂ "wheel", from *Hreth₂- "to run"
- *yugóm "yoke, joiner"
- *h₂eǵ- "to drive"
Something that I noticed for those is that they tend to be missing Hittite descendants, and the exceptions (*yugóm and *kʷékʷlos) are easy to explain as not necessarily associated with the development of axle-and-wheel. So what the article says regarding dating perhaps applies to Late PIE, but not Early PIE.
Their approach, called computational phylogenetics, treats languages as evolving systems, similar to biological organisms.
Worth noting that computational phylogenetics have been used in Indo-European studies since the 90s, by Don Ringe and others. It is not an alternative method for the comparative method, but rather a way to systematise it.