If you go to Google Earth and put in the latitude/longitude coordinates from the article (31 15'15.53N 24 15'30.53W), there is indeed a very odd undersea grid of lines in the seabed that is 100 miles x 80 miles in extent at a depth of 18,000 feet. It's almost certainly either a hughly unusual geological formation (you can find all sorts of similar things if you scan the seabed of the world, though this one is admittedly different) or some smartass at the mapping service has been playing games. I'll put my money on the latter before I attribute the grid to Atlanteans, mermaids, mutant seacows or space aliens. The likely explanation to me is that it's an artifact of defective source data they used for the ocean mapping, which probably came from a ship on the surface that was using sonar while following a grid pattern to map the ocean floor, with GPS to guide it. Probably oil exploration. Any number of things could have caused errors in the depth measurements -- uncalibrated equipment, old equipment, defective equipment, sloppy measurements (imagine if they were taken 40 years ago, say), sea conditions, who knows what.
As for Atlantis, the "experts" who persist in believing it could be in the Atlantic ocean haven't the foggiest clue of the history of pre-antiquity or the way people in ancient times viewed their world, for which there was no "Atlantic Ocean", a modern term. The end of the world in Plato's time was about Gibralter, if even that far, and the historical record wasn't good enough to locate Atlantis precisely. Plato was guilty of a lot of hyperbole, and I doubt he could count well enough to tell the difference between 9000 years and 2000 years before his time based on nothing but hand-me-down stories. His methods of radio-carbon dating were somewhat lacking. There is this interesting bit I found online:
"The oldest known mention of "Atlantic" is in The Histories of Herodotus around 450 BC (Hdt. 1.202.4): Atlantis thalassa (Greek: Ἀτλαντὶς θάλασσα; English: Sea of Atlas); see also: Atlas Mountains. Another name historically used[who?] was the ancient term Ethiopic Ocean, derived from Ethiopia, whose name was sometimes used as a synonym for all of Africa and thus for the ocean. Before Europeans discovered other oceans, the term "ocean" itself was synonymous with the waters beyond the Strait of Gibraltar that we now know as the Atlantic. The Greeks believed this ocean to be a gigantic river encircling the world."
So Plato doubtless picked up on Herodotus' mention of "Atlantis".
In any case, I think we would know the details of a geologic event 9000 years ago that sunk 10,000 square miles of land by over 18,000 feet. I agree with those who place Atlantis on the ancient island of Thera, aka, modern Santorini, which had an extensive Minoan civilization on it, an offshoot of the civilization on Crete. Around 2200BC it suffered possibly the world's most cataclysmic volcanic eruption in human times (say, over a span of 100,000 years or more) creating the "ring-shape" that Plato gave to the Atlantean world, except, after they were all gone. This volcanic eruption was likely the source of all the great flood myths that appeared among many religions, including Noah's, as well as accounts from Jason and the Argonauts. Some estimates suggest that the magnitude of the Tsunami hitting the Greek mainland could have exceeded 1000 feet high. On Crete, 60 miles to the south, there are multi-ton blocks of stone from an ancient seawall that were pushed something like a half-mile inland, toward Knossos. I'm going on memory, so don't beat me up too much over dates and distances.
You see, in any case, to the Greeks of Plato's time that "Atlantis" was simply a place a long way off that something terrible happened, and when your principle mode of transportation was a big rowboat with sails, a hundred miles (the distance from mainland Greece to Thera) is a long way off. Throw in 2000 years of big fish stories and the accuracy is bound to suffer, but probably not so much as you'll read in the London Telegraph.