Apparently my questions (here and here) were mild compared to some that are out there.
The more I've cogitated on arsenic being substituted into DNA backbone, the more I've suspected it's highly unlikely, except for an odd one or two in an odd one or two DNA molecules. My highly-developed, highly-qualitative sense of chemistry says that arsenic bonds are more brittle and aren't going to be able to do the things that phosphorus bonds do in DNA. I've also been wondering if some phosphate hadn't crept into some of the nutrient solutions, even when they were nominally arsenate rather than phosphate. It turns out that others have been wondering the same things.
Thanks to Twitter buddy @jfleck, I've discovered Ivan Oransky's blog and the three most relevant posts, here, here, and here. Carl Zimmer summarizes some of the responses at Slate and has a continuing summary at Discover.
Oransky's blog is called "Embargo Watch," which refers to a practice in science journalism of sending out news releases that are not to be made public ("embargoed") until a certain date and time. This allows reporters to do some background digging before they can all burst out of the gate at once with their stories. More than that, it's tied to the conventions of reporting scientific information and peer review.
There is a protocol to publishing serious scientific papers. First, the paper is submitted to a journal, whose editor sends the paper to reviewers who read and criticize the paper. Then the author(s) (usually plural these days) correct the paper as necessary to get the editor to approve it and publish it in the journal. That publication date is the date to which the news releases are embargoed, because the authors are not supposed to speak publicly about their work until it has undergone the entire procedure, ending in publication.
Like a lot of things, this has become more difficult with the internet and our expectations of instant communication. In this case, word got out early that Science magazine had embargoed a NASA-funded paper, and a couple of the words that got out were "life" and "extraterrestrial." In one of the traditions of the internet, there had to be speculation as to how those words might be related and what the news would turn out to be. Oransky's links give the details.
It occurs to me that that speculation is another downside of this concept of embargo. We already have a deficit in scientific credibility incurred by news stories on medical advances that seem to cancel each other out every few months. Coffee prevents Alzheimers disease. No it doesn't. You should have regular mammograms/PSA tests. Maybe not. Blame for the deficit can be spread around. The scientists (and their funders) want to get publicity, which improves their careers and funding. The reporters may be too credulous or get stuff wrong, or both. An embargo adds that round of speculation to the material that, churned together, confuses the story.