There were some comments about this in a previous posting, but I thought it worth remarking on the unusual situation that papers have started to appear reproducing plots of new data from the PAMELA satellite, even though no such data has been officially released.
This all seems to have started with a July 31 talk at ICHEP 2008, with slides here, but evidently some other slides of preliminary data were flashed on the screen. There was a news story in Nature about this, but still no official release of data. Then, on August 20th there was this talk at idm2008, with no slides made available on-line, but interesting slides again flashed. Evidently some enterprising theorist decided to do some of his or her own data acquisition.
Soon, a preprint on Minimal Dark Matter predictions and the PAMELA positron excess was on the arXiv, complete with PAMELA data, with the notation:
the preliminary data points for positron and antiproton fluxes plotted in our figures have been extracted from a photo of the slides taken during the talk, and can thereby slightly differ from the data that the PAMELA collaboration will officially publish.
There are now at least two other papers on the arXiv featuring PAMELA data, evidently from the same source, here and here. Andrew Jaffe has a new blog posting up entitled Stealing Data? where he expresses discomfort with this situation. I can’t quite see that one is “stealing” data if it is being presented at major conferences.
Are there any of my readers out there who can tell us what’s up with PAMELA?
Update: Nature has a new article about PAMELA being “outed by paparazzi physicists”. One of the paparazzi, Marco Cirelli, is quoted as saying that “we had our digital cameras ready”, and claiming that the PAMELA people at the conference didn’t have a problem with this. On the other hand, a PAMELA PI is quoted as being “very, very upset” about this.
And I should have linked earlier to this posting at Resonaances: Hot Photos of PAMELA.
Update: According to Science News, the problem is not a Nature embargo. They just haven’t finished a paper yet:
“We plan to have final results ready by early October and submit a paper to a peer-reviewed journal,” Boezio told Science News. Until then, he says, the findings remain preliminary, and “We prefer to withhold further comments.”
Well, I have no idea, but a visiting colleague here who works with some Italian PAMELA people said they were ‘very happy’ about new results.
Really, this sort of behavior is just going to make collaborations more reluctant to show preliminary plots at conferences before the paper is ready. (Maybe they should be anyway…)
I’m told by people I work with on SNO that they stressed a lot about this before their first paper. They wanted to show event distributions at a conference, but were worried that someone might try to eyeball-integrate the number of events in the curve. In principle, someone could use that information plus published whitepapers about the detector response to guesstimate the neutrino flux SNO was seeing before the proper analysis by the collaboration would be published. (Never mind that since SNO was doing a blind analysis, the answer would have been wrong anyway.)
They finally decided that hiding the y-axis was sufficient to prevent this. I must admit, years after the fact, this sounded pretty paranoid to me, but I guess I was wrong…
Stan, the issue isn’t that the paper isn’t ready—it’s 100% ready and submitted. The issue that the journal won’t allow the collaboration to release plots officially while it’s waiting for publication.
Stan, probably you remember that when your SNO collaborators presented their first results, they published the slides with the data on the web, together with a web page with extra details of the data (http://www.sno.phy.queensu.ca/sno/results_09_03/howto.ps), and the SNO paper promptly appeared on arXiv, and later on PRL.
If you consider that the PAMELA collaborators did something different while knowing that one cannot be partly pregnant and that results shown at major conferences become public, maybe you can guess the answer to the question by Peter.
Since some people confuse “stealing data” with sharing informations, it is better to wait the papers that will be published by PAMELA and …
I find it embarassing that ambulance-chasing theorists would stoop to this level to beat out rivals for a publication. Everyone know what they did & they do themselves no credit by doing it…even if the experimenter giving the talk doesn’t mind.
Nature is explicitly OK with posting preprints at the arxiv.
Just don’t do press. See http://arxiv.org/abs/0802.1712
I don’t see why a collaboration should protect data from the community. They can just release raw data as such and analyzed ones when ready and people can use whichever is more suitable for their research. Good ideas or valuable experimental data are more important than who delivers them, no?
Depends if anyone else knows what to do with the ‘raw data’ to get a correct analysis of the physics in it. You certainly don’t want people to start putting out wrong analyses before you’ve brought yours to presentable form.
Anyway ‘raw data’ never appear on conference talks except singly as pretty pictures (eg here’s a track in our detector) from which no science follows.
The usual situation is that someone will flash a slide of unpublished results with ‘PRELIMINARY’ over it, which certain favoured theorists/collaborators will already know about, and people in the audience can then write speculative papers at their own risk.
What’s happened here is just this Nature protocol that the journal wants an unusual degree of secrecy of the results over however long it takes to approve them for publication – and that technology has reached the point where people take digital pictures rather than make handwritten notes of what the curve looked like.
Pingback: Not Even Wrong » Blog Archive » Discovery of a New Particle?