7
$\begingroup$

Let's assume you are a group of astronomers that proposed an observation by a telescope.

The observation has been conducted, so you have got the data you wanted. But the data has no proprietary period. So everybody can access it by the internet.

If there are potential discoveries in the data, you would like to maximize the chance that your group will be the first to find them. It's you who proposed the observation. It would be a pity if others were reported the finding ahead of you.

I can guess the data analysis will be more intensive compared to the case when there would have proprietary period when only you can access it. Also you may use new improved pipelines for data analysis which you do not make available for others.

These are my guesses. But in reality - do studies differ much for open astronomical data compared to proprietary data?

$\endgroup$
1
  • $\begingroup$ You might want to provide some examples. I can only imagine "no proprietary period" as an instrument collecting data and feeding it directly to the Internet, with no opportunity for analysis in between. Since such a device would have no opportunity for a proprietary period, studies that want one would use a different device, and the organization would obviously differ because of the different systems and entities involved. $\endgroup$ Commented Aug 19, 2022 at 18:49

2 Answers 2

7
$\begingroup$

I can only tell from my experience, which might be not the ideal type of answer, but I will leave my 2 cents anyway:
For such cases I have seen essentially three different approaches:

  • Be fast: Publish your analysis before another group can scoop you. This is what you want to do if the interesting results are fairly obvious. This will then result in a shorter paper without deep analysis.
  • Be prepared: If you know that this observation will take place early enough and it is somewhat predictable, you can prepare the whole analysis with predicted or "faked" data and have a paper draft already prepared. That way you can just re-run your pipeline with the real data, update your paper draft and publish.
  • Be unique: Do something with the data, which you can be fairly sure nobody else can or will do. This might be the application of some model or algorithm that was developed by and is only available to your team.

So what you sometimes see for data that does public immediately is that instead of one detailed paper, you get a short one (by the same or another group) and detailed one later.

$\endgroup$
5
$\begingroup$

...do studies differ much for open astronomical data compared to proprietary data?

It seems that this one did!

There was a pretty exciting example of (presumably friendly) competition for publishing analysis of nonproprietary data just recently.

The JWST has recently been commissioned and they are being particularly careful to release data in a controlled, evenhanded and fair way. One example of this is discussed in the following question:

Quanta Magazine's Two Weeks In, the Webb Space Telescope Is Reshaping Astronomy highlights two submissions to arXiv soon after the first images were released: "Three days later, just minutes before the daily deadline on arxiv.org..." It certainly sounds exciting!

The Quanta Article begins:

As soon as President Biden unveiled the first image from the James Webb Space Telescope (JWST) on July 11, Massimo Pascale and his team sprang into action.

Coordinating over Slack, Pascale, an astrophysicist at the University of California, Berkeley, and 14 collaborators divvied up tasks. The image showed thousands of galaxies in a pinprick-size portion of the sky, some magnified as their light bent around a central cluster of galaxies. The team set to work scrutinizing the image, hoping to publish the very first JWST science paper. “We worked nonstop,” said Pascale. “It was like an escape room.”

Three days later, just minutes before the daily deadline on arxiv.org, the server where scientists can upload early versions of papers, the team submitted their research. They missed out on being first by 13 seconds, “which was pretty funny,” said Pascale.

$\endgroup$
2
  • 2
    $\begingroup$ Yes, JWST early release data have inspired this question ) Thanks! $\endgroup$
    – Heopps
    Commented Aug 23, 2022 at 5:11
  • 1
    $\begingroup$ @Heopps it did seem like a remarkable coincidence ;-) $\endgroup$
    – uhoh
    Commented Aug 23, 2022 at 6:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .