jump to navigation

Free discussion of public results ? April 3, 2007

Posted by dorigo in internet, physics, politics, science.
trackback

In recent years large scientific collaborations like those operating particle physics experiments have started to be faced with a unexpected problem by the arrival of blogs on the scene of distribution of scientific information. 

Usually these collaborations would love to be the sole authoritative commenters of the results they produce. Some have managed to do so in the past by using confidentiality, strong peer review, careful selection of publishers, micromanagement of speakers at international conferences, and in some unfortunate instances political pressure.

There is nothing really wrong with that attitude, but we should not forget that the precious data these scientists collect and scrutinize are the result of hefty funds granted by institutions that are mostly paid with taxpayers’ money. Taken singularly, each and every one of the scientists belonging to any of the large particle experiments I see around could well be substituted by another worthy scientist with no negative effect on the science. They do not own the data. The data is an asset of humanity.

I belong to two large collaborations, the CDF experiment at Fermilab and the CMS experiment at CERN, and I consider my affiliations a privilege, not a right – and the same goes with the data I can play with just because I am a “member”. I would never consider CDF data as “mine”, just because I strung wires on a muon chamber which is now recording tiny electronic signals as the experiment is taking data. No more will I do with the data of CMS – I did even less for the construction and the understanding of that experiment (but I am catching up quickly).

So, now that blogs are around, these collaborations face a challenge to the way of dealing with the discussion of their results. They have been caught, if you will, with their pants down. What is the solution ? Can some form of censoring the free discussion of physics in the web be an option ?

No, of course not. The best thing they could do – and I really hope CMS and the other future experiments will, but I do not see that happening in the lifetime of CDF – is to play it smart: if you can’t defeat your opponent, join him!

Experiments of the future should have much, much more to offer in the internet than those static, unwelcoming, arid “public web pages” where results are laid bare, with no explanation for the newcomer or for the expert alike. They should create and maintain an experiment-approved blog, with frequent entries explaining what is going on, what is coming out, what is cool, and what one should interpret the results.

I guess this is just another of those good ideas that will never see the light…

About these ads

Comments

1. Not Even Wrong » Blog Archive » Various Stuff - April 3, 2007

[...] Dorigo has an excellent suggestion for experimental collaborations worried about the information that their blogging members are [...]

2. a - April 3, 2007

A related problem is: should collider data be made publicly available?

Usually most of the data are kept within the collaboration (up to “don’t tell that I told” private communications) that only publishes some differential distributions and global fits within selected models (supersymmetry, etc).

Other fields where the contact between experimentalists and theorists is closer than in high energy physics have different habits: the collaboration has the first shot, and next (almost) everything becomes public, together with a simplified version of the data useful for approximate analyses. This was done for example in lambda.gsfc.nasa.gov for CMB data, and it was successful.

Hopefully LHC data will attract a lot of attention, and this will become a hot issue.

3. dorigo - April 3, 2007

Hi A.,

I agree with you – it is due time that particle physics changes attitude toward the publication of their data. However we both know it is not easy to make collider data public – not as easy as ccd images or data from experiments which are high-level by nature. An effort has to be made by the particle physics experiments in that direction, and that is an additional hurdle.

Cheers,
T.

4. Torbjörn Larsson - April 4, 2007

It is really easy for an interested layman or non-expert to agree with the post. The point about the static (and outdated) web presence is especially well taken.

But hopefully the suggestion makes sense for scientists too. It should be easier to get support and funding if there is a public interest, and info from experiments will spread wider, perhaps engage a better science journalism (both being fed the latest info and being outdone and/or criticized by good science blogs) and tickle more people than today.

Other groups are trying out blogs as cooperative tools as well, like “n-category café”, and as a platform to discuss and build the technique for web presence (ibid, distler’s blog, Cosmic Variance, et cetera).

I find it tragic when some experts and especially educators doesn’t find this productive, as in the latest string thread on Cosmic Variance. It seems the topic is so inflamed that some of the theoretical physicists doesn’t feel like they can get their message out. Consequently they find it a waste of time.

Of course there could be something wrong with the format they have used. As some one noted elsewhere, most trolls and nuts go away when the discussion becomes more specific than loosely descriptive.

But experiments blogs could also show the way to a format that encourages expert participation in presenting science for the public.

5. Particle Physics 2.0? « Charm &c. - April 4, 2007

[...] 3rd, 2007 Elsewhere, Tommaso Dorigo makes a plea for particle physics experiments to enter the (free) blogging world. I find the argument intriguing [...]

6. Coin - April 4, 2007

Speaking as someone who’s been trying to vaguely follow physics news by reading blogs, what you propose would rock my world. As things stand right now it’s actually really hard as just a member of the public, even an interested one, to get a good grip on what’s going on with all the different ongoing physics experiments, even if you’re just trying to figure out vaguely what the experiments are and mean and aren’t even looking for details or data or anything.

There’s enough people of various kinds following, say, the LHC that you can assume that any news that happens in or around the LHC will hit a lot of blogs and news sources pretty quickly even without the LHC people themselves having to do do anything themselves (notice how quick the news about the magnet failure last week spread), but relatively smaller experiments may as well not even exist from the perspective of the layman. I have to admit I didn’t even know what the CDF was until I looked it up on wikipedia after seeing the name in your blog post (I had heard of the Tevatron, but not CDF in specific).

7. dorigo - April 4, 2007

Hi Torbjorn,

I find it tragic too, but it is in some instances even worse than that, when people display a malicious attitude, trying to prevent the diffusion of scientific results for a different reason than the one claimed.

These individuals claim they do it for the good name of their experiment and to prevent incorrect science to be circulated, but in fact they do it because their power and their salary depends on the number and quality of papers they publish and citations they get.

So, you see, it is politics.

As for the danger of trolls in a experiment blog, I see no problem at all… A moderated comment column is enough.

Cheers,
T.

8. dorigo - April 4, 2007

Hi Coin,

well, welcome to this blog, where you will find lots of information about CDF and D0. I try to update my readers with the results of these experiments, and explain them – exactly what I think the experiments themselves should be doing.

Cheers,
T.

9. morganusvitus - April 4, 2007

The site looks great ! Thanks for all your help ( past, present and future !)

10. WordPress Wednesday: New Security Release, Updated WordPressMU, More WordPress Plugins, and Time to Get Naked at The Blog Herald - April 4, 2007

[...] Free discussion of public results ? [...]

11. Torbjörn Larsson - April 5, 2007

it is politics

I agree. I see that from my own attitudes and experiences in another area, and of course also from your general descriptions that never the less points to that element.

As for the danger of trolls in a experiment blog, I see no problem at all… A moderated comment column is enough.

Yes. OT here, but I get the feeling it would probably not help in the string controversy. The discussion would just be moved onto other blogs, and it is the malicious attitude that affects some of the experts.

12. Andrea Giammanco - April 5, 2007

What about the “CMS Times”? I guess it’s not very well known outside of the collaboration, but even if it is not in blog style it is a periodic (weekly) page about what’s going on in the experiment.

13. dorigo - April 7, 2007

The CMS times is a step in the right direction, but the blog is a much more direct way for communication with non-experts…

Let´s see what happens!

Cheers,
T.

14. WordPress Wednesday: New Security Release, Updated WordPressMU, More WordPress Plugins, and Time to Get Naked » TechAddress - April 9, 2007

[...] Free discussion of public results ? [...]

15. Markk - April 9, 2007

This is a little late, but I think open reporting of “raw” data from large collaborations is a very important topic. I think this should be done in a modern manner, however there will be some large issues to be overcome. I don’t really see these issues being discussed a lot, but I am probably not looking in the right places.
This is from the outside as it were, and if this is all hashed out somewhere already it would be interesting to see.

Publication of raw data not publicly understood could cost collaborators more effort and cause more confusion among the public than not. Given the fact that all high energy data from accelerators is not really raw at the first level. The design of the detectors themselves filters events. The fact that the information channels are narrow relative to the possible data is even more limiting. There are already decisions being made as to the types of events to collect. The output of these detectors depends tremendously upon their exact configuration and manufactured content. (At least based on my experience with other A/D type design) I am sure that there is expertise buried in the brains of experimenters as to the exact characteristics of particular detectors. This expertise is probably used to design the selection and filtering programs that do things like calculate the pictures of particle tracks. I am sure I could bring up more possibilities.

The point is that I think if the actual raw detector information was published, unless there was something so obvious that it would be published quickly anyway, it would only be useful to approximately the same people that use it now. I would think there would need to be agreement on a level higher for release, where the process for creating this level would be open and reviewed, but perhaps it would be the level of tracks in space, as opposed to voltages on particular instruments. Certainly the types of filters in the detectors ought to be public also. Hmm… the data should be useful to the largest cross section of technically knowedgeable people in the scientific area. That means that there ought to a lot of discussion as to what the right level of release there ought to be, with the minimum effort.

I think this will only grow in importance, and the people it would help most are those at smaller colleges globally who could get a lot of data sets to play with. I would not be so worried about delays that much as I think this data would in the long run be the legacy of these accelerators and I could imagine somebody 100 years from now gleaning some historical tidbit from the data.

16. dorigo - April 10, 2007

Hi Markk,

you raise a few good points in your comment, especially the fact that the datasets are useful also for didactic reasons. I also think the time variable is not so important; but experiments should have a commitment to publishing their data in some form some time, rather than have none and let the availability of their legacy live on the hope that somebody raises the question when nobody cares about ownership any longer.
I guess what I’m trying to say is that it should be clear from the outset when, how, and in what form the data will be released, before the experiment starts collecting it – or at least, well before the experiment closes.

Cheers,
T.

17. Tony Smith - April 11, 2007

Tommaso said “… now that blogs are around, these collaborations face a challenge to the way of dealing with the discussion of their results …”.

According to a 10 April 2007 slashdot post by Zonk entitled “Radical Transparency at NASA Via Second Life”:
“… the NASA CoLab project … is building an infrastructure to encourage and facilitate direct participation from the talented and interested public…’ Apparently, the group holds weekly meetings on their island in the popular online virtual world Second Life. …”.

As to bureacratic resistance to such openness, a commenter on that slashdot thread said:
“… … The space act which formed NASA compells it to release it’s code. But I work for NASA/JPL which tries to keep its code from other NASA centers through tactics like:
1. not documenting the existence of certain tools
2. pretending tools are undocumented when they are released
3. forcing people who request code to be very specific… i.e. if someone just asks for a program… they will get junk like binaries for a UNIVAC (I’m not kidding).
Even internal to my NASA center, it’s impossible to get source code… I fought for years to get source code to a part of a library that was broken and no one would pay to have it fixed… when I finally got the code it was only partial code and all of the comments had been stripped out.
… I’m not even allowed to mail code snippets to mailing lists to answer questions without clearance from an intellectual property lawyer first. In their view, my intellect is their property. …”.

As to their using Second Life for open meetings (rather than just having an open blog), another slashdot commenter on that thread said:
“… So, they want to adopt transparency, but they make themselves accessible to only the fraction of computer users that have Second Life installed? …”.

Tony Smith

18. dorigo - April 11, 2007

Hi Tony,

Interesting! Nasa using Second Life for broadcasting their meetings sounds very strange to me. Why not using the much more meaningful and professional vrvs system (www.vrvs.org) instead ?
This sounds like advertising to me.

About the troubles with getting source code, it is very sad what the NASA employee you quote says. It will take time before openness in scientific research becomes the rule rather than the exception, but I think it’s bound to happen.

Cheers,
T.

19. RedNikki - October 1, 2007

Cameron Diaz takes a big swig of soda and shows the world that she can belch!

Sexy Video of Cameron Diaz


Sorry comments are closed for this entry

Follow

Get every new post delivered to your Inbox.

Join 101 other followers

%d bloggers like this: