Precisely Estimated
-
@the_quiet_one said in Precisely Estimated:
That's answered in the link I added to the post. Basically, if you're only involved in the maintenance of the equipment, it is unlikely to qualify you for inclusion. If you were involved in a more active role in the design of the equipment or experiment, then you are. So, no, cooks and janitors would not be included.
OK, no cooks. What about all programmers of all software used on all computers? All builders who dug the hole in the ground? All finance people who made sure the contractors (and researchers) were paid? All politicians who lobbied in various parliements so that the apparatus was built? And how can you say that an intern who tuned some voltmeter had a more direct involvment than the janitor who ensured the tunnels were cleaned every day?
Thousands of people spent years of their lives designing and building one of the largest superstructures ever constructed for science. Are you saying we should arbitrarily omit who and who doesn't get credit for discoveries from that? I don't think you really realize just how big stuff like the LHC and LIGO are, and how many people had to use their scientific and engineering prowess to make it happen.
I am not minimizing their contributions. I'm saying that it is pointless to list everyone at the same level as every other person.
Like I said, the most significant and important contributors are on the top of the list, and it goes down from there. Should Tom Hanks be offended that the right-most boy in the fourth row of the schoolbus be listed in the credits for Forrest Gump, too?
Tom Hanks would certainly be offended if the boy in the fourth row was on the movie poster with his name written in the same size as Tom Hanks'.
You like the movie analogy, but it's actually one that very well reinforces my position. You have a few names at the start of the movie, you have a few names on the movie poster. And then you have thousands of names in the credits at the end. This is exactly the same here.
A movie star is listed at the start. A gaffer boy is at the end. They both contributed to the movie. They both get to boast that they were involved in it.
That's what recommendation letters are for.
Which are only read if you've passed the first filter of "that guy has no published paper".
You're treating this as if The Paper is the only sacred document that anyone can ever use. That there's nothing else in the world that can distinguish someone's role.
Do you know how academic evaluation and employment works? Have you ever had to fill a list of published papers? The number of published papers is a metric used in many evaluations (of people, of research institutes, of projects...), whether you like it or not. By having papers with huge authors lists, you devaluate that metric. Which will result in only one thing, which is to require even more publications (if everyone is author in 5000+-authors publications, everyone will easily be author of many more publications than if they are single authors). Which means everyone will have to find ways to be authors on more publications to fullfill the criteria. And it gets worse and worse. Huge authors lists are pointless, and they may even be harmful.
(also, "competent employer" and academic world... yeah, good luck with that)
No it's not. It's not a precise measurement. If you're 500th from the bottom versus 600th, the difference is negligible anyway.
And what about 50th against 2000nd? Negligible or not? Again, who knows? So, pointless.
In an experiment, you have different roles. One of these roles are writers. Saying that an intern should have equal say in how the document is written is as incorrect as saying an intern should have equal say in determining how many amps should be fed into one of the power couplings for the LHC.
And yet the intern should have the right to express his opinion about how the paper is written, and to remove his name if he feels that doesn't fit his work. With a few authors, this is easy to do (and again, how it is done). With 5000, it's becoming a huge work and you need to have many people whose only role will be to coordinate other people's opinion. Do they get to be co-authors? And what about the secretary who wrote the mass-email list of 5000 names?
It's just silly all the way down.
-
@remi said in Precisely Estimated:
@the_quiet_one said in Precisely Estimated:
That's answered in the link I added to the post. Basically, if you're only involved in the maintenance of the equipment, it is unlikely to qualify you for inclusion. If you were involved in a more active role in the design of the equipment or experiment, then you are. So, no, cooks and janitors would not be included.
OK, no cooks. What about all programmers of all software used on all computers? All builders who dug the hole in the ground? All finance people who made sure the contractors (and researchers) were paid? All politicians who lobbied in various parliements so that the apparatus was built? And how can you say that an intern who tuned some voltmeter had a more direct involvment than the janitor who ensured the tunnels were cleaned every day?
Again, it depends if they had an active role in the research for the experiment. If you dug a hole in the ground, that's not sufficient enough. If you determined the depth and girth of the hole (
aside) then it is sufficient.That's what recommendation letters are for.
Which are only read if you've passed the first filter of "that guy has no published paper".
Umm... yes, which is why you should include all contributors in papers. You've just made my point. Otherwise, the 500th contributor to a large-scale experiment doesn't have any chance in advancing in their career if they aren't even considered.
You're treating this as if The Paper is the only sacred document that anyone can ever use. That there's nothing else in the world that can distinguish someone's role.
Do you know how academic evaluation and employment works? Have you ever had to fill a list of published papers? The number of published papers is a metric used in many evaluations (of people, of research institutes, of projects...), whether you like it or not. By having papers with huge authors lists, you devaluate that metric. Which will result in only one thing, which is to require even more publications (if everyone is author in 5000+-authors publications, everyone will easily be author of many more publications than if they are single authors). Which means everyone will have to find ways to be authors on more publications to fullfill the criteria. And it gets worse and worse. Huge authors lists are pointless, and they may even be harmful.
If I was consulted on an experiment to provide a figure that contributes to an experiment, then I am obligated to be credited for that figure. If I'm omitted, then you've just fraudulently discredited me for my contribution. It doesn't get any simpler than that.
No it's not. It's not a precise measurement. If you're 500th from the bottom versus 600th, the difference is negligible anyway.
And what about 50th against 2000nd? Negligible or not? Again, who knows? So, pointless.
Again, if you're just reading a list of names and not doing anything else to determine one's contribution to an experiment, you're a fucking idiot.
In an experiment, you have different roles. One of these roles are writers. Saying that an intern should have equal say in how the document is written is as incorrect as saying an intern should have equal say in determining how many amps should be fed into one of the power couplings for the LHC.
And yet the intern should have the right to express his opinion about how the paper is written, and to remove his name if he feels that doesn't fit his work. With a few authors, this is easy to do (and again, how it is done). With 5000, it's becoming a huge work and you need to have many people whose only role will be to coordinate other people's opinion. Do they get to be co-authors? And what about the secretary who wrote the mass-email list of 5000 names?
If the secretary contributed figures for the experiment, they should be a co-author. Otherwise not.
-
@the_quiet_one said in Precisely Estimated:
Again, it depends if they had an active role in the research for the experiment. If you dug a hole in the ground, that's not sufficient enough. If you determined the depth and girth of the hole (
aside) then it is sufficient.So the guy who spent 2 minutes saying "dig a 12.42 m hole" and flew away to another building site is co-author, but not the guy who actually spent months laying down all the concrete, making sure it is perfectly aligned to specs, 'cause you know, he wasn't "active". It's getting better and better.
I mean, listing the people who had a direct contribution to a project is hard, no doubt about it. Even more so for large projects. This is no reason for throwing your hands in the air and saying "add everyone!".
If I was consulted on an experiment to provide a figure that contributes to an experiment, then I am obligated to be credited for that figure. If I'm omitted, then you've just fraudulently discredited me for my contribution. It doesn't get any simpler than that.
Credited doesn't mean "being an author". Many papers have acknowledgement sections. Movies have endless end-credits where everyone is credited. Only a few names make it to the opening titles.
Again, if you're just reading a list of names and not doing anything else to determine one's contribution to an experiment, you're a fucking idiot.
Please show me how citation indexes and the like account for contributions (actually, they don't even take into account the order of the list, in most cases).
-
@remi said in Precisely Estimated:
And why does the lab assistant get to be co-author and not the IT technician who set up the computer used to write the article, or the programmer who wrote the software to analyse the results? After all, he may have more training and put in more effort and knowledge than the lab assistant. Because you think setting up a computer is not related to the paper? Compared to what? He performed some work without which the paper could not have been written, and that the researcher would not have been able to do himself. Are you able to write that software yourself? Do you know of other data analysis software with comparable features involving 10 people only? See how your line of thinking can be reversed and applied to any case?
You're mixing up "writing the paper" and "performing the experiment". The lab technician took part in the experiment, the IT guy that set up the computer did not. Generally, performing the experiment starts from a common ground like "you need a lab, beakers, materials", etc. Some of those steps may produce papers in their own right, but for the experiment being described anyone involved in those steps wouldn't be credited.
In the case of the LHC, even if you start your paper with "turn on your previously purchased and assembled 27 km long particle accelerator", the process of turning it on and running the experiment may require those 5000 people. They were an active part of the experiment. The experiment requires 5000 active participants, and so they all become co-authors. The janitor may not be, unless his contribution is noted in the paper, in which case whatever the janitor did has to follow a protocol and be reproduced by the next team trying to validate the results by redoing the experiment.
-
@remi said in Precisely Estimated:
@the_quiet_one said in Precisely Estimated:
Again, it depends if they had an active role in the research for the experiment. If you dug a hole in the ground, that's not sufficient enough. If you determined the depth and girth of the hole (
aside) then it is sufficient.So the guy who spent 2 minutes saying "dig a 12.42 m hole" and flew away to another building site is co-author, but not the guy who actually spent months laying down all the concrete, making sure it is perfectly aligned to specs, 'cause you know, he wasn't "active". It's getting better and better.
2 minutes? Seriously? That's what you think it takes to figure out dimensions to build an apparatus as finely tuned as the LHC?
What separates the concrete layers from the person writing measurements is one is merely implementing a design while the other came up with the design.
I mean, listing the people who had a direct contribution to a project is hard, no doubt about it. Even more so for large projects. This is no reason for throwing your hands in the air and saying "add everyone!".
Okay, I'm an expert in a certain scientific field. Let's say I'm at my desk, and I get a phone call from someone in some state university who says, "Hey, Dr. Expert, I need you to figure out the angle that this particle gun should be aligned with, precise to a thousandth of a second." I spend 40 hours doing the calculations and triple checking my answer, and get back to them. Because this is an experiment involving 5 people, they add me to the co-author list.
Next year, CERN approaches me with a similar question for their LHC. I spend 40 hours doing the calculations and triple checking my answer, and get back to them. Because this is an experiment involving thousands of people, I get zero credit for my work.
WTF
Again, if you're just reading a list of names and not doing anything else to determine one's contribution to an experiment, you're a fucking idiot.
Please show me how citation indexes and the like account for contributions (actually, they don't even take into account the order of the list, in most cases).
I don't know. Maybe getting off your ass and talking to someone who was integral to the project?
-
@remi said in Precisely Estimated:
But clearly, 5000 is too many.
I'm sorry, are you now Dictator Of Science Papers?
Where do you get off telling someone else how many authors is too many?
-
@the_quiet_one said in Precisely Estimated:
Okay, I'm an expert in a certain scientific field. Let's say I'm at my desk, and I get a phone call from someone in some state university who says, "Hey, Dr. Expert, I need you to figure out the angle that this particle gun should be aligned with, precise to a thousandth of a second." I spend 40 hours doing the calculations and triple checking my answer, and get back to them. Because this is an experiment involving 5 people, they add me to the co-author list.
Next year, CERN approaches me with a similar question for their LHC. I spend 40 hours doing the calculations and triple checking my answer, and get back to them. Because this is an experiment involving thousands of people, I get zero credit for my work.
Or you could, I don't know, write a paper with 5 people explaining how you did so, like in the first case. You seem to assume that something like the LHC, which provides enough innovative work for 5000 people, only allows writing one single paper. One summary paper in Nature or Science, yes, maybe, but probably tens, or hundreds of smaller papers in other publications (or maybe the same, doesn't matter) that lead to the big one. And then, you know what, we even have a way to acknowledge all these other papers in the main one, that's the big fat References section at the end. Where you can go wild and cite 5000 papers if you want, and these papers will get the bonus in citation indexes of being cited by a very recognised paper, so they get proper recognition with the current system.
I cannot believe that 5000 researchers can work for years on something where they all have a significant research input, and yet there is absolutely no way to write smaller stuff. That's bullshit. You did some research on how to calibrate part I-alpha-d of the particle beam? Write a paper on that. You invented a way to measure the energy levels with the required accuracy? Write a paper on that. I'm no particle physicist but really, I would be hugely surprised if that was not already the case -- if nothing else, because some of these 5000 scientists belong to organisations that will have required them to write a certain number of papers.
(you want a precise example? Some universities, if not most, require you to have published a certain number of papers before defending your PhD. So PhD students who worked on that project have necessarily published other papers than the main one who was longer in the making than their PhD)
Please show me how citation indexes and the like account for contributions (actually, they don't even take into account the order of the list, in most cases).
I don't know. Maybe getting off your ass and talking to someone who was integral to the project?
Can you explain to me how a citation index, which is an automated system built on publications lists, can get off his ass and talk to someone?
-
@blakeyrat oh yeah, I forgot that we're not allowed to express an opinion on a topic without being The Authoritative Autority on the topic.
-
@remi Correct.
-
@blakeyrat Which, I guess, means that you are the Official Dictator of Git?
That would make a nice sig...
-
@remi said in Precisely Estimated:
OK, no cooks. What about all programmers of all software used on all computers? All builders who dug the hole in the ground? All finance people who made sure the contractors (and researchers) were paid? All politicians who lobbied in various parliements so that the apparatus was built? And how can you say that an intern who tuned some voltmeter had a more direct involvment than the janitor who ensured the tunnels were cleaned every day?
It depends on the paper. I work for a company that builds equipment for particle accelerators. If a paper is about an experimental result that used a piece of equipment we built, then none of us are listed as authors, because we didn't generate the data. If the paper is about the equipment itself and how it works, especially if it's a new invention, then the physicists and engineers at my company are listed as authors. Plus, this latter paper will be cited in the former. As to the order of names, it depends on the institution. Some groups list project leaders first, then institutional leaders, then engineers and operators. So, there is prestige to be had in being listed first. CERN opts for alphabetical order, which is why there are an inordinate number of papers that cite Aad, et al.
As another example, Geant is a simulation library extensively used to simulate particle interactions with detectors. If a paper uses results from a Geant simulation, then the name of the software will be listed in the references section. If the paper is about new changes to the software, then the programmers and physicists working on it will be authors. Programmers who use Geant to predict results and analyze experiments will be listed as authors on papers about that experiment.
As for digging holes, I wouldn't be surprised if there was a paper on the seismic activity in the surrounding land, which may or may not have input from the construction firm hired to lay the foundation of the lab. Even the phase of the moon has an effect on beam operation.
Physics simply has a tradition of being generous with who gets listed as a paper author. Mostly because there is very little that is routine about operating the machines to make the measurement. Pretty much everyone listed on a paper will have a Ph.D. or equivalent experience. Everyone listed as an author should feel obligated to read the paper to check that it accurately represents the work they did. Everyone listed as an author has veto power over the contents and the option to remove their name if they feel there is some inaccuracy.
@remi said in Precisely Estimated:
I cannot believe that 5000 researchers can work for years on something where they all have a significant research input, and yet there is absolutely no way to write smaller stuff. That's bullshit. You did some research on how to calibrate part I-alpha-d of the particle beam? Write a paper on that. You invented a way to measure the energy levels with the required accuracy? Write a paper on that. I'm no particle physicist but really, I would be hugely surprised if that was not already the case
That is precisely the case. Thousands of papers have come out of the LHC detailing every bit of the machine, from the detectors to the cooling systems for the superconducting components. But, even in the Higgs boson discovery paper, the experiment took years to run, which required thousands of people to operate the 18-mile-long machine. All of those people are originators of the data, calculations, and conclusions. Even beam operators, who "merely" steer the beam around the ring, have to report on the beam intensity, the focusing, the collision rate, and many other tuned aspects that are essential for the calculation of results. So, they helped write the paper and are credited with authorship.
-
@the_quiet_one said in Precisely Estimated:
Thousands of people spent years of their lives designing and building one of the largest superstructures ever constructed for science. Are you saying we should arbitrarily omit who and who doesn't get credit for discoveries from that? I don't think you really realize just how big stuff like the LHC and LIGO are, and how many people had to use their scientific and engineering prowess to make it happen.
Yes, but it's not arbitrary. Designing a piece of lab equipment (which is what these are, even though they are huge) does not automatically justify inclusion on papers about discoveries made while using that equipment.
Designing the equipment does qualify one for one thing, though: being recognized as the designer of said equipment.
The ones who built it are even less justified for inclusion, unless they, in addition, have significance as the equipment's designers or better, actually run experiments on it.
If you include the builders and designers of the equipment, then the employees and designers at TI should be listed on every paper that has an average that was calculated using a TI calculator. Listing the builders and designers of the LHC or the LIGO is exactly the same sort of silliness.
-
@kian said in Precisely Estimated:
@remi said in Precisely Estimated:
And why does the lab assistant get to be co-author and not the IT technician who set up the computer used to write the article, or the programmer who wrote the software to analyse the results? After all, he may have more training and put in more effort and knowledge than the lab assistant. Because you think setting up a computer is not related to the paper? Compared to what? He performed some work without which the paper could not have been written, and that the researcher would not have been able to do himself. Are you able to write that software yourself? Do you know of other data analysis software with comparable features involving 10 people only? See how your line of thinking can be reversed and applied to any case?
You're mixing up "writing the paper" and "performing the experiment". The lab technician took part in the experiment, the IT guy that set up the computer did not. Generally, performing the experiment starts from a common ground like "you need a lab, beakers, materials", etc. Some of those steps may produce papers in their own right, but for the experiment being described anyone involved in those steps wouldn't be credited.
I think that's exactly the point that @remi is making.
In the case of the LHC, even if you start your paper with "turn on your previously purchased and assembled 27 km long particle accelerator", the process of turning it on and running the experiment may require those 5000 people. They were an active part of the experiment. The experiment requires 5000 active participants, and so they all become co-authors. The janitor may not be, unless his contribution is noted in the paper, in which case whatever the janitor did has to follow a protocol and be reproduced by the next team trying to validate the results by redoing the experiment.
You are directly contradicting your first paragraph. If we're including everything necessary to run an experiment (instead of the actual experiment itself), then why not include the local power company's employees for making sure the electricity was being produced consistently so that the electrically-powered lab equipment could function properly?
In other words, those involved with making the lab equipment function do not implicitly deserve inclusion on a paper. Those who contributed the configuration and settings for the particular experiment (pardon the pun) do.
-
@the_quiet_one said in Precisely Estimated:
If the secretary contributed figures for the experiment, they should be a co-author. Otherwise not.
What if the secretary has a good figure?
-
@remi said in Precisely Estimated:
Because this makes it impossible to judge the individual merit of the authors.
Which should not, primarily, be the purpose of a scientific paper. You're looking at this like a University Adminstrator - "Scientific Papers are a mechanism by which we determine who's earned a promotion." In fact, Scientific Papers are the mechanism by which the people who performed an experiment describe the experiment and the results the measured. Their role in grading scientists is just a (sometimes unfortunate) byproduct.
If you want to judge the individual merits of the authors, look to see who wrote the papers interpreting and extending the results of the experiment.
-
@mzh said in Precisely Estimated:
CERN opts for alphabetical order, which is why there are an inordinate number of papers that cite Aad, et al.
That's also what I vaguely remembered, yes, and it breaks down the line of argument of @The_Quiet_One about how the order of authors tells you who did what (which doesn't hold water in my opinion anyway, but this clearly makes it entirely invalid).
@mzh said in Precisely Estimated:
That is precisely the case. Thousands of papers have come out of the LHC detailing every bit of the machine
Which is normal and expected and how science works, if anything because otherwise the Only Paper would have to be thousands of pages and totally unreadable.
And which, again, reinforces my opinion that it is silly to list all the authors of all those other papers as authors of the main one. They had a contribution, sure, their papers should be cited, of course, but putting them as authors of the main one is pointless.
-
@gwowen said in Precisely Estimated:
@remi said in Precisely Estimated:
Because this makes it impossible to judge the individual merit of the authors.
Which should not, primarily, be the purpose of a scientific paper. You're looking at this like a University Adminstrator - "Scientific Papers are a mechanism by which we determine who's earned a promotion." In fact, Scientific Papers are the mechanism by which the people who performed an experiment describe the experiment and the results the measured. Their role in grading scientists is just a (sometimes unfortunate) byproduct.
Did you read my previous posts? I made exactly this argument at one point, let me say it again: in my opinion, a paper has (at least) two roles.
One is the advancement of Science and in that aspect authors don't matter. Not a tiny bit. An equation is true (or false) whoever wrote it, an experimental result should be reproduceable simply by reading what's in the paper (I know, this is not always true in practice, but we're talking Science here, in the abstract sense). So in that regard, we might as well make anonymous papers.
The other is the recognition of the contribution of the individuals. There are two sub-roles here: one is simply the moral right (basically, your place in history) and for this one you can put 5000 authors if you want, but then you are putting on the same level both the lead researcher and the guy who did a tiny part of it (the movie analogy was brought up many times and it works perfectly here: you don't put the catering team on the same footing as the director in the film credits!). Yes, that means cutting the list at some point and deciding that some contributions were "too small" to make them co-authors of that specific paper and instead simply cite their own smaller papers. But I cannot understand how you can get to 5000 without thinking "hmmm, that's a tad bit too many".
The other sub-role is the nitty-gritty of career management in the academic world, i.e. judging the individual merit of authors. And having a paper with 5000 names is pointless for that (for the same reason as above), and it may even be indirectly harmful to the rest of the community because it sets some expectations for the number of papers one is co-author of, which groups with smaller participation cannot hold (i.e. if as part of the LHC you get to be co-author of all main papers with 5000 authors that come out of it, you might end up co-author of many high-visibility papers, even if you actually did less work than someone who's working on smaller scale experiment and thus will only have published a couple of papers).
If you want to judge the individual merits of the authors, look to see who wrote the papers interpreting and extending the results of the experiment.
Yes, but in essence here you are saying "when judging someone, you need to exclude some papers (such as this 5000 authors one)". Which is not how academic evaluation works, what matters is that you're co-author of a paper in a top category journal, or that has been cited a number of times etc.
The system of using publications to evaluate individual researcher is definitely not perfect, but you cannot deny that it is (one part of) the system that is currently used. And a 5000 authors paper breaks down this system, for no good reason since there are already other ways to acknowledge the contribution of these 5000 people (they have their own publications and they should be cited in the main one).
-
@remi said in Precisely Estimated:
There are two sub-roles here: one is simply the moral right (basically, your place in history) and for this one you can put 5000 authors if you want, but then you are putting on the same level both the lead researcher and the guy who did a tiny part of it
For a vast project where would you put the cut-off? Would three authors be 'acceptable'? 6? 10? 30? 100?
Everyone in the field will understand how CERN citations work - no-one's being mislead.
@remi said in Precisely Estimated:
it may even be indirectly harmful to the rest of the community because it sets some expectations for the number of papers one is co-author of, which groups with smaller participation cannot hold (i.e. if as part of the LHC you get to be co-author of all main papers with 5000 authors that come out of it, you might end up co-author of many high-visibility papers, even if you actually did less work than someone who's working on smaller scale experiment and thus will only have published a couple of papers).
I don't see the harm. The Higgs paper could be seen as a marker signifying 'contributed to a prestigious project at a prestigious institution' - no more and no less. That's the way it will be understood.
It's not as though the leading theoreticians on the project will be desperate to keep their publication count high enough - whereas for a postdoc near the start of their career, a citation as one of the 5000 is a nice thing to have, a bit of fun and a kind gesture by the project.
The vast citation count acknowledges the reality of major scientific endeavours more honestly than the alternative pretence that the senior prof did most of the work.
I'm peripherally involved in a project that should legitimately have ~ 150 authors, but won't because the people leading it are zealously guarding their position - resulting in substantial ill-will and creating quite a caustic environment at times.
-
@remi said in Precisely Estimated:
The other sub-role is the nitty-gritty of career management in the academic world, i.e. judging the individual merit of authors.
My guess would be that the goal of large author lists is precisely to prevent this use. Political fights over who is important enough to be listed do not contribute to scientific advancement. This is especially important in international collaborations where conflicts between the home nations of authors can make establishing a working relationship difficult.
-
@remi said in Precisely Estimated:
And a 5000 authors paper breaks down this system, for no good reason since there are already other ways to acknowledge the contribution of these 5000 people (they have their own publications and they should be cited in the main one).
The problem is that the system already broke down way before papers with 1000s of authors.
There are plenty of three to five author papers, where one of the authors is on the paper because she's/he's the boss who had the money. In a smaller group, that person probably knows what the paper is about (and probably actually supervised the research), but it's easy to find people that appear on hundreds of publications a year. I doubt they know what half of the papers they "coauthored" actually are about (I know Ph.D. students who said they had to book a meeting two weeks in advance to meet their main supervisor, and then got maybe 20 minutes of time with that person, maybe once per half-year).
I know a few really good researchers who've added each other to their papers for much more mundane reasons, like needing a certain number of publications a year to pass performance evaluations. (The persons I'm thinking of specifically tend to actually try to contribute to the paper in question in whatever ways they can to justify coauthorship, but it's not hard to imagine instances where that might not the case.)
Apart from that. At what point is it justified to be a coauthor? Participated in in-depth discussions that maybe helped solve a problem? Wrote a piece of code that did a specific thing? Spent a few hours measuring and verifying results? Had a clever idea that got used? Fetched the main author's coffee for half a year, thus clearly enabling the research? Implemented exactly half of the experimental prototype?
tl;dr: The system was broken long before papers with 5000 authors. If, however, papers with 5000 authors are needed to make the brokenness obvious, then we need more papers with 5000 authors.
Finally (and FWIW), there are journals and venues that need each coauthor to briefly list their individual contributions to the paper (and apparently, on occasion, somebody even gets told that, no, they can't be a coauthor on a paper). I think that's a good initiative that more places should adopt, even if that means expanding the 24 page author list by another 2-3x in volume.
-
@japonicus said in Precisely Estimated:
For a vast project where would you put the cut-off? Would three authors be 'acceptable'? 6? 10? 30? 100?
There's only one sane criterion I can think of.
Did "Person X" do contribute significantly to a task requiring scientific knowledge which, had they screwed it up, cause the experiment not to run or cause the results be invalid?
So the man who raised the money and assembled the scientific team - absolutely an author on the paper.
Woman who organised the building of the staff accomodation - not an author.
Woman who finely adjusted and calibrated the position of the magnetron - author.
Man who delivered the concrete to Switzerland - not an author.
Made coffee, did the typing - not an author.
Solved a particular scientific problem in the setup - author.
-
@gwowen yes and far too often in the past most of those deserving to be authors were excluded, because they were 'only technicians'; were women; didn't have the right academic credentials or because there were already 'too many' authors.
-
@japonicus said in Precisely Estimated:
For a vast project where would you put the cut-off? Would three authors be 'acceptable'? 6? 10? 30? 100?
I have no idea, but let me make an car analogy (that was missing from the thread, don't you think?). Someone going 5000 mph is going too fast (OK, that's impossible... I'm just keeping the same numbers for the sake of the argument, feel free to replace them by something more realistic). Someone going 2 mph is not going too fast (he may be going too slow, but that's another topic). Where do you put the cut-off? Would 3 mph be 'acceptable'? 6? 10? 30? 100? And then your answer is "5000 mph is fine".
You're saying "there is no way to put a cut-off, so let's accept anything". I am saying "it might be impossible to define a single all-encompassing cut-off, but when you get to 5000, you know that you've crossed that cut-off".
Obviously, since these papers exist and did not create a huge uproar because of their number of authors, people in that field are OK with it. Still, that doesn't make it less pointless to me.
It's not as though the leading theoreticians on the project will be desperate to keep their publication count high enough - whereas for a postdoc near the start of their career, a citation as one of the 5000 is a nice thing to have, a bit of fun and a kind gesture by the project.
I see the harm for the other postdocs who were not involved in that project. Take two postdocs, doing more or less the same research work. Nothing ground-breaking, just good solid research. One works at the LHC, the other at a smaller experiment. Both will publish the result of their own work in some mid-field specialized journal, a good one but nothing more. Additionally, the first one will get to be one of 5000 co-authors in a paper in a hugely prestigious journal, not the second one. A couple of years later, the first one will have a larger publication record, a better citation index, only on the virtue of having done his postdoc in the LHC and not another lab.
Really, you don't see how that could be a problem?
I'm peripherally involved in a project that should legitimately have ~ 150 authors, but won't because the people leading it are zealously guarding their position - resulting in substantial ill-will and creating quite a caustic environment at times.
It looks to me that the LHC and such decided that dealing with humans was too hard and that the only way to avoid the problem you're mentioning is to not put any barrier to who can be an author. I find it sad.
@mzh said in Precisely Estimated:
@remi said in Precisely Estimated:
The other sub-role is the nitty-gritty of career management in the academic world, i.e. judging the individual merit of authors.
My guess would be that the goal of large author lists is precisely to prevent this use.
Maybe. But then they are breaking the main tool that is used to rank scientists between them, without providing any replacement, nor even saying that they do so voluntarily, and in such a way that may be damaging for the other people in other projects, not to themselves. In other words, if this is what they are trying to do, then it's similar to cheating in a game in an obvious way to point out how rules are silly. That is not right, and I hope that this is not what they are doing.
-
@cvi said in Precisely Estimated:
tl;dr: The system was broken long before papers with 5000 authors.
I agree.
If, however, papers with 5000 authors are needed to make the brokenness obvious, then we need more papers with 5000 authors.
So basically, let's screw up the guys who can't be in those 5000 authors until the rest of the world gets its act together. That's a nice set of moral principles you have here.
-
@gwowen said in Precisely Estimated:
There's only one sane criterion I can think of.
Unfortunately it is not as clear-cut as you make it sound.
So the man who raised the money and assembled the scientific team - absolutely an author on the paper.
Is your university president co-author of all papers published by his university? After all, he raised the money and assembled the scientific team. Or maybe the research group director (and here, depending on cases, sometimes he is, sometimes he isn't...)?
Woman who organised the building of the staff accomodation - not an author.
Woman who finely adjusted and calibrated the position of the magnetron - author.And what about the woman who fine-tuned the electricity supply to the whole building so that there are no fluctuations that would perturb the experiment? If the knob she turned is on the apparatus, she's in, but if it's in the power supplier's generator room, she's not?
Man who delivered the concrete to Switzerland - not an author.
Solved a particular scientific problem in the setup - author.Such as, how to arrange deliveries of hundreds of heavy lorries without their vibration perturbing the setup of the other apparatus? (I'm pretty sure there must have be ground studies to decide where to put the various bits)
What I'm saying is that there is no clear-cut way to decide if someone is an author or not. @cvi listed a few examples, there are tons in the academic world, it is well known that the list of authors of a paper may be the more controversial aspect of any paper. And there is one good reason for that, which I've already mentioned when talking about the roles of authors in papers, which is that being an author of a paper has nothing to do with Science (where you can have clear criteria) and all to do with Human Beings and how they see each other.
So my position is that we should forget about trying to find an abstract and all-encompassing criteria. This does not exist, and cannot exist because authors lists are about humans, not equations. Let's accept that, and work from there. Look at what authors list are used for, and whether it is meaningful to put 5000 people for that purpose. I believe it is not.
-
@remi said in Precisely Estimated:
So basically, let's screw up the guys who can't be in those 5000 authors until the rest of the world gets its act together. That's a nice set of moral principles you have here.
I'm not sure what you mean. Who exactly are the people getting screwed over?
The intern/student/external contractor that didn't make it into the list? They wouldn't make it into a list of three either. Some core team that did disproportionately more than the rest? Same problem with three people - the first author may have done much than the slacker in second place. The reverse may be true too - while the first author gets all the credit, it may have been an actual team effort where everybody contributed equally.
Or perhaps this:
I see the harm for the other postdocs who were not involved in that project. Take two postdocs, doing more or less the same research work. Nothing ground-breaking, just good solid research. One works at the LHC, the other at a smaller experiment. Both will publish the result of their own work in some mid-field specialized journal, a good one but nothing more. Additionally, the first one will get to be one of 5000 co-authors in a paper in a hugely prestigious journal, not the second one. A couple of years later, the first one will have a larger publication record, a better citation index, only on the virtue of having done his postdoc in the LHC and not another lab.
Yeah, that's perhaps unfair. It's not a problem specific to papers with 5k authors though. Take two postdocs, one going to a really prestigious group (a.k.a the "we publish in nature, and forget everything else"-nonsense), and the other to an average one. Guess who has a higher chance of getting to be a co-author in a highly prestigious journal, with a better citation index, a higher H-index, on the virtue of having done his postdoc at $A instead of $B? No 1k+ author papers involved.
Thanks for questioning my moral principles. But I don't see why the problem is specific to 5k+ author papers. Yeah, things aren't fair. I don't see why 5k+ author papers make it more unfair, though. (Yes, it sucks to be the guy who didn't make it on a 5k author list, but the suck is just as bad or worse when you don't make it on a 3 or 5 or 15 or whatever people author list.)
-
@remi said in Precisely Estimated:
And what about the woman who fine-tuned the electricity supply to the whole building so that there are no fluctuations that would perturb the experiment? If the knob she turned is on the apparatus, she's in, but if it's in the power supplier's generator room, she's not?
Huh?
-
@gwowen said in Precisely Estimated:
@remi said in Precisely Estimated:
And what about the woman who fine-tuned the electricity supply to the whole building so that there are no fluctuations that would perturb the experiment? If the knob she turned is on the apparatus, she's in, but if it's in the power supplier's generator room, she's not?
Huh?
I'm trying to point out that your criteria doesn't really help in deciding who get to be co-author.
"Finely adjusted and calibrated the position of the magnetron" gets to be author (according to you), but what if they adjusted the computer that controls the magnetron? The electrical circuits that controls the inside of the magnetron? The transformer that brings power to the magnetron? The power room that powers this transformer? The power network that delivered power to the power room?
Trying to give more than general guidelines is hopeless, because a list of authors is not a scientific fact, it is a human thing.
-
@cvi said in Precisely Estimated:
Yeah, that's perhaps unfair. It's not a problem specific to papers with 5k authors though.
The fact that it happens in other cases doesn't mean it's OK to do it in this case. Yes, authors lists are arbitrary choices that are not made in a consistent way and that are sometime unfair. In my book, that doesn't mean that it's OK to contribute to this unfairness (with 5000 authors) when it could be easily avoided.
-
@cvi said in Precisely Estimated:
I know a few really good researchers who've added each other to their papers for much more mundane reasons, like needing a certain number of publications a year to pass performance evaluations.
Or, as a joke:
-
@remi said in Precisely Estimated:
In my book, that doesn't mean that it's OK to contribute to this unfairness (with 5000 authors) when it could be easily avoided.
And my point is that the exactly same argument can be made for a paper with eight (or whatever) authors. They are too contributing to that unfairness, yet you seem to be OK with that.
-
@cvi said in Precisely Estimated:
@remi said in Precisely Estimated:
In my book, that doesn't mean that it's OK to contribute to this unfairness (with 5000 authors) when it could be easily avoided.
And my point is that the exactly same argument can be made for a paper with eight (or whatever) authors. They are too contributing to that unfairness, yet you seem to be OK with that.
I've already said several times that there is no simple rule to how many is too many. I made no claim that any specific number below 5000 was fair and OK. I only said that to me 5000 is blatantly not OK.
I am quite confident in saying that someone driving 200 mph on standard roads is driving dangerously. Some people driving 20 mph on the same road may also be driving dangerously, yet that doesn't justify in any way the guy driving 200 mph. What I'm saying is that 200 mph is dangerous. Not that any other lower speed is safe. Some are, some aren't. 200 mph is not. You are saying that since some people driving 20 mph are dangerous, it's OK to also have people driving 200 mph, 'cause you know, other people are already doing dangerous things, so why not them as well?
-
@remi said in Precisely Estimated:
@cvi said in Precisely Estimated:
@remi said in Precisely Estimated:
In my book, that doesn't mean that it's OK to contribute to this unfairness (with 5000 authors) when it could be easily avoided.
And my point is that the exactly same argument can be made for a paper with eight (or whatever) authors. They are too contributing to that unfairness, yet you seem to be OK with that.
I've already said several times that there is no simple rule to how many is too many. I made no claim that any specific number below 5000 was fair and OK. I only said that to me 5000 is blatantly not OK.
I am quite confident in saying that someone driving 200 mph on standard roads is driving dangerously. Some people driving 20 mph on the same road may also be driving dangerously, yet that doesn't justify in any way the guy driving 200 mph. What I'm saying is that 200 mph is dangerous. Not that any other lower speed is safe. Some are, some aren't. 200 mph is not. You are saying that since some people driving 20 mph are dangerous, it's OK to also have people driving 200 mph, 'cause you know, other people are already doing dangerous things, so why not them as well?
To completely abuse the vehicle analogy, unique institutions like the LHC and LIGO are more like space stations than cars, where 10,000 mph is too slow.
Every experimental group comes up with their own criteria for who gets listed as an author on papers. Like you said, this is a human issue, not a technical one. Physicists understand what a name on a paper means. The proper place for detailing the contributions of each individual researcher is on their own CVs and resumes.
-
@remi said in Precisely Estimated:
I've already said several times that there is no simple rule to how many is too many. I made no claim that any specific number below 5000 was fair and OK. I only said that to me 5000 is blatantly not OK.
Ok, I get that. I still don't think that 5000 is automatically not OK; I think that the problem lies elsewhere. Any problem that a 5k author list may exhibit is also present with fewer authors, leading me to the hypothesis that the number of authors doesn't matter.
Since we're doing dubious analogies, let me extend a bit on yours. I'm not concerned with the speed people are driving. I'm concerned that people might be driving on the wrong side of the road, at which point any speed is dangerous.
-
BOOOOOOOOOOOOOOOOOORING
-
@mzh said in Precisely Estimated:
To completely abuse the vehicle analogy, unique institutions like the LHC and LIGO are more like space stations than cars, where 10,000 mph is too slow.
I get that they are projects out of the usual, but the problem is that the situations where publication lists and indexes are used do not make that difference. When you look at the citation index of a researcher, you don't get two different indexes according to whether he worked in the LHC or in another smaller project.
So it's kind of like saying that both the ISS astronaut and the regular car driver get a paper stating their average speed during the year, with no distinction of the vehicle they were "driving".
I find it particularly irksome since I don't see what, except an habit that grew progressively in that field, justifies such "abuses" (IMO) of the list of authors.
@cvi said in Precisely Estimated:
the hypothesis that the number of authors doesn't matter.
And yet many institutions do use the publication records for various things. I'm not saying they are right to do so (although I'm also not saying they are necessarily wrong in doing so... publications is not a totally irrelevent metric, even if it is very, very far from being a perfect one). I'm saying that they currently use it. So the number of authors does currently matter and your hypothesis is wrong. Whether you like it or not is irrelephant
(*).(*) that's to make it less boring for @Jaloopa -- please estimate precisely how much bored you are now ;-)
-
@remi said in Precisely Estimated:
please estimate precisely how much bored you are now
Precisely about 10 boreds
-
-
@remi yes
-
@remi said in Precisely Estimated:
@japonicus said in Precisely Estimated:
For a vast project where would you put the cut-off? Would three authors be 'acceptable'? 6? 10? 30? 100?
I have no idea, but let me make an car analogy (that was missing from the thread, don't you think?). Someone going 5000 mph is going too fast (OK, that's impossible... I'm just keeping the same numbers for the sake of the argument, feel free to replace them by something more realistic). Someone going 2 mph is not going too fast (he may be going too slow, but that's another topic). Where do you put the cut-off? Would 3 mph be 'acceptable'? 6? 10? 30? 100? And then your answer is "5000 mph is fine".
In the realm of scientific experiments being analogized to modes of transportation, The Higgs Boson is better compared to a space shuttle... which CAN go 5,000 mph.
The Higgs Boson isn't an experiment you can conduct in someone's garage, or even a major university. They had to construct a friggen gigantic underground apparatus. And I am certain that each of the 5,000 people worked just as hard as each of the scientists in a much smaller scale 10-person experiment somewhere else. Thus, each of them deserve credit. Otherwise, why would someone volunteer their hard work only to be given the shaft on credit because "too many people worked on it?"
Discovering the Higgs Boson required 5,000 qualified physicists. And they each deserve credit.
You're saying "there is no way to put a cut-off, so let's accept anything". I am saying "it might be impossible to define a single all-encompassing cut-off, but when you get to 5000, you know that you've crossed that cut-off".
Nope. You can't just have an arbitrary cut-off that fits every experiment. If this were a baking soda and vinegar experiment, sure. 5,000 sounds a bit much for that. Higgs Boson? Not so much.
@remi said in Precisely Estimated:
I see the harm for the other postdocs who were not involved in that project. Take two postdocs, doing more or less the same research work. Nothing ground-breaking, just good solid research. One works at the LHC, the other at a smaller experiment. Both will publish the result of their own work in some mid-field specialized journal, a good one but nothing more. Additionally, the first one will get to be one of 5000 co-authors in a paper in a hugely prestigious journal, not the second one. A couple of years later, the first one will have a larger publication record, a better citation index, only on the virtue of having done his postdoc in the LHC and not another lab.
Oh cry me a river. Next thing you'll tell me snobby professors at Oxford won't look at UCL candidates. If that's a problem, then the solution isn't to revoke credit to people who spent years of their career working on the LHC.
@remi said in Precisely Estimated:
I'm peripherally involved in a project that should legitimately have ~ 150 authors, but won't because the people leading it are zealously guarding their position - resulting in substantial ill-will and creating quite a caustic environment at times.
It looks to me that the LHC and such decided that dealing with humans was too hard and that the only way to avoid the problem you're mentioning is to not put any barrier to who can be an author. I find it sad.
Wait... you're telling me that you're upset that a project you're involved in is unfairly limiting their selection of co-authors on their paper, and now you're hypocritically saying that means the LHC should do the same to their work? Do you even hear yourself right now?
And how do you know they didn't put a barrier on who can be an author? How do you know there weren't, say, 15,000 scientists involved and they decided to cut off at 5,000?
@mzh said in Precisely Estimated:
@remi said in Precisely Estimated:
The other sub-role is the nitty-gritty of career management in the academic world, i.e. judging the individual merit of authors.
My guess would be that the goal of large author lists is precisely to prevent this use.
Maybe. But then they are breaking the main tool that is used to rank scientists between them, without providing any replacement, nor even saying that they do so voluntarily, and in such a way that may be damaging for the other people in other projects, not to themselves. In other words, if this is what they are trying to do, then it's similar to cheating in a game in an obvious way to point out how rules are silly. That is not right, and I hope that this is not what they are doing.
I don't think it is. I think it simply took over 5,000 people to find one of the most elusive particles in scientific history which required a shit-ton of resources.
-
@djls45 said in Precisely Estimated:
Yes, but it's not arbitrary. Designing a piece of lab equipment (which is what these are, even though they are huge) does not automatically justify inclusion on papers about discoveries made while using that equipment.
While what you write is true, the more custom the equipment (and big physics experiments are extremely custom) the more likely it is that the design of the equipment is a critical part of the experiment and so the designers need to be directly credited. There's simply no hard and fast rule.
But in life sciences, the most important author is the last named author. Because…
-
@dkf said in Precisely Estimated:
@djls45 said in Precisely Estimated:
Yes, but it's not arbitrary. Designing a piece of lab equipment (which is what these are, even though they are huge) does not automatically justify inclusion on papers about discoveries made while using that equipment.
While what you write is true, the more custom the equipment (and big physics experiments are extremely custom) the more likely it is that the design of the equipment is a critical part of the experiment and so the designers need to be directly credited. There's simply no hard and fast rule.
But in life sciences, the most important author is the last named author. Because…
And in my particular branch of physics, the last author is conventionally the advisor of the graduate student/post-doc who actually did the work. And means nothing as far as effort contributed. Generally, the 1st author is the one that matters, the middle ones contributed, and the last is a supervisor.
-
@the_quiet_one said in Precisely Estimated:
Discovering the Higgs Boson required 5,000 qualified physicists. And they each deserve credit.
Which is what they got in the shit tons of other papers that were published around the main one. Not all of them did work that, in any other project than the LHC, would have gotten them a paper in a prestigious journal. They get the credit that they deserve in the publications related to the work that they did, and they get referenced in the summary paper that lists the overall results. Having them as authors of that paper is pointless and may even be harmful to other people. That is absolutely zero good reason for doing it.
Oh cry me a river. Next thing you'll tell me snobby professors at Oxford won't look at UCL candidates.
You still have no idea how the academic world works, do you? The publication list and citation index is not just some weird uni-peen. It is what is used to evaluate people.
OK, we've already been around all this. There is nothing that I can say that I haven't already said at least twice.
Wait... you're telling me that you're upset that a project you're involved in is unfairly limiting their selection of co-authors on their paper, and now you're hypocritically saying that means the LHC should do the same to their work? Do you even hear yourself right now?
Wot?? TDEMSYR. Did I ever say anything about a project I might be involved in and how that hypothetical project is limiting co-authors in an hypothetically unfair way??
-
@dkf said in Precisely Estimated:
But in life sciences, the most
importantsenior author is the last named author. Because…First author tends to be the one doing the most practical work, last author is the head of the lab (or possibly the project supervisor). Between those it's a bit of a grey area.
-
@the_quiet_one said in Precisely Estimated:
Next thing you'll tell me snobby professors at Oxford won't look at UCL candidates.
That might actually be true, though that's more likely in the humanities and social sciences.
-
@benjamin-hall said in Precisely Estimated:
And in my particular branch of physics, the last author is conventionally the advisor of the graduate student/post-doc who actually did the work. And means nothing as far as effort contributed. Generally, the 1st author is the one that matters, the middle ones contributed, and the last is a supervisor.
My field tends to either do the same (when there are many authors and a clear supervision link between them) or to put the authors more or less in contribution order, with the junior who did all the work first and the senior who provided the idea/experience just after (and the rest being smaller contributions). About the only consistent pattern is that the first one is usually the one who did most of the grunt work, but even that may not always be the case, sometimes politics or business relationship trump that (e.g. some partner who paid for most of the study has to be put higher in the order than his contribution, or even first author).
I think there was a PhDComics (or maybe XKCD? but that sounds more like PhDcomics) that compared the order and number of authors in various fields. Each one has its own weird habits, which only goes to show that there is no simple way to decide who should be author...
-
@benjamin-hall said in Precisely Estimated:
And in my particular branch of physics, the last author is conventionally the advisor of the graduate student/post-doc who actually did the work. And means nothing as far as effort contributed. Generally, the 1st author is the one that matters, the middle ones contributed, and the last is a supervisor.
My point is that different fields use different rules for this. Also, physics is nowhere near as bitchy as life sciences.
-
@dkf said in Precisely Estimated:
@djls45 said in Precisely Estimated:
Yes, but it's not arbitrary. Designing a piece of lab equipment (which is what these are, even though they are huge) does not automatically justify inclusion on papers about discoveries made while using that equipment.
While what you write is true, the more custom the equipment (and big physics experiments are extremely custom) the more likely it is that the design of the equipment is a critical part of the experiment and so the designers need to be directly credited. There's simply no hard and fast rule.
But in life sciences, the most important author is the last named author. Because…
It's because it follows evolution. You get the single-celled organisms first, and it grows to a fully upright homo sapiens at the end.
-
@dkf said in Precisely Estimated:
Also, physics is nowhere near as bitchy as life sciences.
In terms of long author lists there are a few Life Sciences examples similar to the Higgs Boson paper, e.g. the human genome paper had hundreds of authors, and then an erratum to add some more.
-
@remi said in Precisely Estimated:
a 5000 authors paper breaks down this system
I've never worked in academia, but from where I sit in my fabric box, this does not seem like a bad thing.
@remi said in Precisely Estimated:
let me make an car analogy
Once again demonstrating that car analogies are not always good analogies for some domains.
@remi said in Precisely Estimated:
dealing with humans was too hard
Duh. Is there anyone here who would disagree with that? We're a bunch of nerds and geeks; how many of us would rather deal with people than machines? Show of hands? Anybody? *crickets*...
Alpher–Bethe–Gamow paper - Wikipedia
