Koch Lab:Principles

From OpenWetWare

Revision as of 00:46, 30 December 2008 by Steven J. Koch (Talk | contribs)
Jump to: navigation, search
Add to My Links

Home Research Lab Members Publications Protocols Contact Funding
Principles Data Notebooks Links Meetings Presentations Inventory
Image by wordle.net
Image by wordle.net

Steve Koch 23:39, 29 December 2008 (EST): On this page, I would like to work on our "Lab Principles" as a collaborative effort, including members of our lab and any others who have ideas. As of today, I am going to post my draft version of this document that I wrote on our private wiki (courtesy of OWW). It took me several hours to write the page, but it was pretty much stream of consciousness in a couple late night sittings. This was during the start of my second semester at UNM, almost two years ago, just shortly after joining OWW, and when my lab was still empty and students were just starting to join. It's a somewhat embarrassing but hopefully interesting snapshot into my thoughts about science, open science, and lab management. I feel like a different person now, a lot more experienced, a little more knowledgeable about open science, and a little less worried about open science.

In the spirit of openness, I will post the article pretty much as it's been posted in isolation on our private wiki...only fixing a minor typo here and there and fixing or adding hyperlinks. Once it looks people are ready to contribute, we can archive this embarrassing version and start fixing it up. Please make edits that you think we agree on, or post comments on the "talk" page. Also, if you'd prefer (or don't have OWW access), you can make comments on my blog (I will post the link soon).

Contents

Spring 2007 Draft Version

By "Lab philosophy" I mean what are the general principles, ethics, and similar fuzzy issues that we all do or should agree on. These are things we should talk about a lot, but it's also essential to have this as part of the wiki, so new members and old members have a place to go to get internal guidance and also a place we can go to hash things out when something new comes up. At first here, I am going to put section headings for a few of the "hot button" issues that come up.

Currently there are a lot of "I" statements in this article. The goal will be to edit it as we dicuss the issues and over time convert them to "we statements". Be bold! If something here doesn't make sense or seems wrong, at a minimum bring it up either on the discussion page or in person!

What are the goals of our lab?

Scientific Goals

  • Our main scientific goal should be to make important discoveries / inventions and to quickly and effectively broadcast these discoveries to the world.
  • We should feel good that our discoveries play a small part in helping humanity.
  • We should hope that we will get lucky and make a discovery or invention that can have a big impact on humanity and we should applaud and help those who think they may have done this.
  • One reason Steve K. is drawn to biophysics is because of the seemingly infinite number of important problems in biology, medicine, genomics, and other life sciences. It seems very realistic that students & postdocs in my lab can make a big impact on some of these problems. That does not mean, however, that we won't pursue exciting discoveries that come up and just happen not to have immediate biological impact. We do need to be realistic, though, about our resources.
  • Since our goal is to further science and technology, our goal must be to ultimately share everything we discover as widely as possible. However, there are some concerns that I have not figured out, which are dicussed further below, about how open we can be and patents and intellectual property. In general, I expect our lab to be much more open than the average lab: as open as we can be without being counterproductive.

Mentoring / Career Development Goals

  • As PI of the lab, my main goal is to successfully mentor the younger lab members. This is a higher priority to me than producing superb science, but I don't worry too much about prioritizing them because I think if I work hard on being a good mentor, the good science will be inevitable. (Conversely, it's not always true that good science is paired with good mentoring.)
    • I want all students and postdocs to grow and learn during their times in the lab. I recognize that each person will be different and will all follow different paths. But no person will be in the lab indefinitely, and we need to recognize that things happen AFTER being in the lab and plan for them early. For example, some graduate students may want to pursue a post-doc position, others may want to go into industry, while others may want to start a business.
    • I will measure success in mentoring in how well prepared the student or postdoc is when he or she leaves the lab. And then we will continue to enjoy and celebrate lab alumni's successes after they leave the lab.
  • Members of the lab also have the opportunity to mentor others in the lab! For sure a goal should be to try to be as effective a mentor to people earlier in their career as you can be. This I think will come naturally if we have a healthy lab.
  • OK, this section needs some work in being clear with the goals. How is this: The goal is to have people that come through the lab lead more successful lives because of it!
  • See also, my mentorship statement I submitted when I applied for this job. I've re-read it and I still believe this. So if you see me acting otherwise, you should call me on it!

Enjoyment Goals

  • Also, let's not forget that having a lot of fun and excitement is a major goal, too!

How open can we be?

Our general philosophy should be that we want to be as open and sharing as possible. But at least at first, we are not going to be completely open, for two major reasons that I have thought of:

  1. Risk to students / postdocs in the lab of trying to be too "open" before the world has figured it out and
  2. Risk to collaborators who are not comfortable with the idea

What are people worried about anyway? Mostly, people are worried about getting "scooped." There are definite instances of scientists sharing their ideas before they've been published only to have a speedier, or more powerful lab steal the ideas and rush to publish first. Only slightly less weasely is when a group may not have stolen the idea, but rushes to publish first once they know that the other group is close to publishing. This has the following negative effects:

  • Unlikely to publish in as good of a journal
  • Feeling devastated because your good ideas have been stolen
  • Student's graduation may be delayed if they were required to publish a "great" paper
  • PI may not be able to obtain funding, jeopardizing students and postdocs jobs
  • And probably many other negative effects

People are also worried about looking stupid. Personally, I am going to try to minimize how much I worry about this. So: I may look stupid sometimes, but I will be having fun doing science.

Those are all reasonable fears, but I hope we can be much more open than usual while at the same time minimizing some of those risks. My thinking is as follows:

  • Be as open with ideas as possible!
  • Make a lot of friends / collaborators, tell all of them your ideas!
  • Publish your ideas on the internet and let people know the ones you are working on
  • Ask people to let you have some time to complete your ideas

If we as a lab act like this, then suppose someone completely underhandedly scoops us. A whole lot of people are going to KNOW that whoever did it were pretty much complete a-holes. And if this happens to one of my students or postdocs, I intend to be very aggressive in letting the whole community know what happened. Imagine, though, if we are very secretive and someone scoops us. Nobody would know or believe us! OK, now this doesn't eliminate the possibility that we can get scooped anyway, just by coincidence of someone working in parallel. If this happens, then we just have to cope with it. See OWW thoughts.

Don't we also have to worry about intellectual property?: Yes, and I don't understand yet what we need to do about this. See the section about intellectual property (IP).

Openness on the wikis

There are some labs who have decided to put every part of their lab on the internet, for example, some on OWW. I think this is commendable and totally in the spirit of science. On the otherhand, I do think it is still experimental and while I believe it is very much in the right direction, I don't want to subject my students and collaborators to the experiment. Particularly since I am a new professor and we have so many other hurdles to cross. In some sense, having your lab 100% completely out in the open is like walking around naked. You probably really don't have anything surprising to hide, but on the other hand, you're uncomfortable doing it and most people aren't interested in seeing it. So, it doesn't hurt to wear some clothing. OK, brutal analogy, but I decided we should wear clothes in the lab and so we will have a private wiki and a public wiki. The adminstrators of OWW were very helpful in providing us the private wiki and their hopes are that it will facilitate us ultimately doing more open science.

Private Wiki

The idea is for us to be very very open on our private wiki. As much information as possible should be put on the wiki and we should constantly look for ways to improve how we do this and make it easy to do so.

  • Lab Notebooks: We are going to totally break the mold of the old-school lab notebook. Everything is going to be electronic. Old-school lab notebooks are revered by many labs around the world, but they are also of very limited value, except in courtrooms. I just don't think it makes sense anymore to have pen and paper be the normal archival medium. This will be controversial, but I am pretty sure about this!
  • Other things: Let's fill this in as we go. But really everything should be on or linked from the private wiki.

Public Wiki

At first, the public wiki will include all of the regular stuff you see on most other lab's web pages. This is a good start. However, we will also have the goal of transfering private wiki material to the public wiki, once we feel it is in a form good enough to help the greater community. For example, "tips and tricks" or protocols that we use that are likely helpful to others. Brief little research results that we don't think we will bother to publish, or that won't be hurt by the Creative Commons licensing, (or that are too time critical to hold on to?). Another example: Everyone in the lab should discuss this "lab philosophy" page and improve it. It would be a scary but useful thing to post on our public wiki, I think.

Guidelines for transferring from private to public wiki

It would seem prudent that nothing major should be transferred (cut and paste) without all of us agreeing.

Guidelines for contributing directly to the public wiki

I for sure would like to see everyone in the lab (including me) contributing a lot to OpenWetWare and to our public wiki on it (without first posting it on the private wiki). So I don't want to discourage that. However, there could be tricky situations where two or more people are working on something and don't agree on how private to keep it. These things should be discussed openly and include me as well. I would bet that usually we can come to a sensible compromise. There is also the possibility that a student alone could have an earth-shattering idea and want to put it on OWW right away. I would suggest talking with me first, just in case there are negative impacts of being hasty.

Openness with people we really trust

I think we should be completely open with people we trust a lot. Like active collaborators, classmates, family, whomever. I know paranoid people who won't tell even trusted people "lab secrets" because they fear this trusted person will let it slip to the "enemy." Let's try not to be paranoid. For one thing, we are not working on nuclear secrets. Presumably if we get scooped at the very least it means that even though we got screwed, the world was bettered. Secondly, for the most part almost every scientist in the world is way too busy to pay attention to rumors he hears from some random person. I suppose there is the possiblity that some of us actually WILL work on something classified. Obviously "openness" does not apply to that. That kind of thing will be very rare I think.

Openness with the general public, face to face meetings

Again I think we should be as open as possible, more open than with the wiki, maybe not quite as open as with people we trust. The reason I say "not quite as open" is because everything we do affects each other. So we don't want to say or reveal things that obviously embarrass another lab member. Accidental embarrassment is fine. A related point: often you will want to edit what you say just because you don't want to confuse people that don't know you too well! Like I said somewhere else, giving the entire world access to your entire lab is like walking around naked: probably not very desirable to the people you're interacting with!

Patents & Intellectual Property

One really big problem with "open science" is the impact on intellectual property (IP) and ability to file patents. We should keep in tune with strategies people develop to deal with this. I really don't know the answer. Some thoughts:

  • I think patents are a very good thing. I do not think they are counter to the spirit of science. The spirit of science is to help the world. In some cases, the quickest way to help the world is to have a commercialized product. And having a patent makes it WAY WAY more likely for investors to take a risk with a new product. Many people in science view patents as "dirty" or "evil" or "greedy," but I do not agree.
  • I think we should file disclosures when we think we have inventions worthy of a patent.
  • I think being too open will hurt our ability to patent. So when there are obvious and easy things we can do to preserve a patent then we should do so.
  • Publishing an invention before filing a disclosure forfeits your international patent rights. However, you still have one year to file a US patent. I don't know all the details here, but all is not lost if you act quickly. This is how we filed our patent for our invention at Cornell, although we intentionally published in Biophysical Journal before submitting the disclosure and could easily have done otherwise (at the risk of being scooped).
  • Maybe once we get the lab cooking, we will set up a meeting with the UNM IP and tech transfer people to get solid legal advice. And also observe what other very open labs are doing.

Ethical research conduct

This is one area we really need to talk a lot about. Most people think they are fine on ethics and that it's a pretty simple subject. This leads most PIs (in my opinion) to ignore the subject all together, which I then think leads to a lack of training on the younger people! It's not as simple as "don't fake data"--most people understand that one, but even that one has different meanings to some.

IMPORTANT The absence of mention of some kind of ethical question on these pages does not mean we condone it! We should add a discussion of it! Any ethical question should be hashed out somehow.

Respect

I tried looking this up in the dictionary and they didn't list the meaning I was thinking of. Then tried wikipedia(permalink) and see how complicated of a concept it is. Let's try this for starts:

  • Respect ourselves. That is: we should all have self-respect and know that we're all growing and striving to be better scientists and to make great contributions.
  • Respect other lab members. Lack of respect (whatever the exact definition!) for peers, subordinates, or supervisors makes for a really unpleasant and dysfunctional lab. But how do we achieve respect? I don't know: hopefully the more effort everyone puts into it, the easier it is for everyone to do it. In case this makes no sense, here is an example: If a person thinks that a piece of equipment is not working they have at least two choices (a) move on to something else because they're in a hurry or (b) take time to put a note on the equipment that it may be broken, so that the next person does not get screwed. Choice (b) is more respectful, of course. Of much greater magnitude, a respectful person will make sure whatever they are doing is as safe as possible for the OTHER lab members.
  • Respect our collaborators. I cannot overstate the importance of this. There have been a number of physicists who have made public statements of HUGE disrespect to biologists. And there are still a lot of physicists who have disdain for any other discipline besides physics. Conversely, I am sure there are a lot of egotistical biologists who denigrate physicists. Any of these attitudes are not only worthless, but really set back the progress of science. There really is a very large barrier between pure biology and pure physics. It takes a lot of work between the biologists and physicists to overcome these barriers. And that work simply cannot be done if the people disrespect each other.
  • Respect Science. I guess by this I mean we should respect science as being really important and providing the key to improving humanity. So any lazy or bad science we do is really disrepectful. So we shouldn't "dis" science.

After sketching out all of this about "respect," it seems to me that self-respect, respect of others, and respect of science are key to really all the other ethical areas. Without respect, how can we have trust, integrity, honesty, etc.? Doesn't seem possible.

Trust

We should hope that trust comes naturally. Trust will be easier with lots of mutual respect. Trust does not, however, mean that we won't make mistakes and that we won't constantly question everything: that is science! We just want to operate under the assumption that none of us are intentionally being deceitful. This needs to be talked about.

Honesty & Integrity

  • Being honest with yourself: This is critical for good science and harder than it may seem to a casual observer. As my friend and colleague Richard Yeh put it: "It is all too easy, after many long days, weeks, months in the lab, to fool oneself into believing that weird data points are flukes or that one's research means more than it does."
  • Being honest with each other also sounds easy. However, I think there is a parallel to "self-deception" in science as "covering things up" or being a "weasel." It's "trying to get away with stuff" in publications. It happens a LOT and I think it is crappy and hurts science. We need to help each other avoid this and call each other on it.

A culture of constantly questioning everything

I think respect also leads to this. If we really respect each other and we respect science, then it's really hard to choose not to question each other's work. We need to get really good at challenging everything, but doing it in a way that doesn't lead to nastiness. That's sort of challenging. But we will work at it. And it's the right thing to do.

Authorship on publications

Particularly if you are new to science, you may think that given how many papers are published each year (or even each day) and how long scientists have been publishing papers that the issue of "who is an author" would be worked out by now. Well, surprisingly it isn't! Not even close! Many different labs have different policies and most of the time they are not spelled out. Thankfully, most labs I think have a good PI who is fair and so things tend to work themselves out. Nevertheless significant problems, disagreements, and damaged relationships occur very frequently.

Factors that mitigate or exacerbate authorship disputes

Here are some factors which make authorship problems worse

  • A feeling that "too many" authors dilutes the PIs or lead author's credit.
  • The very common custom that the PI must be an author on any paper that includes results from any piece of his or her lab.
  • Pressure from tenure committees to produce a lot of publications
  • Pressure on graduate students to produce a lot of publications
  • A very common custom of putting authors on an author list who did not really contribute much
  • Lack of real guidance or rules from journals regarding authorship
  • Lack of any communication about this between all parties throughout the research. Usually there is no discussion until someone drafts a paper, and then commonly it gets worse when the PI doesn't consider the author list right away.

Here are some factors which can make authorship disuptes less common

  • Open discussion between all people in the lab and with collaborators. Part of this I hope will include posting our authorship philosophy on our public wiki
  • A PI who is somewhat less concerned than others about producing a ridiculously long list of publications.
  • A PI who thinks he can succeed in getting tenure and grants while trying to follow authorship guidelines provided by parent societies (such as the American Physical Society).
  • A PI who knows what everyone is up to in the lab. This actually may be harder than I think.
  • Mutual respect between all people involved.
  • A common goal shared between all people: recognizing that one major point of science is to get the results out there!
  • Guidance from large societies that clarifies authorship issues. Believe it or not, this has existed for some time, and I do not really know anyone who follows it! See next section.

Guidance from profressional societies.

  • APS GUIDELINES FOR PROFESSIONAL CONDUCT
    • quote: Authorship should be limited to those who have made a significant contribution to the concept, design, execution or interpretation of the research study. All those who have made significant contributions should be offered the opportunity to be listed as authors. Other individuals who have contributed to the study should be acknowledged, but not identified as authors.
      • Comment: A subtle point is that significant contributors should be offered the opportunity to be an author. So, it would be weasely to secretly draft a paper and then say, "but you didn't write anything". It's also weasely to intentionally exclude someone who took data from being able to participate in the interpretation of it.
  • Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication
    • quote: Authorship credit should be based on 1) substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3.
    • quote2: Acquisition of funding, collection of data, or general supervision of the research group, alone, does not justify authorship.
      • Comment: I have NEVER met a PI who obeys quote2 above! It would actually be pretty uncommon for the PI not to have some contribution but there may be circumstances where it arises. For our lab, I would hope that we have a great atmosphere and that students and postdocs will want to co-author papers with me and brainstorm about experiments, data etc.! I put the quote down because it makes sense. Maybe it could arise if you want to publish a little paper from some research you did in a class, for example. And we decide that it's good enough the way you've researched and written and there is no point in me getting involved? Who knows.
Personal tools