The three essential components of a good laboratory website.

synthesisMy colleague, Attila, asked me months ago what are the components of a good laboratory website. However, my thoughts remained incomplete until I read David Crotty’s post at the CSHL blog that I was able to form an answer, and even then it took me two long comments to get my thoughts straight.


The main argument seems to be that it’s too soon for most of the social web applications in science. Of course, everyone has to make that decision for themselves, but I think this is a great way for a beginning scientist to get noticed and to distinguish themselves. Let’s not forget the “founder effect” that accrues to the early adopters. Certainly, it can also be a great way to waste a lot of time unproductively. The key is to get in at just the right point, when the technology is ready to take off, such that the startup effort is minimal. When the startup effort is minimal, the argument that it’s not worth the time to learn a new system becomes a minor objection, and indeed, the startup effort for use of Connotea is about as minimal as it gets.

That leaves us with the question of whether existing systems are “good enough”. Again, this all depends on how long you’ve been using your system and how comfortable you are with it. If you’ve been doing something for a long time, and it’s always worked, the only argument for change is to adapt to coming changes that will break your system. I don’t intend to argue that what has worked in the past isn’t a good system. However, why wouldn’t a new investigator, just starting out, pick a system that was as future-proof as possible?

I admit I’m a bit of an early adopter, and this probably comes from working in the relatively new field of stem cell research, where protocols aren’t really set in stone just yet, and scientists do do a fair amount of technology development. The lack of the tried-and-true in my field has led me to search around for solid methods used in related fields that I can adapt for my own use. The three most solid applications I have found, borrowed from the information technology crowd, are blogs, wikis, and social bookmarking.

Blogs are a great way to open your notebook to the world, or just to the lab. They’re a rich medium(you can’t imbed an animation of cells exchanging mitochondria in your paper notebook), accessible from anywhere, and automatically backed up. They also make you improve your writing skills. From a social perspective, they’re the root of your identity on the web: the place where you communicate your thoughts, your CV, and where you receive communication from others. On a blog, this kind of interaction can happen 24/7 with anyone in the world. In person, you only interact with people in your lab, university, or company.

Wikis, and their propellerhead cousin, CVS, are the sine qua non of modern software development. They’re the single best way to coordinate the efforts of many people working on developing the same protocol, so it shouldn’t matter whether that protocol specifies the input to output transformation of digits or DNA. For a small lab that works in an established field, it’s probably overkill, but for those on the cutting edge, you must have a system for rapidly developing, validating, incorporating, and then distributing changes to protocols and techniques. If you don’t, you’re going to end up lost in the minutiae or slowed down by repetition of mistakes. If you want to move past hierarchical control and its attendant inefficiency, a wiki is the only way forward.

Social bookmarking like that provided by Connotea is relatively new, but the analog in the IT world, del.icio.us, has been widely adopted over the past 4 years of its existence. Social bookmarking allows you to easily keep, share, and discover interesting pages or papers. Sure, you can email a journal club PDF around, filling up everyone’s mailbox, but why send 3MB * 100 emails = 300 MB when you could send 300KB that’s accessible anywhere? If you’re already using Connotea, and everyone should simply for the discovery features, you can have your own journal club group that the presenter simply adds the article to. No having to maintain a list of current email addresses for everyone on the journal club list, and the article is immediately placed in context of related work.

These three applications are the essential components of a good laboratory website. If my experience is anything to go by, the website is the first thing a prospective graduate student sees. Since all three can literally be up and running in minutes with minimal maintenance required, it’s trivially easy for anyone, even professors under tight grant deadlines, to get started. To further illustrate this point, I’ll put a up a series of posts on my blog giving a step-by-step rundown of how to get started from the very basics. By doing so, I’ll simply be returning the favor done by the bloggers who I read as I was starting things up for myself. I call my blog Synthesis, no doubt annoying some synthetic chemist, but the reason behind the name is that I see my blog as a place for the synthetic product of my reading and learning on the web. By doing this, I’m lowering the barrier to entry by taking what I spent 2 or 3 hours learning about and adapting to my needs, and explaining it so that it only takes 15 minutes for the next person whose needs are close to mine. The promise of the web is that someone, somewhere, has asked the same question as you and has solved it(or at least made some headway), and this is one promise that has been kept.

In the end, each lab has to make the decision whether or not to support blogs, wikis, and social bookmarking. However, they owe it to themselves to fully understand the present and future advantages before making a decision. Not everything picked up by early adopters is a winner, as shown by the lackluster penetration of Apple’s iphone into the mainstream(Nokia FTW!), but the time is right for blogs, wikis, and social bookmarking. I’m sure of it.

About Mr. Gunn

Science, Scholarly Communication, and Mendeley

10. October 2007 by Mr. Gunn
Categories: Uncategorized | Tags: , , , , , , , , | 8 comments

Comments (8)

  1. I’m looking forward to your follow-up posts. As someone who relies heavily on lab’s websites for so many things, it’s shocking how poorly designed and constructed so many are. The simplest, most obvious pieces of information (a phone number, a mailing address) are often nowhere to be found. Lists of publications are usually years out of date. However, websites are a technology that labs understand and can use to their own benefit. I’m a lot less certain that many of the other new social networking attempts are as useful.

    I’m not arguing that it’s “too soon” at least as far as anyone being able to grasp the concepts of what these tools offer. I’m arguing that it’s “too soon” because we have yet to see the “killer app” that suddenly makes sense to everyone. When you’re talking about new investigators, early adopters and those on the bleeding edge of technology development, you’re talking about a very small minority. It’s often hard to see this if one is a part of the blogosphere, surrounded by that small minority. Yes, the current tools can be adapted in some interesting ways, but they don’t seem to have caught the interest of Joe Scientist. The original argument was that it’s because Joe Scientist doesn’t know these tools exist. My argument is that the tools just aren’t that compelling yet (key word is “yet”).

    You are right in that there can be benefits to being a founder. At the same time, there are also lots and lots of people who have put in lots and lots of effort into technologies (and companies) that have gone nowhere. Most scientists are already pressed for time, so investing a large effort at this point that may or not pay off is perhaps too risky. Those hours spent blogging or tagging articles could instead be spent doing experiments or actually physically networking with other scientists at meetings/talks, which still is a stronger currency than leaving comments on a blog. Founders can be rewarded, but remember, they’re always going to be a small group–if everyone jumps in at the early stage, then there’s no founder effect. Those who take the risk and do the work will, if they’ve chosen wisely, reap the rewards. You can’t ask everyone to be a founder–it doesn’t work that way.

    I’ve been thinking a lot about this lately, particularly the concept of, as you put it, “opening your notebook to the world.” The question that immediately comes to mind is whether that is antithetical to the way biology is done these days. Theoretically, it’s a great idea, but given the competitive nature of research these days, how open are people really going to be with their unpublished data? While I’m a firm believer that it’s better to have collaborators than competitors, not everyone shares this philosophy. Are you willing to allow for the possibility of being scooped because you put up all your data in a publicly visible forum? Will the editors of Science care that you had the data up on your blog first when someone else has already published a paper showing the same concepts (which they got from reading your blog)? With so many scientists funding their labs/homes with their patents and universities so dependent upon those same funds, how supportive are they going to be toward a premature release of data? Labs at a biotech company obviously don’t want anyone to know what they’re doing, let alone what they’re reading, what protocols they use, etc. As a publisher, we’re often asked for private versions of our journals/book websites that can be run from within a company so no one has any access to the user stats that company generates.

    Clearly this cut-throat manner is not the best climate for doing research. If you haven’t read it, Jim Watson’s “The Double Helix” gives a fascinating portrait of what it was like to be a biologist before the molecular biology revolution. Things were done in a much more “gentlemanly” manner (pun intended, at least as far as the gender emphasis goes). Watson and Crick were told they couldn’t work on the structure of DNA because a different lab at another institution was already working on that problem, and that it wouldn’t be fair to butt in. Now that’s the other extreme, you have problems that would end up not being solved because you didn’t want to step on anyone’s toes. There’s got to be a happy medium somewhere. As computational biology continues to rise, we’re starting to see the influence of the open source movement inching its way into biology. Hopefully this is something that will continue to grow and alter the competitiveness and secrecy of modern research.

    The other main point I was trying to make is that to overcome the inertia we both agree exists, applications need to be easier to use. They need to be obvious, simple and attractive. It’s interesting you bring up the iPhone here, as I used the iPod as my earlier comparison. I wouldn’t call the iPhone’s penetration lackluster at all, given that it’s outselling all other smartphones on the market (at least in the USA). While the Nokia N95 is a sweet phone, it’s overkill for what most people are willing to do on their phone. The iPhone is still rough in its execution, but it has the right idea, a simple, easy to use, compelling package that doesn’t take much futzing about, doesn’t require you to go to the command line, etc. Again I think you’re seeing things from the minority viewpoint–the few thousands of folks out there who modded their iPhones and are now being very vocally negative about it online are a drop in the bucket compared to the million plus phones sold. They’re the equivalent of those currently building and using the new tools for science. Pioneers to be sure, doing interesting work, but there should be no surprise when Joe iPhoneUser refuses to break open his case and take a soldering iron to his hardware.

    It’s a lesson we learned when we spent hundreds of thousands of dollars (ah, the dotcom days) to create the now defunct molecularcloning.com, which was essentially an online version of our best-selling manual (”Molecular Cloning”, AKA “Maniatis”). It never saw the mass uptake we expected, and the reason is that when given the choice of going to the computer, going to the site, signing in, finding the protocol you want, printing it out, then going to the bench, or, grabbing the book off the shelf and going to the bench, most people are going to opt for the latter. There’s no “there” there because the tools are still too clunky. Back to Connotea–most journal clubs are a lot smaller than 100 people, usually just one lab or a few small labs coming together. The scale of tracking 30 e-mail addresses isn’t all that daunting, and bandwidth doesn’t seem to be much of a pressing issue for e-mailing out pdf files to a small group. As far as the Connotea link being accessible anywhere, you’re assuming an always on internet connection, whereas I can read the pdf I’ve been e-mailed on an airplane, or take the printout to sit poolside. Ink on paper, as old an interface as we have, and it’s still one of our most effective and most popular ones.

    Keep in mind that familiarity trumps superiority every time (as shown by this study in particular http://tinyurl.com/ytz8j7 ). To get people to switch, you need to go beyond just “better”, you need over the top compelling. And that’s my point. There’s nothing compelling out there yet, nothing tailored so well to biologist’s needs that switching to it is an obvious decision.

  2. My argument is that the tools just aren’t that compelling yet (key word is “yet”).

    Connotea is one application that everyone, from a big lab to a small one, could benefit from using, and the startup effort is so small that I don’t know why everyone isn’t using it. I mean, you’re already reading articles, and you’re already doing something in order to remember where you read what, so tagging something once you’ve read it seems like a small effort, and being able to easily find it when you want to reference it more than makes up for the minimal one-time effort of registering an account. Then the added value of seeing everything in context, by browsing similarly tagged articles, or users who’ve tagged similar articles, is just lagniappe. You aren’t factoring that in, because you just don’t know how valuable it can be. Try Connotea out for a while next time you are researching a topic and then let me know whether or not you find it compelling. As far me…well, as I was writing this, I found a great post on open science on the Connotea “open science” tag page.

    [not worth] investing a large effort at this point…

    Stay tuned. It’s so much less effort than you think, you’ll be surprised.

    Are you willing to allow for the possibility of being scooped because you put up all your data in a publicly visible forum?

    Yes, and it’s not only me. Do a look around for open science. You’ll find plenty of people. JC Bradley is a huge supporter, as well as Rosie Redfield, and many others. As you can see from the Romeo database of publisher copyright policies, most publishers support some form of self-archiving, and I fully expect deposition of data in a central database to become a requirement for NIH-funded work over the next 5-10 years. The MIBBI project is laying the groundwork now.

    Biotech companies are a different breed. Let ’em be as secretive as they want. Who has historically discovered the important things first, companies or academic researchers? I guess where you come down on this depends on whether you think you’re better off with a smaller percentage of a larger pie. The secrecy and control is for protection of profits, but pharma would do well to look at what’s happening to the record companies.

    The scale of tracking 30 e-mail addresses isn’t all that daunting, and bandwidth doesn’t seem to be much of a pressing issue for e-mailing out pdf files to a small group.

    I’ll have to defer my comments on this until you’ve tried managing an email list and working with Connotea for a while. Connotea makes existing things easier, sure, but the truly compelling benefits are things you just don’t know you need yet.

    Keep in mind that familiarity trumps superiority every time

    This is what all the discussions eventually boil down to when I talk with someone who’s pessimistic about the technology. Yes, cognitive lock-in does exist, even among scientists, and it’s a barrier to entry for the average user. But do you, personally, want to be average or do you want to be great? Remember, those who remain resolutely close-mined and resistant to change are the ones who become more and more irrelevant as time goes on. (q.v. the RIAA) I’m not an indiscriminant tech evangelist, but I really do think that Connotea and open notebook blogs are right for any academic scientist, and wikis are great for the smart and productive ones. Have you every been to Myspace? If those people can write a blog, surely us scientists can figure it out.


    given that it’s outselling all other smartphones on the market

    OK, this is off-topic, but I have to say something.

    Point 1:The iphone isn’t a smartphone.
    Point 2: It had large numbers for the launch month, but sales have drastically dropped since then. Look at the mad crowds in this picture taken on launch day.
    Point 3:In conclusion, this.

    As far as the soldering iron requirement goes, well, there was a software unlock out before the firmware update. My Nokia, OTOH, was unlocked by simply entering a code provided by AT&T on the keypad.

  3. I don’t doubt the usefulness of Connotea. As information seems to be expanding exponentially, tools like this, to manage information are increasingly valuable. But dismissing my argument about familiarity trumping superiority doesn’t make it any less true. The argument I’m trying to make is not one stating that scientists should avoid new tools (I feel just the opposite really, and am doing my best to drag CSHL Press, kicking and screaming, into the blogosphere, to bring us up to the 21st century a few years late). The question is instead why people aren’t adopting those tools. As you seem to feel, these tools are obvious and useful for labs. So why aren’t people using them? My argument is that they’re not obvious enough and they’re not easy enough. I do expect to see more adoption over time, and I do expect to see better tools emerge over time. But as you noted in your comment on another blog a while back, getting people on board is like pulling teeth these days. And I’m trying to understand why that is.

    I may not be the best test case for most of these tools anyway, as an editor, I have different needs than a bench scientist. But I have managed many an e-mail list for various groups. On the single lab scale, it’s fairly trivial.

    As for Open Science, it’s an interesting ideal, and we’ll have to see if it is indeed the future. As one of the articles you linked to notes, “Open science barely exists at the moment — infancy would be an overly optimistic term for its developmental state.” If so few are participating in this philosophy, then we can hardly expect the tools to be widely adopted, at least not at the moment, can we? And there will always be questions of pragmatism that stand in the way of idealism. Very few are going to be willing to risk their careers and make that leap until it’s an established practice. If you’re a postdoc in Cell Biology and you’re about to hit the job market, would you rather publish your big paper in Nature or in PLOS? Which is more likely to improve your job prospects? Is making a stand for that idealism more important than paying the rent? I’m not saying any of this is a bad thing, I’m just saying that it will require a fundamental shift in the culture of the life sciences for it to reach the mainstream.

    It’s hard enough to get a grad student to ask a question during a big talk at a meeting, let alone to start critiquing the work of others online. Careers and reputations are at stake here, so people are going to move very slowly and very carefully. You can certainly call that “closed-minded” or “resistant to change”. I’d call it prudent and cautious. If you have a family to feed or a lab full of employees to pay, it’s hard to live on the edge like that. Those in the mainstream will certainly benefit in the long run from those who are willing to push boundaries. But they’re very different breeds with very different needs. The mainstream will eventually catch up, but don’t expect it any time soon.

    And as for the iPhone, if you don’t like the term “smartphone”, blame it on Reuters, not me:
    http://tinyurl.com/yq3gp7

    Note that those sale numbers were for July, when the price as $200 higher. With the price drop, sales have only increased:
    http://tinyurl.com/26ld43

    And remember, you are an early adopter, you are someone comfortable getting under the hood of your technology. You are part of a tiny minority, and the iPhone is not designed for you. Yes, the soldering comment was an overexaggeration, but given the steps necessary for unlocking, it wasn’t that far off for most people.

  4. The question is instead why people aren’t adopting those tools.

    Inertia is definitely the explanation. I’m of the opinion that people will adopt the technology when they’re ready. Of course, the value of social software grows as the number of users increases(to a point, then things have to change again), so you eventually want everyone to join in. In order to bring the conservative, traditional types on board, it has to reach a critical mass and tip over into the status quo. There’s nothing you or I can do to bring in late-adopters except reach out to the people who are willing to give something a try, and make sure to only recommend the best-of-breed, sure-fire applications to them. Connotea could certainly be better, but I think it’s good enough to start recommending. It’s dead simple to use, and really adds value to your reading.

    Very few are going to be willing to risk their careers and make that leap until it’s an established practice.

    On the whole, I’m not talking about taking any kind of huge gamble. I guess if I was egotistical enough to think that everyone cared about what I was doing and was watching my every move, but too paranoid enough to think I was too far ahead to be caught, then I would fear the scoopage more than I do, but as it is I’m not worried. My problem has been that I’m not getting enough notice, rather than too much. I guess it all comes down to an individual’s tendency towards risk aversion, doesn’t it? I’m not expecting change anytime soon, but I do think there’s a significant upside to understanding the field well enough to know what’s going to work and what lies ahead, and minimal risk. I read something in SEED Magazine recently talking about how humans tend to be more risk-averse than pure rationality would suggest. There are also arguments that the reason America has become a global leader is because of our increased risk tolerance relative to our European brethren. Clearly, I’m fairly risk-tolerant because I’m both young and optimistic. One day, I’m sure I’ll be the one shooing the kids off my digital lawn.

    I think it’s clear now that we’re coming from two different perspectives. You’re trying to see things from the perspective of the average user, whereas I am presenting the arguments that seem compelling to me. Uptake by the majority just isn’t part of how I judge the value of something. I don’t know when Connotea is going to reach name recognition with Joe User, but I don’t think it’s because it’s not idiot-proof enough, and if you can update your facebook or myspace page, you can write a blog. Doing it well, now, that’s another story, and perhaps that’s what keeps some people away, but that’s nothing that can be solved by technology.

    Waiting for adoption is a system of microattribution, below the level of a whole document, that is compatible with the collaborative nature of shared databases, blogs, and wikis, but even now I think a little bit of effort in establishing an online identity can go a long way towards getting you noticed, without taking away from your chances of getting that Nature paper. However, as you say, reputations are at stake, so you’d better be able to write well and present yourself in a professional manner. The only way to learn to do that is trying and failing a few times, so it’s clear to me that you’ll come out looking best if you’re prolific and make sure that the signal that’s the loudest is the one you want to be the loudest. That’s a strong argument for having an online presence, no matter who you are.

  5. I’d say it varies from case to case. For programs that are "good enough", then yes, it’s really a question of inertia and waiting long enough for them to catch on. But as I said, I want that "killer app", that threshold tool that’s so obvious and so easy to use that it’s a no-brainer and it catches on in a rapid manner. I want Napster. I want the iPod. And that is what’s really going to push a lot of the other tools to the forefront, tools that might not catch on by themselves, but once the community is on board with the concept of participating, you can get away with a lot more.
    On scoopage: it really depends on your work, how competitive your field is, how unique your work is. I had the pleasure of working as a postdoc in a lab that had tools that no one else had. We could be very open about our work in progress because it was unlikely anyone else was going to be able to do what we were doing. But labs like that are few and far between. If you’re one of a dozen labs working on a particular step in a signaling pathway you might not want the 11 other labs to see what you’re working on until you’ve figured it out yourself. So I assume that’s going to be a decision made by a PI, which, given the current funding climate, is more likely to fall on the side of being conservative. It’s hard enough to fund a lab without giving everything away to those who are competing with you for the same funds.
    Getting noticed is often an issue for graduate students/postdocs. It will be interesting to see if blogging does indeed pay off in that respect. When I give "alternative career" talks at schools, the one piece of advice I give everyone is to build a network. That usually means going to meetings, having lunch with the speaker, collaborating with other labs, taking courses, etc. Getting your face seen, getting your name known by others. Blogging could certainly play a role in that as well.
    Standing out does come with inherent risks. Scientists are humans, and they play politics and carry grudges as much, if not more than any other group. The student who stands up during a talk and announces "that’s a load of crap" is sure to get noticed, but this may not be a good thing. I’ve seen grants get rock-bottom percentage scores based solely on the name at the top rather than the work described. Recently I inadvertantly ended up in the middle of a long running argument between different factions in a particular field when I sent a paper out for peer review to someone from "the other side". In the long run, I think getting your voice out there is a good thing, but it’s not something that should be approached lightly. It should be done thoughtfully. If, as you’re suggesting, you’re part of a lab blog, then you’re a representative of the lab. What you say reflects upon your fellow students, postdocs and your PI (and what they say reflects upon you). Biology is such a top-down hierarchy. Does the lowly student dare criticize the important man atop the hill? I’ve often wondered if anonymity would make for better commenting tools–there of course you run into the issue of who to trust and why bother reading something if you don’t know the source.
    The other issue is that if blogging becomes a standard thing to do, then the noise ratio goes way up, and you really have to do something exceptional to stand out. Perhaps it’s better if it doesn’t catch on quite so well. Do we need 10,000 biology blogs? It’s hard enough to keep up with the literature already.
    There are so many interesting paradoxes that spring up around these networks–the network is of no use until it hits a certain threshold of participation, but why join it before it’s of use? Or the question of the tyranny of the masses bringing things down to the lowest common denominator–great quote from Cmdr Taco of Slashdot fame here: "When you’re building a system like this you’re balancing the wisdom of the crowds versus the tyranny of the mob. Sometimes a crowd is really smart, but some things don’t work so well by committee. Crowds work when you have a tightly knit group of people with similar interests, but when you have a loosely knit community you get "Man Gets Hit in Crotch With Football" and Everybody Loves Raymond, where it’s just good enough to not suck." So do you build in editorial oversight, and if so, isn’t that just the thing that the democratizing power of the social network is meant to do away with? Does an open door policy lead to mediocrity? Or do you limit who can and can’t participate?
    All questions we’ll continue to struggle to answer. At CSH Protocols, we’re very interested in incorporating new tools that make people’s lives easier but getting buy-in is never an easy thing. Since we’re part of a not-for-profit research institution, we don’t have the kind of cash that a lot of commercial publishers have to throw around hoping something sticks. If we’re going to invest, it has to be a sure thing, because we can’t afford to be wrong. If the big commercial publishers are the only ones who can afford to experiment on such a large and prominent scale, will the results jibe with the interests of the working scientist, the "open" scientist, or will they be geared more toward corporate interests and profit? So yes, you’re right–I have to think on the level of the average user, what’s going to appeal to enough people to be able to support itself, not necessarily what’s the slickest, coolest thing out there.
    Speaking of promoting one’s own interests, time for a shameless plug–we’ve got a nice set of basic HESC protocols coming out in the next few months, but this is an area I’d really like to expand in CSH Protocols, particularly methods for differentiation of various cell types. So consider this an open invitation to submit methods. We do offer a good deal of editorial support for authors, have no page charges, have an open access option and pay authors a royalty based on usage of their articles.
    Looking forward to hearing more about the tools you find useful…..

  6. Pingback: Bench Marks » Blog Archive » Web 2.0 for Biologists–Are any of the current tools worth using?

  7. Pingback: Bench Marks » Blog Archive » Begging for bloggers

  8. Pingback: Web 2.0 for Biologists–Are any of the current tools worth using? « Bench Marks

Leave a Reply

Required fields are marked *