The current buzz over web 2.0 and the buzz I’m trying to create for Research Web is all well and good. But what about getting some of the efficiencies we associate with the Web into the daily life of a scientist?
I’m talkin’ web 1.0 here, people. Basic stuff like Amazon and eBay – tools that allow for low-cost, low-friction connections between buyers and sellers, with feedback systems and fulfillment, secure orders and acknowledgment and ratings and more.
We’re so far away from this in academic life sciences it’s laughable. Taxpayer dollars flow into labs, and enormous amounts of expertise and effort go into making biological materials for research purposes. Plasmids. Cell lines. Mice. Fish.
And most of these materials, this treasure trove of implicit knowledge and encoded experience, live inside refrigerators, never to be shared. At best they’re hoarded to guarantee the owning lab priority for more publications. At worst they simply decay from neglect, left unlabeled and unloved when a grad student moves to a postdoc elsewhere, or a postdoc takes a job in industry.
Can you imagine the increase in our ability to build on published research if the scientist reading the paper didn’t have to recapitulate all the experimental construction of tools? If instead you could search the web, drop in a credit card number, and get a cell line via fedex in four days? If, yaknow, e-commerce worked for science?
It’s possible. The technology is easy. It’s the culture that’s hard. Getting to one-click for this stuff requires a lot of people to change at the same time.
More after the jump.
We take the efficiency of e-commerce and other web tech for granted in our daily lives. If you’re reading this blog, you probably order books from amazon, use eBay when you need Pez dispensers, maybe order food from gourmet shops, art from galleries around the world, hoodies from NYC.
It used to be really hard to get that kind of stuff in most places. I grew up in East Tennessee in the 1980s, and we had to walk uphill in the snow in both directions to get a freaking salad. And when we did, it was buried in ranch dressing. Books were Ludlum and King, by the pound. Pez dispensers, good art, hoodies? Get on a plane and go somewhere with enough people to support storefronts that dealt that stuff. Gooood luck.
It’s different now. There’s chain stores, yes – but you’re also likely to hear a thick Tennessee accent saying “I did my holiday shopping online” when you go home for turkey and yule logs. That is to say, the network has totally transformed the way you get physical goods, not just digital ones.
But in life science, it’s like a pre-internet world – maybe even pre-industrial. It’s like begging an artisan, not placing an order, when it comes to getting tools.
I’m not talking about the well known patented tools, the stem cells or the equivalent. They’re the rock stars. I’m talking about the boring tools from the boring papers, the unloved plasmid or the one-time cell line – the long tail of biology.
These tools are not findable online, not available by digital contract, not fulfilled by anyone other than the creator, and there’s no easy way to give credit other than a citation. You need at least those four elements to get to one-click. And each is eminently doable, right now: we have search engines, we have standard contracts, we have repositories that will host these things and fulfill orders, and we can automate karma and feedback.
All we’re missing is the willpower to do something about it. Studies show that about half of all scientists have been denied access to these kind of materials. Usually for competition and protection purposes. Those are only good reasons to individual labs and grant seekers.
Those aren’t good enough reasons for a society of taxpayers, the people who pay for research. Those aren’t good enough reasons to justify slowing science down for all of us waiting for cures for autism, for arthritis, for ALS, for Huntington’s Disease. For ovarian cancer. For malaria. Are we really going to sit by and let individual competitive forces derail an obvious way to increase efficiency, a way that requires no magical insight into the mechanisms of the cell and the genome?
I’ll come back to each of the four points in future posts, maybe several for each. But I wanted to plant a flag that when I talk about the Research Web, I’m not just talking about semantic web and web 2.0 and so forth. A huge part of this is simply getting the culture of scientific toolmaking out of the 16th century and into the network.
As my economist friends would say, we’re leaving $20 bills scattered all over the research sidewalk by allowing this kind of system to continue.
I thought BioRoot was going to bring science up to e-commerce speed, but uptake seems to have been limited. I still think it’s a great idea though.
John, how can we address the cultural/competitive issues surrounding the sharing of materials? A lab that has worked hard to develop a cell line, knockout mouse, plasmid, etc, won’t want to just hand that over to a competing lab. Can technology really solve that?
Forgot to say earlier: what role do journals and funding agencies have in coaxing biologists to share more?
BioRoot’s a great link – and a great idea. But I think that any one org is destined to run into the collective action problem here. Everyone makes an individually rational choice (not to share) which is really bad for the overall group.
Thus, I think Corie has nailed the crowbar with which this will be opened. Funding agencies and journals have to change the dynamic, not the individual lab, and not any one organization like Science Commons or BioRoot. In SC’s materials transfer work we’ve had to put in place an entire mini-ecosystem of tech transfer offices, via ibridge, plasmid fulfillment via addgene, our own bespoke contracts and technology, and it’s still tough sledding. It’s the funders who can call the tune.
Journals have a history of success in this space with the requirements of genbank IDs for genes in publications. If Nature, Cell, and Science all decided they wouldn’t accept papers dealing with plasmids and cell lines if those papers didn’t carry URLs where you could order the cell line, I figure you’d see things change pretty quickly.
Surely publication in any worthwhile journal this days is already contingent on sharing the reagents generated in the reported work?
I hear synthetic biologists are hoping their field will be able to contribute to this in the future: design various cell cultures, genes, etc. and then when you need to use ceratin genes, gentic combinations or cellular cultures you just order whatever you need online and get it by next day delivery. Like a science version of LoveFilm
Richard: sadly, no. PNAS has a good policy on this but I don’t know of journals actually refusing to publish someone’s article because they failed to deposit in the past. Policy and enforcement are different beasts, I’m afraid. We need some teeth in the policy like we had for the genome.
Richard: Policy at the Nature journals is clear – if you generate a reagent and publish a paper with it, it needs to be available to readers. Every complaint that we receive at any of the Nature journals is dealt with by asking the author to please supply. In 99.99% of the cases the reagent is eventually supplied. In fact, the only case that was unresolved, as far as I am aware, is when the 2nd party who was contracted by the authors to supply their reagent was bought out by a third party. That company was a couple steps and a couple years removed from the publication and had no agreements with the journal. They were willing to supply a similar reagent, but not the actual one. The journal had no choice but to print a notice and attach it to the paper online that the reagent was no longer available.
That being said, the need is real; implementation is the challenge. The system would need to be simple and provide incentive to the reagent-makers. It may never be super simple, what with controlled reagents and the transport regulations surrounding some substances, but for many items, this would be a post-doc’s dream.
Linda is right and what she writes is indeed what happens in practice. Here are our policies—and we do follow up any complaints from readers about availability—which in the event are rare.
I like the idea of sending a reagent to a loved one for Christmas via one’s Amazon wish list, fully gift wrapped!
If there was a company/resource that unilaterally put together the infrastructure to store, maintain, and quickly distribute a variety of reagents, then of course the journals could apply pressure on scientists to relinquish their reagents; it would be little different from the policies that are already in place.
The journals will play no role in the production of this resource, only in the policing of its usage. The researchers will also play no role in the production of this resource. Despite the study you cite (which focuses on a very small community), reagent-sharing is very prevalent and there are few cases in which the journals actually have to step in and enforce their policies with much resistance. On average, labs are happy to provide reagents, but hate the hassle. Trust me, as someone who created several desired reagents throughout my research career, distribution and responding to the emails were a large pain in my…
Most researchers would welcome the chance to remove this bit of trivial nonsense from their everyday work activities (as much an incentive as any). For example, GFP-guru Roger Tsien has fielded so many requests for plasmids over the years that he had to start his own site and staff it with his assistants just to keep track of everything. Other resources do exist for researchers to dump the responsibility of reagent maintenance on somebody else (and indeed they do take advantage of it!), such as Jackson Labs, ATCC, and Cabri. Also, where do you think that a lot of the best commercially-available antibodies and plasmids come from? Academic research labs. Companies buy them (the source or the rights) and maintain the stock, relieving the researcher of such mundane distribution duties.
So that leaves either (both) the public or private sector to fund such a resource, and if they did, I believe it would be embraced by scientists and supported by the journals enthusiastically. I think that you are preaching to the wrong audience. What we need are investors.
I started to reply to Noah’s post in detail and realized it’s better to add this to my list of topics for full blown blog posting.
But in short: Noah’s pretty much right. Journals can mandate and police but can’t produce the resource that provisions reagents.
However, I don’t believe that any one company or resource can make the grade. I think it’s got to be a network of resources. These things aren’t like digital objects – they’re much more like books in an era where it’s hard to print anything, and when nature is constantly threatening to destroy books. We need a network fulfillment model adapted to biological reagents.
Inter-library loan is the right model here, my instincts tell me…I’ll flesh this out in my next post once I get off the road for the holidays.