You say you got a real solution
Well, you know
We’d all love to see the plan
In a recent piece in the Guardian, Dorothy Bishop proposes a new system for scientific communications. The piece is thoughtful and well-written; not a trace of cynicism. A longer piece by Rebecca Kennison and Lisa Norberg also proposes a new system. You can find the white paper here. It also is very well-written and untouched by cynicism. You can feel the presence of superior intelligence in all of these pieces–that is not in question. Google “scholarly communications system” and you will retrieve over 5 million pages, most of them irrelevant, of course, and not all of them well-written. But that there is a “system” there can be no doubt. Once you have a system, of course, it can be improved. Let’s swap one Big Idea for another.
Now, you already know that the “however,” “but,” or “on the other hand” is coming. Exhibit A is an analogy: a walk down any street in Manhattan. I am sitting in my poorly planned, unsystematic suburban home as I write this, but this is almost a coherent algebraic formula in comparison to the spectacle of arriving at Grand Central Terminal and stepping onto the sidewalk. Who could have foreseen this? And even harder to contemplate: Who could have planned it? The grid of streets almost makes it feel systematic, until all the exceptions are brought to mind. The diversity, the vitality, the sheer creativity of a capital city: where is the system in this glorious mess of individual acts and improvisation?
I doubt that Exhibit B, the current practice of scholarly communications, is any different. It wasn’t a system that gave rise to the author-pays (aka Gold Open Access) service at Vitek Tracz’s BioMed Central or the hoovering up of metadata that forms the core of Symplectic. It’s not a system that creates a group of librarians working together in what ultimately became ICOLC. And it’s not a system that has enabled the largest commercial publishers to coopt both Gold OA and library consortia. Things develop a wee bit at at time–first some pressure here, then some pressure there. At some point we then look back and say, “We knew where this was headed all along.” This is the same argument that is made for Intelligent Design.
The problem with cooking up a system is that it trades the creative contributions of thousands of individuals for the more refined and articulate plan of a small number of elite advocates. If the advocates were not as accomplished as they are, it would be easy to dismiss any proposed system out of hand. But intelligence is a great seductress; it slyly leads us to assume that being smart and being right are the same thing. Meanwhile, the evidence to the contrary is messy and contradictory. Although “everybody knows” there are too many journals, new journals appear every year. Although it is assumed that the university press sector is under strain and shrinking, at least two new such enterprises have come to my attention in the past several months. And for all the trumpeted reductions in library budgets, the publishers that sell materials to libraries are reporting modest growth. Now where could that have come from? The answer seems to be that entrepreneurs and intrapreneurs don’t read the headlines.
What we need are not new systems but new services. Services are not top-down comprehensive solutions to all the problems (and some of the merits) of scholarly communications but activities that address specific needs. They usually are conceived by one person, rarely by a committee, and have as their virtue that they come into the world with blinders, never turning their head to the left or right. It is precisely because they do not try to do everything that they are successful. They can be disruptive and unpredictable (who knew that Google would change library discovery services forever?), disruptive and predictable (demand-driven acquisitions), and even something that eventually gets absorbed into the basic infrastructure of publishing and librarianship (Portico and LOCKSS). What all of these things have in common is that they did not set out to change the entire world but to improve one piece of it.
I am itching for new top-down systems to be launched. Every such plan will leave things out, will open up windows where the opportunistic will jump through to their profit. All things grow from the bottom up. Plant a tree in the air and there will be a rush to fill wheelbarrows with dirt to anchor it to the ground. My bet is on the clever and perhaps shortsighted pusher of barrows. It’s a messy lot, running off in several directions and often at odds with itself. Welcome to the real world! Next stop: Bedford Square.
11 Thoughts on "Can We Stop Talking about the “System” of Scholarly Communications, Please?"
While the supply side may not manifest the quality of a system, how about the demand side, viz., the tenure and promotion “system”? That is what drives everything else, after all, and it does seem to be a system that has been long entrenched and is very difficult to change.
I feel the same way about calls for funders to somehow regulate publishing. Scholarly publishing is a system in the sense of having a shared purpose, but it is a highly distributed system, not subject to central control, just like scholarship itself. This is one of the great values.
I think this is why the word “ecosystem” is used so much these days: it suggests something more complex and messy than the word “system” alone does. System suggests something computers do according to clear rules; ecosystem suggests a lot of inter-actors without any central rule book, more socially organized. “Systems thinking” was a big part of organization development back in the 80s and 90s, wasn’t it? Was that just a phase some of us went through? 🙂
I’m working on planning a conference session later this year, and the conference organizer and I are going back and forth on which panelists to invite. I had been thinking about inviting some of the people who are doing the major experiments in an area of scholarly communication; the conference organizer was thinking about inviting people who are representatives of stakeholders. These two approaches represent two different (not conflicting, just different) views of how progress is made: I see progress coming through innovations some of which catch hold, resonate and then diffuse through the (eco)system; the conference organizer might have a view that pressures build up in some areas where there are stakeholders and then change is demanded or driven as a result.
“Once you have a system, of course, it can be improved.” Of course? But “the system” helps some more than others. They end up at the top and this reinforces their positive self-perception. And it is these topsters who become best placed to improve the system. But nothing happens. “I am superior. The system thinks I am superior. Therefore the system is superior. Why change?”
A system does not have to be designed to be a system. The recent brouhaha regarding the fabricated gay marriage survey data has many in the media once again fulminating about how poorly science works. But to my mind, this is a wonderful example of just how well science works! Let’s leave aside that one of the authors appears to have been an intentional a bad actor, and the other duped; the same sort of problem might have obtained from a computer programming error, and any number of other problems could have caused such a result to be obscurely incorrect. (Many have worried about the peer review given the paper. But the folks who discovered the problem actually had to try to replicate it to discover that it was suspect. We certainly don’t expect peer review to go to anything like THAT length, so I doubt that there was any obvious peer review issue here. Hindsight is, as they say, 20:20!) What happened was perfectly normal: An important but erroneous result appeared. Since it was important, someone tried to replicate it. (Another common worry is that there is little replication in science. This is wrong. What there is in science is little that matters enough to bother to replicate. As with everything, there is a cost-benefit trade off; important results are almost always replicated, or cross-checked in some other way.) They failed to replicate. They debugged the problem. End of story. The best quote that I have heard about this sort of thing – I don’t think that it was about this particular case – is attributed to the former editor of Nature, who is said to have been asked how many of the papers in Nature are probably wrong, to which he replied something like: “That’s easy. All of them.” Ours is a system that works, for the most part, yet it was only designed in bits and pieces. The only folks who believe that science pretends to being absolutely correct are priests and journalists.
“And for all the trumpeted reductions in library budgets, the publishers that sell materials to libraries are reporting modest growth.”
Perhaps on the journal side, but not in monograph sales – unless they’re publishing more new titles. This is critical to university presses, some of which are in fact publishing fewer books and launching journals for this very reason. But that’s part of the messy system too –