A few thoughts from last Friday’s British Library event. We should embrace Doug’s idea of facilitated evolution, including looking at how scientific papers live and move.
The amount of scientific papers published today is overwhelming.
It’s important to help the scientific and lay community find the right (relevant, useful quality science) papers.
Adding structure is one part of the solution through such means as by assigning more automatic and more user, author and editor meta-tags. Also of value I think will be Structured Lay Summaries (SLS’s), where the papers contains summaries which conform to best practice Q&A’s, with fields such as the expected impact of the work, what it builds on and so on.
How papers are found is shifting and varying by user needs and should be able to continue to shift and over-structuring the information is not only expensive and time consuming (I would say it takes more energy to organize information then to produce it in the first place), it could also hinder innovation in how we find and interact with papers in the future – it is important to support the best ways to ‘navigate’ the scientific paper corpus today whilst also leaving room for future improvements.
One way to approach this is to think of the papers as ‘food’ and readers as foragers. A forager who happens upon a good paper would like to refer to it, reflect upon it and evangelize it to others. In this approach we can think of the forager as an ant who leaves a path to the paper for others to see (with links, ratings, citations and references to it online and offline). In the ant world the path fades over time so that new discoveries are clearly identifiable. Can we learn from the ant and time stamp tags and recommendations in a way that can help other researcher/forager ants?
As opposed to the ant’s environment of food scraps are eaten and disappear, in the research paper ecology papers shift in value and have different value depending on the needs and interests of the researcher – they ‘look’ at the papers differently. Perhaps we can augment their ability to choose how and what to see the papers?
(Before we drop the ants analogy let’s remember what a free environment they have to work in – completely 3D, 360. We have tiny laptop screens…)
Too much structure would overly enforce the structure maker’s view on the rest of the viewers.
All online information is re-presented, there is no ‘original’ or definitive version in a digital world. Even the simple text on a web page is re-presented differently in different browsers on different Operating Systems and so on. You could say that in cyberspace we only see the maps.
Every time we make a map we have to decide what the salient features of the ‘landscape’ are. And this is the crux of the matter, providing the data (the research papers) with rich meta-information (what can be rendered on a map) & flexible ways for the user to interact with the maps, changing how they are shown, what they show and how relationships within and between maps can be seen and manipulated. That’s the key: rich data & rich interactions.
SLS’s need the weight of The British Library in concert with publishers to take on hardness and be used. Tags and other meta information will likely be more fluid and loose in their assignment and use. But this is no bad thing. Over time standard tags emerge and as software gets more powerful, real-time translation of tagging sets will become trivial.
Hyperwords and hyperwords-like systems can then allow you to pole-valult around these points of contact, rearrange them, see them with different ‘filters’, bring transformations and other information to bear and much more.
Give me something to hold (tags, basic structure) and I’ll let you transform your world. 🙂