Q. Why the name “Textop”?
A. It’s formed from Text Outline Project.
Q. Do you think this project will rival Wikipedia in size and utility? Is that, somehow, the point?
A. It’s definitely not the point of the project. It resembles Wikipedia only in that it will be a collaborative, free reference project. But it is not an encyclopedia, and one of the reasons Wikipedia is as popular as it is is that people want encyclopedic information. Textop will provide an extremely useful compendium of information, but, especially initially, it will probably appeal more to students, teachers, and researchers than the more general public.
Q. Will you focus only on public domain books?
A. At first, yes. Whether and under what conditions we might have access to proprietary works remains to be seen. It does seem likely that, if we succeed in creating an extremely useful resource that allows texts on the precise same issues to be collected all in one place, at least some living authors will be interested in having their work represented as continuous with work that is now in the public domain.
Q. How is it possible that Textop would ever make use of proprietary works?
A. There are at least two ways to do this. First, we might pay publishers for the right to use certain works, and make access to those works via Textop a “premium service.” Second, we might (after carefully considering the practical and legal ramifications of this) make proprietary works themselves accessible only to people actually working on Textop, while the general public sees only the summaries of chunks of the works.
Q. Work has begun only on philosophy. Will you focus only on philosophy?
A. No. We are focusing on a few philosophy books for purposes of the pilot project, only to make the pilot project more tractable. But when the project begins in earnest, we will welcome collation of works in very many different disciplines.
Q. Also, work has begun only in the English language. Is this an English language-only project?
A. No. Again, we are focusing only on books in English for purposes of the pilot project, only to make the pilot project more tractable. When the project begins in earnest, we will welcome collation of works in many different languages. See internationalization.
Q. Why do you have a pilot project going, anyway?
A. There are several different reasons. One central reason is to study the proper requirements of the project before we start off on a larger scale. We would like to show that collaborative work on collating texts is viable, and the results can be coherent and useful. In this way we can both make sure we do not make any strategic mistakes in project planning, and attract future collaborators.
Q. This project seems to involve engaging professionals (especially academics) doing professional-level work. They usually get paid for such work. Yet the project summary maintains that work will be “built by volunteers.” Isn’t that a problem?
A. In short, no–though it is understandable that it looks that way at first.
First, let’s get something on the table that supports but does not prove the point. Academics are constantly giving talks, writing papers and even whole books, without the expectation of significant or even any remuneration.
All this being said, if we find an effective funding system for professionals consistent with our mission, then we will use it to pay them.
Q. Why not just make it a proprietary project? Why insist on making it free? Why not at least run ads?
A. The impact and usefulness of this project would be so much more dramatic if it were available free of charge to all. Textop could have a revolutionary impact on research, education, and decision-making. If it is viewed as anything but fully independent of commercial or other special interests, its impact will be greatly dulled.
Q. But without paying people, can you really expect get people to work on it?
A. You could ask the same question about Wikipedia.
Q. Is there any reason to think that Textop will “take off” exponentially the way Wikipedia did?
A. Yes, there is. As long as the outline’s nodes are spiderable, then Textop can enjoy the same sort of viral development that Wikipedia did. Moreover, once it is better developed, we can fully expect there to be many, many links deeply into the outline from highly-rated academic websites. This will increase the nodes’ Google pageranks. The higher the pagerank, the more traffic; more traffic, more contributors; more contributors, more nodes; more nodes, more traffic.
Q. Why not just use tagging system?
A. Here’s a short answer. Folksonomies, tagging, or bookmarking systems are good for is building communities, talking about web pages, showing other people newly-discovered web pages. But they aren’t very good for search, let alone research.
There could be an increasingly detailed outline of texts, which becomes more useful as it becomes more detailed. Detailed work on categories can be done well only by people with at least some relevant expertise, and then only with great care and difficulty. Devising the appropriate categories is very hard work. It can’t be done on the fly with little thought, a la tagging. Moreover, if you’re creating a unitary reference, you’ve got to have people agree on tags. That means there needs to be a tag repository, and you need to compare related tags–i.e., you need an outline.
Q. Taxonomies are useless. Everybody knows that. Isn’t the idea behind Textop embarrassingly “1990s”?
A. We know everybody says that taxonomies are outdated, but they’re just jumping on a bandwagon. Textop aims to create a different kind of taxonomy, one that will be much more interesting and useful than, say, Yahoo!’s directory. The main reason that taxonomies have come under so much fire are (1) the categories the taxonomists use are not the ones you want to use, and (2) often, they don’t deep-link, so you have to do your own little search once you reach the front page of a website.
Textop will be superior in both respects. First, there will be categories at every level of generality on (almost) every topic that is mentioned in nonfiction books. Most people won’t want any more categories or levels of depth than that. Every way of dividing up the world that you find represented in the paragraph-by-paragraph details of books, you’ll find represented in the outline. Second, the text of books will be extremely deep-linked. It will be like having a detailed, analytical table of contents for all the books in the world.
Q. Wait, we’re not done with this taxonomy business. An outline of human knowledge is highly subjective. So isn’t it highly impractical to expect people to be able to arrive at a single outline collaboratively? For that matter, isn’t a hierarchical representation of knowledge really contrary to the collaborative spirit that dominates online activity today?
A. So long as enough people accept the constraints–namely, the main constraint of compromise–of strongly collaborative work, then there does not seem to be anything very impractical about people working together on a single outline, collaboratively. People have done that, after all, on DMOZ, although with debatable results. So the fact, if it is a fact, that outlining is bound to make all sorts of arbitrary, “subjective” choices hardly means that people cannot make such arbitrary choices together.
As to the notion that an outline of knowledge is contrary to the modern collaborative spirit: that really is a stretch. The criticism presumably rests on the assumption that collaboration requires that, somehow, since the information is arranged hierarchically, so will the people. Of course, that just doesn’t follow. We will have a pretty flat management structure, quite consistent with the requirements of strong collaboration. And actually, isn’t it(if anything) more in the spirit of collaboration that people also have to collaborate on an outline to which they are adding, rather than just using their own idiosyncratic tags?
Q. It’s an interesting idea perhaps, but how feasible is this, really, anyway?
It’s worth pointing out that prototypes already exist. In addition to The Outline as we’re working on it now for the pilot project, Mortimer J. Adler’s Syntopicon was an early, non-digital, more coarse-grained precursor of the Collation Project. Consider a few more points:
- The energy behind various Web 2.0 projects, such as Wikipedia, shows that many, many intelligent people are willing to volunteer to help create a well-designed, well-conceived reference work.
- The value of the outline will become rapidly clear as we add more works. When people “grok” the idea behind the Collation Project (Textop’s flagship project and the one that will be used to build the outline used by the other Textop subprojects), they tend to get quite excited.
- A vigorous, energized group of collaborators on a genuinely free project, as Textop always will be, will almost inevitably attract people willing to create open source software for it.
- Since the output of the project will be many pages loaded with content and interesting titles, we can expect a (viral) “Google effect.”
- Among the earliest participants in the project are a Ph.D. information scientist specializing in ontology, another Ph.D. project manager with much familiarity with the Syntopicon, a manager from Encyclopedia Britannica, and various Ph.D. philosophers. The project also has an impressive Advisory Committee.
- The project director, Dr. Larry Sanger, who co-founded Wikipedia, has said that he is more excited about this project than any–including Wikipedia–that he has ever worked on, and that he would like to make it, along with the larger Digital Universe with which it is affiliated, his life’s work.
This is going to happen. It’s not a matter of whether, it’s a matter of how quickly.