Re: poly: canonical texts

From: Michael Nielsen <>
Date: Tue Jun 15 1999 - 00:13:30 PDT

On Mon, 14 Jun 1999, Robin Hanson wrote:

> At 02:47 AM 6/14/99 -0700, you wrote:
> >What are the canonical texts which define a subject area? Having a
> >structured list of such texts would be of great use to those interested in
> >self-education, and could form part of the core of any good library.
> The way I find this out is to browse through the textbook section of
> a local university bookstore.
> With that in mind, here's a simple
> idea for a web business: Create a place where folks can find out what
> texts big name universities are using for various courses. Maybe call
> it YouU, and market it toward the self-education crowd.

I like your idea. Here's a few critical comments:

> That is, a user types in some keywords, and gets a sorted list of
> related course names, paired with textbook titles. Textbook
> titles are linked directly to a web bookstore where you can buy them,
> and priority on the list might be given to "top" schools.
> Alternatively, the site might collect courses which are "the same"
> under a single label, and give users stats on which books are how
> popular.

A difficulty with popularity ratings is that they tell you how popular a
book is, but not why it is popular. Typically, I am much more interested
in the latter than I am in the former.

One solution to this problem is to offer reviews of books in
the list. However, credibility is a major issue. A major failing of's reader reviews is that they (usually) lack
credibility, as I have little basis on which to judge the author of the

Ideally, brief opinions of a book by fair but opinionated experts will
be available. If the quality of the reviews can be kept uniformly high by
a suitably strong editorial hand, then that will build trust on the part
of users of the site.

A major additional difficulty is getting a global view. One difficulty
with book reviews is that they typically only focus on one or a few books.
Using University Courses as the basis for recommendations may assist in
structuring reading, but there are typically gaps in a University's course
offerings that must be remedied through private reading.

> The income from the site would come from bookstores like
> paying for people forwarded to their site.

Advertising revenues are another possibility.

> And the expenses would
> mainly be finding out what texts are being used in different
> schools. A network of student assistants at the schools, paid in
> textbook discounts, might get you enough info.

Indeed, web searching may provide a substantial amount of info also.

I've been considering some parallel ideas for a website, along a slightly
different track.

Harold Bloom's book "The Western Canon" inspired a lot of thoughts about
how such a site should be structured. TWC is an exceptionally
opinionated book by an intelligent and well-informed critic. Even if one
is not familiar with the works of all the authors Bloom regards as
central to the Canon, one comes away with strong feelings about their

Bloom's work succeeds in part because of its global nature; he
examines the works of not one but many many writers in
considerable depth, drawing useful comparisons between their
works, and giving you some feel for the order in which one ought
to be reading.

I'd certainly love to use an integrated website with competent and
in-depth reviews referencing related material, telling potential readers
what preparatory material they need, what related subjects may be of
interest, and how the book compares with other books of a similar nature.

(Rather like a souped-up University Course catalog, now that I think about

Finally, one last comment: the site should be _small_. I'll quickly read
several thousand books over the next 20 years, I'm sure. However, I will
only have time to read a few hundred books, at most, in depth. A
site claiming to be canonical should contain, in my opinion, no more
material than could be absorbed by a curious individual over their

Michael Nielsen
Ph: 626 395 8431 Email:
Fax: 626 793 9506 Web:
Received on Tue Jun 15 00:12:49 1999

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST