16

By Richard Herley

What e-reading really, really needs right now, what would give it a major boost, is a Web site dedicated to book reviews. As far as I know, there is no central, obvious place where readers can go for recommendations.

My first thought is that this should be based on Joomla! or something similar. Most of the reviews, especially in the beginning, would need to be written and uploaded by account-holding readers. Other readers would then vote, in two ways: on the quality of the book, and on the quality of the review. Reviews should be sorted in order of their ratings, best first.

Books should be classified according to genre, with no limit on the number of genres involved. For example, 1984 by George Orwell might be classified as fiction, science fiction, political fiction, dystopian fiction, British fiction, fiction of the 1940s, novels which have been filmed, novels which have been made into TV dramas, etc., and searches on multiple genres should be allowed.

Despite what I have just said, the notion of “genre” is hopelessly inexact. One can only characterize a book by its place in the swarm of other books.

A book’s place in the swarm may be established by means of association tables. Readers should be asked to submit a list of their favorite books (not just the books in their library, as at LibraryThing). The list can be of any length, as long as there are at least two entries. Order within the list does not matter. Provided the set of lists is large enough, associations between books can be measured by the frequency with which other titles occur together with the “target title”.

Suppose I have just read 1984 and liked it very much. This becomes my “target title”, which I enter in a search dialog. The software returns a table. I imagine it might look something like this:

1984, Orwell – 100%
Animal Farm, Orwell – 83%
The Handmaid’s Tale, Atwood – 75%
Riddley Walker, Hoban – 71%
One Day in the Life of Ivan Denisovich, Solzhenitsyn – 64%
Heart of Darkness, Conrad – 56%
A Clockwork Orange, Burgess – 42%

etc., down to

Bend Sinister, Nabokov – 8%
Keep the Aspidistra Flying, Orwell – 7%
Ulysses, Joyce – 7%
Madame Bovary, Flaubert – 6%
The Time Machine, Wells – 4%

followed by large numbers of titles at 3% and below.

The table tells me the number of lists, expressed as a percentage, in which each of these titles occurs together with 1984. Thus, Animal Farm occurs in 83% of all those lists submitted by other readers which include 1984. The presence of Flaubert, Joyce, Nabokov, etc., suggests that 1984 is a piece of world-class fiction; and that much else in the table is likely to appeal to lovers of good writing.

If I am inclined to read more of Orwell’s work, I can go straight to Animal Farm, noting that Keep the Aspidistra Flying has less immediate appeal, but is still worth investigating.

On the other hand, if I have never heard of Margaret Atwood, I might click on The Handmaid’s Tale, so generating another association table for that book, which would give me a better idea of whether it was for me. If I liked what I saw, I could go to the reviews page for The Handmaid’s Tale, discover that this was also a dystopian novel, read what others think of it, and from there click on “get prices” (or whatever). This would generate a price-comparison table, and links to places where the title can be found.

No book should be excluded, in any language; fiction or not; in print or not; in copyright or not; however published; with an ISBN or not; and whether available only as a p-book, as an e-book, or both. The price-comparison would almost always favour e-books; people without an e-book display might then think more seriously about acquiring one.

Searches which yielded only a p-book could include a radio button with a legend like: “Would you like this title made available as an e-book?” Anonymous data from this button could be made available, for a fee, to publishers, so encouraging them to adopt E.

Further revenue could come from advertising and a commission on all sales originating from the site. This would not only pay the overheads, but allow professional critics to be engaged. Payment could also be made to amateur critics whose reviews were highly rated by readers.

Above all, the reviews should be intelligent, considered, and grammatical. If it is anything at all, the site must be authoritative: the first port-of-call for those needing independent opinion about books. A forum would be a useful addition.

I would like to get involved with such a project. If you would too, especially if you know a lot about computers, e-commerce and content management systems, please email me or add your voice to the voluminous commentary which I hope appears below!

 
16