a book proposal by Philip Greenspun author of Database Backed Web Sites
Despite Macmillan's abandonment of the title, it became a best seller at amazon.com, making their "Amazon 500" list (subsequently eliminated), though not quite popular enough to make their "Computer 50" list. Note the wealth of reader reviews at the amazon site.
I nagged Macmillan to either promote the book like they promised to or give it back to me (ergo, we can reuse any material from this book in a new book). They decided to give it back to me. I think this is because they'd rather focus on the "books about desktop apps" market. The people within Macmillan who've read my book like it.
Reaching this market should not require rocket scientists or huge budgets. My personal Web site gets 18,000 visitors/day (and 500,000 hits/day). We have great reviews and blurbs from all kinds of readers, some of them with big executive titles (e.g., V-P Internet IBM or Chairman, Perot Systems; Edward Tufte loves my site and apparently says so when he gives seminars). I have personal relationships with executives at Hewlett-Packard, Oracle, and other large companies that could buy stacks of the book for training or promotion (HP bought a few hundred copies of the first book and would have bought more if it had more FlashPix content, which this new book will). HP has asked me to team with them and Cambridge Technology Group in giving talks every month to 50 or 100 managers from big companies. This book could be a text for the "building image-rich Internet servers" course.
As far as traditional publicity goes, Macmillan never arranged a book signing or a magazine review or anything. However, some aspect of my site gets featured at least once/week in the traditional media. I spent a whole hour being interviewed by the NPR station in Alaska at the end of October; they were specifically interested in the book and my thoughts on Web publishing. A week later, a station in Detroit had me on to talk about the Bill Gates Personal Wealth Clock. I of course mentioned that I wrote a book about how and why I built this and that folks should read Chapter 9. The Wall Street Journal featured photo.net as one of its three favorite photo sites on the Internet (though they dissed as "plain" the same user interface design that was lauded by Tufte). For what it is worth, my site has been written up in the New York Times (several occasions), the International Herald Tribune, on CNN, and on lots of radio stations.
My personal belief is that there is a big untapped market for my book among smart people. Most computer publishers try to sell books to stupid people. I guess the theory is that there are more stupid people than smart people. Or maybe that the average product quality of the computer trade press is now so low that only stupid people are willing to enter the computer section of a bookstore. I think my book could appeal to smart people because it respects the reader by (1) not taking too many pages to cover its topics, (2) not being filled with user interface widgets, (3) not giving contrived examples but only real-life examples from sites with 20 hits/second, (4) looking at problems from a high level first before getting down to source code or step-by-step instructions.
A couple of guys here at MIT had big hits with their computer books. One is Negroponte with Being Digital. The other is Dertouzos with What Will Be. At some level you could say that these were targeted to stupid people because they didn't say anything technical and/or anything that would surprise the average AOL user. But if you went into a bookstore, you found that these books were out front with "real books" rather than in the back with Learn to be an Unleashed Dummy in 21 Days. I guess I'd like to break out of the computer trade book ghetto though I'm not sure how. Tufte is my hero. He's produced a sort of practical book that nerds (including me) like. But most bookstores stock it near books that literate people might want to read. I want to be like Tufte, although I don't want to self-publish!
This is a long-winded way of saying that I think my book might do well at conferences and other places where smart people gather. I know that academic publishers spend a lot of time, effort, and money schlepping their product to conferences. This ought to be a similar audience to that trooping through the MIT Press Bookstore where people who saw my book out front ended up buying it.
I will insert more tutorial threads in this book, so that people can't say "this is just background; it doesn't teach anything specific" (the first book actually does, btw, but a casual leaf-through won't reveal that Chapter 11 is a step-by-step SQL tutorial). For some of these threads, I'd like to work in popular product names, e.g., "step-by-step planning for a new static Web site", "making Oracle ConText work", or "building a comprehensive db-backed magazine publishing system."
I've been thinking a lot about what keeps "the average person" from developing a Web site. It has nothing to do with the mechanics of writing HTML. It has everything to do with thinking formally and understanding how to use the file system intelligently. I've put some of this down in writing in my site development tutorial, which I'll be expanding significantly.
I've been thinking a lot about what are the appropriate software products for Web publishers to buy and/or build. I've concluded that a publisher who even has an opinion about operating system, RDBMS brand, or Web server program has already failed. What matters are the data models and the apps. The important thing is to buy or build "SAP for the Web". I wrote about this in the future of the Web software industry and I will expand on the article dramatically.
My photo.net Q&A forum has become much more popular and this has gotten me thinking about on-line community. I'm going to have a whole chapter on different kinds of forums and moderation policies and what these mean. I'm going to put in the source code that implements these different policies.
XML addresses many of the complaints about HTML that I raised in 1994 and raised again in the first edition. But it isn't a solution by itself. In fact, for publishers it means that they have to make a whole new round of publishing decisions and technology choices.
I spent five months thinking about image databases and image presentation for HP, one of the authors of the FlashPix format. I've rethought my approach to presenting images and I have a bunch of new ideas and software to share with the world. Chapter 4 was good and very tutorial and very specific. But it could be a lot better and, more importantly, HP would probably consider buying thousands of copies of a book with a good treatment of its pet technologies.
I have 6000 photos scanned to Kodak PhotoCD imagepacs. There are several programs specifically designed for automatic conversion of .pcd to CMYK. We should figure out how to use one of these effectively so that the cost of color separations for the 10th or 100th photo is practically nil.
It would be nice to have photos/graphics on the front and back that will appear familiar to people who've seen my Web site either live, on television, or in print media. My click server cookied out about 600,000 unique browsers in 10 months during 1997. So I think it is worth considering simply putting the Elsa Dorfman photo of me and Alex from http://www.photo.net/philg/ on the cover. Note that this photo by itself will give us an excuse to put in a little paragraph; Elsa is pretty famous and has one of the world's 5 Polaroid 20x24" cameras. Also note that Elsa used this photo in the December 1, 1997 New Yorker magazine to advertise her services. So that means it will have some subconscious familiarity for New Yorker subscribers.
Alternative titles: "Philip and Alex explain Web publishing"; "Philip and Alex's Thinking Person's Guide to Web Publishing" (I like these because they are informal).
Here are some other advantages of putting a photo on every page: (1) we are better able to attract people who like my Web site for the photography; (2) readers will be much more likely to leave the book out on their coffee tables and/or drag it out to show someone a photo (thus giving them the opportunity to recommend it).
But if you insist on something traditional, here's what I give out to journalists:
Philip Greenspun won Best of the Web '94 for Travels with Samantha, his 210-page story about a trip from Boston to Alaska illustrated with 300 photographs. Those were the good old days for Greenspun. Now he can't take three and a half months off to drive around because he is chained to his desk by his Web fame.Greenspun was the architect of the database-backed Web services for Hearst Corporation, a $3 billion publisher. Then Macmillan persuaded him to write a book on everything he'd learned from building about 50 Web sites. You'll find it at your local Borders or at amazon.com wearing the rather ungainly title "Database Backed Web Sites" but really the book is almost entirely geared to a nontechnical audience.
Greenspun's personal site has a lot of bizarre facets. The greatest fame-to-effort ratio was achieved by his Bill Gates Personal Wealth Clock. More people probably know him because of photo.net, his photography page that has grown to the point of requiring a 3000 pound Hewlett-Packard Unix box to serve all the requests.
The Web boom hasn't changed certain things about Philip Greenspun. He's still at the same email address he's had since 1976 (philg@mit.edu). He is still doing research at the Massachusetts Institute of Technology, which he describes as "regarded by people worldwide as the finest engineering school in East Cambridge."