« J2EE Security Presentation available for download | Main | The death of brick-n-Mortar bookstore »

A Failed Experiment

As an experiment, I uploaded few chapters of work-in-progress book J2EE Security for Servlets, EJBs and Web services for review by anyone who agreed to look at the chapters for review and personal use only and was willing to provide a contact email address. Besides announcing it at my weblog, I sent out a mail to an HP internal mailing list. My expectations were modest, but still I thought that I would get some good comments.

More than 50 people registered for review. The interesting thing was that only around five were HP internal. I was quite surprised that remaining came by reading my blog entry. But something was amiss. The number of folks who read the blog entry was much smaller. So who did the promotion?

This mystery got resolved by sheer chance. I was going through my website usage reports and there I noticed that the top search keywords for my site were "j2ee security". I entered j2ee security in my Google bar, and lo -- the third entry was my announcement page, out of a total resultset of 308,000 entries. I must have done something right in my message!!

This explained everything. Surfers were being directed to my chapters by none other than Google, Yahoo and perhaps others.

I sent out an email reminder to all those who had registered after six week of the original post. Four of the email addresses turned about to be non-operational (or with typos). This was quite encouraging (I expected half the folks to give bogus email addresses -- My registration s/w was a free s/w that did not validation but perhaps this was not known to downloaders).

My enthusiam turned out to be short-lived. Comments came from only one person. And this was a colleague !! No luck with random reviewers.

Okay. I learnt my lesson. No more upload for random reviews.

Luckily, the reviewers recruited by my publisher liked the manuscript snd have come up with some great feedback. The editor was excited reading those reviews and so was I. Let us hope the book comes out on schedule and does some business!

Comments (2)

Sheldon Hearn:

While your disillusionment is understandable, consider that you're drawing conclusions from a VERY small sample.

You may have had more registrants without requiring registrations, which still seems to turn a lot of people off. True, you can't nag them for feedback without an email address, but then how much did that help this time?

Hang in there.

Pankaj Kumar:

Anonymity on the Web may have its usefulness, but not for my experiment. The normal practice for book reviews is that first you get a legally binding confidentiality agreement. What I was asking for was a milder consent with just identification. If that bothered someone, they were not my target audience.

For sample size -- I learn't in my statistics class that any sample of size > 30 is large enough to get statistically significant results.

I agree that I may not have promoted it well enough. May be, next time!


This page contains a single entry from the blog posted on April 6, 2003 11:57 PM.

The previous post in this blog was J2EE Security Presentation available for download.

The next post in this blog is The death of brick-n-Mortar bookstore.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.33