Robert Asprin - Wikipedia, the free encyclopedia
[Ann Arbor's own] Robert Asprin passed away [at 62] 22 May 2008, dying quietly in bed where he had been reading a Terry Pratchett novel.

Asprin was important to me because he was a very successful SF writer from my home town.  I enjoyed the first few books in the THIEVES WORLD and MYTH series. Too bad his life and career was cut short.

Please check back later as new posts are added regularly. To discover more authors, visit our Author Blog Directory

Official Gmail Blog: A need for speed: the path to a faster loading sequence
We want Gmail to be really fast, and we keep working on ways to make it faster. Gmails architecture eliminates many of the delays in reading mail by employing techniques like prefetching, but recently we decided to take a close look at some other key parts of Gmail to see if we could speed things up.

So much so that I recently went into the settings and changed maximum threads per page from 50 to 25. I also tried the basic html version. It is too frustrating.

Gmail team: why not give us a clock?

When I was at LexisNexis, I was product manager for the web version of Lexis, and we spent a tremendous amount of time on performance improvement.  As the gmail team’s post suggests, it is very hard work.It’s hard to tell what’s even happening without extensive experimentation and instrumentation, and then when you make changes, there are always people who see zero or negative improvement. I feel for the gmail team, but I have to be honest — gmail really does seem slower to me.

I wonder if gmail has advertently or inadvertently taken a step away from managed performance.  When I was at LexisNexis, we were absolutelly appalled to learn that the backend search engine had a “governor” on it which slowed down any search that came back too fast.  The idea was to manage customer experience  by making every search take the same amount of time, no matter whether it was busy time or not.    Perception is reality: it’s important whether response time is consistent.

• Google Analytics page views 1/1/2008 - date: 26032
• Associates clicks 1303
• Associates unit sales 14
• conversion rate 0.0107
• Associates referral revenue $10.04
• revenue/month $2.26
• NB after royalties revenue for referred titles 40
• Total imputed revenue (referrals + publisher comp)= 50.04
• Imputed revenue/month $11.29
• imputed $/page view $0.0019
• imputed $/1000 page views $1.9222
• imputed $/click $0.0384
• imputed $/conversion $3.5743
• bandwidth used (GB) 56.259
• GB/month 12.69
• max GB/month 200
• days covered 133
• months 4.433
• Imputed revenue/GB $0.89
• overage charge per GB $1.99
• Cost of High Volume Plans per GB $0.19
• possible ^ in usage w/o incurring additional bandwidth charge 15.76044129
• possible ^ imputed revenue by ^ usage x15 w/o additional BW charge =15 x imputed rev /month = $177.89 /month
• possible ^ imputed revenue by ^ conversion rate x 5 = $889.46

what all this tells me:

* I could increase the usage of my website by a factor of 15.7 before I hit the bandwidth threshold for my account at Pair
* assuming that revenue increased in linear proportion, and that I could improve the conversion rate, I could increase my Associates & Associates-driven revenue into the several hundreds of dollars per month
* increasing usage by x 15 would be hard work, there are other things I could do that would increase revenue more (i.e. publish more or better books)

My website has always been a bit of a mess because I use it as a combined blog and bookstore. It would be better to have a single purpose store. Right now the website is under construction as I experiment with new Amazon Associates widgets.

The first chapter of this book, which describes the inevitable death of a father and his eight-year-old son at the hands of a hunter-killer robot, is as perfect as everything I have ever read in fiction.