IMHO it is much more useful to have a more complete set of data, even if it shows shortcomings, than to have a "redacted" set that leaves out some not-so-good areas. marketing is good, but we are not micro$haft, after all. i don't think we have to stoop to that level of "being economical with the truth". jens On Wednesday, Mar 5, 2003, at 08:47 US/Eastern, Paul Winkler wrote:
On Wed, Mar 05, 2003 at 12:14:14AM -0600, J Cameron Cooper wrote:
Although benchmarks are a very small part of the story as far as I'm concerned, there is a need for such numbers. I think I'll try to find time Thursday to do some tests with the usual suspects (at least so far as my work is concerned): Zope, Apache, Tomcat, Jetty, and maybe a little JBoss.
The results will certainly be published.
yay! one request: please don't do anything that skeptics can point to and say e.g. "but zope wasn't serving as much data". there were a number of comparisons on the BwanaZulia page where zope had a slight advantage in terms of bytes delivered. It may seem trivial, but marketing zope vs. java-app-server-X is an uphill battle, so every little bit matters.
--
Paul Winkler