Mike;

I think we all have seen benchmarks that do not adhere to what would be termed “good practice”. We only have to take a look at Oracle, db2 and SQL Server benchmarks to see one industry example. A bench mark improperly done or looking at one element and not others may be flawed and could cause a serious problem for an entire product line or industry.

Some industries use the National Institute of Science and Technology (NIST) to establish standards. As a Metrology engineer we used the NIST as our representative and emulated the practices they had established. These practices were agreed upon by many international agencies throughout the world. Perhaps the software world should consider something like this as in the physical world it has worked for over 100 years.

Tom


>>So, if the PR hype is that product X will do such and such, but the reality is that it won't do it well, but to relate your experience in a public forum is illegal because of a clause in a EULA. Such clauses violate the 1st Amendment, which the Declaration states to be "unalianable" -- meaning, they can not be alianated, or made illgal by federal, state, local or commerical law.
>
>Yep, I agree. And like I said earlier, this is not new light.
>
>The review part of the license is plain ludicris, I think we both agree there.
>
>The benchmarks, thats a greyer area. I think that benchmarks should be allowed to be published, but... it almost seems to me, that there should be some regulations in that area. Making the published numbers more credible somehow seems like ... hmmm, dunno yet. My jury's still out on that one.