I have not read the license, but some of the RDBMS
licenses have similar spiel. I think the vendors
request is reasonable, since no one is going to agree
to a standard by which these products can be tested. I
mean look at TPC-x. No one pays attention to these
suckers anymore because all the vendors pull their own
little tricks, no one uses the same hardware, more
than half of them use Tuxedo and the result is that
you have apples, oranges and pomegranates to compare
to each other.
I think it is a reasonable request for the vendor to
ask for some involvement/consultation when performance
numbers/features are going to be publicly compared so
that they can at least feel that their product has a
fair shot. Consider the sham that was the comparison
of NT vs. Linux and it turned out that NT has the heck
tuned out of it and Linux was practically retarded.
This was an alleged "fair and objective" comparison
which NOBODY believes is valid because it was made
amply evident that the idiots never consulted with ANY
Linux experts at all.
Furthermore, it was also determined that Microsoft
paid for the study. Under no circumstances am I
contending that this study was paid for nefariously,
but at the same time, I do not really have any way to
determining that it was not. With vendor involvement
(particulalry in a study of this size and detail) it
would make me more comfortable that all the cards were
on the table.
Which brings me to my last point which is that NONE of
these products is error/defect/problem free, so surely
a critical metric is the vendor's ability to directly
or indirectly address any issue that comes up in the
study. If the vendor is sticky or unhelpful, say so in
the study. If they helped make the product run better,
say so in the study. In my book, support is a feature
just like the latest bitchin' EJB optimizer, so why
evaluate a product without it, any more than you would
evaluate an app server running without all of its best
software components ? Without the support, it is an
incomplete product.
We are having trouble with one vendor (not in the
study) where it will not work with Oracle under some
very ordinary conditions and we cannot get the darn
issue escalated to a point where it gets fixed. So do
I care if it's EJB compliant book, chapter and verse ?
Nope.
In conclusion, in my view, it is the height of egotism
to assume that the definitive best measure of a
product is how well YOU got it run without consulting
the authors of that product on issues that arise.
(breath)
//Nicholas
--- Tim Endres <[EMAIL PROTECTED]> wrote:
> WebSphere,NetDynamics,GemS tone)
> In-Reply-To: <[EMAIL PROTECTED]>
> Mime-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 7bit
> X-Mailer: ICEMail (rel 2.8.4)
> <http://www.ice.com/icemail/>
> Organization: ICE Engineering, Inc.
>
> > Tim Endres wrote:
> > >
> > > I would ignore their license and force them to
> sue me.
> > >
> >
> > This has been said more than once, and strikes me
> > as absolutely wrong.
> >
> > It's perfectly okay for vendors to restrict the
> testing
> > of their product. Rather idiotic, imho, because it
> then
> > leads to the sort of unprovable back-room whispers
> that
> > only benefit the least-honest vendor.
>
> And the argument that this is only to protect the
> vendor
> doesn't cut it. If they wish to protect their
> product,
> they can publish their own "tweaked" results, with
> the
> appropriate disclaimer. Or they can join forces to
> fund
> an objective third party to compare all products on
> an
> equal playing field, as one poster already
> suggested.
>
> > But, nonetheless, a particular vendor may see
> things
> > differently and is perfectly within their rights
> to do
> > so.
> >
> > And, if you sign a license that promises not to do
> > something, well, you ought not to do it.
> >
> > That's pretty straightforward, isn't it ? Entering
> an
> > agreement knowing that you intend to violate it is
> > not entirely ethical behavior.
>
> Depends on whether I believe that the license is
> legally
> founded and will hold up in court. Such a license as
> was
> intimated in this case would not hold up. No more
> than a
> license that said I couldn't post to the Internet
> giving
> my opinion of the product!
>
> tim.
> Tim Endres [EMAIL PROTECTED]
> ICE Engineering, Inc. http://www.trustice.com/
> "USENET - a slow moving self parody." - Peter
> Honeyman
>
>
===========================================================================
> To unsubscribe, send email to [EMAIL PROTECTED]
> and include in the body
> of the message "signoff EJB-INTEREST". For general
> help, send email to
> [EMAIL PROTECTED] and include in the body of the
> message "help".
>
>
=====
"Suzie, Suzie, Suzie, Suzie
Suzie, Suzie Greenberg"
Suzie Greenberg, Phish
--
Nicholas Whitehead
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
