Lists Home |
Date Index |
This isn't new news. Both are done. One
gathers opinions and contracts. Remember,
contracting is a formal process because of
protests. If one does things promiscuously,
the bid results are protested. Protests are
expensive for the contracting parties and
can result in rebids. Now all of that money
spent on process is lost and one has to start
over in some cases. In any case, the only
benefits go to the lawyers.
Most contract-based processes do have a
provision for customer references. What
one wants to do is get a more complete
list than the cheery cherries that the
vendor will suggest (so called "reference
If I want to hunt down a reliable vendor, I will check
their references, but when it comes down
to the procurement, it has to be based on
the assertions of the contracting parties.
In many cases, discovery doesn't work like
either of these cases. One goes to conferences,
shows, etc, inspects product, interviews, etc.
Then an RFP is published. What Google is good
for is finding published RFPs and RFIs.
Yes, ranking should be based on multiple criteria,
in other words, question the sources. I think
that what Google does is useful if worked alongside
other sources and multiple means to rank. All
opinions and all searches are not equal. For example,
all opinions are not "small". To sell a system
in the business we are in here, one has to prove
that one has sold systems where the customer is
of a given size, has a given rate of incidents,
and so on. So to Google that and it be contract
worthy, Google would have to keep a lot of metadata
about the business and customer types. This starts
being more like UDDI.
From: Elliotte Rusty Harold [mailto:firstname.lastname@example.org]
At 10:21 AM -0500 10/10/02, Bullard, Claude L (Len) wrote:
>Thanks for making the point. Worse is worse,
>but better needs requirements to insure it is better.
>Better than what is an operative question.
>Do we want Google to "evolve" into a business
>registry where the opinions of the competitors
>determine the ranking instead of the registered
>assertions of the vendor so registered?
I give much more weight to what a business registry where the
opinions of the *customers* determine the ranking of a vendor than
one that depends on the registered assertions of the vendor so
registered. That, in effect, is what Google does today for the Web.
The more happy customers a site has, the higher it ranks. Google's
algorithm for determining customer happiness is heavily based on
linking, not perfect, but the best we've got so far. I trust this
much more than search engines that sell placement, contract or no.
No, this isn't exactly what UDDI does, but I still think there's a
lesson to be learned here about the emergent behavior of gathering
many small opinions vs. trying to rely on a few centralized systems,
experts, and contracts.