Clinging to sanity, pgsql@mohawksoft.com ("Mark Woodward") mumbled into her beard:
>> Mark,
>>
>> First off, I'm going to request that you (and other people) stop
>> hijacking Simon's thread on hypothetical indexes. Hijacking
>> threads is an effective way to get your ideas rejected out of hand,
>> just because the people whose thread you hijacked are angry with
>> you.
>>
>> So please observe the thread split, thanks.
>>
>>> Well, if it would get rejected if it looked like Oracle, assuming
>>> you would probably be one of the people rejecting it, what do you
>>> envision as not being rejected?
>>
>> Something "better than Oracle".
>>
>> Since you're the one who wants hints, that's kind of up to you to
>> define. Write a specification and make a proposal.
>
> What is the point of writing a proposal if there is a threat of
> "will be rejected" if one of the people who would do the rejection
> doesn't at least outline what would be acceptable?
If your proposal is merely "let's do something like Oracle," it should
be obvious why that would be rejected. There is considerable legal
danger to slavish emulation. Further, since PostgreSQL isn't Oracle,
slavish emulation wouldn't work anyways.
If a proposal is too fuzzy to be considered a source of a
specification, it should be obvious that that would be rejected.
If you have an idea clear enough to turn into a meaningful proposal,
put it in for the usual "to and fro"; that generally leads to enormous
improvements.
I'm not sure what a good hinting system ought to look like; what I
*do* know is that a fuzzy proposal won't be much good.
--
(format nil "~S@~S" "cbbrowne" "acm.org")
http://linuxfinances.info/info/postgresql.html
The quickest way to a man's heart is through his chest, with an axe.