Re: Joint user survey with EnterpriseDB?

Excuse me for showing my noose again where noone called me:

Josh Berkus escribió:
> Simon,
>
>> The discussion was about whether it was OK to send out things via
>> pgsql-announce, which will have more people, not less people in one
>> month's time.
>
> Yes, I'm thinking this survey low-key, just News and a link in place of
> community/survey at this time. That way we capture some of the increased
> website traffic, but save the real survey blitz for when we have a more
> complete survey we can use several times. This one is a one-off.
>
> Then when we get the full, complete community survey done we can blast
> it through the various e-mail channels and elsewhere. I do think
> getting it done by August 1 is ambitious though, especially if we're
> talking about a July 15 CommitFest.
>
>> Multiple surveys would be strange to say the least. The
>> questions on the current survey have obviously not been well considered,
>> though some of those questions are useful.

The whole idea is great... having one for the release, even if it a 5
question survey... it will give great data to analyze... and to play
with... on the other hand... having a second one... *after* the release
(say about one month), would give *us* the chance to tune up the 2nd
survey, based on the behavior of the first one...

> We'll learn some stuff by seeing what we collect off this survey and how
> people respond to it which will improve future surveys. Release early,
> release often, you know?

My point.

>> I'm disappointed that you asked for opinions and then when you didn't
>> like them, asked for them again.
>
> How many times do I need to repeat this? I got several people who
> approved, and several specific objections to specific things in the
> survey. I changed the survey to address those objections, and asked if
> there were still problems. *just* like I would with a piece of code.
>
>> The subsequent name calling is a great way
>> to have nobody else join the debate on either side.
>
> What name calling?

I join the question...

>> As Rob says, if you don't have clear standards about what is and is not
>> acceptable then it will, possibly has, eroded away to nothing. Nobody
>> has the time for strong and lengthy debates on minor points. But where
>> do we draw the line? We just continually step backwards in small enough
>> steps that no single step is worthy of debate.

Not when debates looks pointless (Personal opinion)... if i'm not lost
on the olap's hyper (cubic) space... the last part of this debate, is
all about the *¿rush?* of having a survey as soon as the 8.4 release
comes out, right?... from what i see, and IME, sometimes is better to
have a small sample, and develop something bigger... by doing some
tuning based on the content/results/information that sample dropped...
(If you want to get useful information off a target... you *have* to
experiment a little bit... and i belive... this is the moment for that...)

> When did doing a survey become a moral point? It's a survey. Don't
> make it more than it is.
>
>>> However, to allieviate some of your concerns, JD has duplicated the
>>> survey on postgresql.us so that we can have the survey on a site we
>>> control.
>>
>> Yes, I think the reason for JD's change of heart is clear.
>
> So where the survey is hosted is not a consideration for you? If that's
> the case, I'd just as happily use SurveyMethods. They have a nice console.
>