Обсуждение: Your RDBMS Survey ...
A couple of days ago, one of our brethren noticed and pointed us to your
survey asking which RDBMS we prefered/were using ... pride in our choice
prompted alot of us to pop over to your site and register our vote
... when I put mine in, the results were at something like 1461 for PgSQL
and now they are down to 510 ...
one of our brethren has seen it be reset back to 450 at least twice so far
...
I'm curious ... is this survey supposed to have any accuracy to it, or are
the results fixed? *raised eyebrow*
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
On Mon, Dec 04, 2000 at 12:31:06AM -0400, some SMTP stream spewed forth:
>
> A couple of days ago, one of our brethren noticed and pointed us to your
> survey asking which RDBMS we prefered/were using ... pride in our choice
> prompted alot of us to pop over to your site and register our vote
> ... when I put mine in, the results were at something like 1461 for PgSQL
> and now they are down to 510 ...
\begin{Devils Advocate}
Maybe they are attempting to compensate for a spike in the number of
votes in favor of Postgres. Was there a group of votes from a single ip
or some such?
\end{Devils Advoate}
do
{
sleep();
} while(tired())
gh
>
> one of our brethren has seen it be reset back to 450 at least twice so far
> ...
>
> I'm curious ... is this survey supposed to have any accuracy to it, or are
> the results fixed? *raised eyebrow*
>
> Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> Systems Administrator @ hub.org
> primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
>
On Mon, 4 Dec 2000, mycgiserver wrote:
> Hi,
>
> that's an idea, however this approach would fail if 2 clients in an INTRANET
> would like to cast a vote, only the gateway IP would be visible to our
> server and that would cause the second client to be rejected.
>
> See, it ain't that easy.
okay, so instead you are proposing to make the vote only accessible to
users of your system (re: original email response) ... so now, instead of
losing the *occasional* person that is sharing an IP, you lose everyone?
> -----Original Message-----
> From: The Hermit Hacker <scrappy@hub.org>
> To: mycgiserver <general@mycgiserver.com>
> Cc: pgsql-general@postgresql.org <pgsql-general@postgresql.org>
> Date: 4. Dez 2000 14:35
> Subject: Re: Your RDBMS Survey ...
>
>
> >
> >Good point, about the proxy ... we run Squid at work, and you are corect
> >that REMOTE_ADDR will show up the proxy server's IP, but, at least with
> >Squid, a second environment variable will be setup in that case, called
> >'HTTP_X_FORWARDED_FOR', that contains the true IP of the visitor ...
> >
> >A simple:
> >
> >if(HTTP_X_FORWARDED_FOR) {
> > use this IP
> >} else {
> > use REMOTE_ADDR
> >}
> >
> >would work around that limitation/problem ...
>
>
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
This is on their website:
2000-12-04 07:56:51 CET
As some people contacted us what was going on with the current survey (the
PostgreSQL vote significantly decreased from one second to another), we'd like
to inform you, that, after some additional research, this correction from
our side turned out to be unjustified.
The decrease was initially done because of a suspicion of mass-voting for the
item in question. However, the tremendious vote increase for PostgreSQL was
caused by a group of PostgreSQL supporters rather than abusive behavior.
The number of votes is now reset to the all-time high (1524 votes). We're
sorry for any excitement this may have caused!
On Mon, 4 Dec 2000, The Hermit Hacker wrote:
>
> A couple of days ago, one of our brethren noticed and pointed us to your
> survey asking which RDBMS we prefered/were using ... pride in our choice
> prompted alot of us to pop over to your site and register our vote
> ... when I put mine in, the results were at something like 1461 for PgSQL
> and now they are down to 510 ...
>
> one of our brethren has seen it be reset back to 450 at least twice so far
> ...
>
> I'm curious ... is this survey supposed to have any accuracy to it, or are
> the results fixed? *raised eyebrow*
>
> Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> Systems Administrator @ hub.org
> primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
>
>
--
==========================================================================
Vince Vielhaber -- KA8CSH email: vev@michvhf.com http://www.pop4.net
128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================
Good point, about the proxy ... we run Squid at work, and you are corect
that REMOTE_ADDR will show up the proxy server's IP, but, at least with
Squid, a second environment variable will be setup in that case, called
'HTTP_X_FORWARDED_FOR', that contains the true IP of the visitor ...
A simple:
if(HTTP_X_FORWARDED_FOR) {
use this IP
} else {
use REMOTE_ADDR
}
would work around that limitation/problem ...
On Mon, 4 Dec 2000, mycgiserver wrote:
> Hi,
>
> we're storing the voter's IP which gives us at least a vague hint on whether
> a group of votes was casted by a single or multiple visitors.
>
> As for limiting the voting to one per IP, this would be a bad idea as two
> actually different visitors wouldn't be able to vote (one of them at least)
> in the case they're using the same proxy for example although they would
> have the right to as they haven't yet voted.
>
> Thanks for the idea though!
>
> cheers,
>
> mycgiserver team
>
> -----Original Message-----
> From: The Hermit Hacker <scrappy@hub.org>
> To: mycgiserver <general@mycgiserver.com>
> Cc: pgsql-general@postgresql.org <pgsql-general@postgresql.org>
> Date: 4. Dez 2000 14:05
> Subject: Re: Your RDBMS Survey ...
>
>
> >On Mon, 4 Dec 2000, mycgiserver wrote:
> >
> >> Hi,
> >>
> >> well, actually we noticed that some person voted hundreds of times for
> ONE
> >> item (which has to be the PostgreSQL one). That was a single person
> though,
> >> not multiple individuals.
> >
> >Just curious, but how did you determine this? As there is no login
> >information required to complete the survey ... if based off of IP, why
> >not setup a database for logging the votes, including the IP of the person
> >making the vote. Using something like PHP, you could easily save:
> >
> >REMOTE_ADDR <database>
> >
> >and if REMOTE_ADDR already exists, don't add another record for it ... so
> >that now you've restricted voting to be one per IP ...
>
>
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
On Mon, 4 Dec 2000, mycgiserver wrote:
> Hi,
>
> well, actually we noticed that some person voted hundreds of times for ONE
> item (which has to be the PostgreSQL one). That was a single person though,
> not multiple individuals.
Just curious, but how did you determine this? As there is no login
information required to complete the survey ... if based off of IP, why
not setup a database for logging the votes, including the IP of the person
making the vote. Using something like PHP, you could easily save:
REMOTE_ADDR <database>
and if REMOTE_ADDR already exists, don't add another record for it ... so
that now you've restricted voting to be one per IP ...
>
> After noticing that, we removed the this person's votes by removing this
> single group of votes. All other votes are still there, thus the survey is
> still accurate.
>
> This so-called quick survey cannot really be compared to form-based surveys
> in respect to accuracy as the approach of showing the result in advance
> influences the decision of the voter in a significant way, at least in some
> cases. Thus, this survey should primarily present a tendency instead of 100%
> accuracy (which is not possible given the design).
>
> As preventing multiple votes (and that is what caused the problem and
> subsequent vote decrease) is nearly impossible (unless one uses cookies or
> sessions [which would solve the problem partly]) for open surveys, we'll
> most likely move this survey to the user area in order to be able to control
> the accuracy to the full extent. One idea would be presenting the result on
> the front page (as it is currently done) and allowing voting in the user
> area only.
>
> Thanks for bringing this to our attention.
>
> cheers,
>
> mycgiserver team
>
> -----Original Message-----
> From: The Hermit Hacker <scrappy@hub.org>
> To: marketing@mycgiserver.com <marketing@mycgiserver.com>
> Cc: pgsql-general@postgresql.org <pgsql-general@postgresql.org>
> Date: 4. Dez 2000 05:32
> Subject: Your RDBMS Survey ...
>
>
> >
> >A couple of days ago, one of our brethren noticed and pointed us to your
> >survey asking which RDBMS we prefered/were using ... pride in our choice
> >prompted alot of us to pop over to your site and register our vote
> >... when I put mine in, the results were at something like 1461 for PgSQL
> >and now they are down to 510 ...
> >
> >one of our brethren has seen it be reset back to 450 at least twice so far
> >...
> >
> >I'm curious ... is this survey supposed to have any accuracy to it, or are
> >the results fixed? *raised eyebrow*
> >
> >Marc G. Fournier ICQ#7615664 IRC Nick:
> Scrappy
> >Systems Administrator @ hub.org
> >primary: scrappy@hub.org secondary:
> scrappy@{freebsd|postgresql}.org
> >
>
>
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
Woo hoo!! We are back in the lead ...
thanks Vince :)
On Mon, 4 Dec 2000, Vince Vielhaber wrote:
>
> This is on their website:
>
>
> 2000-12-04 07:56:51 CET
>
> As some people contacted us what was going on with the current survey (the
> PostgreSQL vote significantly decreased from one second to another), we'd like
> to inform you, that, after some additional research, this correction from
> our side turned out to be unjustified.
>
> The decrease was initially done because of a suspicion of mass-voting for the
> item in question. However, the tremendious vote increase for PostgreSQL was
> caused by a group of PostgreSQL supporters rather than abusive behavior.
>
> The number of votes is now reset to the all-time high (1524 votes). We're
> sorry for any excitement this may have caused!
>
>
> On Mon, 4 Dec 2000, The Hermit Hacker wrote:
>
> >
> > A couple of days ago, one of our brethren noticed and pointed us to your
> > survey asking which RDBMS we prefered/were using ... pride in our choice
> > prompted alot of us to pop over to your site and register our vote
> > ... when I put mine in, the results were at something like 1461 for PgSQL
> > and now they are down to 510 ...
> >
> > one of our brethren has seen it be reset back to 450 at least twice so far
> > ...
> >
> > I'm curious ... is this survey supposed to have any accuracy to it, or are
> > the results fixed? *raised eyebrow*
> >
> > Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> > Systems Administrator @ hub.org
> > primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
> >
> >
>
> --
> ==========================================================================
> Vince Vielhaber -- KA8CSH email: vev@michvhf.com http://www.pop4.net
> 128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
> Online Campground Directory http://www.camping-usa.com
> Online Giftshop Superstore http://www.cloudninegifts.com
> ==========================================================================
>
>
>
>
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org
I figured as much on this issue (they were trying to compensate for
perceived ballot-box stuffing from the same individual (or IP address,
perhaps).
I'm pretty certain that I couldn't see MY way clear to hosting on
mycgiserver.com, because according to the news item on the front page, they
no longer support Perl. So from their perspective, if my vote didn't get
counted, it's no big deal.
Cheers!
--
David Hancock | dhancock@arinc.com | 410-266-4384
-----Original Message-----
From: GH [mailto:grasshacker@over-yonder.net]
Sent: Monday, December 04, 2000 12:06 AM
Cc: pgsql-general@postgresql.org
Subject: Re: [GENERAL] Your RDBMS Survey ...
On Mon, Dec 04, 2000 at 12:31:06AM -0400, some SMTP stream spewed forth:
>
> A couple of days ago, one of our brethren noticed and pointed us to your
> survey asking which RDBMS we prefered/were using ... pride in our choice
> prompted alot of us to pop over to your site and register our vote
> ... when I put mine in, the results were at something like 1461 for PgSQL
> and now they are down to 510 ...
\begin{Devils Advocate}
Maybe they are attempting to compensate for a spike in the number of
votes in favor of Postgres. Was there a group of votes from a single ip
or some such?
\end{Devils Advoate}
do
{
sleep();
} while(tired())
gh
>
> one of our brethren has seen it be reset back to 450 at least twice so far
> ...
>
> I'm curious ... is this survey supposed to have any accuracy to it, or are
> the results fixed? *raised eyebrow*
>
> Marc G. Fournier ICQ#7615664 IRC Nick:
Scrappy
> Systems Administrator @ hub.org
> primary: scrappy@hub.org secondary:
scrappy@{freebsd|postgresql}.org
>