Thread: http://www.pgsql.com/register/submit.php
I've always been uneasy about the notice at the end of the installation process that asks the user to register at http://www.pgsql.com/register/submit.php Now some users have also complained to me about that, so I'm wondering what others think about it. First, I think collecting addresses from users is something that an open-source project should never do. It's what makes them stand out from packages with annoying license requirements. Second, do we really get anything useful out of those registrations? Maybe in the early days, anyone who was insane enough to try PostgreSQL was worth rejoicing about, but now I think we get a pretty good overview over the uses in other ways. Third, the method is flawed, because only users that install source packages manually are invited to register. I believe that the majority of users use prepackaged versions. Fourth, the registration is run not by the project but by a commercial company. That seems inappropriate for all the reasons you can imagine. Comments? -- Peter Eisentraut peter_e@gmx.net
> -----Original Message----- > From: Peter Eisentraut [mailto:peter_e@gmx.net] > Sent: 12 October 2003 14:02 > To: PostgreSQL Development > Subject: [HACKERS] http://www.pgsql.com/register/submit.php > > I've always been uneasy about the notice at the end of the > installation process that asks the user to register at > > http://www.pgsql.com/register/submit.php > > Now some users have also complained to me about that, so I'm > wondering what others think about it. > > First, I think collecting addresses from users is something > that an open-source project should never do. It's what makes > them stand out from packages with annoying license requirements. Agreed. > Second, do we really get anything useful out of those registrations? > Maybe in the early days, anyone who was insane enough to try > PostgreSQL was worth rejoicing about, but now I think we get > a pretty good overview over the uses in other ways. No, even I can't get to the data. See your fourth reason. > Third, the method is flawed, because only users that install > source packages manually are invited to register. I believe > that the majority of users use prepackaged versions. Agreed. > Fourth, the registration is run not by the project but by a > commercial company. That seems inappropriate for all the > reasons you can imagine. Agreed. > Comments? I think it should be removed. Regards, Dave.
On Sun, Oct 12, 2003 at 03:01:39PM +0200, Peter Eisentraut wrote: > First, I think collecting addresses from users is something that an > open-source project should never do. It's what makes them stand out from > packages with annoying license requirements. Couldn't agree more. > Third, the method is flawed, because only users that install source > packages manually are invited to register. I believe that the majority of > users use prepackaged versions. Even being a developer does not prevent me from using prepackaged version for some or even most needs I have. So, yes, I agree completely here too. > Fourth, the registration is run not by the project but by a commercial > company. That seems inappropriate for all the reasons you can imagine. And another "me too". I'd say let's remove it unless someone has some good reasoning. Michael -- Michael Meskes Email: Michael at Fam-Meskes dot De ICQ: 179140304, AIM/Yahoo: michaelmeskes, Jabber: meskes@jabber.org Go SF 49ers! Go Rhein Fire! Use Debian GNU/Linux! Use PostgreSQL!
Peter Eisentraut <peter_e@gmx.net> writes: > I've always been uneasy about the notice at the end of the installation > process that asks the user to register at > http://www.pgsql.com/register/submit.php There was discussion of rewording that message some time ago, but nothing happened. (I think Vince volunteered to prepare a new draft, and then disappeared...) I agree that there's no need to ask people to register. I think it'd be okay to have a message that suggests visiting www.postgresql.org. Want to propose some new text? regards, tom lane
On Sun, 12 Oct 2003, Peter Eisentraut wrote: > Fourth, the registration is run not by the project but by a commercial > company. That seems inappropriate for all the reasons you can imagine. To be honest, I don't know why it was every put on pgsq.com in the first place, and think the whole thing should be re-written ... the original purpose was to get an idea of who was using what version, as well as to get a feel for the size of dataases ppl are using ... but, there is no way of changing ones data after the fact, so, of course, when you upgrade you can't reflect that ... Personally, I think the work on case studies going on right now are of more value ... at they provide more info, and more an understanding of why ...
Tom Lane writes: > I agree that there's no need to ask people to register. I think it'd > be okay to have a message that suggests visiting www.postgresql.org. The text already suggests www.postgresql.org, so there's nothing to do besides removing the register bit, if we agree to do so. However, I think the message is a bit pointless altogether. First, only those that install from source see it, so it's not the place to communicate vital information. Second, how did they ever get the source code if not via www.postgresql.org? Third, that same piece of information is contained in README and INSTALL, so it's impossible to miss. So the only group of people we could target with this text is those that 1) install random source packages that they got from somewhere they don't remember, 2) don't read README and INSTALL files even if they have questions, and 3) don't know how to use a search engine, if they really have trouble finding us. -- Peter Eisentraut peter_e@gmx.net
Peter Eisentraut <peter_e@gmx.net> writes:
> However, I think the message is a bit pointless altogether.
You have a point. Anyone against ripping it out entirely? If so,
what needs to be said in this message?
regards, tom lane
On Mon, 2003-10-13 at 00:23, Tom Lane wrote:
> Peter Eisentraut <peter_e@gmx.net> writes:
> > However, I think the message is a bit pointless altogether.
>
> You have a point. Anyone against ripping it out entirely? If so,
> what needs to be said in this message?
>
Someone could have downloaded the package from sourceforge/freshmeat, so
I don't think it hurts to have a pointer to postgresql.org. What might
be better than a link for the submit page would just be a blurb
mentioning we're always looking for new case studies from folks who are
using postgresql, and perhaps point them some place for that. What I'd
like to see added to the message is a reminder to run initdb...
Robert Treat
--
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
Robert Treat writes: > Someone could have downloaded the package from sourceforge/freshmeat, so > I don't think it hurts to have a pointer to postgresql.org. Those two places already have prominent links to the home page. > What might > be better than a link for the submit page would just be a blurb > mentioning we're always looking for new case studies from folks who are > using postgresql, and perhaps point them some place for that. See my last post: This is not the place to print any project-related messages. You're not going to get a representative audience. > What I'd > like to see added to the message is a reminder to run initdb... Anyone who is completely clueless about PostgreSQL will have to read INSTALL anyway, which gives more complete information. It's a bit more involved than just entering "initdb": you need to set up users and directory permissions. I don't think duplicating that information would be useful. -- Peter Eisentraut peter_e@gmx.net
Le Lundi 13 Octobre 2003 14:45, Robert Treat a écrit : > Someone could have downloaded the package from sourceforge/freshmeat, so > I don't think it hurts to have a pointer to postgresql.org. What might > be better than a link for the submit page would just be a blurb > mentioning we're always looking for new case studies from folks who are > using postgresql, and perhaps point them some place for that. What I'd > like to see added to the message is a reminder to run initdb... In France, according to the "Loi informatique et liberté", users have a right to access their personal data. As a result, every web form must display a warning which reminds users that they can access their personal data. This is not a potential problem for PostgreSQL.inc, but for Linux vendors who release PostgreSQL (RedHat, Mandrake, etc...) in France. The fine can be quite large. Also, a judge may stop immediately the release of products including PostgreSQL. Cheers, Jean-Michel
Tom Lane wrote: > Peter Eisentraut <peter_e@gmx.net> writes: > > However, I think the message is a bit pointless altogether. > > You have a point. Anyone against ripping it out entirely? If so, > what needs to be said in this message? Let's remove it. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001+ If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania19073
Jean-Michel POURE kirjutas E, 13.10.2003 kell 16:20: > Le Lundi 13 Octobre 2003 14:45, Robert Treat a écrit : > > Someone could have downloaded the package from sourceforge/freshmeat, so > > I don't think it hurts to have a pointer to postgresql.org. What might > > be better than a link for the submit page would just be a blurb > > mentioning we're always looking for new case studies from folks who are > > using postgresql, and perhaps point them some place for that. What I'd > > like to see added to the message is a reminder to run initdb... > > In France, according to the "Loi informatique et liberté", users have a right > to access their personal data. As a result, every web form must display a > warning which reminds users that they can access their personal data. Even if no personal data is entered or queried ? > This is not a potential problem for PostgreSQL.inc, but for Linux vendors who > release PostgreSQL (RedHat, Mandrake, etc...) in France. The fine can be > quite large. Also, a judge may stop immediately the release of products > including PostgreSQL. Do we need special NOTICE for French users telling them "In case you have entered some personal data you can access it by doing a SELECT query" after each INSERT/UPDATE to be legally sound in France ? This could then be automatically enabled by LC_ALL=fr . ---------------- Hannu
> -----Original Message----- > From: Hannu Krosing [mailto:hannu@tm.ee] > Sent: 13 October 2003 18:00 > To: jm@poure.com > Cc: PostgreSQL Development > Subject: Re: [HACKERS] http://www.pgsql.com/register/submit.php > > Do we need special NOTICE for French users telling them "In > case you have entered some personal data you can access it by > doing a SELECT query" after each INSERT/UPDATE to be legally > sound in France ? No more so than would be required under the Data Protection Act in the UK, but as the project and pgsql.com servers are neither in France or the UK, the content on them doesn't have to comply with those laws - though in the interests of privacy and general good practice we should probably try to do so. Of course, the mirrors are a different issue altogether. Currently I guess it's up to the mirror operators to let us know if anything doesn't comply with their local laws. Regards, Dave.
Le Lundi 13 Octobre 2003 21:28, Dave Page a écrit : > No more so than would be required under the Data Protection Act in the > UK, but as the project and pgsql.com servers are neither in France or > the UK, the content on them doesn't have to comply with those laws. RedHat is a French company in France (as well as Mandrake). The law does not care for the location of servers, mirrors, etc... If a French company asks to register, it shall obey the laws. PostgreSQL inc. is not at stake here. > though in the interests of privacy and general good practice we should > probably try to do so. Agreed. Good practice is the beginning of democracy. Cheers, Jean-Michel
Bruce Momjian writes: > Let's remove it. It's gone. -- Peter Eisentraut peter_e@gmx.net