Thread: [Fwd: Questions on 7.3 version for LinuxWorld]
Hi everyone, Nadia Cameron, a journalist with IDG, is looking to do a brief story on PostgreSQL and has a couple of questions for us. So, being newly motivated to set up the collaborative area for the Advocacy site, it's now ready for general usage (http://advocacy.postgresql.org/documents/) Nadia's questions (from below) have been added into a page in the new collaborative area, so people can make notes/comments/etc directly, and we should be able to get the results to her by about this time tomorrow if it's as effective as I'm hoping... http://advocacy.postgresql.org/documents/NadiaCameron :-) Regards and best wishes, Justin Clift -------- Original Message -------- Subject: Questions on 7.3 version for LinuxWorld Date: Mon, 9 Dec 2002 12:04:10 +1100 From: Nadia_Cameron@idg.com.au To: justin@postgresql.org Hi Justin, Just trying to put together a brief story on the release of PostgreSQL version 7.3, but had a couple of questions I'd like to put forth to you (bear with me, I'm not very techie!) - What is the 92 scheme specification, and why has support for it been included in the latest version of postgreSQL? - In the release, Neil Conway comments that the new version contains a dependency tracking systems "that allows PostgreSQK to 'safely' support many more subtle enhancements like the ability to drop columns". I'm unclear on the use of "safely" here - would you be able to elaborate on this comment further? - What are the conditions of using PostgreSQL? - How long did it take to develop the latest 7.3 release? - When do you expect to release the next version of the system? thanks Nadia Cameron Journalist Online division IDG Communications 88 Christie st St Leonards NSW 2060 Phone: (02) 9902 2782 -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Sun, 2002-12-08 at 22:52, Justin Clift wrote: > Nadia's questions (from below) have been added into a page in the new > collaborative area, so people can make notes/comments/etc directly, and > we should be able to get the results to her by about this time tomorrow > if it's as effective as I'm hoping... I was going to add comments to that page, but since this is just my opinion, it's probably easier to put it out for discussion on this list first. Also, I wasn't sure if you wanted a "quotation and response" to the original email, or an entirely new response... > - What is the 92 scheme specification, and why has support for it been > included in the latest version of postgreSQL? The "92 scheme specification" you refer to is the SQL92 standard, which specifies the query language used by most major relational databases. Schema support is part of this standard. Implementing more features of the standard makes it easier for PostgreSQL to interoperate with other standards-compliant databases. [ ... anyone care to contribute a blurb on why schemas are great? ] > - What are the conditions of using PostgreSQL? PostgreSQL is free to use, modify or distribute for any purpose, provided the copyright notice is included. In other words, it's licensed under the standard "BSD" terms. > - How long did it take to develop the latest 7.3 release? By my count (judging from the release of 7.2.0 until 7.3.0), about 10 months. > - When do you expect to release the next version of the system? Since we haven't discussed a release date on -hackers (or even have a firm idea of what we'd like to see in 7.4, and when those features are going to be implemented), I don't think we should commit to a release date with the media. Cheers, Neil -- Neil Conway <neilc@samurai.com> || PGP Key ID: DB3C29FC
Neil Conway wrote: > On Sun, 2002-12-08 at 22:52, Justin Clift wrote: > >>Nadia's questions (from below) have been added into a page in the new >>collaborative area, so people can make notes/comments/etc directly, and >>we should be able to get the results to her by about this time tomorrow >>if it's as effective as I'm hoping... > > > I was going to add comments to that page, but since this is just my > opinion, it's probably easier to put it out for discussion on this list > first. Also, I wasn't sure if you wanted a "quotation and response" to > the original email, or an entirely new response... Have phoned her to find out if the 24 hours delay is cool, and she's ok about it. Before making it clear that the reason for the call was to ask if a 24 hour delay is ok (and also telling her why... "it's a good "quick" trial project for the new Advocacy and Marketing group and the collaboration software"), she grabbed her notes and started reading the questions off to me. Gut feeling says that she's asking these questions because she's not sure what else to ask. So we should probably answer them directly, but also it might be a good idea to give her some other stuff so she can write something that sounds interesting for the readers. ;) The readers may be reasonably technical, as it's for the linuxworld.com.au site, but she isn't. >>- What is the 92 scheme specification, and why has support for it been >>included in the latest version of postgreSQL? > > > The "92 scheme specification" you refer to is the SQL92 standard, which > specifies the query language used by most major relational databases. > Schema support is part of this standard. Implementing more features of > the standard makes it easier for PostgreSQL to interoperate with other > standards-compliant databases. Maybe a bit too formal? How about: The "92 scheme specification" is the American National Standards Organisation (ANSI) SQL92 standard, which specifies the SQL query language used by most major relational databases (PostgreSQL, Oracle, DB2, Sybase, etc). Schema support is one of the parts of this standard, making it easier for [blurb], and to migrate from other (often commercial) databases to PostgreSQL. PostgreSQL has actively been adding ANSI SQL features for many years and is now one of the most ANSI SQL compliant databases in existence. [is this true?] This makes it easier for PostgreSQL to interoperate with other ANSI SQL compliant databases. > [ ... anyone care to contribute a blurb on why schemas are great? ] > >>- What are the conditions of using PostgreSQL? > > > PostgreSQL is free to use, modify or distribute for any purpose, > provided the copyright notice is included. In other words, it's licensed > under the standard "BSD" terms. Needs a bit more explanation, really spelling it out for people. i.e. "There are no license fees of any kind for PostgreSQL, and you are free to modify it however you want, and don't even have to share your modifications with anyone. The only thing you have to do is make sure the BSD license and copyright notice are included with any versions you do distribute." >>- How long did it take to develop the latest 7.3 release? > > > By my count (judging from the release of 7.2.0 until 7.3.0), about 10 > months. Cool. :) >>- When do you expect to release the next version of the system? > > > Since we haven't discussed a release date on -hackers (or even have a > firm idea of what we'd like to see in 7.4, and when those features are > going to be implemented), I don't think we should commit to a release > date with the media. Maybe we should use this opportunity to build anticipation? Something like "Because we'll be adding native support for Windows, Point in Time recovery, and other advanced features in PostgreSQL 7.4, our initial target date is around April 2003". Does this sounds like an ok approach? :-) Regards and best wishes, Justin Clift > Cheers, > > Neil -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Sun, 2002-12-08 at 23:54, Justin Clift wrote: > The "92 scheme specification" is the American National Standards > Organisation (ANSI) SQL92 standard, which specifies the SQL query > language used by most major relational databases (PostgreSQL, Oracle, > DB2, Sybase, etc). "SQL query language" is redundant. Not sure why there's a need to mention ANSI, as it's not as if there are competing SQL92 standards :-) And name-dropping standards bodies is a bit pretentious, IMHO. The "92 scheme specification" isn't SQL92 (as your first sentence implies), it's a part of SQL92. > PostgreSQL has actively been adding ANSI SQL features for many years > and is now one of the most ANSI SQL compliant databases in existence. > [is this true?] Sounds pretty dubious to me. > Maybe we should use this opportunity to build anticipation? Something > like "Because we'll be adding native support for Windows, Point in Time > recovery, and other advanced features in PostgreSQL 7.4, our initial > target date is around April 2003". I still think mentioning a release date, even as an"initial target", is not appropriate, particularly since it hasn't been discussed yet on -hackers. But I'm fine with mentioning some of the possible features in 7.4, *provided* that you make it clear that those are tentative plans. Cheers, Neil -- Neil Conway <neilc@samurai.com> || PGP Key ID: DB3C29FC
Neil Conway wrote: > On Sun, 2002-12-08 at 23:54, Justin Clift wrote: > >>The "92 scheme specification" is the American National Standards >>Organisation (ANSI) SQL92 standard, which specifies the SQL query >>language used by most major relational databases (PostgreSQL, Oracle, >>DB2, Sybase, etc). > > > "SQL query language" is redundant. > > Not sure why there's a need to mention ANSI, as it's not as if there are > competing SQL92 standards :-) And name-dropping standards bodies is a > bit pretentious, IMHO. Rats' Thought it was better to go for name-recognition here. > The "92 scheme specification" isn't SQL92 (as your first sentence > implies), it's a part of SQL92. Ok, my misunderstanding. Sorry about that. Haven't yet gotten around to touching schema's yet, although it's going to have to be soon. :-) >>PostgreSQL has actively been adding ANSI SQL features for many years >>and is now one of the most ANSI SQL compliant databases in existence. >>[is this true?] > > > Sounds pretty dubious to me. Wasn't sure either. Some time last year someone made a comment like that, and I'm wondering if there's any way of testing it's validity. >>Maybe we should use this opportunity to build anticipation? Something >>like "Because we'll be adding native support for Windows, Point in Time >>recovery, and other advanced features in PostgreSQL 7.4, our initial >>target date is around April 2003". > > > I still think mentioning a release date, even as an"initial target", is > not appropriate, particularly since it hasn't been discussed yet on > -hackers. Didn't Bruce say that 7.4 was 6 months away or less? From memory we were hoping for a 3-4 month period until releasing 7.4, mostly just adding in the Windows native port and PITR, but that is delayed due to the wait for PeerDirect's new version of the Windows code. > But I'm fine with mentioning some of the possible features in 7.4, > *provided* that you make it clear that those are tentative plans. Ok. :-) Regards and best wishes, Justin Clift > Cheers, > > Neil -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
Hi everyone, What we have so far for this is at: http://advocacy.postgresql.org/documents/NadiaCameron Am also wondering how we should arrange for future stuff like this. Don't want to be the only one presenting info to the press when it's really a whole group of us creating it. That wouldn't be giving due credit to people. Any suggestions for something a bit more workable? :-) Regards and best wishes, Justin Clift -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
Hi Neil, One of Nadia's points is: "In the release, Neil Conway comments that the new version contains a dependency tracking systems "that allows PostgreSQL to safely support many more subtle enhancements like the ability to drop columns". I'm unclear on the use of "safely" here - would you be able to elaborate on this comment further?" It's that we now have the ability to drop columns properly isn't it? Perhaps we should just mention that one of the more requested features for PostgreSQL has been the ability to drop columns using the SQL92 compliant syntax, so that was added this release? Not sure. Might be stretching things a bit far. http://advocacy.postgresql.org/documents/NadiaCameron Regards and best wishes, Justin Clift -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Mon, 2002-12-09 at 02:08, Justin Clift wrote: > "In the release, Neil Conway comments that the new version contains a > dependency tracking systems "that allows PostgreSQL to safely support > many more subtle enhancements like the ability to drop columns". I'm > unclear on the use of "safely" here - would you be able to elaborate on > this comment further?" > > It's that we now have the ability to drop columns properly isn't it? Well, that and the fact that (a) dropping columns is transaction-safe (something a lot of the big commercial DBs don't even do, I believe) (b) the system takes care of accounting for all the various dependancies the column might have -- so it's difficult (not impossible) to break the database by dropping a column that something else in the DB depends upon. I was trying to think of a response to the query, but couldn't find a way to phrase the above in a way that was sufficiently non-technical. > Perhaps we should just mention that one of the more requested features > for PostgreSQL has been the ability to drop columns using the SQL92 > compliant syntax, so that was added this release? Well, I think you can mention that, definately. Cheers, Neil -- Neil Conway <neilc@samurai.com> || PGP Key ID: DB3C29FC
Justin Clift wrote:
> Hi Neil,
>
> One of Nadia's points is:
>
> "In the release, Neil Conway comments that the new version contains a
> dependency tracking systems "that allows PostgreSQL to safely support
> many more subtle enhancements like the ability to drop columns". I'm
> unclear on the use of "safely" here - would you be able to elaborate on
> this comment further?"
Well since the comment was really mine (and I hid behind Neil), I'll tell you
what I was thinking and others can agree or shoot it down in flames ;-)
The point is that in the absence of dependency tracking, there is nothing to
prevent you from, for example, dropping a column that is used as a foreign key
reference, or is included in a view. With dependency tracking, DROP COLUMN is
a safer feature than it would have been, because you are prevented from
shooting yourself in the foot. E.g.:
test=# create table dep1(f1 int, f2 int);
CREATE TABLE
test=# create view vwdep as select f1,f2 from dep1 ;
CREATE VIEW
test=# alter table dep1 drop column f2;
NOTICE: rule _RETURN on view vwdep depends on table dep1 column f2
NOTICE: view vwdep depends on rule _RETURN on view vwdep
ERROR: Cannot drop table dep1 column f2 because other objects depend on it
Use DROP ... CASCADE to drop the dependent objects too
HTH,
Joe
Hi guys, Cool. Thanks heaps. :) We should be able to come up with something decent out of both your and Neil's comments. :) There's a lot of info between both of them, and it would be interesting to know if the other main commercial databases track this properly. It could be useful here and into the future as an example of where we do stuff that's obviously safer+better than our commercial counterparts. :-) Regards and best wishes, Justin Clift Joe Conway wrote: <snip> > Well since the comment was really mine (and I hid behind Neil), I'll > tell you what I was thinking and others can agree or shoot it down in > flames ;-) > > The point is that in the absence of dependency tracking, there is > nothing to prevent you from, for example, dropping a column that is used > as a foreign key reference, or is included in a view. With dependency > tracking, DROP COLUMN is a safer feature than it would have been, > because you are prevented from shooting yourself in the foot. E.g.: > > test=# create table dep1(f1 int, f2 int); > CREATE TABLE > test=# create view vwdep as select f1,f2 from dep1 ; > CREATE VIEW > test=# alter table dep1 drop column f2; > NOTICE: rule _RETURN on view vwdep depends on table dep1 column f2 > NOTICE: view vwdep depends on rule _RETURN on view vwdep > ERROR: Cannot drop table dep1 column f2 because other objects depend on it > Use DROP ... CASCADE to drop the dependent objects too > > HTH, > > Joe > -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Mon, 9 Dec 2002, Justin Clift wrote: > Hi everyone, > > What we have so far for this is at: > > http://advocacy.postgresql.org/documents/NadiaCameron > > Am also wondering how we should arrange for future stuff like this. Okay. This is the question I was waiting for. I know Nadia Cameron. She is a good example of your average IT journalist. This is how it came to be that she's writing an article. 1) Last Friday I sent her boss the PostgreSQL 7.3 press release. Why, he didn't receive it. Neither did his US counterparts. 2) He passed it on to Nadia. 3) Nadia probably went looking for some information about PostgreSQL. Didn't find a media kit and so had to send off *too many questions* Things got interesting from there. Correctly, she emailed Justin -- since Justin's details are on the press release. The problems we're suffering were then exposed. Justin did not have the information, skills, confidence or support -- which ever -- to reply to her independently. See why media kits are so useful? They contain everything you need to write 'product news'. If Nadia was to undertake a more significant article, perhaps she would ask for some different quotes. That need not have been the case here. > > Don't want to be the only one presenting info to the press when it's There's nothing wrong with that. That's why product news only ever quotes the Marketing manager or PR manager. We need two people to do this. We can do it on a rotating basis. We can supply email addresses at the bottom of the press release 'for quotes, email ...'. > really a whole group of us creating it. That wouldn't be giving due > credit to people. Any suggestions for something a bit more workable? Its not a group effort. Its too hard then. Too slow. Nadia's goal would have been to get that article out in a few hours. The end. The PostgreSQL team needs to be able to handle 100 of these at once. Doing this 'collaboratively' will just not work. Who wants to put their hand up for quotes? I am happy to. Who wants to play press contacts(s)? It should work as follows: Contact receives media request. Contact drafts a response and sends it to someone else in advocacy -- anyone who can spell in the English language -- to check what they write back. It gets sent back. Very quick turn around. Then it is finished. Gavin
Gavin Sherry wrote: > On Mon, 9 Dec 2002, Justin Clift wrote: <snip> > Who wants to play press contacts(s)? It should work as follows: Contact > receives media request. Contact drafts a response and sends it to someone > else in advocacy -- anyone who can spell in the English language -- to > check what they write back. It gets sent back. Very quick turn > around. Then it is finished. Sounds like a decent idea to start with. If we have a well developed Press Kit, then that will give a lot of consistency to the information going out too won't it? So, every time we improve the Press Kit, the information going out is improved too? Is that a correct way to look at possible strategic benefits for the Press Kit in addition to lightening the number of contact requests for basic info? :-) Regards and best wishes, Justin Clift > Gavin > -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
On Mon, 9 Dec 2002, Justin Clift wrote: > Gavin Sherry wrote: > > On Mon, 9 Dec 2002, Justin Clift wrote: > <snip> > > Who wants to play press contacts(s)? It should work as follows: Contact > > receives media request. Contact drafts a response and sends it to someone > > else in advocacy -- anyone who can spell in the English language -- to > > check what they write back. It gets sent back. Very quick turn > > around. Then it is finished. > > Sounds like a decent idea to start with. > > If we have a well developed Press Kit, then that will give a lot of > consistency to the information going out too won't it? So, every time > we improve the Press Kit, the information going out is improved too? Naturally. > > Is that a correct way to look at possible strategic benefits for the > Press Kit in addition to lightening the number of contact requests for > basic info? This is why I'm suggesting it. I've organised these things before :-). Now: who wants to be the press contact or shall I point and see if people object? As for people for quotes, I'm organising it. Gavin
Gavin Sherry wrote: <snip> > This is why I'm suggesting it. I've organised these things before :-). > > Now: who wants to be the press contact or shall I point and see if people > object? Hmm... now I'm trying to decide if I'm an attention seeker or not. ;-) Um, probably. :) > As for people for quotes, I'm organising it. Ok. Lucas is probably going to organise the PR kit too isn't he? It sounds like you'll both be able to add a lot of good stuff to it. :-) Regards and best wishes, Justin Clift > Gavin > > -- "My grandfather once told me that there are two kinds of people: those who work and those who take the credit. He told me to try to be in the first group; there was less competition there." - Indira Gandhi
Neil, Joe, Justin, Oliver, etc.:
I've taken a stab at answering the first two questions. See how you
like it.
-Josh
______AGLIO DATABASE SOLUTIONS___________________________
Josh Berkus
Complete information technology josh@agliodbs.com
and data management solutions (415) 565-7293
for law firms, small businesses fax 621-2533
and non-profit organizations. San Francisco
Neil, > PostgreSQL has actively been > adding ANSI SQL features for many years and is now one of the most ANSI > SQL compliant databases in existence. [is this true?] Depends on how you measure compliance. For example, FrontBase only implements about 65% of the standard, but what it does implement is 100% compliant. So, does one call them 65% or 100% compliant? Or what? However, by *whatever* measure, we are more compliant than MSSQL, Oracle, Informix, or MySQL. -- -Josh Berkus Aglio Database Solutions San Francisco
Gavin, > Now: who wants to be the press contact or shall I point and see if people > object? This is why I suggested a mailing list alias. If press@postgresql.org hits you, me, justin, bruce, etc., there is bound to be someone available at the time it is sent to turn around answers quickly. -- -Josh Berkus Aglio Database Solutions San Francisco
On Mon, 9 Dec 2002, Josh Berkus wrote: > > Gavin, > > > Now: who wants to be the press contact or shall I point and see if people > > object? > > This is why I suggested a mailing list alias. If press@postgresql.org hits > you, me, justin, bruce, etc., there is bound to be someone available at the > time it is sent to turn around answers quickly. Done. Can we see it up? There will of course need to be some liason to ensure that all contact comes from a single person (so as to not confuse the journos). Gavin
Gavin,
> > This is why I suggested a mailing list alias. If press@postgresql.org
hits
> > you, me, justin, bruce, etc., there is bound to be someone available at
the
> > time it is sent to turn around answers quickly.
>
> Done. Can we see it up? There will of course need to be some liason to
> ensure that all contact comes from a single person (so as to not confuse
> the journos).
Marc set this up, but I've lost my admin password and web address due to a
mail client error. Prepare for some delay.
--
-Josh Berkus
______AGLIO DATABASE SOLUTIONS___________________________
Josh Berkus
Complete information technology josh@agliodbs.com
and data management solutions (415) 565-7293
for law firms, small businesses fax 621-2533
and non-profit organizations. San Francisco