Re: Time to start drafting the 8.4 release

On 3/30/09 11:49 PM, Rob Napier wrote:
> Josh
>
> Is there anything I can do to help?
>
> If you give me the raw details, I'd be happy to write it for you.

See the wiki page.

But the first thing to do is start with a theme. Suggestions?

We used "rapid development" last year, so we can't use it this year.
But this release doesn't have any central "theme" in terms of what's
been added; it's just a whole buncha features.

The one thing I thought of which is somewhat new is that we added
several features or partial features designed to pave the way for larger
features later.

--Josh

Josh Berkus wrote:
> On 3/30/09 11:49 PM, Rob Napier wrote:
> > Josh
> >
> > Is there anything I can do to help?
> >
> > If you give me the raw details, I'd be happy to write it for you.
>
> See the wiki page.
>
> But the first thing to do is start with a theme. Suggestions?
>
> We used "rapid development" last year, so we can't use it this year.
> But this release doesn't have any central "theme" in terms of what's
> been added; it's just a whole buncha features.
>
> The one thing I thought of which is somewhat new is that we added
> several features or partial features designed to pave the way for larger
> features later.

I used this text for the top of the 8.4 release note; it highlights
targeted development and recently-adopted SQL-standard features.

After many years of development, <productname>PostgreSQL</> has
become feature-complete in many areas. This release shows a
targeted approach to adding features (e.g. authentication,
monitoring, space reuse), and adds capabilities defined in the
later SQL standards; the major areas of enhancement are:

You will notice from the release notes that the feature additions are
more focused in specific areas than in previous releases, and we have
added some cool SQL-standards stuff (windowing, recusive queries).

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +

Bruce,

> I used this text for the top of the 8.4 release note; it highlights
> targeted development and recently-adopted SQL-standard features.
>
> After many years of development,<productname>PostgreSQL</> has
> become feature-complete in many areas. This release shows a
> targeted approach to adding features (e.g. authentication,
> monitoring, space reuse), and adds capabilities defined in the
> later SQL standards; the major areas of enhancement are:

Yeah, I don't know that "maturity" is a sexy message for a press
release, though.

--Josh

On Tue, 2009-03-31 at 10:36 -0700, Josh Berkus wrote:
> > I used this text for the top of the 8.4 release note; it highlights
> > targeted development and recently-adopted SQL-standard features.
> >
> > After many years of development,<productname>PostgreSQL</> has
> > become feature-complete in many areas. This release shows a
> > targeted approach to adding features (e.g. authentication,
> > monitoring, space reuse), and adds capabilities defined in the
> > later SQL standards; the major areas of enhancement are:
>
> Yeah, I don't know that "maturity" is a sexy message for a press
> release, though.

Just brainstorming:

WITH RECURSIVE and window functions could both fall into an "analytics"
theme of some kind. That doesn't tell the whole story, though.

Some of the other major features, like the Visibility Map, are very
important, but I can't think of any way to make them sound exciting
(other than a "performance" theme, which is not exciting without
numbers).

Regards,
Jeff Davis

Jeff Davis wrote:

> WITH RECURSIVE and window functions could both fall into an "analytics"
> theme of some kind. That doesn't tell the whole story, though.
>
> Some of the other major features, like the Visibility Map, are very
> important, but I can't think of any way to make them sound exciting
> (other than a "performance" theme, which is not exciting without
> numbers).

VM can also be categorized as "analytics", because it makes vacuuming
huge tables so much convenient.

--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.

Jeff Davis wrote:
> Some of the other major features, like the Visibility Map, are very
> important, but I can't think of any way to make them sound exciting
> (other than a "performance" theme, which is not exciting without
> numbers).

Space reuse?

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ If your life is a hard drive, Christ can be your backup. +

Expanding the base [set of functionality]? A theme like, "PostgreSQL
8.4: Expanding the foundation"?

---Michael Brewer
mbrewer(at)gmail(dot)com

On Tue, Mar 31, 2009 at 02:10:13PM -0400, Alvaro Herrera wrote:
> Jeff Davis wrote:
>
> > WITH RECURSIVE and window functions could both fall into an
> > "analytics" theme of some kind. That doesn't tell the whole story,
> > though.
> >
> > Some of the other major features, like the Visibility Map, are
> > very important, but I can't think of any way to make them sound
> > exciting (other than a "performance" theme, which is not exciting
> > without numbers).
>
> VM can also be categorized as "analytics", because it makes
> vacuuming huge tables so much convenient.

Is, "analytics" the latest in the DSS/OLAP/BI series of wacky terms
for, "system need not respond within a second?"

Cheers,
David (having trouble keeping up with this term-of-the-day merry-go-round).
--
David Fetter <david(at)fetter(dot)org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david(dot)fetter(at)gmail(dot)com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate

On Tue, 2009-03-31 at 11:56 -0700, David Fetter wrote:
> On Tue, Mar 31, 2009 at 02:10:13PM -0400, Alvaro Herrera wrote:
> > Jeff Davis wrote:
> >
> > > WITH RECURSIVE and window functions could both fall into an
> > > "analytics" theme of some kind. That doesn't tell the whole story,
> > > though.
> > >
> > > Some of the other major features, like the Visibility Map, are
> > > very important, but I can't think of any way to make them sound
> > > exciting (other than a "performance" theme, which is not exciting
> > > without numbers).
> >
> > VM can also be categorized as "analytics", because it makes
> > vacuuming huge tables so much convenient.
>
> Is, "analytics" the latest in the DSS/OLAP/BI series of wacky terms
> for, "system need not respond within a second?"

WITH RECURSIVE and window functions both increase the expressiveness of
PostgreSQL's dialect of SQL substantially, allowing users to ask
interesting questions in a single query that would have been impossible
to ask before. These queries work on small data sets as well, so I don't
think that these features are only useful for long-running queries.

Do you have a word in mind that captures this benefit better than
"analytics"? "Expressiveness" might work, but I don't think that people
really associate that word with database systems.

Regards,
Jeff Davis

On Tue, Mar 31, 2009 at 02:03:41PM -0700, Jeff Davis wrote:
> On Tue, 2009-03-31 at 11:56 -0700, David Fetter wrote:
> > On Tue, Mar 31, 2009 at 02:10:13PM -0400, Alvaro Herrera wrote:
> > > Jeff Davis wrote:
> > >
> > > > WITH RECURSIVE and window functions could both fall into an
> > > > "analytics" theme of some kind. That doesn't tell the whole
> > > > story, though.
> > > >
> > > > Some of the other major features, like the Visibility Map, are
> > > > very important, but I can't think of any way to make them
> > > > sound exciting (other than a "performance" theme, which is not
> > > > exciting without numbers).
> > >
> > > VM can also be categorized as "analytics", because it makes
> > > vacuuming huge tables so much convenient.
> >
> > Is, "analytics" the latest in the DSS/OLAP/BI series of wacky
> > terms for, "system need not respond within a second?"
>
> WITH RECURSIVE and window functions both increase the expressiveness
> of PostgreSQL's dialect of SQL substantially, allowing users to ask
> interesting questions in a single query that would have been
> impossible to ask before.

Indeed. I've been going around and boring people with this fact for
months :)

> These queries work on small data sets as well, so I don't think that
> these features are only useful for long-running queries.
>
> Do you have a word in mind that captures this benefit better than
> "analytics"? "Expressiveness" might work, but I don't think that
> people really associate that word with database systems.

Maybe that's the hook, then :) POSTGRES, and INGRES before it, were
both from the "active database" camp, and we've taken it much further
with the PL zoo of languages to make PostgreSQL a highly capable
programming environment.

Cheers,
David.
--
David Fetter <david(at)fetter(dot)org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david(dot)fetter(at)gmail(dot)com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate

> Maybe that's the hook, then :) POSTGRES, and INGRES before it, were
> both from the "active database" camp, and we've taken it much further
> with the PL zoo of languages to make PostgreSQL a highly capable
> programming environment.

Hmmm. Seems like a rather difficult idea to get across to a general
tech blogger or MySQL-using developer. Is there any way we could phrase
it that would work?

--Josh

On Tue, Mar 31, 2009 at 02:38:21PM -0700, Josh Berkus wrote:
>
>> Maybe that's the hook, then :) POSTGRES, and INGRES before it,
>> were both from the "active database" camp, and we've taken it much
>> further with the PL zoo of languages to make PostgreSQL a highly
>> capable programming environment.
>
> Hmmm. Seems like a rather difficult idea to get across to a general
> tech blogger or MySQL-using developer. Is there any way we could
> phrase it that would work?

PostgreSQL gets smarter and more flexible. Again!

Cheers,
David.
--
David Fetter <david(at)fetter(dot)org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david(dot)fetter(at)gmail(dot)com

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate

On Tuesday 31 March 2009 2:38:21 pm Josh Berkus wrote:
> > Maybe that's the hook, then :) POSTGRES, and INGRES before it, were
> > both from the "active database" camp, and we've taken it much further
> > with the PL zoo of languages to make PostgreSQL a highly capable
> > programming environment.
>
> Hmmm. Seems like a rather difficult idea to get across to a general
> tech blogger or MySQL-using developer. Is there any way we could phrase
> it that would work?
>
> --Josh

Postgres, Not your fathers database
Postgres, You have the itch we have the scratch

Blame it on the caffeine :)

--
Adrian Klaver
aklaver(at)comcast(dot)net

Actually, we have four basic theme ideas:

1) Smarter PG: more ability to develop "smart" applications on PG (David)
Pro: It's what we do as a project
Con: very similar to our theme last year

2) Fastest Developing Database: putting out a release *every* year, no
matter what (Josh)
Pro: timely, given issues with other DBMSes
Con: very similar to our theme last year

3) Maturity: After many years of development, PG has become
feature-complete in many areas. (Bruce)
Pro: accurate
Con: "maturity" ~ "stagnation"

4) BI/OLAP: focus on Windowing/Recursive.
Pro: industry buzzword-compliant
Con: don't have rest of toolset (Cube,etc.), ignores several major
features.

Given these, I'd think that combining (1) and (2) would probably be our
best route. Thoughts?

--Josh

On Wednesday 01 April 2009 00:43:33 David Fetter wrote:
> PostgreSQL gets smarter and more flexible. Again!

I like it -- because it's true!

2009/4/1 Josh Berkus <josh(at)agliodbs(dot)com>:
> Actually, we have four basic theme ideas:
>
> 1) Smarter PG: more ability to develop "smart" applications on PG (David)
>    Pro: It's what we do as a project
>    Con: very similar to our theme last year
>
> 2) Fastest Developing Database: putting out a release *every* year, no
> matter what (Josh)
>    Pro: timely, given issues with other DBMSes
>    Con: very similar to our theme last year

Making a combination:

Releases every year, gives to developers more tools and advantages as
the (market|needs)? grows, with a wide array of features

> 3) Maturity: After many years of development, PG has become feature-complete
> in many areas. (Bruce)
>    Pro: accurate
>    Con: "maturity" ~ "stagnation"
>
> 4) BI/OLAP: focus on Windowing/Recursive.
>    Pro: industry buzzword-compliant
>    Con: don't have rest of toolset (Cube,etc.), ignores several major
> features.
>
>
> Given these, I'd think that combining (1) and (2) would probably be our best
> route.  Thoughts?
>
> --Josh
>
> --
> Sent via pgsql-advocacy mailing list (pgsql-advocacy(at)postgresql(dot)org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-advocacy
>