Horton

Thread

  1. Numbering of the next release: 8.0 vs 7.4

    Justin Clift <justin@postgresql.org> — 2003-03-12T05:29:52Z

    Hi everyone,

Thinking about the numbering further.

Would it be cool to decide on the version numbering of our next release 
like this:

  + If it looks like we'll have Win32 and/or PITR recovery in time for
    the next release, we call it PostgreSQL 8.0

  + If not, we call it 7.4

Win32 and PITR are great big features that will take us a long way to 
the goal of Enterprise suitability.  They're worth making some specific 
marketing/branding efforts about and making a big fuss, that why I'd 
like to see them in an 8.0 release.

Sound feasible?

:-)

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

  2. Re: Numbering of the next release: 8.0 vs 7.4

    Paul Ramsey <pramsey@refractions.net> — 2003-03-12T05:40:27Z

    Justin Clift wrote:

> Win32 and PITR are great big features that will take us a long way to 
> the goal of Enterprise suitability.  They're worth making some specific 
> marketing/branding efforts about and making a big fuss, that why I'd 
> like to see them in an 8.0 release.

 From a marketing point of view, wouldn't it be better to skip that 
risky "point-O" release and go straight to version 8.1? :)

-- 
       __
      /
      | Paul Ramsey
      | Refractions Research
      | Email: pramsey@refractions.net
      | Phone: (250) 885-0632
      \_

  3. Re: Numbering of the next release: 8.0 vs 7.4

    Justin Clift <justin@postgresql.org> — 2003-03-12T05:43:33Z

    Paul Ramsey wrote:
> Justin Clift wrote:
> 
>> Win32 and PITR are great big features that will take us a long way to 
>> the goal of Enterprise suitability.  They're worth making some 
>> specific marketing/branding efforts about and making a big fuss, that 
>> why I'd like to see them in an 8.0 release.
> 
>  From a marketing point of view, wouldn't it be better to skip that 
> risky "point-O" release and go straight to version 8.1? :)

Err... lets not get into deceptive marketing.  ;-)

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

  4. Re: Numbering of the next release: 8.0 vs 7.4

    Tom Lane <tgl@sss.pgh.pa.us> — 2003-03-12T06:26:47Z

    Justin Clift <justin@postgresql.org> writes:
> Would it be cool to decide on the version numbering of our next release 
> like this:
>   + If it looks like we'll have Win32 and/or PITR recovery in time for
>     the next release, we call it PostgreSQL 8.0
>   + If not, we call it 7.4

Works for me: release schedule is solid, what we call it gets decided
at the last minute ;-)

			regards, tom lane

  5. Re: Numbering of the next release: 8.0 vs 7.4

    Sander Steffann <sander@steffann.nl> — 2003-03-12T12:33:01Z

    Hi,

> Would it be cool to decide on the version numbering of our next release
> like this:
>
>   + If it looks like we'll have Win32 and/or PITR recovery in time for
>     the next release, we call it PostgreSQL 8.0
>
>   + If not, we call it 7.4

Wouldn't a new FE/BE protocol be a better reason to call it 8.0? Raising the
major version number together with introducing a new protocol which causes
incompatibilities between new clients and older servers seems like a logical
combination...

Just a thought... :)
Sander.

  6. Re: Numbering of the next release: 8.0 vs 7.4

    Mark Woodward <pgsql@mohawksoft.com> — 2003-03-12T13:58:25Z

    
Justin Clift wrote:

> Hi everyone,
>
> Thinking about the numbering further.
>
> Would it be cool to decide on the version numbering of our next 
> release like this:
>
>  + If it looks like we'll have Win32 and/or PITR recovery in time for
>    the next release, we call it PostgreSQL 8.0
>
>  + If not, we call it 7.4
>
> Win32 and PITR are great big features that will take us a long way to 
> the goal of Enterprise suitability.  They're worth making some 
> specific marketing/branding efforts about and making a big fuss, that 
> why I'd like to see them in an 8.0 release.
>
> Sound feasible?

Sounds reasonable, but from a "change" perspective, the FE/BE protocol, 
Win32, and PITR, I would say that this is a "new" PostgreSQL, thus 
should be 8.0. I thought when WAL was added that warrented a different 
major version, but hey, that's me.

But, if the decision is to go for an 8.0, then it should be reasonable 
to be a little bit more aggresive about adding features and perhaps a 
few wish list items. What I mean is, if it is just a minor release, one 
just expects minor improvements and bug fixes. If it is a major release, 
then one expects an update of the "PostgreSQL vision."

So, if the decision is to go with an 8.0, what would you guys say to 
having a roll call about stuff that is "possible" and "practical" and 
really design "PostgreSQL 8.0" as something fundimentally "newer" than 
7.x. "8.0" could get the project some hype. It has been 7x for so many 
years.

  7. Re: Numbering of the next release: 8.0 vs 7.4

    Robert Treat <xzilla@users.sourceforge.net> — 2003-03-12T14:55:35Z

    On Wed, 2003-03-12 at 01:26, Tom Lane wrote:
> Justin Clift <justin@postgresql.org> writes:
> > Would it be cool to decide on the version numbering of our next release 
> > like this:
> >   + If it looks like we'll have Win32 and/or PITR recovery in time for
> >     the next release, we call it PostgreSQL 8.0
> >   + If not, we call it 7.4
> 
> Works for me: release schedule is solid, what we call it gets decided
> at the last minute ;-)
> 

Personally I think Justin is a little off base with his criteria, since
I see the FE/BE protocol changes as the real differentiator between an
8.0 and 7.4. Everyone is effected by a FE/BE protocol change, not nearly
so many are effected by either win32 or PITR. And think of this crazy
scenario: We release an 8.0 with PITR, then need either a 8.1 or a 9.0
with a FE/BE overhaul, then need a possible 10.0 because we've added
win32... yuk. 

That said, I'll take Tom's position on this that we might as well worry
about whether it's going to be 7.4 or 8.0 once we hit feature freeze; by
then the whole discussion could be irrelevant. 

Robert Treat

  8. Re: Numbering of the next release: 8.0 vs 7.4

    Justin Clift <justin@postgresql.org> — 2003-03-12T15:00:45Z

    mlw wrote:
<snip>
> So, if the decision is to go with an 8.0, what would you guys say to 
> having a roll call about stuff that is "possible" and "practical" and 
> really design "PostgreSQL 8.0" as something fundimentally "newer" than 
> 7.x. "8.0" could get the project some hype. It has been 7x for so many 
> years.

Sounds great.  I just don't want it to take _ages_ to accomplish.

:)

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

  9. Re: Numbering of the next release: 8.0 vs 7.4

    Lamar Owen <lamar.owen@wgcr.org> — 2003-03-12T15:15:30Z

    On Wednesday 12 March 2003 09:55, Robert Treat wrote:
> Personally I think Justin is a little off base with his criteria, since
> I see the FE/BE protocol changes as the real differentiator between an
> 8.0 and 7.4. Everyone is effected by a FE/BE protocol change, not nearly
> so many are effected by either win32 or PITR. And think of this crazy
> scenario: We release an 8.0 with PITR, then need either a 8.1 or a 9.0
> with a FE/BE overhaul, then need a possible 10.0 because we've added
> win32... yuk.

FWIW, the 6.4 protocol change didn't force a move from 6.3.2 to 7.0.
-- 
Lamar Owen
WGCR Internet Radio
1 Peter 4:11

  10. Re: Numbering of the next release: 8.0 vs 7.4

    Tom Lane <tgl@sss.pgh.pa.us> — 2003-03-12T15:34:57Z

    Lamar Owen <lamar.owen@wgcr.org> writes:
> FWIW, the 6.4 protocol change didn't force a move from 6.3.2 to 7.0.

True, but that was a much smaller change than what we're contemplating
here.  AFAIR, those changes did not affect the majority of applications
--- they only needed to relink with a newer client library, and voila
they spoke the new protocol perfectly well.  The planned changes for
error handling (error codes, etc) will be something that will affect
almost every app.  They won't *need* to change, maybe, but they'll
probably *want* to change.

But let's wait till feature freeze to have this discussion; we'll know
better by then exactly what we're talking about.

			regards, tom lane

  11. Re: Numbering of the next release: 8.0 vs 7.4

    Justin Clift <justin@postgresql.org> — 2003-03-12T15:38:22Z

    Tom Lane wrote:
> Lamar Owen <lamar.owen@wgcr.org> writes:
> 
>>FWIW, the 6.4 protocol change didn't force a move from 6.3.2 to 7.0.
> 
> 
> True, but that was a much smaller change than what we're contemplating
> here.  AFAIR, those changes did not affect the majority of applications
> --- they only needed to relink with a newer client library, and voila
> they spoke the new protocol perfectly well.  The planned changes for
> error handling (error codes, etc) will be something that will affect
> almost every app.  They won't *need* to change, maybe, but they'll
> probably *want* to change.
> 
> But let's wait till feature freeze to have this discussion; we'll know
> better by then exactly what we're talking about.

Yep, that sounds like the best idea.

:-)

Regards and best wishes,

Justin Clift


> 			regards, tom lane


-- 
"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

  12. Re: Numbering of the next release: 8.0 vs 7.4

    Peter Eisentraut <peter_e@gmx.net> — 2003-03-12T15:43:31Z

    Justin Clift writes:

>   + If it looks like we'll have Win32 and/or PITR recovery in time for
>     the next release, we call it PostgreSQL 8.0

To me, those sound fairly unspectacular as reasons for 8.0.

-- 
Peter Eisentraut   peter_e@gmx.net

  13. Re: Numbering of the next release: 8.0 vs 7.4

    Felipe Schnack <felipes@ritterdosreis.br> — 2003-03-12T16:23:40Z

      I think the first thing we should do about that is to define what are the
reasons for a major version change. The way this discussion is being taken
will not take us anywhere... is just too much about personal opinions. 
  Anyway, for most users a win32 port is not a big deal (after all,
practically all of us are using pgsql in an unix-like system)... but a lot of
Windows users that doesn`t try pgsql because mysql is just so easy to install
on windows machines...
  and they`re much better in marketing too. Well, they have a company behind them.

Felipe Schnack
Analista de Sistemas
felipes@ritterdosreis.br
Cel.: (51)91287530
Linux Counter #281893

Faculdade Ritter dos Reis
www.ritterdosreis.br
Fone/Fax.: (51)32303328


---------- Original Message -----------
From: Peter Eisentraut <peter_e@gmx.net>
To: Justin Clift <justin@postgresql.org>
Sent: Wed, 12 Mar 2003 16:43:31 +0100 (CET)
Subject: Re: [HACKERS] Numbering of the next release: 8.0 vs 7.4

> Justin Clift writes:
> 
> >   + If it looks like we'll have Win32 and/or PITR recovery in time for
> >     the next release, we call it PostgreSQL 8.0
> 
> To me, those sound fairly unspectacular as reasons for 8.0.
> 
> -- 
> Peter Eisentraut   peter_e@gmx.net
> 
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
> 
> http://archives.postgresql.org
------- End of Original Message -------