Thread: My new job
As many of you know, several businesses are involved in providing PostgreSQL support. After careful consideration, I have decided to accept a job with Great Bridge. There will be a press announcement tomorrow (Tuesday) with more details. I will post a URL here when I have it. Interestingly, I am the last core member to become officially attached to a PostgreSQL company. I have been assisting Great Bridge for some time, but always in an unofficial capacity. I realize my signing on with any company will make some of you uncomfortable. I hope my value to the group does not suffer, and that I continue to be a positive influence for all involved. I expect my new job will give me even more time to continue doing the things I have done in the past with PostgreSQL. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Bruce Momjian writes: > After careful consideration, I have decided to accept a job with Great > Bridge. Whatever happened to this: Date: Tue, 09 May 2000 15:19:48 -0400 From: Tom Lane <tgl@sss.pgh.pa.us> To: Ross J. Reedstrom <reedstrm@wallace.ece.rice.edu> Cc: PostgreSQL-general <pgsql-general@postgresql.org> Subject: Re: [GENERAL] Steering committee responce to Great Bridge LLC : One thing we have agreed to is that there must not be an unseemly fraction : of core members working for the same company. With six people on core, : probably about two working at the same company would be a reasonable : limit. -- Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/
> Bruce Momjian writes: > > > After careful consideration, I have decided to accept a job with Great > > Bridge. > > Whatever happened to this: > > Date: Tue, 09 May 2000 15:19:48 -0400 > From: Tom Lane <tgl@sss.pgh.pa.us> > To: Ross J. Reedstrom <reedstrm@wallace.ece.rice.edu> > Cc: PostgreSQL-general <pgsql-general@postgresql.org> > Subject: Re: [GENERAL] Steering committee responce to Great Bridge LLC > > : One thing we have agreed to is that there must not be an unseemly fraction > : of core members working for the same company. With six people on core, > : probably about two working at the same company would be a reasonable > : limit. Excellent question. I suggested leaving core, but that would still mean more than 1/3 of core people would be in one company. Our short-term solution is to keep going until we see some problems. Our long-term strategy is to increase the size of the core group. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Peter Eisentraut <peter_e@gmx.net> writes:
> Bruce Momjian writes:
>> After careful consideration, I have decided to accept a job with Great
>> Bridge.
> Whatever happened to this:
> From: Tom Lane <tgl@sss.pgh.pa.us>
> : One thing we have agreed to is that there must not be an unseemly fraction
> : of core members working for the same company. With six people on core,
> : probably about two working at the same company would be a reasonable
> : limit.
I knew someone was going to bring that up ;-).
There's already been discussion of this point among core. What we
now have is three core members employed by Great Bridge and the
other three either fully or partly employed by PostgreSQL Inc.
In one sense that's a stable situation, but on the other hand it does
not agree with our original informal goal of keeping any one company
to a minority position of the core membership.
None of the core members are interested in giving up their new
positions. En masse resignation from the core committee would preserve
our high moral standards, perhaps, but it wouldn't do the project any
good that I can see. So it seems like the choices are to accept the
status quo, or to appoint some more core committee members to bring
the numbers back where we said they should be.
While I can think of a number of well-qualified candidates for core
membership, I don't much like the notion of appointing core members
just to meet some kind of numerical quota. Also, suppose we do appoint
more members, and then some of them accept positions with GB or PgSQL
Inc; do we repeat the exercise indefinitely? (This is not an unlikely
scenario, since the sort of people who'd be asked to join core are
exactly the sort of people whom both companies would love to hire.)
Bottom line is we're not sure what to do now. Opinions from the
floor, anyone?
regards, tom lane
What is the main concern? That Great Bridge or PostgreSQL Inc will try to influence development? This is just my lowly opinion but it seems to me that this could be a storm brewing in a tea cup, it just doesn't seem to be that threatening a situation at a glance. Congrats to everyone on their new positions. *hats off* -Mitch ----- Original Message ----- From: "Tom Lane" <tgl@sss.pgh.pa.us> To: "Peter Eisentraut" <peter_e@gmx.net> Cc: "PostgreSQL-general" <pgsql-general@postgresql.org>; "PostgreSQL-development" <pgsql-hackers@postgresql.org> Sent: Tuesday, October 10, 2000 10:02 AM Subject: Re: [HACKERS] My new job > Peter Eisentraut <peter_e@gmx.net> writes: > > Bruce Momjian writes: > >> After careful consideration, I have decided to accept a job with Great > >> Bridge. > > > Whatever happened to this: > > > From: Tom Lane <tgl@sss.pgh.pa.us> > > : One thing we have agreed to is that there must not be an unseemly fraction > > : of core members working for the same company. With six people on core, > > : probably about two working at the same company would be a reasonable > > : limit. > > I knew someone was going to bring that up ;-). > > There's already been discussion of this point among core. What we > now have is three core members employed by Great Bridge and the > other three either fully or partly employed by PostgreSQL Inc. > In one sense that's a stable situation, but on the other hand it does > not agree with our original informal goal of keeping any one company > to a minority position of the core membership. > > None of the core members are interested in giving up their new > positions. En masse resignation from the core committee would preserve > our high moral standards, perhaps, but it wouldn't do the project any > good that I can see. So it seems like the choices are to accept the > status quo, or to appoint some more core committee members to bring > the numbers back where we said they should be. > > While I can think of a number of well-qualified candidates for core > membership, I don't much like the notion of appointing core members > just to meet some kind of numerical quota. Also, suppose we do appoint > more members, and then some of them accept positions with GB or PgSQL > Inc; do we repeat the exercise indefinitely? (This is not an unlikely > scenario, since the sort of people who'd be asked to join core are > exactly the sort of people whom both companies would love to hire.) > > Bottom line is we're not sure what to do now. Opinions from the > floor, anyone? > > regards, tom lane >
> Peter Eisentraut <peter_e@gmx.net> writes: > > Bruce Momjian writes: > >> After careful consideration, I have decided to accept a job with Great > >> Bridge. > > > Whatever happened to this: > > > From: Tom Lane <tgl@sss.pgh.pa.us> > > : One thing we have agreed to is that there must not be an unseemly fraction > > : of core members working for the same company. With six people on core, > > : probably about two working at the same company would be a reasonable > > : limit. > > I knew someone was going to bring that up ;-). > > There's already been discussion of this point among core. What we > now have is three core members employed by Great Bridge and the > other three either fully or partly employed by PostgreSQL Inc. I should mention that the Great Bridge hires are full-time employment, while not all the PostgreSQL Inc.'s are, so the Great Bridge group is more in voliation of the original plan. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
* Tom Lane <tgl@sss.pgh.pa.us> [001010 10:03] wrote: > > From: Tom Lane <tgl@sss.pgh.pa.us> > > : One thing we have agreed to is that there must not be an unseemly fraction > > : of core members working for the same company. With six people on core, > > : probably about two working at the same company would be a reasonable > > : limit. > > I knew someone was going to bring that up ;-). > > There's already been discussion of this point among core. What we > now have is three core members employed by Great Bridge and the > other three either fully or partly employed by PostgreSQL Inc. > In one sense that's a stable situation, but on the other hand it does > not agree with our original informal goal of keeping any one company > to a minority position of the core membership. > > None of the core members are interested in giving up their new > positions. En masse resignation from the core committee would preserve > our high moral standards, perhaps, but it wouldn't do the project any > good that I can see. So it seems like the choices are to accept the > status quo, or to appoint some more core committee members to bring > the numbers back where we said they should be. > > While I can think of a number of well-qualified candidates for core > membership, I don't much like the notion of appointing core members > just to meet some kind of numerical quota. Also, suppose we do appoint > more members, and then some of them accept positions with GB or PgSQL > Inc; do we repeat the exercise indefinitely? (This is not an unlikely > scenario, since the sort of people who'd be asked to join core are > exactly the sort of people whom both companies would love to hire.) > > Bottom line is we're not sure what to do now. Opinions from the > floor, anyone? I think anyone with doubts should take a good look at the initial companies backing Linux, (Redhat, VA, Debian) to see what a boon this can be to project. It is open source, so if you guys do happen to piss us off too much we can always fork off our own version no? :) So instead of panicing, it makes much more sense to ride it out and get a feel for where things are going, there's never going to be anything terribly binding that will come out of this because it is an opensource project. It's much more important to continue on with the rapid pace of developement than to fear black helicopters that haven't even shown up as blips on the radar. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk."
At 01:02 PM 10/10/00 -0400, Tom Lane wrote: >Bottom line is we're not sure what to do now. Opinions from the >floor, anyone? Yeah, quit worrying and work your collective butts off on 7.1 and 7.2 :) Seriously...the core group is obviously committed to PG, and appear to be folks of integrity. We all will benefit by your working on PG full time while being paid enough so you can eat, drink, and be merry, too. - Don Baccus, Portland OR <dhogaza@pacifier.com> Nature photos, on-line guides, Pacific Northwest Rare Bird Alert Service and other goodies at http://donb.photo.net.
> At 01:02 PM 10/10/00 -0400, Tom Lane wrote: > > >Bottom line is we're not sure what to do now. Opinions from the > >floor, anyone? > > Yeah, quit worrying and work your collective butts off on 7.1 and 7.2 :) C'mon, Don. Stop beating around the bush. Tell us what you really think. :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Alfred Perlstein <bright@wintelcom.net> writes: > I think anyone with doubts should take a good look at the initial > companies backing Linux, (Redhat, VA, Debian) to see what a boon > this can be to project. I'll just clarify that Debian is not a company, it is an non-profit all-volunteer effort. I agree with everything else that Alfred says. Mike.
On Tue, Oct 10, 2000 at 10:15:03AM -0700, Alfred Perlstein wrote: > I think anyone with doubts should take a good look at the initial > companies backing Linux, (Redhat, VA, Debian) to see what a boon I certainly get your point but I have to correct this as Debian is not, never has been and never will be a company. Sorry, couldn't resist. :-) Michael -- Michael Meskes Michael@Fam-Meskes.De Go SF 49ers! Go Rhein Fire! Use Debian GNU/Linux! Use PostgreSQL!
Is the following expected behavior for a UNION query with ORDER BY:
executing this query:
+++++++++++++++++++++++++++++++++++++++++++
SELECT a.attnum as number, a.attname as attribute, CASE WHEN t.typname = 'varchar' THEN
t.typname|| '(' || a.atttypmod - 4 || ')' ELSE t.typname END as type, CASE WHEN
a.attnotnull= 't' THEN 'not null '::text ELSE ''::text END || 'default ' || CASE WHEN
a.atthasdef= 't' THEN substring(d.adsrc for 128)::text ELSE ''::text END as modifier
FROM pg_class c, pg_attribute a, pg_type t, pg_attrdef d
WHERE c.relname = 'tblplayer' AND a.attnum > 0 AND a.attrelid = c.oid AND a.atttypid = t.oid
AND c.oid = d.adrelid AND d.adnum = a.attnum
UNION ALL
SELECT a.attnum as number, a.attname as attribute, CASE WHEN t.typname = 'varchar' THEN
t.typname || '(' || a.atttypmod - 4 || ')' ELSE t.typname END as type, CASE WHEN
a.attnotnull= 't' THEN 'not null '::text ELSE ''::text END as modifier
FROM pg_class c, pg_attribute a, pg_type t
WHERE c.relname = 'tblplayer' AND a.attnum > 0 AND a.attrelid = c.oid AND a.atttypid = t.oid
AND a.attname NOT IN (SELECT a.attname FROM pg_class c,
pg_attributea, pg_attrdef d WHERE c.relname = 'tblplayer' AND
a.attnum > 0 AND a.attrelid = c.oid AND
a.atttypid = t.oid AND c.oid = d.adrelid AND
d.adnum = a.attnum)
ORDER BY a.attnum;
+++++++++++++++++++++++++++++++++++++++
yields
number | attribute | type | modifier
--------+---------------+-------------+-------------------------------- 1 | play_id | int4 | not null
defaultnextval('... 2 | play_name | varchar(30) | not null 3 | play_username | varchar(16) | not null
4| play_password | varchar(16) | not null 5 | play_online | bool | default 'f'
However, if I execute the same query and drop "a.attnum as number" from
the select part, it returns the following:
attribute | type | modifier
---------------+-------------+--------------------------------play_id | int4 | not null default
nextval('...play_online | bool | default 'f'play_name | varchar(30) | not null play_username | varchar(16)
|not null play_password | varchar(16) | not null
which is incorrect accoring to the initial query. It appears to be
ordering the individual selects and then appending the second query to
the first -- is this correct?
Thanks
Michael Fork - CCNA - MCP - A+
Network Support - Toledo Internet Access - Toledo Ohio
Bruce Momjian writes: > Excellent question. I suggested leaving core, but that would still mean > more than 1/3 of core people would be in one company. Our short-term > solution is to keep going until we see some problems. Our long-term > strategy is to increase the size of the core group. In the end, PostgreSQL is still controlled by those who contribute the work, so it doesn't matter who's in core and who's not. Your leaving core would certainly have been the worse "fix". But I'm pleased that this agreement was at least remembered. Those who hang around GNU toolchain/build tools development lists may recall occasional annoyances that design or implementation decisions are apparently made on Cygnus-internal mailing lists. Even if those cases might have been unintended in hindsight, this is the sort of stuff that one needs to be aware of. -- Peter Eisentraut peter_e@gmx.net http://yi.org/peter-e/
Michael Fork <mfork@toledolink.com> writes:
> However, if I execute the same query and drop "a.attnum as number" from
> the select part, it returns the following:
> ...
> which is incorrect accoring to the initial query. It appears to be
> ordering the individual selects and then appending the second query to
> the first -- is this correct?
I believe that this query should not be considered valid --- and, in
fact, current sources will return an error if you try to ORDER a UNION
result by something that's not one of the output columns of the UNION.
The issue is that if you are union'ing arbitrary queries together,
how do you decide what the ORDER BY expression means in the context
of each component query? Consider
select a, b from tab1UNIONselect c, d from tab2ORDER BY z;
Even assuming that there are columns named z in both tables, the ORDER
BY would be exceeding its authority to assume that those columns are
what is meant. Furthermore, since we're doing a UNION here, the result
will be reduced to just the unique output rows, meaning that there might
be more than one possible z value for each output row; so the sort order
wouldn't be well-defined.
It seems to me that for UNION-type queries we need to stick to the
letter of the SQL standard and only allow ORDER BY an output column
name. In my example you'd be allowed to do "ORDER BY a" or equivalently
"ORDER BY 1", but not "ORDER BY z".
Existing releases fail to defend against this situation, and produce a
plan that does who-knows-what. In current sources you'll get an error:
regression=# select q2 from int8_tbl union all select q2 from int8_tbl
regression-# order by int8_tbl.q1;
ERROR: ORDER BY on a UNION/INTERSECT/EXCEPT result must be on one of the result columns
regards, tom lane
On Tue, 10 Oct 2000, Michael Meskes wrote: > On Tue, Oct 10, 2000 at 10:15:03AM -0700, Alfred Perlstein wrote: > > I think anyone with doubts should take a good look at the initial > > companies backing Linux, (Redhat, VA, Debian) to see what a boon > > I certainly get your point but I have to correct this as Debian is not, > never has been and never will be a company. Sorry, couldn't resist. :-) and last I heard, RedHat doesn't necessarily have the best name ...
The Hermit Hacker <scrappy@hub.org> writes: > On Tue, 10 Oct 2000, Michael Meskes wrote: > > > On Tue, Oct 10, 2000 at 10:15:03AM -0700, Alfred Perlstein wrote: > > > I think anyone with doubts should take a good look at the initial > > > companies backing Linux, (Redhat, VA, Debian) to see what a boon > > > > I certainly get your point but I have to correct this as Debian is not, > > never has been and never will be a company. Sorry, couldn't resist. :-) > > and last I heard, RedHat doesn't necessarily have the best name ... Red Hat-bashing doesn't change the fact that Red Hat employees is by far the largest corporate (or other single entity) contributor open source projects. Project on which we contribute a lot include gcc, gdb (through former Cygnus and other employees, we are by far the biggest there), rpm, XFree86, glibc, gtk+, gnome, the Linux kernel and apache. We also try hard to feed patches back to the original authors when we fix something generic. Anyway, flamewars never serve any particular purpose - followups should go to /dev/null -- Trond Eivind Glomsrød Red Hat, Inc.
Bruce Momjian <pgman@candle.pha.pa.us> writes:
> more in voliation of the original plan.
>
violation ? Or is this just another gap in my knowledge of the English
language ?
Gunnar
As promised, the press release is at:
http://www.greatbridge.com/news/p_101020001.html
--
Bruce Momjian | http://candle.pha.pa.us
pgman@candle.pha.pa.us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Bruce Momjian wrote: > > As promised, the press release is at: > > http://www.greatbridge.com/news/p_101020001.html Wow. Nice. The combined qualifications of the six steering committee members are staggering. Me, I'm just a lowly broadcast engineer with ten years experience and a measly Bachelor's degree in Electronics Engineering Technology. Oh well. Even I have a job to do! -- Lamar Owen WGCR Intermet Radio 1 Peter 4:11
Gunnar R|nning <gunnar@candleweb.no> writes:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
>> more in voliation of the original plan.
> violation ? Or is this just another gap in my knowledge of the English
> language ?
You're right, he's wrong. We native English speakers are notoriously
poor spellers of our own language ;-)
regards, tom lane
> Gunnar R|nning <gunnar@candleweb.no> writes: > > Bruce Momjian <pgman@candle.pha.pa.us> writes: > >> more in voliation of the original plan. > > > violation ? Or is this just another gap in my knowledge of the English > > language ? > > You're right, he's wrong. We native English speakers are notoriously > poor spellers of our own language ;-) Oh, sorry. I didn't even see the spelling error. Yes, violation. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
On Tue, Oct 10, 2000 at 01:02:52PM -0400, Tom Lane wrote: > > Bottom line is we're not sure what to do now. Opinions from the > floor, anyone? > First, I include my voice with those congratulating Bruce, and wishing him the best of luck in his new position. I concur that no one who has paid any attention at all to how the core developers interact with the user base on the mailing lists can have any thing but the highest regard for all of your integrity and devotion to the project. Given that, there is little fear of overt actions by Great Bridge that could harm the project, short of firing you all and leaving you destitute on the street (which wouldn't last long, I'm sure ;-) As I mentioned in thread that followed Ned Lilly's first post here, the real threat to code quality will be management pressures, such as schedule pressure: when Management wants a new release so that Marketing can better sell against a competitor with a higher release number, what do you do? There are a dozen scenarios I could spin, each more fantastic than the last, but all grounded in someones real world experience. All the core members have more experience than I in the world of corporate coding, so can probably spin worse ones. How each of you handles those kind of pressures is up to your own internal compass: just let me remind you that, unlike most situtations where the individual is alone, the user community here can be a _personal_ resource, standing up for you, and providing an outside voice, if needed. Enough with worst case scenarios. As you said, Tom, the real problem is about the preception of the community. How to avoid misunderstandings? I think Peter's point about transparency of development _process_ is crucial. As it is, there have been in the past occasional back channel communications where design decisions get made, via IRC or phone calls. This in and of itself is not a problem: some problems are just easier to thrash out that way. The problem comes when the decision is presented as a fait accompli, without a clear public statement of the reasoning behind the decision. Sometimes it's easy to forget if a particular point got made on te phone, in a public email list, or a private, core list. This could easily spin out of control, if decisions get made over the water cooler, as it were. To date, the core developers have served as steering committee, as well. This is only natural on an all volunteer project: in that case, no one can order anyone to do anything they don't want to, so only the developers can direct the project. The Debian project runs into this all the time: Herding kittens, it's called. ;-) Now, the problem is that it is perceived that some one _can_ order the developers: analogous to the criticisms of electing John Kennedy as U.S. President, since he was Catholic, and therefore preceived to be under the Pope's control. That's, of course, extreme, but Tom himself has said that'd he'd work on bug fixes for paying customers over mailing list submissions. That's his right, and no different than a volunteer developer deciding that work or school assignments take precedence. It's happened to me, enough. But it's the perception that matter here, not the fact. What to do? Make as much communication as possible public. When in doubt, err on the public side. Develop in the fish bowl. If you feel there still a need for private channels, perhaps include some outside representitive, trusted by the community, who can serve as a witness of record, if you will, vouching for the intent of the communications, without having to reveal the content. Well, there's my nickel. Do with it what you will. Ross Ross J. Reedstrom, <reedstrm@rice.edu> NSBRI Research Scientist/Programmer Computer and Information Technology Institute Rice University, 6100 S. Main St., Houston, TX 77005 -- Open source code is like a natural resource, it's the result of providing food and sunshine to programmers, and then staying out of their way. [...] [It] is not going away because it has utility for both the developers and users independent of economic motivations. Jim Flynn, Sunnyvale, Calif.
> What to do? Make as much communication as possible public. When in doubt, > err on the public side. Develop in the fish bowl. If you feel there still > a need for private channels, perhaps include some outside representative, > trusted by the community, who can serve as a witness of record, if you > will, vouching for the intent of the communications, without having to > reveal the content. > We were talking about this today down here at Great Bridge, and I mentioned that there is very little that happens in the core group. Up until Great Bridge arrived, and we had to secretly communicate with them, there really wasn't much going on in core. Occasionally they will get upset with me about accepting too many patches, but other than that, months go by with nothing happening in core at all. So, I am really saying that core doesn't do much. You non-core folks aren't missing anything. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
On Fri, Oct 13, 2000 at 04:10:48PM -0400, Bruce Momjian wrote: > > We were talking about this today down here at Great Bridge, and I > mentioned that there is very little that happens in the core group. Up > until Great Bridge arrived, and we had to secretly communicate with > them, there really wasn't much going on in core. Occasionally they will > get upset with me about accepting too many patches, but other than that, > months go by with nothing happening in core at all. > > So, I am really saying that core doesn't do much. You non-core folks > aren't missing anything. Yeah, that's what you say in public ... There is no cabal! Seriously though, we're talking about perceptions. The once or twice over the last two years I've noticed sign of out-of-band decision making, it's usually been a mention of IRC or a phone call. No great shakes, just if someone's already (irrationally) upset about their 'great' design idea not getting in, and it's not clear why, but something happened in person (or via IRC or phone or core) ... I'd rather _not_ see the slashdot thread. Ross -- Open source code is like a natural resource, it's the result of providing food and sunshine to programmers, and then staying out of their way. [...] [It] is not going away because it has utility for both the developers and users independent of economic motivations. Jim Flynn, Sunnyvale, Calif.
At 16:10 13/10/00 -0400, Bruce Momjian wrote:
>
>So, I am really saying that core doesn't do much. You non-core folks
>aren't missing anything.
>
The issue here is transparency: doing the right thing as well as being seen
to do the right thing.
When someone devotes hours of time to PGSQL for no recompense, their
motives are generally not questioned. So when someone makes a design
decision, the motive is assumed to be because it is best in the long term
for the project. As soon as someone is paid to do work, their motive is (at
least partly) to get paid. As Tom has already said, this has the potential
to distort scheduling priorities.
The fear is that this may distort other priorities - hence why increased
transparency in decision making is important. If Bruce, Tom & Jan make a
design decision, then chances are it's going to be pretty good. The problem
is it will/may be seen as a GB decision.
As I said in an earlier post, any and all drive for increased transparency
has to come from GB; they must now be aware of the issues and potential
risks. Certainly planning to contribute *all* source is a very good sign
since it will drastically reduce the chances of making compromising design
decisions.
----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.B.N. 75 008 659 498) | /(@) ______---_
Tel: (+61) 0500 83 82 81 | _________ \
Fax: (+61) 0500 83 82 82 | ___________ |
Http://www.rhyme.com.au | / \|
| --________--
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/
> > So, I am really saying that core doesn't do much. You non-core folks > > aren't missing anything. > > Yeah, that's what you say in public ... There is no cabal! > > Seriously though, we're talking about perceptions. The once or twice over > the last two years I've noticed sign of out-of-band decision making, it's > usually been a mention of IRC or a phone call. No great shakes, just if > someone's already (irrationally) upset about their 'great' design idea not > getting in, and it's not clear why, but something happened in person (or > via IRC or phone or core) ... I'd rather _not_ see the slashdot thread. You should know that phone calls are my secret weapon. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
> At 16:10 13/10/00 -0400, Bruce Momjian wrote: > > > >So, I am really saying that core doesn't do much. You non-core folks > >aren't missing anything. > > > > The issue here is transparency: doing the right thing as well as being seen > to do the right thing. > > When someone devotes hours of time to PGSQL for no recompense, their > motives are generally not questioned. So when someone makes a design > decision, the motive is assumed to be because it is best in the long term > for the project. As soon as someone is paid to do work, their motive is (at > least partly) to get paid. As Tom has already said, this has the potential > to distort scheduling priorities. The majority of core discussions are closed because either we need to decide on a central direction for the project (release date) or we need to discuss something that would embarrass someone if it were publically known. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
At 22:12 13/10/00 -0400, Bruce Momjian wrote:
>
>The majority of core discussions are closed because either we need to
>decide on a central direction for the project (release date)
These are the things that you should consider making more transparent.
>or we need
>to discuss something that would embarrass someone if it were publically
>known.
Personal opinions are of course private. Can you think of an example of a
secret embarrassing item that has affected the direction of the project?
I'd be fascinated!
----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.B.N. 75 008 659 498) | /(@) ______---_
Tel: (+61) 0500 83 82 81 | _________ \
Fax: (+61) 0500 83 82 82 | ___________ |
Http://www.rhyme.com.au | / \|
| --________--
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/
> At 22:12 13/10/00 -0400, Bruce Momjian wrote: > > > >The majority of core discussions are closed because either we need to > >decide on a central direction for the project (release date) > > These are the things that you should consider making more transparent. > > >or we need > >to discuss something that would embarrass someone if it were publically > >known. > > Personal opinions are of course private. Can you think of an example of a > secret embarrassing item that has affected the direction of the project? > I'd be fascinated! You know, we have to take people aside once and a while and get them back on course. Of course, we do that for core members too. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
On Sat, 14 Oct 2000, Philip Warner wrote: > The fear is that this may distort other priorities - hence why > increased transparency in decision making is important. If Bruce, Tom > & Jan make a design decision, then chances are it's going to be pretty > good. The problem is it will/may be seen as a GB decision. I don't know ... the recent discussion in -hackers on the whole ALTER TABLE DROP COLUMN tends to show that even those "in the pay of" don't necessarily agree :)
"Ross J. Reedstrom" <reedstrm@rice.edu> writes:
>> So, I am really saying that core doesn't do much. You non-core folks
>> aren't missing anything.
> Yeah, that's what you say in public ... There is no cabal!
It's true that very little goes on on the private core mailing list,
and we try to keep it that way. I think that most of the power that
core has (such as it is) is that people on pghackers are willing to
defer to us on decisions like what the release schedule should be.
There are a dozen or more non-core people with CVS write access,
so it's not like core is tightly controlling what happens to the code.
I think ideally our role is one of cat herders, as you put it ---
making the kinds of decisions that a group of dozens or hundreds
can't make effectively. But the long-term direction of the project
is largely determined by what the individual CVS committers choose to
work on. In that sense, a core member has no more power than any
non-core committer. (Case in point: Peter E. has had more influence
on what 7.1 will look like than most of core ;-).)
When you look at it from that point of view, power comes from having
time to work on the code. In that sense, now that Great Bridge is
paying me to work full-time on Postgres, I personally may be the
most dangerous loose cannon on the deck. (Jan is less dangerous
right at the moment only because he's distracted by moving concerns.
Once he's settled again in Norfolk, look out...) Outer joins will be
in 7.1 because *I* decided that would be a good thing to work on ---
this wasn't a core decision, nor one imposed on me by Great Bridge.
I doubt anyone will complain too hard about that particular choice,
but further down the road I might make more debatable choices about
how to spend my time.
I agree 100% with your comments that openness of decision-making
is a critical element in keeping the trust of the community. But
looking at it as just an issue of core vs non-core is missing some part
of the problem. Everyone who contributes code has a responsibility,
proportionate to how much work they're doing, to ensure that the rest
of the community understands and approves of what they're doing.
regards, tom lane
> > At 22:12 13/10/00 -0400, Bruce Momjian wrote: > > > > > >The majority of core discussions are closed because either we need to > > >decide on a central direction for the project (release date) > > > > These are the things that you should consider making more transparent. > > > > >or we need > > >to discuss something that would embarrass someone if it were publically > > >known. > > > > Personal opinions are of course private. Can you think of an example of a > > secret embarrassing item that has affected the direction of the project? > > I'd be fascinated! > > You know, we have to take people aside once and a while and get them > back on course. Of course, we do that for core members too. In fact, we built a shed outside especially for Jan, who is in Poland giving a speech and can't possibly respond in a timely manner. :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
> On Sat, 14 Oct 2000, Philip Warner wrote: > > > The fear is that this may distort other priorities - hence why > > increased transparency in decision making is important. If Bruce, Tom > > & Jan make a design decision, then chances are it's going to be pretty > > good. The problem is it will/may be seen as a GB decision. > > I don't know ... the recent discussion in -hackers on the whole ALTER > TABLE DROP COLUMN tends to show that even those "in the pay of" don't > necessarily agree :) Man, Tom, our cover is working perfectly. Let's disagree on something again. That will really convince them. :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
> I think ideally our role is one of cat herders, as you put it --- > making the kinds of decisions that a group of dozens or hundreds > can't make effectively. But the long-term direction of the project > is largely determined by what the individual CVS committers choose to > work on. In that sense, a core member has no more power than any > non-core committer. (Case in point: Peter E. has had more influence > on what 7.1 will look like than most of core ;-).) Jan says that if I start coding more, GB will have to hire more developers to clean up after me. Now, is that supposed to make me feel valued? :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Philip Warner <pjw@rhyme.com.au> writes:
>> or we need to discuss something that would embarrass someone if it
>> were publically known.
> Personal opinions are of course private. Can you think of an example of a
> secret embarrassing item that has affected the direction of the project?
> I'd be fascinated!
There have been a couple of cases where core has decided that a committer
needed to be admonished ("yo, mon, why you committing new features
during beta freeze?" or something like that). Marc has generally done
the admonishing with a cc to core, but we don't embarrass people in
public. I don't propose to name names here for obvious reasons.
Another class of properly-private discussions have been reports of
security-related bugs; that sort of thing seems best not mentioned
too widely on the public lists until a fix is available. (BTW, if
you ever have a security bug report that you don't think ought to be
mentioned in the public archives, send it to pgsql-core.)
Dunno about "affecting the course of the project". I don't think that
core as core has all that much influence on the course of the project.
Timing (release schedule) yes, because people allow us to decree that,
but direction no.
regards, tom lane
> > I think ideally our role is one of cat herders, as you put it --- > > making the kinds of decisions that a group of dozens or hundreds > > can't make effectively. But the long-term direction of the project > > is largely determined by what the individual CVS committers choose to > > work on. In that sense, a core member has no more power than any > > non-core committer. (Case in point: Peter E. has had more influence > > on what 7.1 will look like than most of core ;-).) > > Jan says that if I start coding more, GB will have to hire more > developers to clean up after me. Now, is that supposed to make me feel > valued? :-) OK, why is no one laughing at my crafty jokes? Did core tell you guys not to laugh? Is that what they are doing? :-) -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Philip Warner <pjw@rhyme.com.au> writes:
> When someone devotes hours of time to PGSQL for no recompense, their
> motives are generally not questioned. So when someone makes a design
> decision, the motive is assumed to be because it is best in the long term
> for the project. As soon as someone is paid to do work, their motive is (at
> least partly) to get paid. As Tom has already said, this has the potential
> to distort scheduling priorities.
A side comment here: generally committers' motives are not questioned,
but what makes you think they're doing it for no recompense? I know
that when I first started getting involved with PGSQL, the first fixes/
changes I sent in were directly related to problems my then company
was having. Since most uses for databases seem to be business-related,
I suspect that most people who are involved with PGSQL have at least
some connection to a business need.
The real issue is how much control does any one entity exert, and if
it's a lot, is that entity driving things in a direction that other
people don't like?
regards, tom lane
On Fri, Oct 13, 2000 at 10:54:31PM -0400, Bruce Momjian wrote: > > Man, Tom, our cover is working perfectly. Let's disagree on something > again. That will really convince them. :-) > This was supposed to go to -core, right? (I see Bruce got my joke. I wasn't so sure Tom did ...) Ross -- Open source code is like a natural resource, it's the result of providing food and sunshine to programmers, and then staying out of their way. [...] [It] is not going away because it has utility for both the developers and users independent of economic motivations. Jim Flynn, Sunnyvale, Calif.
> On Fri, Oct 13, 2000 at 10:54:31PM -0400, Bruce Momjian wrote: > > > > Man, Tom, our cover is working perfectly. Let's disagree on something > > again. That will really convince them. :-) > > > > This was supposed to go to -core, right? (I see Bruce got my joke. I > wasn't so sure Tom did ...) Tom chuckles, even if he doesn't get the joke. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
Peter Eisentraut wrote:
> Bruce Momjian writes:
>
> > After careful consideration, I have decided to accept a job with Great
> > Bridge.
>
> Whatever happened to this:
>
> Date: Tue, 09 May 2000 15:19:48 -0400
> From: Tom Lane <tgl@sss.pgh.pa.us>
> To: Ross J. Reedstrom <reedstrm@wallace.ece.rice.edu>
> Cc: PostgreSQL-general <pgsql-general@postgresql.org>
> Subject: Re: [GENERAL] Steering committee responce to Great Bridge LLC
>
> : One thing we have agreed to is that there must not be an unseemly fraction
> : of core members working for the same company. With six people on core,
> : probably about two working at the same company would be a reasonable
> : limit.
Been in Poland for a week, so pardon for the delay.
Initially it was (if I recall correct) Vadim's and my idea.
The main reason behind it wasn't to avoid influence from
commercial entities into core. We've all been working
together for years as a group with great honour and trusting,
so the aims of all core members where never questioned. We
just decided this "fraction" to avoid any hireing to look
like a takeover.
This world is spinning a little fast at the moment. Let me
repeat what I said to a person I met last week in Poland,
during a PosrgreSQL conference in Wierzba. We have a BSD
license and now I know a good reason why we kept it all the
time. With that license in place, there's absolutely no
reason to panic right now. Nothing can be taken away, and if
things go wrong in the future, those left in the "OpenSource"
corner can start from our last official release again - be
sure I'll be somewhere in that corner, even if it might take
some time before I can surface again. And I'm sure I'll meet
all those I loved to work with together in that corner again.
Never underestimate the power of Open Source.
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== JanWieck@Yahoo.com #