Thread: MySQL Con 2008
Hello,
I have submitted a talk to MySQL Con 2008:
What MySQL Can Learn from PostgreSQL
This talk, was accepted. I would like to solicit professional (non
flaming) information on where people think MySQL can learn from PostgreSQL.
Here is the description:
Designed to deliver pointed information on the flaws in the MySQL and
PostgreSQL communities, what they can learn from each other and how they
both can improve the Open Source database landscape.
Here is the abstract:
A presentation delivering detailed examples of where MySQL has gone
wrong in its development of its most powerful resource, the community.
Practical examples will be provided including what PostgreSQL is doing
today in grass roots efforts to build, solidify and grow not only its
community but its contributor base as well.
Counter points will be provided to show where PostgreSQL has learned
from MySQL and is taking what it has learned further, through its world
wide network of supporters.
Specific topics will include:
* User group initiation
* Advocacy efforts
* The feature game
* The right way
* Building a community through coopetition (competition + co operation)
* How commercial friction invigorates the community
* Induction of newbies
Sincerely,
Joshua D. Drake
Josh, > This talk, was accepted. I would like to solicit professional (non > flaming) information on where people think MySQL can learn from > PostgreSQL. Well, frankly, I'm not much help. The reason why I never submitted a talk to MySQLCon, despite invitations, is that I couldn't think of content which would be both of interest to MySQLers and yet not improve MySQL's competitive position against PostgreSQL. You've chosen an even tougher row to hoe ... how does a corporate project create a real community when nobody in the community can ever be more than a 2nd-class citizen? I don't know what to suggest. -- --Josh Josh Berkus PostgreSQL @ Sun San Francisco
Josh Berkus wrote: > Josh, > >> This talk, was accepted. I would like to solicit professional (non >> flaming) information on where people think MySQL can learn from >> PostgreSQL. > > Well, frankly, I'm not much help. The reason why I never submitted a talk > to MySQLCon, despite invitations, is that I couldn't think of content > which would be both of interest to MySQLers and yet not improve MySQL's > competitive position against PostgreSQL. You've chosen an even tougher > row to hoe ... how does a corporate project create a real community when > nobody in the community can ever be more than a 2nd-class citizen? > > I don't know what to suggest. > Well you did just suggest. :)... Perhaps their community should no longer be considered 2nd class citizens. Remember... this is me you are talking about.. I can dodge any tomato that a dolphin lover is going to throw ;) Joshua D. Drake
Hi, On Tue, 2007-12-11 at 13:53 -0800, Josh Berkus wrote: > how does a corporate project create a real community when nobody in > the community can ever be more than a 2nd-class citizen? This may be a good item for the talk: Accept patches from community (even though I know that it will never happen because of dual licensing , but still...) Regards, -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/
Attachment
Devrim GÜNDÜZ wrote: > Hi, > > On Tue, 2007-12-11 at 13:53 -0800, Josh Berkus wrote: >> how does a corporate project create a real community when nobody in >> the community can ever be more than a 2nd-class citizen? > > This may be a good item for the talk: Accept patches from community > (even though I know that it will never happen because of dual > licensing , but still...) My understanding is they do accept patches from the community (I could be wrong). It is just that they require you sign over the rights to the patches, which is not all that uncommon. Joshua D. Drake
Hi, On Tue, 2007-12-11 at 13:17 -0800, Joshua D. Drake wrote: > What MySQL Can Learn from PostgreSQL They don't have Tom Lane, so they can't compete with us. -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/
Attachment
Hi, On Tue, 2007-12-11 at 14:06 -0800, Joshua D. Drake wrote: > My understanding is they do accept patches from the community I don't think so. Maybe Tom knows something about that -- he's applying some patches to Red Hat's MySQL which are not in upstream. Regards, -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/
Attachment
Devrim =?ISO-8859-1?Q?G=DCND=DCZ?= <devrim@CommandPrompt.com> writes:
> On Tue, 2007-12-11 at 14:06 -0800, Joshua D. Drake wrote:
>> My understanding is they do accept patches from the community
> I don't think so. Maybe Tom knows something about that -- he's applying
> some patches to Red Hat's MySQL which are not in upstream.
I've sent patches upstream by filing bugs in their bugzilla with
suggested fixes, and AFAIR they've adopted them (or some other fix)
... eventually. They've not been amazingly responsive. The patches I'm
still carrying are for things that I think they won't care to fix ---
either because they're obsoleting the code, such as their local copy of
bdb, or because it addresses Red-Hat-specific issues. It's not worth
my time to try to get them to adopt such stuff.
None of those patches have been large enough that anyone would think
there was an interesting intellectual-property issue in them. I don't
know what would happen if someone sent a large patch ... maybe they ask
for a copyright assignment?
regards, tom lane
On Tuesday 11 December 2007 18:04, Tom Lane wrote: > Devrim =?ISO-8859-1?Q?G=DCND=DCZ?= <devrim@CommandPrompt.com> writes: > > On Tue, 2007-12-11 at 14:06 -0800, Joshua D. Drake wrote: > >> My understanding is they do accept patches from the community > > > > I don't think so. Maybe Tom knows something about that -- he's applying > > some patches to Red Hat's MySQL which are not in upstream. > > I've sent patches upstream by filing bugs in their bugzilla with > suggested fixes, and AFAIR they've adopted them (or some other fix) > .... eventually. They've not been amazingly responsive. The patches I'm > still carrying are for things that I think they won't care to fix --- > either because they're obsoleting the code, such as their local copy of > bdb, or because it addresses Red-Hat-specific issues. It's not worth > my time to try to get them to adopt such stuff. > > None of those patches have been large enough that anyone would think > there was an interesting intellectual-property issue in them. I don't > know what would happen if someone sent a large patch ... maybe they ask > for a copyright assignment? > The do, although it's of questionable legal value in some locales (not ours though). Basically they just want rights to include the stuff in thier commercials package; the license agreement is pretty straightforward: http://forge.mysql.com/contribute/cla.php -- Robert Treat Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
Hello,
On Tue, 11 Dec 2007 13:17:21 -0800 Joshua D. Drake wrote:
> A presentation delivering detailed examples of where MySQL has gone
> wrong in its development of its most powerful resource, the community.
i have never seen a MySQL booth on any event (except MySQL events of
course) to talk with users, communiy, customers. They can't get in
touch this way with their user base.
> Counter points will be provided to show where PostgreSQL has learned
> from MySQL and is taking what it has learned further, through its world
> wide network of supporters.
Do more public relation. Make PostgreSQL known by ppl who might be
interested.
Kind regards
--
Andreas 'ads' Scherbaum
Failure is not an option. It comes bundled with your Microsoft product.
(Ferenc Mantfeld)
Not sure if people have seen this: http://www.kaltenbrunner.cc/blog/index.php?/archives/21-guid.html But it's showing 8.3 being 2.2x faster than 8.2. An interesting result, even if not generally applicable. Brian
Joshua D. Drake wrote: >...What MySQL Can Learn from PostgreSQL > ...information on where people think MySQL can learn from PostgreSQL. * Perhaps a bit on specific areas where both companies could cooperate for the benefit of both. One that comes to mind is that good and fair benchmarks would be good for everyone. Customers and end users would obviouslybenefit by seeing what use cases are better suited for each product; and developers of both systems would benefitby quickly seeing low-hanging fruit and existence proofs of optimizations. * Perhaps a bit on BSD vs (dual CPL/Proprietary) licensing. Note that MySQL's been on both sides on this since they rely on the owned-by-Oracle-but-available-by-GPL innoDB; and thepain they felt when Oracle bought Innobase mirrors the pain that MySQL-the-company's customers feel when dealing withMySQL's licenses. > Specific topics will include: > * User group initiation > * Advocacy efforts And to hopefully have such efforts focused on facts rather than spreading outdated FUD against the other products. > * The feature game Does standards compliance and the importance of SQL200X fit under features; or should it be a separate item? One more related area to features is identifying useful mysqlisms or postgre's QL's idiosyncrasies that both systems might want to adopt. This might make life easier on customers who use both products. I think the "replace/upsert" feature is one that some postgresql users ask for. > * How commercial friction invigorates the community Seems they understand commercial friction - their relational engine's owned by their biggest competitor. :-) > * Induction of newbies Ah yes - perhaps they'd like to adopt some of our community's welcoming initiation/hazing rituals such as "thou must abbreviate MySQL as Mys and pronounce it as Mice Q. L."
On Wednesday 12 December 2007 17:27, Ron Mayer wrote:
> One more related area to features is identifying useful
> mysqlisms or postgre's QL's idiosyncrasies that both systems
> might want to adopt. This might make life easier on customers
> who use both products.
I think the majority of this probably would fall under the sql standards bit;
ie. promoting things like use of information schema over show tables?
> I think the "replace/upsert" feature
> is one that some postgresql users ask for.
>
i think the only ones who do are mysql converts, given none of the other
commercial db's offer that syntax (afaik). note there are sql standard ways
of accomplishing it; though currently neither of the two systems support it.
--
Robert Treat
Build A Brighter LAMP :: Linux Apache {middleware} PostgreSQL
On Dec 13, 2007 12:56 AM, Robert Treat <xzilla@users.sourceforge.net> wrote: > > I think the "replace/upsert" feature > > is one that some postgresql users ask for. > > > > i think the only ones who do are mysql converts, given none of the other > commercial db's offer that syntax (afaik). note there are sql standard ways > of accomplishing it; though currently neither of the two systems support it. Huh? We run into this all the time from non-MySQL people. Oracle had UPSERT and now Oracle, DB2, and SQL Server all support MERGE. FWIW, I'd say it's a pretty big missing feature that can currently only be worked around in a hackish PL sort of way. I know Jan was kicking around some ideas for it, but I don't know what happened after that. -- Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324 EnterpriseDB Corporation | fax: 732.331.1301 499 Thornall Street, 2nd Floor | jonah.harris@enterprisedb.com Edison, NJ 08837 | http://www.enterprisedb.com/
Robert Treat wrote: > ie. promoting things like use of information schema over show tables? +1. I think it'd be good for the whole database world if the F/OSS communities encouraged use of standards wherever possible; since that'd result in users at large using the standards and pushing the commercial products to do so as well; reducing the friction between the systems for everyone. Spun in a "What can MySQL learn from Postgres", it seems to me they should by default use their most standard mode; http://dev.mysql.com/doc/refman/5.0/en/server-sql-mode.html and use the standard mode for their examples and docs. Spun in a "What can Postgres learn from MySQL", perhaps we should err on the side of following the standard even when it's insane (timestamp/timezone stuff) or inconvenient (case folding stuff); and make our own types/modes for people who want the non-standard (tho arguably better) modes.
Joshua D. Drake wrote: > Devrim G?ND?Z wrote: > > Hi, > > > > On Tue, 2007-12-11 at 13:53 -0800, Josh Berkus wrote: > >> how does a corporate project create a real community when nobody in > >> the community can ever be more than a 2nd-class citizen? > > > > This may be a good item for the talk: Accept patches from community > > (even though I know that it will never happen because of dual > > licensing , but still...) > > My understanding is they do accept patches from the community (I could > be wrong). It is just that they require you sign over the rights to the > patches, which is not all that uncommon. Yes, or they rewrite the patch. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://postgres.enterprisedb.com + If your life is a hard drive, Christ can be your backup. +