Обсуждение: Is the MySQL situation an opportunity or a distraction?
The recent discussion of opportunities among the new orphans in the MySQL
community raises another issue in my mind-
Do we really want the MySQL users?
Will they be satisfied with us? -I'm not sure it is a good match.
In comparing the two databases, I've often likened MySQL to an economy car
and PostgreSQL to a reliable large truck:
MySQL is not equipped to handle large loads, but is simple to install and
operate. MySQL makes sense for folks who need a lightweight database that
allows them to ignore database administration and concentrate on developing
low-volume applications quickly.
On the other hand, PostgreSQL is the obvious choice if you have a
high-volume application- But you have to be willing to do a little learning
before you can operate it.
Given my perception of these niches, I've never really felt that the
databases were serious rivals- Each had a niche to fill and each fit
perfectly in that niche.
I've been rather glad to be in the company of the folks who have "learned to
drive the truck" because we're thus not distracted by folks who aren't ready
to take that step. My attitude has always been that we should let MySQL have
those folks- when their databases get big and they get serious about
administering them, they will naturally migrate over. I'm not putting down
this group- just noting that their needs and skills are different.
My fear is that too much outreach to the MySQL crowd will lead to PostgreSQL
acquiring a split personality and loss of momentum- If we work too hard on
the features that make PostgreSQL simple to install and administer, will it
suck resources away from the efforts to maintain PostgreSQL as the clear
leader in the areas of reliability and load-handling?
I'd suggest that early in the advocacy process, there should be some clarity
about the niche(s) that PGSQL is aiming for. This will help us sort out the
opportunities from the distractions.
I'm not necessarily concluding that MySQL is a distraction, but I think the
question is worth asking. (I have similar reservations about supporting MS
Windows.)
Josh- this thought is an outgrowth of the short thread we had on the order
of .conf parameter groups- The best order depends heavily on who we decide
the target user group is. Without that decision made, the whole discussion
is moot.
-Nick
"If you don't know where you want to go, any road will get you there."
---------------------------------------------------------------------
Nick Fankhauser
nickf@doxpop.com Phone 1.765.965.7363 Fax 1.765.962.9788
doxpop - Court records at your fingertips - http://www.doxpop.com/
On Thu, Jun 26, 2003 at 14:13:42 -0500, Nick Fankhauser <nickf@ontko.com> wrote: > > My fear is that too much outreach to the MySQL crowd will lead to PostgreSQL > acquiring a split personality and loss of momentum- If we work too hard on > the features that make PostgreSQL simple to install and administer, will it > suck resources away from the efforts to maintain PostgreSQL as the clear > leader in the areas of reliability and load-handling? That kind of work has already been going on. The default postgresql.conf is more likely to be suitable to what people have on their desk than it was previously. The slowness of IN has been fixed. Automatic vacuum is a contrib entry now. The problem with index growth on ever increasing columns has been fixed. Explicit joins no longer restrict join order by default. There are still some gotchas for new people (e.g. max and min), but it is getting easier to have reasonable performance without having to know anything.
Nick Fankhauser wrote: The recent discussion of opportunities among the new orphans in the MySQL community raises another issue in my mind- Do we really want the MySQL users? <snip> I think that is up to the users themselves to decide. It's hard to argue against a larger user base because that means more exposure which in turn means more money and contributions to the project. What is the downside besides posting stupid questions on the lists? Until recently there were still some valid reasons to choose mysql so a lot of people had to go that route, regardless of their skill level. Mysql still has a strong niche in the web publishing sector which is an attractive group because they put a lot of content out on the internet. Regards, Merlin
Nick, > Do we really want the MySQL users? > > Will they be satisfied with us? -I'm not sure it is a good match. This is a good question, and one I have often asked myself. On technical merits alone, I would be thrilled to have MySQL have their niche and us have ours. Unfortunately, the sales strategy of MySQL AB has made this impossible; either we "fight back" by expanding our user base at the expense of MySQL, or we risk becoming the BetaMax of OSS databases. Particularly since in 2 years or so MySQL will have significantly expanded "enterprise" capabilities, thanks to the merge of SAP-DB. The merge might kill them, but then again, it might not. > My fear is that too much outreach to the MySQL crowd will lead to PostgreSQL > acquiring a split personality and loss of momentum- If we work too hard on > the features that make PostgreSQL simple to install and administer, will it > suck resources away from the efforts to maintain PostgreSQL as the clear > leader in the areas of reliability and load-handling? As long as we maintain the perspective that "ease of use" is a third or forth priority, I don't see this as a problem. Our current list of priorities goes something like: 1) reliabilty/stability/efficiency 2) features 3) adherence to standards 4) ease of use But making it the 4th priority doesn't put it off the map ... far from it. > I'd suggest that early in the advocacy process, there should be some clarity > about the niche(s) that PGSQL is aiming for. This will help us sort out the > opportunities from the distractions. Sounds good to me. Your proposal? > I'm not necessarily concluding that MySQL is a distraction, but I think the > question is worth asking. (I have similar reservations about supporting MS > Windows.) Like it or not, Windows as a server OS will be with us for some time (about 8 years, my estimate). And do we really want to deny 20,000 DBAs who are confined to Windows by company policy the opportunity to use PostgreSQL? > Josh- this thought is an outgrowth of the short thread we had on the order > of .conf parameter groups- The best order depends heavily on who we decide > the target user group is. Without that decision made, the whole discussion > is moot. Well, given that Feature Freeze is Tuesday, I've already picked an order. You're welcome to re-hash it for 7.5. However, it looks likely by then that we will have one or more GUIs to configure PostgreSQL.conf. -- -Josh Berkus Aglio Database Solutions San Francisco
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday 26 June 2003 12:13, Nick Fankhauser wrote: > The recent discussion of opportunities among the new orphans in the MySQL > community raises another issue in my mind- > > Do we really want the MySQL users? > Yes. More users => more testers and more developers. More testers and developers => More users. We must crave new users as we crave new testers and developers. This is an exponential growth cycle. We will one day literally become "THE" database, just like Linux is fast becoming "THE" unix kernel. - -- Jonathan Gardner <jgardner@jonathangardner.net> (was jgardn@alumni.washington.edu) Live Free, Use Linux! -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQE++1ihWgwF3QvpWNwRAnhEAJ0aLCK0kOwPYyXBwShxZ34C9HdB+ACg0dXn 6+GR2ZAFiSAbBQNq0UwlgYM= =EVA0 -----END PGP SIGNATURE-----