[Zak]

I wonder why MySQL became super-popular and Postgres didn't. I remember reading for years about how Postgres had a bunch of stuff (e.g. transactions) that MySQL didn't (until recently) and that anybody who knew anything about databases would choose Postgres, yet people still chose MySQL.

What did MySQL have during the last decade that made it better for certain applications than Postgres?

[buro9]

MySQL was simple.

You didn't have to deal with schema/tablespaces, or the chance that someone used stored procs, or security (almost everyone just used the equivalent of a root account).

MySQL just worked.

For the most part, with zero monitoring, it stayed working until the server ran out of disk space or died.

MySQL was always available.

No matter which hosting company you looked at, no matter how little you wanted to spend, MySQL was pre-installed and ready for use.

I never thought it was the better choice, but I understood why it dominated. Once things like Wordpress, vBulletin, phpBB and other stuff was written only for MySQL the choice was then made for you.

I've long been a fan of applications that give you a choice of DB, but it's now unrealistic as many applications have designed schemas to cope with the quirks and limitations of MySQL specifically. These would now be harder to migrate than simply changing a database string, and most application devs don't invest in such things when what they have works well enough and is all anyone is asking for.

[Zak]

The second half of your explanation, which boils down to "it's everywhere" makes sense now, but is only true because it was already in demand.

That it was faster (for certain applications), easier and more reliable does make sense. I understand, for example why PHP got popular even though there were other options available at the time that I believe were better from a software engineering[0] perspective. Things get popular by being better at the things the majority of users care about, even if they're awful at everything else.

[0] Did I really just use that term? I can't think of a better one to describe the issue I'm talking about.

[sjs]

For many people, including hosting admins, MySQL was the path of least resistance. MySQL required less maintenance than Postgres back in the day. It's pretty much exactly the same situation as with PHP. The tool created by mere mortals is usable by mere mortals, so they use it. The tool created by scholars is technically superior at core functions but is slightly more difficult or cumbersome to use so is used by fewer mortals.

One thing people seem to forget is that Postgres has not been around longer than MySQL, both were released to the public in 1995. Postgres skipped a few version numbers so maybe that makes it seem older.

[stonemetal]

Counting public releases of non university projects sure, but at least according to the wiki page it has been under active development since 1986, with first release in 1989. Where Mysql started development in 1994.

[igorw]

Actually, phpBB 2.0.0 which was released in 2002 supported: IBM DB2, MS Access, MS SQL, MySQL3, MySQL4, Oracle... and Postgres 7.

But you are right, many of the other popular projects did not make this effort.

[epo]

MySQL is a triumph of branding and fanboy self-delusion.

MySQL made it easier initially by omitting lots of difficult stuff (data integrity, management, transactions) thus making easier to get something working quickly. It also had a reputation for being faster then PostgreSQL, a reputation spread by fanboys (in truth people who were bigging up the only database they had a faint comprehension of).

Then it became ubiquitious. Now, like some nasty rash, we can't get rid of it.

[masklinn]

> It also had a reputation for being faster then PostgreSQL

It was indeed much faster when reading (probably still is on MyISAM) due to the lack of everything (especially transactions and any concept of data integrity).

[lamby]

> MySQL made it easier initially by omitting lots of difficult stuff (data integrity, management, transactions) thus making easier to get something working quickly.

Remind anyone of NoSQL? :)

[jacques_chester]

MySQL (MyISAM, really) was faster for loading chunks of text and spitting them into Perl and PHP. Back in the 133Mhz days, people cared about that. Given that MyISAM was pretty much flat files in a clown suit, you didn't get much in the way of serious database functionality.

[Udo]

I can say from a web dev's perspective that the ubiquitousness of MySQL is only part of the allure (an important part though). MySQL makes simple stuff very easy to implement, it's a great solution if you just want your database to store your stuff and get out of the way. Sure, like any other DB it requires some expertise if you want performance, but it makes this incredibly simple. It just works and you don't spend a lot of time arguing with the software.

It also has a single feature that I find incredibly cool: it supports automatically generated row IDs. It's been a while since I last tried Oracle and Postgres, but at the time you had to create a second table just for a sequence number and it was extremely awkward to handle.

Could the auto-increment feature in MySQL be better? Sure, for example it would be nice to have an option where the row is identified by a hash value instead of a continuous sequence. But even in its limited form today it beats having to wrestle with the glued-together equivalent as implemented in "real databases".

[masklinn]

> It also has a single feature that I find incredibly cool: it supports automatically generated row IDs. It's been a while since I last tried Oracle and Postgres, but at the time you had to create a second table just for a sequence number and it was extremely awkward to handle.

Must have been a while indeed. Postgres's SERIAL datatype has been available at least since 6.4. Postgres 6.4 was released in 1998.

[wlll]

Postgres has had auto-incrementing row ids for a while, see this example: https://gist.github.com/1178237

[Udo]

I may be wrong, but isn't that just a syntactical shortcut for creating a separate sequence table?

[masklinn]

Yes (that's noted in the gist). But the end result's the same, instead of writing AUTOINCREMENT you write SERIAL. That's about it, the only issue may happen if you drop the table (the sequence will not be dropped, not a huge issue in practice)

[Udo]

Did you notice any differences in performance?

Edit: I'm not saying there are, but it feels like there might be some repercussions when you take the underlying implementation into account. It would be interesting to know if someone conducted comparative tests about this.

[wisty]

MySQL was popular for the same reason No-SQL is - it lacked features, but it was fast as hell with no tuning. Postgres did more, so it had to be slower, and it was tuned to work on minimal resources (and not steal all your RAM). Like MongoDB today, MySQL was tuned to work fast, at the costs of both reliability, and being a poor citizen on a shared host.

There are other factors - you could limit its use of disk space (so hosts liked it), but mostly it was a good key-value store with some database functionality.

People who wanted a "real" database used Postgres. People who wanted to write a web app used MySQL.

[Tichy]

I seem to faintly remember that in the beginning Postgres was Unix only, and MySQL was available for Windows, too.

[Nelson69]

Well, MySQL has been easier to set up, that cannot be underestimated; I've basically saved off a pair of configuration files I've just reused a dozen times to set up PostgreSQL. MySQL also works pretty fast right out of the box, if you data is smaller than physical memory, it might be nearly as fast as you can make it by default. PostgreSQL has fairly conservative performance settings when first installed, you can make it really sing but it takes some tuning and in some cases that tuning requires a fairly deep understanding of your data. Think most MySQL users are not DBAs, all the data specific settings for a PostgreSQL like database might frighten them some; it's a really odd juxtaposition when you think about it, stereotypically, geeks are drawn to the knobs and dials you can play with but I honestly think in this case it turned them away.

Then a little bit of support from ISPs and corporations and it's a no-brainer for most folks. All the oss CMS and blogs and everything support MySQL really really well. ALl the ISPs pretty much use MySQL as the "database" solution.

[smackay]

The query browser and admin tools were easier to use. They weren't really better, rather the presentation was simpler. The Postgres tools (pgAdmin) shows a little too much detail right away (namespaces, etc.) and looks a lot more complicated to use which might put off people who just need to get it working and build from there.

[enobrev]

When I first got into full-time pro web development (about a decade ago), I started learning PHP. Just about every tutorial I read or book I wished I could afford had 'MySQL' next to 'PHP' as though they were one in the same.

From there forward, any time I was looking into a new host, they ALL had both PHP and MySQL. Any job I was bidding, discussion I was having in IRC or on Forums, anywhere - the two came as a matching set. I can't say that's WHY it became more popular, but I've always assumed their popularity were closely aligned.

[sandGorgon]

Right now - the biggest problem is Wordpress which has ZERO postgres support. A PG-supported version of Wordpress will go a long, long way in popularizing Postgres.

[jacques_chester]

It'll never happen. Wordpress core might be ported, but there's thousands of plugins and themes that assume MySQL.

[dramaticus3]

MySQL got the wrong results faster.

[davidw]

I've always likened it to riding a bicycle downhill with no brakes. Yes, it's faster, but...

Of course, now they've converged a lot, with Postgresql getting significant speed upgrades, and Mysql having InnoDB, with a few 'minor details' like transactions.