We have bigger targets than beating Oracle, say open source DB pioneers MySQL pioneer Peter Zaitsev, an early employee of MySQL AB under the original open source database author Michael "Monty" Widenius, once found it easy to identify the enemy. "In the early days of MySQL AB, we were there to get Oracle's ass. Our CEO Mårten Mickos was always telling us how we were going to get out there and …
>He cited Greenplum, which used PostgreSQL as the basis of its data warehousing system, which he said limited success. "They were going to be so good, they said that, 'We don't need to stay with what the community is doing'."
I'd argue Greenplum failed because the company behind it had limited experience in building and selling data platforms and were never particularly well-funded, so they struggled to eat into Teradata's business. They were then were very *very* rapidly overtaken in price-performance first by the Hadoop-adjacent query engines and then ultimately the cloud native database platforms. All of which were and are "single vendor" products.
Lack of community had nothing to do with it.
-> but Oracle has been doing pretty good engineering in MySQL,"
Yeah, fixing a shed load of bugs and massively improving it over all.
-> "They don't do development in public," Zaitsev said.
MySQL was not always open source, was it?
Quite a bit of sour grapes in this article.
I wouldn't quite say that. The truth is that MySQL was never going to replace Oracle. Not only was it too buggy, it just didn't do the big things well. And, then there was the dual-licensing. Oracle definitely has improved the engineering and fixed many long-standing bugs. But the licensing has got more complicated, as with all the open source stuff that Oracle has kept. The real winner in that fight was Postgres which picked up developers and customers while MySQL's attention was elsewhere, being bought by Sun and then Oracle. I suspect existing commercial customers were happier with the changes. Elsewhere, the deficiencies in MySQL have offered opportunities for companies like Percona to help set up and run environments.
But the article is really a collection of quotes. And, yes, SaaS does make some comparisons more difficult but Oracle will still want to earn as much in rent as it did with licensing.
Fair enough. Your comment is much more reasoned than my mildly sarastic one.
-> The truth is that MySQL was never going to replace Oracle.
Yes, this was just hyperbole. It was never going to happen. MySQL has some good niches, and getting it up and running is a doddle. But it wasn't and isn't a drop in for Oracle.
Postgresql is a great pleasure to use.
I remember talking to an EA hiring rep at an event around 2010. She said they were thinking about Mongo. I told her they were WAY too big to be thinking about MySQL or Mongo at the time. I did not know then how badly folks wanted to avoid Oracle, or why. Still don't think I was wrong.
MySQL in 2010 was fine for a lot SMBs. Not fine for big businesses, which is why Bezos went with Oracle before he had the chops to build his own.
If you just need a database, just a transactional database to do a general job then Postgres or MariaDB.
MySQL carries the risk of Oracle coming after you with lawyers.
Postgres has spatial extensions, it's excellent and I love it. Even though I've made a lot of career out of commercial databases.
I chose my (shared) hosting provider in one aspect because it has PostgreSQL where most only seem to offer MySQL or maybe the MariaDB equivalent.
MySQL has some odd quirks with respect to embedded quotes. They do NOT follow the SQL standard for that, and it kinda bugs me to have to escape quote marks when it is legit to have "Joe's Computer" in a text field, or double quotes for that matter - the 3" valve, or the 1" pipe. Deaing with this for MySQL adds unnecessary complexity. (the standard SQL way is to double the quote mark if it is an actual embedded quote mark, within whatever context you use it).
So for that, at least, I prefer a database that follows the standard closely so as to be more predictable.
Also PostgreSQL seems to be extremely compatible with others, generating SQL statements (for a dump) that are generic enough for pretty much any DBMS to be able to import the data.
And MySQL (to me) has always been difficult to set up out of the box. Too much security diddly-fiddly crap, or at least was when i tried it a few years ago. Contrast to setting up many PG databases from scratch, and it "just works".
Interesting, The EDB perspective takes examples on what partially failed (Greenplum not well-funded, CockroachDB not able to innovate in the open source model). Yugabyte is building another distributed SQL database, fully open source, with fast innovation, but not mentioned here. Yes, the PostgreSQL API is more and more popular, especially as an alternative to Oracle, but we should not ignore the open source projects to distribute storage (like Neon) or storage and transactions (like YugabyteDB)
It seems written by and for managers.
In particular, that Olofson guy, you can see from a league away that he has no idea about software development.
That's the problem with these "marketing intelligence" companies: they think that a degree in marketing (which can be useful) or an MBA (which cannot) is a substitute for actually having some domain expertise, which is the only way you can properly do qualitative research and find out what is it that people really want.
The article is also not clear on *how* MySQL intended to compete with oracle.
It wasn't about offering all the bells and whistles that oracle has. It was about targeting the large sector of the database market which didn't need those bells and whistles.
That's why MySQL was hugely successful even though it (by design) never was a very good RDBMS. Their strategy paid off in that respect.
If technical excellence is what you're looking for, you can't go wrong with PostgreSQL. It's rock solid and a pleasure to use.
The one that isn't mentioned here, and probably should be, is SQLite.
Not only it's hugely popular: it's popular for a good reason.
In many common cases you get the benefits of relational databases without the inconvenience of RDBMs (Android, car entertainment systems and aircraft systems are good examples).
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
