Slashdot Mirror
Mickos Says MySQL Code Better Than Ever Under Oracle
jbrodkin writes "Oracle hasn't done much to foster a community around open source projects, but the former CEO of MySQL said Oracle's expertise has helped boost the database to new heights from a technology perspective. 'Many in the community will ... feel that it's not as open and open source as it used to be and that's true,' Marten Mickos said. 'But the core product, the actual code, is in better shape than ever. And I think they will keep it that way.' Mickos, now head of Eucalyptus, left Sun before the Oracle merger because he correctly predicted that the company could not survive on its own."
I think the founder of MySQL would disagree, since he forked MySQL and started MariaDB. MySQL 5.5 was a long time coming and added quite a bit, but much of what it added was already in the stable MariaDB by the time it came out. Some of the linux distros such as debian are looking to add or switch to mariadb. I switched to MariaDB a while ago and development in MySQL looked like it was starting to stagnate. Not to go dragging out things, but look into Maria, it has quite a few bug patches, performance enhancements, features and such that MySQL lacks and may never have if Oracle splits off community development features from the "enterprise" version.
PostgreSQL will stay open, and stay strong.
Until Oracle buys them up too.
Unlike MySql, there seems not to be a single entity that owns the copyright for PostgreSQL. Meaning: Oracle would have trouble to buy all the copyrights, probably it will think twice before doing trying to do it (and at the secind round of thinking, will actually stop of even attempting).
Questions raise, answers kill. Raise questions to stay alive.
Oracle Enterprise is actually $47,500 per core base rate. Where it gets interesting is the fact that different processors are given different "core factors" based on their performance and/or Oracle's business interests that you multiply by the base price; so, for instance, Power7 and Itanium 9300 both have a core factor of 1.0 (full price) whereas various SPARC chips have 0.5 or 0.25.
I've been through lots of version upgrades in the 5.1 series with a couple of our managed hosting customers, and they simply don't appear to be able to make a stable release. One customer's car loan system segfaulted after 600-odd days (surpriiiise!), another seems to break it every 100 or so. The latter had a support contract with MySQL AB and I dealt with them personally - what seemed really worrying was, even though this customer was paying £6000 per year, it *still* didn't seem important to anyone at the other end that a "stable" / "general availability" release of their flagship database was segfaulting. I had filed bugs, with backtraces and sample data, offered them them root passwords so they could do whatever they needed to catch the bug, but no thanks, we can't take control of your server.
To anyone that might say "but why not use 5.5, surely 5.1 is ancient history!" I'd say - this customer has been through 4, 5.0 and 5.1 and not found a single release that will stay up for more than a couple of hundred days. This customer is a MySQL "power user" who got burned on every new feature that was introduced. Stored procedures, the geospatial functions, massive sub-SELECTS - anything new tends to crash it even more often than before, and he's often had to back out and rewrite features as a result. So major version upgrades aren't considered lightly.
MySQL is going to need more than a press release to convince me that they have a commitment to high-quality code. I'll continue to plan my installations around the assumption that it dies under heavy traffic.
Matthew @ Bytemark Hosting
it's not as open and open source as it used to be and that's true,' Marten Mickos said. 'But the core product, the actual code, is in better shape than ever
if it's not open source then the code is in worse shape then before