Thread

Ugh. Does this news give anyone else the Fear?

http://www.infoq.com/news/2009/04/oraclebuyssun

Always wanted to get into PostgreSQL, looks like I now have the motivation to do so. That said, Oracle may keep MySQL running under the same licenses… either way it’s dead from now on I think.

I don’t think this will affect the average MySQL user much, to be honest: MySQL is dually licensed, if I remember, and one of them is the GPL. You can’t take that away, really, but if they insist on changing it, I see a code fork in their immediate future.

I reckon development on MySQL will come to a halt, and Oracle will use some of the technology in MySQL. OR…. MySQL development will continue but very slowly and just become a community project.

I reckon development on MySQL will come to a halt, and Oracle will use some of the technology in MySQL. OR…. MySQL development will continue but very slowly and just become a community project.

I couldn’t agree more. Unfortunately I think we can expect MySQL development to slow to a snails pace.

How depressing eh? Seems I’ve been waiting forever for MySQL to finally get GIS right as well, don’t see that happening anymore. While we all have to play wait-and-see for now, I wonder how the ExpressionEngines of the world will plan their reaction…

Only plus side is a big reason behind some of the alternative db engines that were being written is because Oracle purchased the rights to InnoDB a few years ago, and it was feared it would impact its availability in MySQL.  Times like these you really want to punch the people at Sun for purchasing MySQL a year or so ago…  Kills my hope that Sun was going to implement better, stronger clustering support within MySQL, unlikely Oracle is going to create a strong clustered open-source product to compete with Oracle RAC…  Scary to say probably the easiest to cluster DB server without having to change coding or SQL engine requirements is SQL Server from MS and does not look like that will change any time soon now.

A very smart strategic move for Oracle to kill MySQL.

Ugh. Does this news give anyone else the Fear?

Yep.

Oracle is an evil beast of a company which is singularly devoted to one thing. Huge profits. Secondarily, they’re committed to squashing competition. The MySQL we know and love is competition.

Sun hasn’t really been profitable since the dot com days, and their purchase of MySQL didn’t help much. It’s unlikely that Oracle will do much to help MySQL GPL (open source) grow and prosper. For the immediate future there won’t be much notable change to the MySQL we know and use. It’s the soon-to-arrive future that matters. Oracle would incur a great deal of worldwide wrath and legitimate ill will by mucking with MySQL’s licensing terms, but the snake is unlikely to change its spots (to mix a few metaphors).

There are already several forks or independent MySQL development efforts: drizzle, MariaDB, by Monty Widenius (blog), and an improved storage engine: Percona XtraDB. I think there may be others as well. Google also has a patch set for improved performance, so they may be in this game as well. These were all underway before the Oracle announcement. So I am not too worried.

There are already several forks or independent MySQL development efforts: drizzle, MariaDB, by Monty Widenius (blog), and an improved storage engine: Percona XtraDB. I think there may be others as well. Google also has a patch set for improved performance, so they may be in this game as well. These were all underway before the Oracle announcement. So I am not too worried.

The biggest concern is which of the branches do you trust once the primary branch of a project withers?  Larger players like Google will not feel anything as unlikely they are running stock releases from MySQL anyway.  But it takes some time to build trust to forked versions of projects, so if MySQL itself stagnates, which group do you then trust?  That would be the driving factor for people to move to other database platforms if/when that time comes.

There are already several forks or independent MySQL development efforts: drizzle, MariaDB, by Monty Widenius (blog), and an improved storage engine: Percona XtraDB. I think there may be others as well. Google also has a patch set for improved performance, so they may be in this game as well. These were all underway before the Oracle announcement. So I am not too worried.

The biggest concern is which of the branches do you trust once the primary branch of a project withers?  Larger players like Google will not feel anything as unlikely they are running stock releases from MySQL anyway.  But it takes some time to build trust to forked versions of projects, so if MySQL itself stagnates, which group do you then trust?  That would be the driving factor for people to move to other database platforms if/when that time comes.

I’m not a guru with MySQL so I don’t know the big picture very well, but I think MySQL will be fine.  The founder of MySQL left Sun and is working on Maria for instance.  I don’t know how much value there is in the lower end but I don’t think Oracle really isn’t much of a competitor in the space where most developers here work in.  I don’t know who uses Oracle products, but it’s not anyone I know, and their products don’t generate any buzz in my development circles.

What I think is more interesting is what will happen with Solaris.  Will Oracle push Solaris as their enterprise alternative to Redhat?  My feeling is that Solaris is probably superior to Linux but Linux is “good enough” so it will be interesting to see if Solaris can see increased adoption under Oracle.

Actually the major point is that MySQL as a company was making money selling support services to enterprises for $600 - $5000 per server annually, as a cheaper competitor higher prices of Oracle.  Will Oracle bother trying to sell two competing products to the same market? Not likely, they would be better off lowering slightly the entry point price wise to use Oracle, and axe funding for MySQL.  Sun did not buy MySQL because it wanted to simply support open source projects, you don’t spend a billion dollars to do that, MySQL has a revenue stream, and Sun could use it to sell servers as well.

I don’t think much at all will happen to Solaris, and actually almost every large scale Oracle installation I know of are running on systems like Sun e2900’s (ie: 1/4 of a million per server for a small configuration) and larger systems.  I don’t think I have seen or met anyone running Oracle RAC on a cluster of Linux servers.  Also Oracle is really not advocating the use of Redhat, rather they simply did the same thing CentOS did, take the Redhat Enterprise source and compiled it into their own version.  Another point is that Solaris has been open sourced for some time, http://www.opensolaris.com/ and Oracle could have used Open Solaris in much the same way they used Redhat Enterprises open source packages as well.

But Solaris has a defined value to Oracle as likely their biggest installations are using Solaris, and probably have been for decades.  Now they can sell them the hardware, OS support and database software.  Which again is why MySQL becomes sort of a bastard project, and why Oracle has yet to make any mention of it in any press releases I have seen, but they sure love their java (which is used in a lot of Oracle middleware stuff.).

If you notice the most viable and highly used open source tools all have businesses working on or have an existing revenue model.  The stronger that model is, the more likely businesses will base their projects on those open source projects, unless the business is large enough to simply take over their own development of those open source projects in house.

Again it will take some time to be apparent, but with Oracle’s track record… yeah well.

Nevin, thanks for weighing in.  The world of big enterprise just isn’t something I know much about but it still has it’s trickle down effects.

By the way, Here is a post from Monty (founder of MySql) about his thoughts on the whole situation.

The biggest threat to MySQL future is not Oracle per se, but that the MySQL talent at Sun will spread like the wind and go to a lot of different companies which will set the MySQL development and support back years.

Exactly my point, all Oracle has to do is pull funding from MySQL, no staff, no in-house development, people look for jobs elsewhere…  bad day indeed.  Or even better simply reassign them to projects revolving around Oracle db itself, making them leave even faster. 😉