IBM Invests In MySQL/Oracle Competitor

stoolpigeon writes "IBM has made a move to support open source RDBMS PostgreSQL by investing in EnterpriseDB, a company that supports PostgreSQL as well as selling their own proprietary extensions to the database product. IBM participated in a $10 million funding round, though the article doesn't say how much they invested. In the past EnterpriseDB has primarily advertised itself as an Oracle competitor, though the article says, 'Derek Rodner, EnterpriseDB's director of product strategy, explained that Postgres Plus 8.3 also adds in new application quick starts which are supposed to help with installation issues. They will also help in EnterpriseDB's battle against MySQL for open source database supremacy.'"

Comments

geeks want to do it right

[Anonymous Coward]

Question: how do you properly pronounce "PostgreSQL"?

Re:geeks want to do it right

[totally bogus dude]

http://www.postgresql.org/docs/faqs.FAQ.html#item1.1

PostgreSQL is pronounced Post-Gres-Q-L. (For those curious about how to say "PostgreSQL", an audio file is available.)

Re:geeks want to do it right

[hunteke]

Or, the rest of us lazy folks just say Postgres (POST-grez)

Re:geeks want to do it right

[X0563511]

Yes, because Linux is so very hard to say. Same number of syllables as XP, and it sounds just like it reads.

Re:geeks want to do it right

[FatdogHaiku]

Here it is from the source. http://www.youtube.com/watch?v=5IfHm6R5le0

Re:geeks want to do it right

[drsmithy]

Yes, because Linux is so very hard to say. Same number of syllables as XP, and it sounds just like it reads.

There are at least three distinctly different pronunciations I've experienced:

Lee-nooks (Linus)
Linnicks (Americans)
Ly-nucks (Commonwealth English speakers)

And I'm sure everyone using them believes "it sounds just like it reads".

Re:geeks want to do it right

[slack_prad]

post greS sequel

EnterpriseDB also has Cloud Database service

[1sockchuck]

Interesting. EnterpriseDB was also in the news today for its partnership with Elastra, a startup that announced a "cloud server" that lets companies quickly create database applications on Amazon's utility computing platform. "In the future, enterprises will view massive capital investment in on-premise server infrastructure to support database applications as entirely optional," said Bob Zurek, chief technology officer of EnterpriseDB, which uses Elastra to run its EnterpriseDB Cloud Edition. Maybe all that IBM money has their head in the clouds.

Re:EnterpriseDB also has Cloud Database service

[TheLink]

Yeah, if I'm the owner of a company and not just some "Slash and Burn" CEO, I wouldn't want to have my core assets hostage to some third party _company_.

Having it in the hands of a trusted _person_ is different. If that person works for a different company, it's harder to ensure it's always that same trusted person who manages it.

Whereas if that trusted person works for you and the assets are in your company, it's a bit easier eh?

Re:EnterpriseDB also has Cloud Database service

[phantomfive]

I don't know, isn't that the whole premise of Akamai? They are a third party company who host a lot of major companies, and they seem to be reliable. I think the key issue here is *trust*. In either case you need to find an entity you trust, whether it is a person or a company. Arguably a company is better than a single person, since a truck factor of one is never a good thing.

Re:EnterpriseDB also has Cloud Database service

[totally bogus dude]

Akamai don't generally host the data, they just mirror it. Although they are the public face of your site and therefore you need to trust them, if you do start to get nervous about them you can just adjust your DNS so nobody uses their servers -- you're still in control of the first link in the chain, and you're still the original source of the content.

Having your data on Amazon's servers is more like having your email in a Gmail account. The best you can do is frequently back it up so you have a local copy, but since it's the live data you're always going to be slightly behind, and if the company hosting it decides to deny you access to it for some reason (legal, technical, bullying, incompetence) you're pretty much screwed until you can get the courts to force them to give you access to your data.

Another difference is that Akamai are caching data which is intended to be public (or at least semi-public), which may not be the case with a hosted database app. If you've got private data you won't be putting it on your website for Akamai to cache in the first place, and if you have a secure "members only" area there's a good chance that content will only be on your own servers, and not served by Akamai. But if all of your data (both public and private) is on someone else's servers, then you're trusting them to a) keep it secure and b) respect your privacy.

Re:EnterpriseDB also has Cloud Database service

[TheLink]

Banks are a better counter example to my OP. Companies have to trust banks with their cash and financial transactions.

So yeah I'm wrong.

Re:EnterpriseDB also has Cloud Database service

[AlecC]

I don't think it is company vs. individual, it is main focus vs peripheral interest. If looking after your data is either the person's or the company's priimary task, you are much more likely to get good, focussed service from that individual. If your data is only one of 47 differnt things the company, or the indiividual, does, and is not regarded as the main one, then inevitably you will get poor service, and may be dropped or otherwise inconvenienced when they regard other aspects of their business as more important.

Obviously, if you employ someone, it is easy to tell them that their primary job, on which their salary depends, is storing and searching your data. But you can say the same about a company whose sole focus is managing other people's data: if they screw you around, you can werck them by telling the truth about their service. That would not be true if your data were stores by some giant such as Google or Amazon whose main business is elsewhere.

A small, hightly specialised, company is likely to perform as well as, or better than, your single employee, but with a higher truck number. But keep an eye on their financials - if they go bust you could be truly screwed. It is in your interest for them to be profitable. And what you do when tney get takenover by Google, I don't know. But what you do when your one in-hous expert leaves is no easier.

Re:EnterpriseDB also has Cloud Database service

[rs79]

Yeah, if I'm the owner of a company and not just some "Slash and Burn" CEO, I wouldn't want to have my core assets hostage to some third party _company_."

Ya mean like with DNS?

Re:EnterpriseDB also has Cloud Database service

[LordLucless]

Any company that doesn't run their own hosting is in this situation, though. If you use shared hosting, managed hosting, or even run your own dedicated server in a third-party colo plant, someone else has access to it.

The only difference in this case is what data is being stored.

Re:EnterpriseDB also has Cloud Database service

[Forbman]

Hmm... Well, I work at a rather large Fortune 500 company, and our entire enterprise e-mail (Exchange, feh) infrastructure is outsourced to HP, and we're definitely not the only F500 company doing so...

Most larger companies DO outsource some core IT functionality to 3rd-party companies...

Gotta remember: most CEO, COO and CFO types view IT as a necessary, non-profit-generating, evil. Sure, most of us know that IT tends to be a profit enabler, but the PTM (Powers That Matter) see it only as an expensive function, even if their entire business model is wrapped around and 100% dependent on it. They're more than happy to give up a little bit of "3am control" to get a lot of control on expenses, especially if they can reduce IT headcount.

Re:EnterpriseDB also has Cloud Database service

[ckaminski]

And no less than three times in the span of a century, look how well THAT'S turned out.

MySQL databae supremacy

[scorp1us]

Now there's an oxymoron!

MySQL, while it has come a long way, still has a ways to go to rival PostgresSQL, technically speaking. By the time you enable all the atomicity, and PostgreSQL feature set, you arrive at worse-than PostgreSQL performance.

MySQL, while it has come a long way, still has a ways to go to rival PostgresSQL, legally speaking. PostgreSQL is BSD. MySQL is anything but. Sure, the community edition is free, but it cannot be used with commercial software. In fact, there's a special "open source exception" to the license. That's not really open source. Open Source would never make you pay server licensing fees for use in commercial software, it would only make you distribute your source at worst.

Re:MySQL databae supremacy

[chromatic]

What suggests to you that the terms "open source" and "commercial" are antonyms?

Re:MySQL databae supremacy

[thedarknite]

I think part of the reason for MySQL being so prevalent is the fact that it is probably the easiest backend database you can use for a web app. I've seen quite a few that have started life with MySQl and added in additional database support as the application has grown and matured.

Re:MySQL databae supremacy

[Daimaou]

I completely agree. PostgreSQL used to be a bit slow, but for the last few years, there just isn't a reason to choose MySQL over PostgreSQL. PostgreSQL's license is certainly better and all the great features and SQL standards compliance makes it a no brainer, I think.

Re:MySQL databae supremacy

[MrNaz]

2002 called. It wants its MySQList retort back.

MySQL is no longer easier to use than PostgreSQL. PG is now availabe for Windows with a nice packaged msi installer. It is as easy or easier to install under Linux/BSD/other POSIX, and is (if you assume the same level of experience with both system) far easier to administer.

Not only that, MySQL's community consists of many newbies, which makes getting help on complex issues difficult. PG on the other hand has a vibrant community consisting of highly skilled DBAs and the PG core developers themselves. I've often had help from the PG core dev team members. Finding similarly skilled MySQL help is like trying to find Dodos in Manhattan.

Re:MySQL databae supremacy

[Omnifarious]

MySQL, while it has come a long way, still has a ways to go to rival PostgresSQL, legally speaking. PostgreSQL is BSD. MySQL is anything but. Sure, the community edition is free, but it cannot be used with commercial software. In fact, there's a special "open source exception" to the license. That's not really open source. Open Source would never make you pay server licensing fees for use in commercial software, it would only make you distribute your source at worst.

This is a blatant distortion bordering on an outright lie.

MySQL community edition is licensed under the GNU GPL (version 2 even) and may be used inside any random commercial entity without having to distribute anything. And if you want to use MySQL in commercial software you may use it as long as you distribute source code for MySQL and any changes you have made to it.

It is true that the interface libraries are also covered by the GPL. But this can be gotten around easily enough by writing your own interface libraries, or having a GPLed thunk which speaks a proprietary protocol to your proprietary application and then uses the MySQL GPLed interface layer to talk to MySQL.

Personally, I consider PostgresSQL's license to be less free, and I'm disappointed in IBM for supporting it in any way. Ultimately IBM is throwing away their money by doing this. The article even tells you why PostgresSQL's license is less free. The company distributes proprietary extensions to PostgresSQL. They've essentially forked the code. If their proprietary extensions even become widely accepted and relied upon they have essentially rendered PostgresSQL no longer Free Software.

I'm surprised that IBM would be throwing away money on such a thing. They are essentially encouraging the development of another little monopolistic company making secret software that could ultimately hurt them very badly.

Re:MySQL databae supremacy

[jo42]

MySQL, while it has come a long way, ... Yep. It even supports "--i-am-a-dummy" startup option.

Joke you not.

Re:MySQL databae supremacy

[timmarhy]

your a dud.

http://www.postgresql.org/about/licence it uses the fucking BSD license, it doesn't get anymore free.

Re:MySQL databae supremacy

[thedarknite]

That may well be true now, because the last time I tried using PostgreSQL was nearly two years ago and it was a configuration nightmare that was poorly documented for windows. But if PostgreSQL is so much easier to use now than MySQL, why is MySQL still that default database for hosted websites, and why do most open source web applications that I've looked at recommend a LAMP/WAMP stack?

Incidentally, you do know that Slashdot runs on MySQL don't you?

Re:MySQL databae supremacy

[Spy der Mann]

It is true that the interface libraries are also covered by the GPL. But this can be gotten around easily enough by writing your own interface libraries

Yeah sure... we all do that (/sarcasm).

Re:MySQL databae supremacy

[Neil Hodges]

Good luck using MySQL for your next commercial product. If you run into problems there, perhaps Oracle or MS's SQL server will suit your needs better.

If you want truly free, you might as well go with SQLite. Sure, it'll be a little slower, but nothing's more free than public domain.

Re:MySQL databae supremacy

[Jonner]

MySQL is released under the GNU GPL. This means that it can be used and incorporated into commercial software as long as that sofware is compatible with the GPL. It is just as much Free Software and Open Source as any other code released under the GPL, like GCC and Linux. You don't have to pay anyone anything.

Any software linking with the MySQL client library must be compatible with the GPL, but I don't see any reason why a different implementation of the protocol would necessarily be bound by the GPL. So if you wanted to use MySQL to hold data for a GPL-incompatible application, such as a proprietary one, the most you should have to do is write a new client.

Re:MySQL databae supremacy

[Niten]

Sadly, there's still the whole WordPress thing -- the darn program was never intended to work with anything other than MySQL at the back end. At one point there was an effort to "port" WordPress to PostgreSQL, but that fork has long since stagnated. And adding support for other databases is not on the WordPress team's short list.

I wouldn't know the actual numbers any better than the next guy, but it's clear that WordPress is one of the top reasons MySQL retains such a dominant market share in the Web segment. Until WordPress adds support for multiple back-ends, MySQL will always be, at minimum, just as entrenched a product as WordPress is.

I hope that Movable Type's recent open-sourcing will eventually help effect more widespread adoption of PostgreSQL. Unlike WordPress, MT was designed from the ground up with forward-thinking features like database abstraction; it currently supports the Berkeley Database format, SQLite, PostgreSQL, and MySQL, and adding support for additional back-ends is relatively easy. Perhaps if Movable Type can chip away at WordPress's market share a bit, it will in turn help relax MySQL's stranglehold on the Web market.

Re:MySQL databae supremacy

[Zarluk]

If the purpose of the database is mainly for to answer some queries (i.e., mainly reads) and you are not using transactions, MySQL might be the right answer, as long as you don't need to use more sofisticated stuff, like triggers or transactions.

In theese cases, MySQL tend to give the user faster response times then PostgreSQL but, again, if you use PostgreSQL you are probrably also taking advantage of PostgreSQL more sofisticated fetures.

Once most websites just use a database to store data that is seldon updated, it's natural that most websites continue to use MySQL.

Re:MySQL databae supremacy

[MrNaz]

why is MySQL still that default database for hosted websites, and why do most open source web applications that I've looked at recommend a LAMP/WAMP stack?

Inertia.

Incidentally, you do know that Slashdot runs on MySQL don't you?

Yes, which is why they need to do such a large number of crazy voodoo tricks to scale.

Re:MySQL databae supremacy

[Ed Avis]

I think the EnterpriseDB extensions are for companies currently using Oracle who want a cheaper alternative. They do not affect the core Postgres development.

Note that MySQL AB is also free to distribute proprietary extensions to MySQL, since they own the copyright. And this is much more likely to affect MySQL core development, since you have the same company maintaining the free version and trying to sell proprietary addons.

Re:MySQL databae supremacy

[LurkerXXX]

As the other poster said, inertia.

I run a few mysql servers in addition to postgresql and mssql servers. I LOATHE mysql. Yet I use it in a few cases. Why? Because there are a few applications I need to run which were unwisely written to only support mysql. If postgresql or any other database support is ever added to them (or I ever find the extra time to add it myself) I'll switch in a heartbeat. But for now, since I need to run those applications, I am stuck using mysql.

So don't think every mysql server running out there is running it because the admin thinks it's the best or even just-as-good of a database. (It isn't)

Re:MySQL databae supremacy

[asc99c]

MySQL is a database that is very well suited to running a web application. That's why. Many web apps are generally going to have many more reads than writes. Additionally, the writes tend to be pretty simple and not require complex locking mechanisms.

E.g. my photos website uses gallery2 which works easiest on the LAMP stack. The main database queries are simple - what albums are there, does the user have permission to see them, what photos are in this album etc. The updates are similarly easy - add a new album, add a new photo, add a new user. For this, MySQL is perfect, simple and fast.

However once you start writing more complex systems with tougher constraints things become a lot harder. Typically, you have a record in a state that needs notifying to another system. You need to read that record, send a message to the other system and then modify that record to say the message is sent and the other system then has control over it. For this you need the various locking mechanisms and guarantees from the database or the two systems get out of line. MySQL just isn't the database to use for that sort of application.

Re:MySQL databae supremacy

[GooberToo]

This used to be true but no longer. The most recent release series of PostgreSQL stomps MySQL in almost every area of performance. I say, "almost" because it's possible some corner cases still exist. Now that PostgreSQL properly (native binaries vs Cygwin and fast/east installation) supports Windows, only a fool would use PostgreSQL for new projects.

MySQL uses lots of non-ANSI SQL, teaching poor SQL habits. MySQL is feature poor compared to PostgreSQL, requiring involved work arounds to do what is easy in most other RDBMs. PostgreSQL's performance now completely rocks across the performance and scalability (PostgreSQL always was ahead here) spectrum.

The only thing preventing MySQL users from migrating to a superior platform is poor, non-ANSI SQL learned from using a crappy MySQL platform in the first place.

Re:MySQL databae supremacy

[0xdeadbeef]

It is true that the interface libraries are also covered by the GPL. But this can be gotten around easily enough by writing your own interface libraries, or having a GPLed thunk which speaks a proprietary protocol to your proprietary application and then uses the MySQL GPLed interface layer to talk to MySQL.

Wow, freedom sure sounds complicated.

In other words, "The GPL of MySQL is awesomer because, with a lot of work, you can violate its intent without violating the license!"

Re:MySQL databae supremacy

[Daniel832US]

The other good thing they provide is commercial support for PostgreSQL. I've been tinkering with some software of my own to publish. I've decided to stick with postgres for the fact that you can distribute it with for-profit software without reservation, but I don't want to get into the database configuration business. This way I can say "I'm giving you PostgreSQL and EnterpriseDB to install. If you're a small business then postgres should work fine. If you think this is mission critical enough, then get the support from EnterpriseDB". This way I can concentrate on the application and not field calls about how to do detail configuration.

I just looked at their web site and it looks like they're changing their theme from "Oracle compatable for less" to "Postgres Plus" and it looks like they've bringing Mac compatability back :) I like software that is available everywhere.
http://www.openoak.com/

Re:MySQL databae supremacy

[MrMunkey]

The only feature that has been keeping me from using PostgreSQL in work projects is replication. My company requires that there be multiple backups at all times, and MySQL's replication has done a pretty decent job (mysqldump -u user -p -master-data -databases db1 db2 -create-options > file.sql allows me to create a new slave quite easily). I do know there are projects out there, but they're just not quite ready yet: http://edoceo.com/liber/db-postgresql-replication

Feel free to let me know if there's another way to do this, because foreign keys would be great :)

Re:MySQL databae supremacy

[encoderer]

Uhh... I'm using triggers. And transactions.

On MySQL.

And I have been, for 3 years.

Re:MySQL databae supremacy

[ShieldW0lf]

That may well be true now, because the last time I tried using PostgreSQL was nearly two years ago and it was a configuration nightmare that was poorly documented for windows. But if PostgreSQL is so much easier to use now than MySQL, why is MySQL still that default database for hosted websites, and why do most open source web applications that I've looked at recommend a LAMP/WAMP stack?

You get MySQL instead of PostgreSQL by default on hosted websites for the same reason you get a WYSIWYG config panel instead of root access. Power comes with responsibility. The person selling you those services wants you crippled to the point that you can't screw things up on their server because they don't trust you. Therefore, they give you a toy database that you can't hose their server with. By extension, the reason most of those cutesy little web apps recommend MySQL is because they were written to go on cheap hosting, and hit critical mass because there are a lot of cheap people looking for cutesy web apps.

You do know that Slashdot data has less requirements for atomic operation than your average mom-and-pop e-commerce site, right? It's easy to set up fast multi-master replication across a cluster if you decide to relieve the system of the requirement to honor data integrity constraints, but it's not much use if your problem can't be solved with a "close enough" database.

Re:MySQL databae supremacy

[untouchable]

In what universe is the BSD license less free than the GPL, of any version???

BSD - Basically, do whatever you want, just include the credits of previous contributors. http://www.freebsd.org/copyright/license.html/

GPL - Basically, as basically as it can be stated, if you modify the code source, " . . .However, nothing other than this License grants you permission to propagate or modify any covered work." If you start in gpl, you're stuck in gpl. http://www.gnu.org/copyleft/gpl.html/

Which one is more free?

Re:MySQL databae supremacy

[rbanffy]

"but it cannot be used with commercial software"

That's incorrect.

It's incorrect because "Free" and "Commercial" are not mutually exclusive. You can pay for free software, if you are so inclined. In comparison with closed software, you get so much more (you get the complete source for it) it's surprising it doesn't cost more. The folks at Red Hat may also enlighten you about this.

It's also wrong because you can use MySQL with commercial software. What you can't is to embed MySQL code within a non-free application unless you pay MySQL AB for the different license.

Re:MySQL databae supremacy

[rbanffy]

What worries me with BSD licenses is that anyone can pick up the software you helped to build, improve it (or not) a little bit, and start competing and fragmenting the market with a closed-source alternative product.

The GPL strikes a balance between what you give to a given project and what you get in return. You give work and, in return, you get other people's work. It's a very good deal and it's responsible for a lot of the progress in free software.

Re:MySQL databae supremacy

[plopez]

have you assessed this for your needs?

http://slony.info/

Re:MySQL databae supremacy

[Omnifarious]

The GPL is more free because it preserves freedom. The BSD license is less free because it allows that freedom to be taken away.

Re:MySQL databae supremacy

[MSG]

The release notes for 8.3 mentioned that both Bucardo and Slony-I replication packages are included. Both of those are also usable on releases older than 8.3.

http://www.postgresql.org/about/press/features83.html

Re:MySQL databae supremacy

[portnoy]

And yet, I can take the PostgreSQL code, create some new extensions under the GPL, and then license the result as a GPL work. So, um, how again is the BSD license less free than GPL, if I can create a GPL'd work from it?

Re:MySQL databae supremacy

[eric2hill]

Now that PostgreSQL properly (native binaries vs Cygwin and fast/east installation) supports Windows, only a fool would use PostgreSQL for new projects.

Did you mean MySQL? :)

Re:MySQL databae supremacy

[CrazedWalrus]

I work for a large investment bank, and I've got PG handling a departmental portal, documentation wiki, and production statistics collection for publication to the portal. When I wanted a CMS to run the portal, I first looked to Wordpress because I was using it on my personal web site. When I found out they didn't support PG, I found Drupal and haven't looked back. Drupal kicks ass in every way, and I'm glad that WP's lack of support for PG forced me to give it a try.

Re:MySQL databae supremacy

[nxtw]

I think part of the reason for MySQL being so prevalent is the fact that it is probably the easiest backend database you can use for a web app. I've seen quite a few that have started life with MySQl and added in additional database support as the application has grown and matured.

No, the easiest right now would be SQLite. On Windows, the JET Engine (MS Access) has been prevalent in the past, but is being overtaken by SQL Server (which can be used like SQLite or JET - by "connecting" to a file instead of a database on a server).

Re:MySQL databae supremacy

[MrMunkey]

Great. Thanks for that info. I haven't been keeping a close watch on the project.

Re:MySQL databae supremacy

[h3]

Serendipity supports Postgres and has for a long time (though I can't say for certain that it did from birth). Which is why I use it. As with other projects of this nature (Drupal comes to mind), though the core supports Postgres quite well, 3rd party plugins sometimes fall into MySQL-only habits, but that hasn't been too problematic for me.

Frankly, in this day and age, I'm leery of projects that are written to MySQL specifically. To me, it smacks of amateurdom: if you don't know enough to use an abstraction layer (of which there are jillions of options), what does that say about the quality of your code as a whole?

Re:MySQL databae supremacy

[Rich0]

That and the fact that software inevitably uses the mysql runtime libraries and you'd need to patch half an application to make it work in postgresql as a result.

For all that MS does wrong, one thing they did right is getting everybody to use ODBC for everything (well, almost everything). If you stick to ODBC and ANSI SQL your app will work with just about any backend. Sure, similar solutions might exist for unix, but nobody uses them.

If mythtv officially supports anything other than mysql then maybe I can consider switching. However, I really don't want to run two databases for everything!

Re:MySQL databae supremacy

[ShieldW0lf]

Perhaps it says that you know your way around a DB enough to know not to use database abstraction layers, but rather to marry your application to a proper database.

A good web application stores all its data logic in stored procedures, denies the middle tier user access to anything except those stored procedures, and builds the application to use the procedures.

A bad web application sticks to the minimal SQL featureset, has embedded SQL commands that are dynamically generated in code and moves a lot of unnecessary data forth from the database to the middle tier.

I would go so far as to say that if your application is capable of working with MySQL5, it was poorly designed.

Re:MySQL databae supremacy

[untouchable]

The GPL enforces what it considers 'freedom'. It traps whatever gets into its license, like a venus flytrap. That is not freedom.
It's like holding someone hostage at Buckingham Palace and then stating it's not a hostage situation since the palace is so big and nice and extravagant. It may be a nicer situation than being held hostage than in someone's basement, but it's still a hostage situation. That's the GPL.
Not the greatest analogy, but close.

Re:MySQL databae supremacy

[Omnifarious]

I consider it more like the constitution of the US which restricts the actions of government in order to keep people from voting themselves censorship or other such stupidities. It too restricts freedom in the strict sense. But in the broader sense it does more to guarantee freedom than to restrict it.

I find it really interesting that the most vocal proponents of the BSD license seem to be people who want to sell my own work back to me at an exorbitant markup because they've made a few secret special changes to it. The BSD license seems to me to be a great detriment to the continued freedom of my work.

Re:MySQL databae supremacy

[HAWAT.THUFIR]

by this logic, the BSD license is better than the GPL. Says who?

Re:MySQL databae supremacy

[DragonWriter]

The GPL is more free because it preserves freedom. The BSD license is less free because it allows that freedom to be taken away.

So, the GPL is "more free" because it gives the licensee less freedom, and imposes more of the licensor's ideology on the licensee while the BSDL is "less free" because it gives the licensee more freedom, and imposes less of the licensor's ideology on the licensee.

Re:MySQL databae supremacy

[chromatic]

Yes, which is why they need to do such a large number of crazy voodoo tricks to scale.

Are you seriously comparing PostgreSQL's replication and clustering support to MySQL's?

Re:MySQL databae supremacy

[smellotron]

MySQL might be the right answer, as long as you don't need to use more sofisticated stuff, like triggers or transactions.

Or complex joins, multi-level joins, or functions that return recordsets (essentially "efficiently-parametrized views"). My last attempt at creating views within views resulted in a LEFT OUTER JOIN somehow transforming itself into an INNER JOIN, forcing me to "inline" the entire SQL query into a single view. Oh yeah, and the performance wasn't so great, and the MySQL "query explanation" capabilities suck, so I had a much harder time improving said performance.

Re:MySQL databae supremacy

[Omnifarious]

Yes. Would you rather live in a country with laws or without them? In which do you have more freedom?

Re:MySQL databae supremacy

[smellotron]

I'm leery of projects that are written to MySQL specifically...if you don't know enough to use an abstraction layer...

It's more than just the abstraction layer. There are two levels to deal with:

• API functions used to access the database. For example, mysql_*() and pgsql_*() functions in PHP should be replaced with PDO for consistent access regardless of database.
• SQL syntax and functionality used. ADOdb solves this by providing PHP functions for building specific SQL portions (e.g. using LIMIT for MySQL and Postgres, but something else for other database engines).

I'm all for using PDO, but I looked at ADOdb and decided I'd rather use my database rather than attempt to hide it under the covers. Plus, something like PDO doesn's support advanced database features, like Postgres's LISTEN and NOTIFY support, which can be very valuable for basic asynchronous communication between e.g. a web interface and any particular backend daemon.

Re:MySQL databae supremacy

[Zarluk]

You misunderstood me, I choosed PostgreSQL as the DBMS for my company 8 years ago, and I don't regret a single bit!

But f you stick to the ISAM engine, MySQL has to perform better than a true DBMS. I've used both, and that's my private statistics ;-)

Once we're developing a ERP system, it would be insane to use MySQL (no stored procedures, no triggers, strange SQL sintax...).

Re:MySQL databae supremacy

[MrNaz]

Yes, I am. Granted, to get a large scale operation running requires more skill with PG than it does with MySQL, but if you're running a large scale operation and you're short on skills, a complex replication solution is the least of your worries.

Re:MySQL databae supremacy

[GooberToo]

Should have read, "wouldn't use PostgreSQL". I make /. posts in such a hurry such mistakes are common for me.

Interesting...

[jonaskoelker]

Here's a few random thoughts:

Having recently seen Sun buy MySQL, this looks a lot like a "me too"-move. That's not to say that it doesn't make business sense.

Last I checked, IBM makes its money from two things: hardware and support. Note that software is not one of them; the software is (to them) merely what enables them to sell their bread and butter. It's also costing them money to develop and maintain software that drives sales.

That's why they've invested money in Linux, and that's why they're investing money in Postgres: offering software with a good track record and a good reputation drives sales better, and cost is driven down as the software is open source.

Re:Interesting...

[Cosmic Debris]

No money to be made on software, eh? Don't tell Steve Mills that. He's been working under the assumption that IBM is one of the world's largest software companies and that it's quite profitable, thank you.

I know this for a fact. And btw, when did you last check your figures? Take a look at IBM's 2007 annual statement and get back to me.

Since your thoughts are random, I'll assume you're using Microsoft's Random Number Generator.

Re:Interesting...

[TheLink]

Yep.

I think IBM makes more money from customers having more choices than they can cope with. Then those customers pay IBM to help them decide :).

That's why they are happy to provide the market with tons of different choices. Java, .Net, Linux, Windows, x86, RISC, Mainframes etc. Something has got to be pretty crap/loss-making for IBM to drop support for it ;).

And then as you say IBM provide consulting+support services and the hardware to handle all the zillions of combinations of choices ;).

Re:Interesting...

[perlith]

"Last I checked, IBM makes its money from two things: hardware and support. Note that software is not one of them; the software is (to them) merely what enables them to sell their bread and butter. It's also costing them money to develop and maintain software that drives sales." Check your facts. 20% of IBM's income is from software and growing. Annual reports are good things to read if you are an investor: http://www.ibm.com/annualreport/

Re:Interesting...

[JAlexoi]

Sun also has some Postgres core developers on payroll.

Re:Interesting...

[khakipuce]

Not only did Sun buy MySQL, Oracle is after BEA (Weblogic, amongst other things). Now, from a Java perspective, Sun used to be the langauge provider, Oracle seemed to have become the de-facto database to run heavy weight Java applications against and IBM Websphere, or BEA Weblogic was the app server.

I just wonder if these guys are all about to explode the Java App server space (watch out for shrapnel), and try to drive customers down either:
Sun - Glassfish, MySQL
Oracle - Weblogic, Oracle database
IBM - Websphere, PostgreSQL

What is more concerning is that it seems likely that such fragmentation will cause non-technical management to run for cover under .NET

BTW whilst Sun has Solaris and IBM has AIX they all have Linux on which to run, and I guess that Oracle just isn't bothered about getting into hardware.

Re:Interesting...

[DannyO152]

Let me add, I don't believe that supremacy, mentioned in the summary, is neither necessary nor sufficient for a product's profitability.

Re:Interesting...

[jadavis]

Having recently seen Sun buy MySQL, this looks a lot like a "me too"-move.

"Astor [Andy Astor, CEO of EnterpriseDB] added that the discussions with IBM pre-dated the $1 billion acquisition of MySQL by Sun (NASDAQ: JAVA) earlier this year."

        -- http://www.internetnews.com/software/article.php/3736336

db2...

[fatp]

Why isn't this a competitor of db2?

Re:db2...

[DougReed]

Because DB2 isn't really a competitor. It's not really a bad database, but I think only three people in the world is running it. ...my wife being one.

Re:db2...

[afidel]

DB2 just doesn't scale down as well as some of the others so it doesn't get as much exposure to the masses, if you check out things like the TCP-H results you'll notice at the 10TB level DB2 is #1 and #3, it's typically used for very large databases running on IBM big iron. It's yet another IBM technology that kind of sits in the corner running some of the largest financial systems in the world without getting a lot of exposure.

Re:db2...

[TheLink]

It's not rare for IBM to sell stuff which competes against products its subunits make.

You can have one IBM unit recommending/selling Cisco products which compete against more expensive IBM products by another IBM unit. You need some Sun stuff to work with some Microsoft stuff? IBM will say they'll do it.

From what I see, IBM is about providing choice, and helping customers make that choice for $$$$ :).

If there isn't much choice you don't need as much "consulting" and support. For example if your choices are: reinstall, or format and reinstall, I don't think you'll want to pay a lot.

Re:db2...

[firefly4f4]

Why isn't this a competitor of db2?

For the same reason PostgreSQL and MySQL aren't really competitors to Oracle.

There might be SOME crossover, but one database system (MySQL, PostgreSQL) is aimed at user performing simpler tasks (web forums, home users) which don't necessarily need all the features of the larger products (transactions, large numbers of simultaneous users, data integrity checking), whereas the other (DB2, Oracle) is aimed at business users who require those full-fledged features.

Now, I'm not saying that PostgreSQL and MySQL don't provide some of the features, but people will choose what best suits their needs. You won't see banks running PostgreSQL/MySQL on for their financial transactions, just like you most likely won't see DB2/Oracle running as the backend of your (run-of-the-mill) web forum.

Disclaimer: I work for IBM. These thoughts are my own, and may not represent those of the company.

Re:db2...

[Heembo]

IBM is all about cramming the absolutely most asinine complex and difficult to support solutions into your company. After IBM consultants had had their way with your project, you will need to pay them 250-350/hr utilizing their professional services division to support the crap that crammed into your infrastructure. IBM is the IT mafia - absolute bastards!

Re:db2...

[firefly4f4]

I believe DB2 manages more data than Oracle, whereas Oracle has more installations. Disclaimer: I work for IBM. The thoughts posted are my own, and do not necessarily reflect those of IBM.

Re:db2...

[ryszard99]

..which don't necessarily need all the features of the larger products (transactions, large numbers of simultaneous users, data integrity checking...

I'm a postgres user and i know your generalising here, but postgres has supported transactions for some time, and foreign keys. I also used to work in one of the major 4 banks in australia, and altho' i wasnt privy to all the purchasing details, i believe one of the major reasons oracle was chosen, (and infact most business critical software and hardware), is support from vendors and direct access to hardcore mofo's that know what to do when "once in a decade" bugs are found.

as others have mentioned postgres replication sucks, but other than that, my experience is that postgres is a more than capable database for the enterprise.

Re:db2...

[odourpreventer]

These thoughts are my own

Fortunately not. Oracle and MySQL operate in two different worlds. Postgres operates somewhere in between, but much closer to MySQL than Oracle. I really don't see the point of trying to compete with both products, it's just not cost effective.

Re:db2...

[JAlexoi]

Any IT service company is IT mafia. It's just a result how deep you let them reach into you company.

Re:db2...

[JediTrainer]

Because Informix is going to take it all.

Re:db2...

[ceswiedler]

Postgres, and especially EnterpriseDB, are basically open-source clones of Oracle. They're designed along the same lines and support very similar features as Oracle. EnterpriseDB for example has added support for Oracle's PL/SQL language, which means that many apps written for Oracle can be ported to EnterpriseDB without too much effort.

There are a lot of people out there using Oracle because that was one of the only real options five years ago, but don't really need it's high-end features, and certainly don't want its high-end price. EnterpriseDB is a very good choice for them.

Re:db2...

[Ctrl V]

There are a lot of people out there using Oracle because that was one of the only real options five years ago, but don't really need it's high-end features, and certainly don't want its high-end price. EnterpriseDB is a very good choice for them. In the last couple years, Oracle has been going after the smaller markets to some extent... I saw the release of their Express Edition database as targeting those who would otherwise defer to MySQL/PostgreSQL because of $$$. Express Edition is free, supports all the core Oracle features (though with some limitations like 4GB db size and 1GB mem usage max).

It's a pretty good option in many applications.

Then (before?) they did a licensing option (Standard Edition One) that seems to be targeted at MS SQL I think. Much cheaper than the $40k/CPU Enterpise Edition.

Re:db2...

[kcbrown]

Disclaimer: I work for IBM. The thoughts posted are my own, and do not necessarily reflect those of IBM.

Just once, I'd like to see someone state in their disclaimer "I work for IBM. The thoughts posted are those of IBM, because what IBM thinks, I think. IBM owns me."

Or how about "I work for IBM. The thoughts posted are those of IBM, because what I think, IBM thinks. IBM is my biatch."

I like the latter better, though the former might be more realistic. :-)

Re:db2...

[Shados]

Indeed. Postgres was my open source DB of choice years ago because it had all these features when MySQL didn't.

I think the grandparent had issues understanding why people go for DB2/Oracle/SQL Server/Whatever... transactions, data integrity check, etcs are the basic of the base, and even MS Access has that (not the large amount of simultaneous users though, hahaha).

Replication, OLAP cubes, table partitioning, load balancing, data mining orchestration, complex backup and failover scenarios. advanced diagnostic solutions, ETL solutions, etc etc etc etc etc is where the "big enterprise" market is at. The only "databases" (and I use that term lightly...very lightly) that are "allowed" to not support transactions and such basic features are spreadsheets, in my opinion :)

Postgres has always been a fairly nice "in the middle" solution... When Spreadsheets and MS Access/JET are failing you, but Oracle would be overkill. Now that most of the big names have free "Small installation" editions, it lost a bit of its niche, but its still definately there when you have large databases with lots of users that don't need the hardcore tools (which is actually very, VERY common).

And using any number of database abstraction tools when building your applications (Hibernate, LLBLGEN, whatever) means that the day you swap RDBMS, the only thing that needs to go is the one trick pony DBA :)

Re:db2...

[anomalous cohort]

DB2 just doesn't scale down...it's typically used for very large databases running on IBM big iron

I believe the original poster is most probably asking about DB2 UDB and not the mainframe version.

Still waiting for a decent GUI

[bogaboga]

I love all these Open Source databases but what troubles me most is the absence of a decent [fully] programmable GUI to "slap" onto the actual database back-end. I would like readers to think of Access which is on top of Microsoft's jet database engine. It works and works beautifully but I loath Microsoft's products.

Can one tell me why we (in the open source world), do not have a single product that competes with Access in terms of functionality, ease of use and ease of programming business logic?

Re:Still waiting for a decent GUI

[timmarhy]

it's simple. no one in the open source world is dumb enough to WANT to be known as the access db of the open source world.

postgresql has a couple of brillant gui tools that hold their own easily against sql server managment 2005.

Re:Still waiting for a decent GUI

[InlawBiker]

Maybe because nobody wants to complain about a missing GUI when the product is free. But anyway I've found 4 GUIs for Postgresql in a quick search, not counting Navicat. I've never used it but it looks very nice. I've used PG Admin, which is great for simple work. Most of these are better than Access, which is just a toy, but not as good as Microsoft's query analyzer (now called "server management studio" I believe).

I have specialized in database applications with a web front-end for a while now. While they can't touch Oracle yet (or even MS SQL), Postgresql and MySQL are rapidly improving, and beat some of the expensive commercial offerings of not too long ago. A lot of medium-sized applications can exist just fine on either one. Eventually they will find the limitations of either system too limiting and switch, but for starting up they are both good back ends. With an expert at the helm, a serious application running either MySql or Postgresql is very possible.

At the rate they're improving MS and Oracle should be very concerned. IBM and Sun throwing serious money into the mix is a very interesting development.

Re:Still waiting for a decent GUI

[ashridah]

Uh. Because we actually do have a product like this?

OpenOffice.org has support for pulling data from a database.
It also has support for a forms-like interface.
It also has it's own vb-alike language. (Still in development perhaps, by the looks of it)

There are also plenty of other tools. RealBasic, etc.

Re:Still waiting for a decent GUI

[timmarhy]

I'd like to add that open source does need a simple access style db for simple low brow jobs, but please not something that has a jet style access built in

Re:Still waiting for a decent GUI

[kylehase]

Probably because people in the open source world write their own GUIs specifically for the task at hand in whatever language they are comfortable. I write my own PHP GUIs for MySQL databases all the time. If you're talking about GUI admin interfaces you may want to check out phpMyAdmin or HeidiSQL.

Re:Still waiting for a decent GUI

[QuantumRiff]

Most DBA's don't use the included tools anyways, they are pretty much crap compared to others.. Look for DBArtisan (One I have seen and used, not sure if its the best, there are many others out there.. Works with every major database, Oracle, Sybase, MSSQL, MYSQL, Postgres, etc. Last I looked it up for our DBA, it was about $7500 per database type.. IE, manage as many Oracle Databases as you want for one price..

Re:Still waiting for a decent GUI

[scorp1us]

Check out KDE's Kexi

Kexi is a Free/Libre and Open-Source integrated data management application, a long awaited Open Source competitor for products like Microsoft Access. Kexi can be used for creating database schemas, inserting data, performing queries, and processing data. Forms can be created to provide a custom interface to your data. All database objects - tables, queries and forms - are stored in the relational database, making it easy to share data and design.

I also like the (commercial) Maestro tools

Re:Still waiting for a decent GUI

[tacocat]

Access is not a database. Its a toy.

Re:Still waiting for a decent GUI

[fatp]

I basically agree with your comment. Actually, I did a lot of search recently for application builder (which run on windows) for PostgreSQL and have the following findings:

1) OpenOffice.org: Very poor scripting ability... wait for 3.0
2) pgaccess: Access for PostgreSQL. Looked promising, with Form Editor, Report Editor etc & scripting with TCL. But the project is dead and website is recycled.
3) bond (http://www.treshna.com/bond/): Looked interesting but I could not run the windows version... never tried linux version :P
4) rekall, knoda: both are database frontends for KDE. Looked interesting. I hope they will come with KDE on windows

Also, if your objective is to find something free (as beer), and have plenty of resources, you can have a look on Oracle XE+Application Express. Looks a bit strange at first (with a non-WYSIWYG form builder), but much better after one understand its design concept.

Re:Still waiting for a decent GUI

[BrentH]

Say what you want, but together with Excel's pivot tables these two it's the centerpillar of small to medium sized business' administration. Not having a competing opensrouce product is just too bad, because there's money to be saved and it would further the use of OSS.

Re:Still waiting for a decent GUI

[jaseuk]

Maybe because nobody wants to complain about a missing GUI when the product is free. But anyway I've found 4 GUIs for Postgresql in a quick search, not counting Navicat. I've never used it but it looks very nice. I've used PG Admin, which is great for simple work. Most of these are better than Access, which is just a toy, but not as good as Microsoft's query analyzer (now called "server management studio" I believe).

You've missed the point with Access. Access is a very simple to use application development environment. Someone with minimal database and programming experience can cobble together straightforward applications. Discounting Microsoft Access as a toy really shows ignorance of the power of the platform, the database engine may have been limited particularly with scaling and multiuser performance, but it's SQL feature set was far superior to MySQL for many years supporting features such as subselects, constraints and foreign keys.

I'm not an Access fan by any stretch of the imagination, but it has a niche that is not filled by pretty much any other commercial or open-source database in that a "power user" can put together an application.

Re:Still waiting for a decent GUI

[SCHecklerX]

The whole point of sql in the first place was so that you could have, say, management do some queries without HAVING to have somebody code a userland tool for them.

Do people forget this? SQL itself is the 'gui'.

Re:Still waiting for a decent GUI

[Ilan Volow]

One reason why you really don't have a plethora of good Open Source GUI apps for PostgreSQL is that there's really no good, reusable Open Source framework for writing PostgreSQL applications. Yeah, you've got basic client interface libraries that let you run basic queries, but to write an effective GUI DB app you need execute lots of those queries in one pass and to save the information in some structured way via objects which can easily be manipulated and turned back into further collections of complex queries. Writing this layer is probably one of the largest and most time-consuming parts of writing a database GUI app, and Microsoft has no problem doing it because they're so gosh-darned huge.

Put in terms of MVC, we really don't have good, documented M, and the M of all the current OSS PostgreSQL clients is so hopelessly intertwined with V and C and so undocumented that reuse of M is impossible. My solution is to build an effective, well-documented and well-tutorialed M in python that many OSS projects can effectively reuse to build their own GUI apps.

Re:Still waiting for a decent GUI

[bogaboga]

In my experience all features of OpenOffice.org suck big time!

Re:Still waiting for a decent GUI

[ashridah]

That's okay. He wanted a tool similar to access....

Re:Still waiting for a decent GUI

[ashridah]

And he, I meant "you". Didn't realise you were the same.

So what, exactly, is your problem with openoffice? it's worked well enough for my purposes in the past. It edits documents, does spreadsheet stuff, and can be made to suck data out of a database or create presentations.

Sure, its memory usage is high, it's an older style interface, and sometimes it's a pain finding things, but for the most part, it's pretty capable. We sure don't have much else that's as complete. TeX is nice, but doesn't fill everyone's cup and only does documents, abiword is nice, but feature-poor, gnumeric is nice, but feature poor. The two don't embed well enough yet, and you'd have to go out to realbasic or equivalent to start building access-forms-alike apps at a similar skill level.

Re:Still waiting for a decent GUI

[bogaboga]

Some of my problems with OpenOffice.org include the following: -

It's almost archaic report authoring tools. Heck I cannot even put logic in the report itself! I mean, for example deciding what data to display and how.

It is almost impossible to design input masks to form data. This helps to for example catch errors at form level instead of waiting for the database back-end to flag the error. Think of valid zip codes like those of Canada which are in the form A9A9A9. Where A's represent letters of the alphabet and the 9's represent numbers 0 to 9.

I found it hard to define what [form] controls are activated depending on what has been selected. For example, making a control ready for "how many years one has been married" if a "married" control has been selected, and demanding that some input be supplied. This is possible but so cumbersome. More complex logic is not possible.

There are many others.

Re:Still waiting for a decent GUI

[ashridah]

Ah, see, now that makes much more sense than "it sucks". With issues like that, I'm inclined to agree. :S

Re:Still waiting for a decent GUI

[FictionPimp]

You can do something very similar with netbeans and sqlite. In fact it is very easy to make a very access like system using that combo. I think sun even has tutorials on it.

Re:Still waiting for a decent GUI

[nurb432]

That is rather short sighted, as there is a need for a end user 'access like' front end creator out in the business world. Much larger then most people ever realize. Having a OSS choice that was useable would go a long way to creating inroads into small and medium sized businesses that cant afford a regular developer. ( the alternatives I've seen so far, are not yet useable )

And just for the record, Access, when linked to a real DB back end and run by a knowledgeable individual is much more capable then people give it credit for. Of course its also really easy to make a mess of things but that isn't the tool's fault.

IBM needs to make up their mind

[k1980pc]

They have DB2 and DWE for enterprises.
Informix for certain niche areas.
Cloudscape for Websphere applications
They use Derby in Rational suite.

Not even counting the different versions for each of these.

Re:IBM needs to make up their mind

[ip_vjl]

Cloudscape == Derby

http://en.wikipedia.org/wiki/Apache_Derby

Commercial open source software

[Per Abrahamsen]

That's not really open source. It is open source, according to the people who invented the term.

Open Source would never make you pay server licensing fees for use in commercial software, it would only make you distribute your source at worst. MySQL doesn't make you pay a license fee in commercial software, if you distribute your software under an open source (as defined by the people who invented the term) license. Like, e.g., Sun does with their very commercial MySQL product.

Re:Commercial open source software

[HAWAT.THUFIR]

who says you have to pay a commercial license? you can *install* the community version for free, you just can't *distribute* it.

Postgres clusters?

[the_greywolf]

What I want to know is can I run a Postgres DB on a cluster of servers? We want to add some failover capabilities to our server cluster, and our current solution is incapable of five-nines availability. Is there a way to cluster them to provide both load-balancing and redundancy with a single database?

I've heard Oracle has some capability to that end, but I'm not clear on just what it can do.

Re:Postgres clusters?

[jsse]

IBM has its own (real) enterprise-level database that does everything and all way before Oracle ever exists.

Putting 10m into PostgreSQL's vendors could be an marketing strategy. If you ask an IBM sales executive the same questions, he'd introduce their own line of DB2 for your needs - depends on how deep your pocket, that is. ^^

Disclaimer: I were programming with DB2 during my time in IBM, therefore I may not be too objective in comparing DB2 and PostgreSQL. Please bear with me. ^^

Re:Postgres clusters?

[greg1104]

The subject of this article, EnterpriseDB, is trying to target this market with GridSQL. As it's new in it's current form, impossible to say how reliable systems built with it will be quite yet. Those looking for reasons behind the IBM investment might consider whether GridSQL might one day talk to DB2 databases as well.

The closest fully open-source PostgreSQL solution to your requirements that's been around a bit is pgpool-II. It think it's still too immature to be considered five-nines quality though, and there are some restrictions you have to observe. A PostgreSQL replication solution that is very robust and proven is slony but it's not a load-balancing solution in the way I suspect you want.

There's also the Greenplum Database, which isn't free or open-source but is rooted in PostgreSQL technology.

Good enterprise-grade clustering with load-balancing is still on the PostgreSQL work in progress list rather than being here right now. I expect the core infrastructure piece needed to really make it work well (support for read-only warm-standby slaves) will make it into PostgreSQL 8.4 and be released around a year from now. I started a comparison page of the replication solutions currently available that's on the PostgreSQL wiki now that is trying to track progress in this area. Much like core PostgreSQL support for enabling replication, it still needs some work .

Re:Postgres clusters?

[Tablizer]

What I want to know is can I run a Postgres DB on a cluster of servers?

You simply imagine a cluster of them. That's what I learned on slashdot.

Re:Postgres clusters?

[plopez]

this may work:

http://www.pgadmin.org/docs/1.6/slony/failover.html

http://slony.info/

20% revenue, 40% profit

[Per Abrahamsen]

The software division of IBM accounts for 20% of their revenue, and 40% of their profit.

See http://www.ibm.com/annualreport/2007/md_4rco.shtml

Key applications are WebSphere, "Information Management" (db2?), Lotus, Tivoli, Rational, and operating systems.

Some of this is probably tied to the success of their hardware and service departments, I doubt many people buy IBM operating systems (2% of their total revenue, 12% of their software revenue) without IBM hardware.

But the non-disclosed revenue from Rational is probably pretty much standalone.

Re:20% revenue, 40% profit

[ePhil_One]

The software division of IBM accounts for 20% of their revenue, and 40% of their profit.

As a former customer, I can attest to that. A few years ago, we effectively got a "Free" server upgrade by updating our software maintenance contract; we saved enough by only buying 4 CPU licenses to purchase a 4x faster new Quad processor 2/ 2x more RAM to replace our old 6-way server. The old box then became a nice Test server, running much discounted Developer editions of all software.

Of course, I imagine 55% of the rest of the profit comes from services/consulting.

Re:20% revenue, 40% profit

[Per Abrahamsen]

Of course, I imagine 55% of the rest of the profit comes from services/consulting. 55% of their revenue, but only 37% of their profit, comes from services. Follow the link in the GP.

Re:20% revenue, 40% profit

[ePhil_One]

Follow the link in the GP.

I prefer to get my stats from "PODOMA"*

_____
*PODOMA - Pulled Directly Out of My Ass

di bi tu

[Per Abrahamsen]

I asked my local IBM sales representive since they now sell PostgreSQL, and he gave the pronunciation in the subject line.

PostgreSQL ROCKS

[mcrbids]

As the CTO of a rapidly growing, million-dollar company that provides ASP-model information management software, I can attest that PostgreSQL is just... awesome.

It quickly and easily scales into the hundreds of millions of records with good support on commodity hardware and incredible reliability. It provides excellent data-integrity checks - it's like programming with a safety net built in! Its license is open to commercial development, the support is great, and rarely needed. We rely HEAVILY on foreign keys, constraints, and the like to ensure clean data, with a schema now at almost 200 tables, fully normalized. PostgreSQL handles 12-table joins with flair. Bonus - its syntax is highly compatible with ANSI SQL, meaning that porting a project developed on PG will easily port to Oracle or DB2, even when you use a rich database schema!

Could it be better? Yeah - replication options are weak, especially in our environment, where we have a database schema that changes daily. But even in this case, this is mitigated by hourly database snapshots created a la cron - the performance hit is minor, and the recovery time in the (very rare) event of a failure is quick. And as a former sysad, I can attest to the number of times MySQL replication got it all wrong and had to be rebuilt from scratch.

Really, I just don't understand why MySQL still gets all the press - in nearly every metric that matters, PostgreSQL wins hands-down.

Re:PostgreSQL ROCKS

[sirket]

I love postgres- use it to handle millions of queries per day- but let's not kid ourselves: postgres replication in the form of slony blows. I have a master DB and it plublishes to a dozen read only databases. Managing that with slony just plain sucks. The simple fact is that setting up replication with mysql is dirt simple and that's part of why people use it.

I'm thinking of giving EnterpriseDB and their custom replication engine a try.

-sirket

Re:PostgreSQL ROCKS

[mysqlrocks]

Sorry, just couldn't resist responding based on the title of your comment.

Nothing more to say, carry on ;-)

hint: read my slashdot username

Re:PostgreSQL ROCKS

[tuxbond]

I've never understood why the "MySQL" buzzzzzz everywhere. PostgreSQL has been a __real__ DBMS from its beginning. It had support for really huge tables and it has an ingenious way to work around the 2GB file size limit on the old file systems.

The replication is its weakest point (it was non-existent for a while). The commercial and free packages to bring replication to PostgreSQL were sometime good and others don't. Many projects started it but were down or dorman in a year or so. Hopefully new efforts will finally bring that feature with the same rock solid standing as the other PostgreSQL features.

Happy to see IBM around!

Re:PostgreSQL ROCKS

[sirket]

Ahh well, I had been led to believe they had written their own engine. -sirket

Postgres: Best Intro of All Worlds

[Doc Ruby]

IBM buying into Postgres through EnterpriseDB is clearly a response to Sun's buying into MySQL. But what's really exciting about the move is that Sun also bought into Postgres, shipping it with Solaris 10 and integrating it with its Java App Server, as an entry-level database. Since Sun is also supporting and bundling MySQL (and therefore using it to drive sales of Sun machines), tools for porting between Postgres and MySQL are likely in the works.

Now IBM will follow suit, probably offering Postgres as an intro to selling its DB2 database, which will mean IBM tools for upgrading from Postgres to DB2. Meanwhile, EnterpriseDB already offers tools to port Oracle apps to Postgres.

The next move will probably come from Oracle. To continue the head-to-head competition, Oracle will probably offer tools for porting Postgres (and maybe MySQL) apps to Oracle. It's surprising that Oracle didn't buy a Postgres or MySQL company before Sun or IBM got them, but maybe that's why Sun bought one of each: to keep them from Oracle. Though Oracle did buy the InnoDB corp that makes the MySQL engine with serious DB features, and SleepyCat, the BerkeleyDB corp.

So as the dust settles, there could finally be a grand unification at work. IBM, Sun and Oracle each have incentive and in-house teams for producing tools to port between Postgres, MySQL and their proprietary high-end RDBMS'es. And since the lower-end (though Postgres competes well with them all) DBs are all open source, there is a good chance the upgrades will be available for freely porting among all of them.

The age of database lockin might finally be falling behind us. We might finally be free to use whichever DB is best for the job today, not determined by which DB was best for some other job yesterday.

Re:Postgres: Best Intro of All Worlds

[Doc Ruby]

As I said, Postgres is fairly high-end itself.

But the licensing, and the support it brings, is worth a lot more than just whichever DB platform happens to be running. For running enterprises, not only does DB2 have major features that Postgres doesn't, but you can get DB2 experts from IBM available at the drop of a table (;). For major applications, that's worth the very expensive prices IBM charges.

And with Postgres as a path to DB2, IBM would be able to give lots more people that pitch. And a lot more people who now have to choose not to use DB2 because of the upfront expense, would instead be able to include DB2 in their plans as they deploy a Postgres app designed to be eventually upgraded to DB2.

And of course that's also true about Oracle.

Which, as I explained, is why each of them are buying into the free platforms. They didn't get rich for nothing. They're smart, and they know what they need to sell DBs to a largely saturated market.

Re:Postgres: Best Intro of All Worlds

[Doc Ruby]

That's why the prospect of tools to easily port between them (and perhaps good middleware that allows selecting among any of them at deployment time without rewriting code, sacrificing performance or features, or jumping through hoops, is so exciting. Because lots of apps get written to one or another corp's DB precisely because of a perceived asset in the DB vendor corporation itself, but then learn too late to switch that the corp is actually a liability.

If IBM, Oracle and Sun are each applying their proprietary expertise to make Postgres and MySQL easily portable to each other, DB2 and Oracle's DB, to reduce the cost of switching so they can poach on each other's customers for "upgrades" across the different DBs, we might actually get that. Once they stop seeing lockin as a net benefit, they'll work to remove that mostly arbitrary barrier. And since the MySQL and Postgres code is GPL, those improvements are likely to be available to the public.

Even if these buyins produce only easy portability between Postgres and MySQL, especially to the point where it's easy to cluster a mixed group without real performance hits, we'll finally be able to disregard which DB an app was written for, by installing its native DB as an API to the other if we prefer it, and pool all our data in all our apps for direct access each by the other.

Re:Postgres: Best Intro of All Worlds

[illumin8]

The age of database lockin might finally be falling behind us. We might finally be free to use whichever DB is best for the job today, not determined by which DB was best for some other job yesterday.

It's interesting to note that what is driving all of this is the rise of multi-core processors. Oracle charges entirely too much for licensing ($$$ per CPU) and only discounts 50% if that second CPU is another core. So, now that companies like Sun and IBM are starting to push quad core, octo-core, and in Sun's case, the Niagara architecture which has something like 16-32 threads, you're starting to see Oracle licensing for a single database server jump from $$(tens of thousands) up to $$(hundreds of thousands).

Both IBM and Sun see a market opportunity in selling open-source databases that are not tied to archaic per-CPU licensing schemes like Oracle. Unless Oracle adapts their licensing and drops costs dramatically (something like 5-10% per additional core, not 50%) they stand to lose a lot of market share on the low end to MySQL and PostgreSQL.

I say this as a sysadmin that supports lots of Oracle database servers running on Linux. Oracle should see the writing on the wall and fix their licensing... We spend way more than the hardware costs on Oracle licenses alone, and that just isn't right.

Re:Postgres: Best Intro of All Worlds

[Doc Ruby]

Well, per-CPU pricing is a way to charge for how much work the DB is doing, which (when the DB is good) is roughly proportional to CPU capacity. So they're roughly charging the more profitable businesses per operation as much as they're charging those who make less off of each operation. That's why the less profitable customers will chafe, as the average pricing works against them, while subsidizing their more profitable "fellow" customers (some of whom can be competitors).

Oracle and IBM would do well to go the Red Hat model on their software, and charge for support and custom development, not the SW itself. Probably the best model would be to open the source code for reading, but keep their copyrights (not GPL), while giving it away free, but charging for developing apps and customized DB implementations (eg. stripping unneeded bloat). They could charge customers what each would bear from their operating profits, which would maximize their own profits, not just the minimum CPU capacity the customer can get away with.

Since IBM and Sun are supporting MySQL and Postgres, they're already doing that at the low end. Oracle is nearly doing that by supporting InnoDB. Once the "low" end (especially Postgres) eats more sales from the high end proprietary DBs than they generate in entry level customers upgrading, the tail will probably wag the dog. The trend is now towards more openness and integration, platform neutrality. They'll hold on as long as they can, but the writing is on the wall.

Re:Postgres: Best Intro of All Worlds

[Shados]

That said, Microsoft for example charges per socket, not per core/cpu/whatever in their "per cpu" softwares such as SQL Server. I foresee Oracle changing its way eventually.

Re:Postgres: Best Intro of All Worlds

[mounthood]

The age of database lockin might finally be falling behind us. We might finally be free to use whichever DB is best for the job today, not determined by which DB was best for some other job yesterday.

I find it strange that nobody on Slashdot talks about the fact that Sun and IBM both made investments in companies that keep private control of the source. MySql does it with dual licensing by not allowing people to contribute unless you assign all rights to them, a tactic Sun uses with Open Office and others. EnterpriseDB keeps control directly by making additional software thats separate from the distribution.

Both companies make great contributions (Thank you!) and will now make more contributions, but isn't there a clear danger from companies controlling the code? If it was Microsoft that made either investment wouldn't people be predicting:

1. Embrace by investing
2. Extend by making a bunch of widely used additions that are not free software
3. Extinguish by moving the users to a non-free version (re-written or a different product)

Re:Postgres: Best Intro of All Worlds

[Doc Ruby]

I agree. But Slashdot is a completely reactive "journalism" publisher, and its public commenters are notoriously shortsighted.

But I think that these higher profile moves will spur some confrontations. What really surprises me is that Stallman or some other FOSS activist hasn't forced a confrontation over these essential SW packages. Especially since Debian is so popular, and a modern OS should really bundle a SQL server in it for all apps to access if they need one. Maybe if Microsoft does that with Vista and SQL Server (Access wasn't really comparable), it will encourage perhaps Ubuntu to try the same, and finally push the conflicting interests into the public eye.

Anyone working on a modern comparison?

[pembo13]

Is anyone working on a clear, modern comparison of MySQL vs. Postgres? I believe the old arguments for and against both are no longer accurate: Postgres has gotten faster, and MySQL has gotten stricter.

Re:Anyone working on a modern comparison?

[greg1104]

Why PostgreSQL Instead of MySQL: Comparing Reliability and Speed in 2007

I'll have a 2008 update out soon now that PostgreSQL 8.3 has been released.

PostgreSQL runs factories

[Colin Smith]

MySQL runs web sites.

Almost literally. I know of at least one large multi billion dollar semiconductor manufacturer which basically runs it's fabs on postgresql.

MySQL vs PostgreSQL

[xtracto]

The reason why MySQL wins over PostgreSQL for me is because my el-cheapo web host provides me only with MySQL database backend.

Really, show me a host which is competitive to ICDSoft (which have very nice support service) for $6 a month for 1000 MB / 20 GB-traffic with php-perl-python-ruby-tcl and whatnot.

Looks to me as if Sun and IBM have ....

[chris_sawtell]

Just learnt about one of the oldest sales gimmicks in the book.

It's called 'bait and switch'. The vendor introduces the cheap version and then sells the still somewhat confused purchaser the (much - in these cases, infinitely) more expensive model. For these two vendors it makes sense because the free software moves the lever to open the door to generate the hardware sale.

I'm amazed they didn't think of it earlier.

If they had only invested back in 2001

[jocknerd]

I might still be working at Great Bridge.

MySQL license clarification: free as in freedom

[Wiseman1024]

"PostgreSQL is BSD. MySQL is anything but."

Both the community server and the enterprise edition of MySQL are licensed under the GPL, which is freer than BSD (BSD vs GPL holy war, go!). You just can buy it under a different license for embedding purposes if you want to, but you don't have to.

"Sure, the community edition is free, but it cannot be used with commercial software."

WTF? It's GPL. It can be used for commercial purposes with any commercial software. The only restriction is that it cannot be derived to or distributed as an indivisible part of a commercial product that's not released under the GPL. This means the embedded server. If you don't use the embedded server (like 99.99% of MySQL developers/users), you can very well build any commercial product that uses MySQL and license it with any terms.

Re:MySQL license clarification: free as in freedom

[TheSunborn]

The problem is that the client lib, which applications link with when they need to talk to mysql is also gpl. So I i write a
c++ program which connect to mysql, I need to release my application under a gpl compability license.

Re:MySQL license clarification: free as in freedom

[Wiseman1024]

Again, that's wrong. There's an exception to use libmysqlclient with commercial software.

Re:MySQL license clarification: free as in freedom

[TheSunborn]

Are you sure about that? I did a search on mysql.org and all I found was
(From http://www.mysql.com/about/legal/licensing/commercial-license.html)

Quote:
If you include one or more of the MySQL drivers in your non-GPL application (so that your application can run with MySQL), you need a commercial license for the driver(s) in question. The MySQL drivers currently include an ODBC driver, a JDBC driver and the C language library.
---------------
And if you look at
http://www.mysql.com/about/legal/licensing/opensource-license.html

Quote:
FLOSS License Exception. We have created a license exception which enables Free/Libre and Open Source software ("FLOSS") to be able to include the GPL-licensed MySQL client libraries despite the fact that not all open source licenses are compatible with the GPL (this includes the PHP license version 3.0). Read more about the FLOSS License Exception.

So they have made a special exception that allow opensource software to link with the mysql client library. That would not be needed if anyone were allowed to link to it.

Re:MySQL license clarification: free as in freedom

[Wiseman1024]

This exception allows you to do what you want to do and use non-embedded MySQL for all commercial purposes. It's the best of both worlds, as MySQL remains free as in freedom. This "free" is not just "free of charge" or "do whatever you want". Free means that the software guarantees a series of freedoms that cannot be taken away, ruined or in any way jeopardized by companies. You should be thankful that MySQL is released under the GPL and not a BSD-like license.

Disgusting corporations like Apple, for example, are benefiting from the open source community without giving anything back to it, and what's even worse, they are using previously open source software to backstab their own customers with their Digital Restrictions Malware (DRM) and outrageously immoral business practices.

Re:MySQL license clarification: free as in freedom

[TheSunborn]

Yes for commercial purpose, but not for non-opensource software, because the exception are only for OSI license approved software;

Quote:
If your application is licensed under GPL or compatible OSI license approved by MySQL AB, you are free to ship any GPL software of MySQL AB with your application ('application' means any type of software application, system, tool or utility).

So I still can't make a close sourced application that includes the mysql client library and link with it.

IBM Invests In MySQL/Oracle Competitor

[ponraul]

Is it called DB2?

license Re:MySQL databae supremacy

[leuk_he]

The fact how they market the community version as not for commercial usage is a bit misleading. The community version is available under a gpl version, and they cannot limit usage. The only limit that apply is if you want to change the code, and mix some commercial code in the database server. That is something that happens very rare.

You can pay for support of a version with extra features, but you don't have to pay for normal commercial usage.

[ikspi:] || [eggspeh] || [igz:peeh]

[lanc]

Why is "linux" hard to pronounce? If you have problems with that probably it isn't for you. Oh, and btw the Linux pronunciation is documented.

What would you say about the grml project? Or about the overlengthy and oversimplified GNU Is Not Unix Image Manipulation Program Toolkit?

DB -- ODBC -- Access

[Dareth]

Best of both worlds can be had with the right use of ODBC as a database abstractor.

Use Access for the forms and user interface, keep the data in whatever DB floats your boat.

PostgreSQL outperforms MySQL on FreeBSD 7

[brlav35]

http://www.scribd.com/doc/551889/Introducing-Freebsd-70 a presentation of the SMP in FreeBSD 7.0 using PostgreSQL and MySQL to produce benchmarks. Notable quotes - a) MySQL degrades after utilizing all CPUs, while PostgreSQL does not (the explanation is that MySQL has scalability problem). b) PostgreSQL is in general 35%-45% faster. I can't tell if all is true, just wanted to make it public, if it hasn't been already made. Iv

Re:Check out EnterpriseDB's tools

[TheBracket]

You should try out their admin tools - methinks they're pretty damned good The first screenshot looks suspiciously like PGAdmin III, available for free http://www.pgadmin.org/here. It is really nice to work with; handles all of our needs very nicely anyway.

Open Office Base.

[LWATCDR]

The new version of Open Office does offer the Open Office Base program. I think it uses SQLLite as the back end and can attach to real databases like Postgres if you need to.
I do agree with you that FOSS needs something like Access. I hate Jet with a passion because too many people abuse it and use it for tasks that I just don't think it is well suited too.
The standard FOSS solution is to us LAMP but for something like a CD collection that is just massive overkill. I don't want to have to set up a SQL server and apache just to keep track of my CD collection.
How well does base work compared to access? I don't know I have not tried it yet. I think I will give it a shot.

Re:The hierarchy of that statement

[YoureaClown]

"I believe DB2 manages more data than Oracle, whereas Oracle has more installations." Here is the hierarchy of that exact statement. Teradata-->DB2-->Oracle-->SQL Server-->MySQL With Teradata being front line on data warehouses only

Re:Smart move for IBM

[CrazedWalrus]

VERIZON is tired of getting ripped off?!?! Holy shit.

My guess is that its customers will still want to keep the lube handy when invoice time rolls around.

MySQL - PostgreSQL migration

[ChristTrekker]

I've been wanting to reimplement a small web project I'd done in MySQL back in the day, and I've also wanted to learn more about PostgreSQL as well. I have a dump of the MySQL db. Can I pull this into PostgreSQL somehow? Do I need to look for converter tools?

Re:Check out EnterpriseDB's tools

[CrazedWalrus]

That's probably because Dave Page, who leads PGAdminIII dev, works for EnterpriseDB. EnterpriseDB contributes quite a lot to Postgres, along with Red Hat and Command Prompt. I don't work for any of those companies, but the most active people on the PG mailing lists have those companies in their email sigs.

My guess is that this product is PGAdmin3 plus some additional goodies.