MySQL Prepares To Go Public

prostoalex writes "MySQL CEO Marten Mickos told Computer Business Review the company plans to go public: 'Now entering its twelfth year, the company has built up just less than 10,000 paying customers, and an installed base estimated to be close to 10 million... When it does go public, MySQL will be one of only a handful of open source vendors to do so. Red Hat, VA Linux (now VA Software), and Caldera (now SCO Group) led the way in 1999 and 2000...'"

Comments

two out of 3 aint bad

[macadamia_harold]

When it does go public, MySQL will be one of only a handful of open source vendors to do so. Red Hat, VA Linux (now VA Software), and Caldera (now SCO Group) led the way in 1999 and 2000...

Well, as long as Darl McBride doesn't get his hands on the company they should be ok.

Re:two out of 3 aint bad

[jack_csk]

Wait... you are giving ideas to Ralph Yarro and his army...

amen

[battery111]

I think it's awesome when open source companies go public. It allows them to get enough capital to truly innovate, and help prove to the unbelievers that open source IS a viable, successful way to make outstanding software. I hope more open source companies continue this trend.

Re:amen

[Anonymous Coward]

Really? What wild new innovations have come from those you listed?

Going public means that the companies primary goal becomes to please the stockholders rather than employees and customers. It's nice that the folks who started it up usually get rich, but it doesn't tend to do good things for anyone else.

Re:amen

[egreshko]

Sorry to disagree with you...but...

      What makes stockholders happy?
      Rising stock prices.

      How do companies like Red Hat make their revenue?
      Maintenance subscriptions.

      If customers are unhappy they will stop subscribing to maintenance.
      If customers stop their subscriptions, Red Hat's revenue declines as do
      their stock price. This makes stock holders unhappy.

So, there is a linkage.

Re:amen

[PodBayDoor]

Right, and open-source and open standards are a desperate attempt to get ancillary or support revenue a) from a product that is either a commodity (e.g. JBoss) or is under serious threat in its market (e.g. OOo, Adobe), or b) for a company that hasn't the momentum to achieve volume by other means. Occasionally an open-source project is simply an ego extension of it's original or primary developer/author (Linux being the archetypal example in more than one respect). Some successful open-source projects attract enough parasites or symbionts (e.g. Red Hat)

Or more bluntly, most open-source is cheap, a knock-off or both.

Re:amen

[Anonymous Coward]

Uh oh! Sounds like someone didn't get his red stapler.

Is that 10,000 customers total over 12 years?

[QuantumG]

Or is that 10,000 customers that regularly renew their MySQL licenses?

What's the average license cost? $40,000?

Re:Is that 10,000 customers total over 12 years?

[Duncan3]

$595 - $4995/server/year. Most in the low end I'm sure.

$4995 is still a heck of a lot less then a full time DBA.

Re:Is that 10,000 customers total over 12 years?

[timmarhy]

most real dba's have a lot more then a single server to look after. try about 10 or more. a dba's a heck of a lot cheaper.

Re:Is that 10,000 customers total over 12 years?

[Anonymous Coward]

You can get a DBA for $50K/year?? I'm in!

Re:Is that 10,000 customers total over 12 years?

[Anonymous Coward]

In Canada, depends where. In India, definitely. In US, probably not. DBA is not really that skillful of a job when it comes to MySQL.

Re:Is that 10,000 customers total over 12 years?

[ScentCone]

$4995 is still a heck of a lot less then a full time DBA

Um... how does paying for the license get around needing a DBA? It's not exactly an either-or.

Re:Is that 10,000 customers total over 12 years?

[Anonymous Coward]

I can use Access for like $300.

and I don't need a full time DBA with that either!

Re:Is that 10,000 customers total over 12 years?

[QuantumFTL]

4995 is still a heck of a lot less then a full time DBA

Um... how does paying for the license get around needing a DBA? It's not exactly an either-or.

I'm not the parent poster, but it seems to me that the argument goes along the lines of "if you hire someone expensive to do a job, why not spend a small amount more to allow them to do it properly."

I hate it when I see expensive, skilled employees being forced to use outdated technologies in their tech-sensitive jobs. It seems like such a waste.

10,000 customers?

[mcrbids]

I'm not sure how to take this.

1) They managed to acquire 10,000 customers? Who are these customers, and why would they pay MySQL for a product that's not only free, but has better competitors available for free?

2) 10,000 customers, with 10 MILLION installs? So the odds are 1 in 1,000 that a user of your product would actually pay you anything? Those are TERRIBLE numbers....

Ahgh. Conflict. Partly because I just don't like MySQL - I'm a Postgres user and shrug my shoulders as to why anybody would use something with all the warts of MySQL...

Re:10,000 customers?

[timmarhy]

mysql has some nice things put into it recently that postgresql could learn from eg. paritions and it's distributed abilities. (i use postgresql personally and i'd love to see pg do similar things) I'd guess that they have 10,000 business customers who would want to either modify the code themselfs and not gpl it or have mysql write them customised versions of the db for them. those licenses aren't cheap, they'd be making a healthy profit on them to be sure.

Re:10,000 customers?

[GMontag]

Who are these customers, and why would they pay MySQL for a product that's not only free, but has better competitors available for free?

Apparently they are customers who don't really need a database but they have employees who convinced them to spend money on a cutsie file system.

Re:10,000 customers?

[martenmickos]

Thanks for the questions!

The customer count is over several years. Yes, the majority of our users choose not to pay. The current ratio is something like 1 in 1,000. But as you probably know as an open source user, there is great benefit to a project also from the ones who don't pay.

Those who pay do it for the value-add they receive: production support, scheduled binaries with only bug fixes, the monitoring and advisory servce, etc. From a business perspective the great thing is that the ratio of paid to non-paid is changing and our business is steadily growing.

We are proud at MySQL to build something that has great value to the FOSS communities and is a great business at the same time.

Sorry to hear that you don't like MySQL, but great to see that you nevertheless take time to read /. postings about us and to post your own. Let us know what "warts" you see in our product and help us improve it. Then perhaps one day you will find that it serves your needs.

Marten Mickos, CEO, MySQL AB

Re:10,000 customers?

[suv4x4]

I'm a Postgres user and shrug my shoulders as to why anybody would use something with all the warts of MySQL

Don't you feel the burning irony of posting this on Slashdot, one of the more prominent MySQL users?

While you're busy with your tiny holy war, people take MySQL for what it offers and builds useful services and sites with it, among those Google, Yahoo, Digg, Apple...

Re:10,000 customers?

[drix]

Well, instead of trolling for Postgres, let's mosey on over to the MySQL website and see if we can figure out why someone might want to pay, hrm? Ahh yes, here we go, MySQL Enterprise. Mmm. Let's click that. Iiiinteresting. Says here you get 24x7 web and phone support plus 30 minute emergency response time. Eat that, pgsql-bugs. You also get consultative support from people who spend all day tuning MySQL installations for max performance and reliability. I can't even find the Postgres analogue of that to make fun of. Lots of other goodies too numerous to mention that might be worth paying for.

If you're tossing Wankr 2.1 together in your bedroom then MySQL free, pgsql, or even sqlite is more than enough to meet your needs. If you run a large business that relies on MySQL to actually make some $$, then purchasing support is a rational choice. Especially since TCO is still about an order of magnitude less than competition.

Re:10,000 customers?

[flyingfsck]

Sounds about right. In my own experience with free software, about 1 in 1000 downloads pay up. The large companies almost always pay. Small one man businesses sometimes pay. Europeans usually pay. Russians and Indians never pay...

Re:10,000 customers?

[nagashi]

1) An admin utility (no, phpmyadmin doesn't count for crap) that doesn't suck. Please, just take pgadmin and make it connect to mysql. PLEASE. MySQL Administrator and MySQL Query Browser work very poorly.

2) fix Unicode. UTF8?

3) How about stored procedures/functions with the same name, but different # of parameters? Works great in postgres.

4) Character truncation when inserting into char fields. (maybe this is fixed now? Last version I used was 5, just before it went GA)

5) Real standard TIMESTAMP data types.

6) Get rid of myisam and make InnoDB the standard. MyISAM is a joke.

Of these, 1-2 are very serious issues which will prevent me from working with it. 3-4 make my life more difficult, but I can get around them. 5-6 just make it much more of a serious database. Something where if people ask me what database I recommend for a project, I can honestly say 'MySQL!' and not have every other developer in the room give me odds looks (currently I usually say Postgres).

Re:10,000 customers?

[Reality Master 201]

Don't you feel the burning irony of posting this on Slashdot, one of the more prominent MySQL users?

I'd bet not; nor would I. I'd imagine that the grandparent poster is merely someone who likes writing applications against a more featureful database.

Re:10,000 customers?

[codepunk]

You mean warts like not having to run a vacuum on the frigging database every week so it does not grind to a halt. I love the functionality of postgres but I don't see it as being no where near as fast or as stable as mysql.

Re:10,000 customers?

[mcrbids]

Don't you feel the burning irony of posting this on Slashdot, one of the more prominent MySQL users?


Not at all. You *can* build great things with marginal technologies. It's just harder to do so.

Slashdot doesn't face a number of problems that MySQL would fail them on. Slashdot has a rather simple database schema - complex queries and joins are few to none. They don't rely on 100% ACID compliance. They don't use the database to help enforce data integrity.

So MySQL is sufficient for their needs.

But PostgreSQL matches in *all* these areas, and still manages to offer solid performance on complex queries/joins. It offers robust and mature ACID compliance. It offers excellent integrity constraints for your data.

It's not whether or not you can get something to work with MySQL - just like you can build a house with a dollar-store hammer. But why use the dollar-store hammer if both it and the $20 hammer are available to you for free?

Furthermore, the license behind PostgreSQL is MORE FREE than the one behind MySQL. You can build a commercial, shipping product with PostgreSQL and not be beholden to per-sale fees, as you'd see with MySQL.

So, again I ask.... Why would anybody use something with all the warts of MySQL?

Re:10,000 customers?

[Hohlraum]

Marten how do I get in on the IPO? :)

Re:10,000 customers?

[codepunk]

Marten, he wants you to add table vacuum so we have to spend our weekends like we do on postgres running it just to keep the database from grinding to a halt.

Nope, just keep doing what you do best...

Re:10,000 customers?

[slamb]

Sorry to hear that you don't like MySQL, but great to see that you nevertheless take time to read /. postings about us and to post your own. Let us know what "warts" you see in our product and help us improve it. Then perhaps one day you will find that it serves your needs.

I don't like that MySQL does not keep my data safely and securely out of the box. Some examples:

• I need to flip a whole set of knobs to make MySQL return failure on invalid data. Apparently TRADITIONAL, ERROR_FOR_DIVISION_BY_ZERO, NO_AUTO_VALUE_ON_ZERO, NO_ENGINE_SUBSTITUTION, NO_UNSIGNED_SUBTRACTION, NO_ZERO_DATE, NO_ZERO_IN_DATE, ONLY_FULL_GROUP_BY, and STRICT_ALL_TABLES. No other RDBMS even has these knobs, much less has the defaults wrong.
• There's no way (that I can find) to completely turn off non-transactional tables. As I understand it, if I forget to tell it when creating a table to make it transactional, it's silently not. If a transaction involves even a single non-transactional table, the whole thing is non-transactional. This makes me nervous.
• I don't know if it does an fdatasync() at the right times out of the box on all table types. I need ACID, not doubt.
• When users have no password set, anyone can connect without a password. Contrast to PostgreSQL: no one connects without authentication unless you explicitly say so in the configuration file. But it's unobtrusive because local users can authenticate via Unix domain sockets / SO_PASSCRED.

I can't take MySQL seriously until this changes. I understand that you have backward compatibility concerns, but that's life - you pay a price for the poor decisions you've made in the past. You might have to go through a long deprecation period before you can get rid of these knobs. At the very least, don't have them flipped this way unless I start mysqld with the --treat-my-data-as-garbage command-line option.

If you fix this fundamental problem, I'll be impressed. I may not use your product, but I will stop laughing at it.

Re:10,000 customers?

[Anonymous Coward]

Partly because I just don't like MySQL - I'm a Postgres user and shrug my shoulders as to why anybody would use something with all the warts of MySQL...

Well congrats! This post marks your 1933-th time you used MySQL over here.

Re:10,000 customers?

[dfetter]

No more ironic than that slashdot runs on the .org TLD, which in turn runs atop PostgreSQL.

Re:10,000 customers?

[B3ryllium]

I disagree with your slam against phpMyAdmin. I find that it's much easier to use for quick and simple projects than something bloated and ugly like pgadmin3.

That said, pgadmin3 is a cut above MySQL Administrator. But it's all open-source, so there's always room for improvement :)

Re:10,000 customers?

[Anonymous Coward]

How about you start caring about data integrity? Going for years before no longer accepting February 31st as a valid date isn't really inspiring about how much you care about the integrity of my data.

You can start by making strict mode actually work. You know, so that say when someone tries to insert 0-0-0000 as a date, it not only throws an error, but doesn't enter the bad data in the table. It does now. That's not how "REAL" databases work. If you enter invalid data, it throws the error, and refuses to insert the bad data into your table to corrupt it's integrity.

Re:10,000 customers?

[grcumb]

In my own experience with free software, about 1 in 1000 downloads pay up. The large companies almost always pay. Small one man businesses sometimes pay. Europeans usually pay. Russians and Indians never pay...

In Soviet Russia, YOU pay the customers!

(Shit, I just made a Soviet Russia joke. Now I feel dirty....)

Re:10,000 customers?

[timmarhy]

you do realise mysql requires vacuum just like postgresql, and all other dbs for that matter? postgresql just has AUTO vacuum so that you don't need to do it manually....

Re:10,000 customers?

[timmarhy]

postgresql has auto vacuum which does this in the background 99.999% of the time without you even noticing. all db's require vacuum, bar none. this is because when you update or delete a record the space that record consumes is not removed only it's data. if you update or delete a lot of records this can result in a huge database filled with nothing, slowing your queries down.

Re:10,000 customers?

[Anonymous Coward]

MyISAM is certainly not a joke. Sure, it has its limitations, but for its niche, it works quite well for what we use it for. Bulk load performance makes for a great staging area when you load 250+ million records per day (and that includes everything else we throw at MySQL to process the data). The ability to choose engines optimized for the task at hand is powerful and we make use of a number of them in our design including MyISAM and INNODB. I feel no need to defend MySQL; it works.

Re:10,000 customers?

[darekana]

Why would anybody use something with all the warts of MySQL?

Probably for out-of-the-box replication...

Re:10,000 customers?

[Anonymous Coward]

Let us know what "warts" you see in our product and help us improve it.

Well I'd love it if a vendor picked up a book on the relational model and created a product that implements it. Yeah, I guess *all* SQL databases are enormous "warts" when compared to the theory buy MySQL is particularly egregious.

A DBMS should do three things: store data, manipulate/query data, and maintain the integrity of data. Can we get a little more help on #3 please? Would it be possible to create a "MySQL for non-dummies" that is actually strict? That allows arbitrary declarative constraints, and uses them to optimize queries when appropriate? That requires all tables to be transactional?

For extra bonus points (which means I would actually *pay* for such a product), let's just toss SQL baggage out the window. How about a DBMS that allows me to create custom types and operators? So I can store my customer objects directly in the DB, instead of decomposing them into fields? How about storing data in set-theoretic relations, instead of "rows" that allow duplicate entries? How about a simpler query language so that queries can be expressed as formulas in relational algebra instead of weird, verbose pseudo-english sentences? How about fully updateable views that are stored on disk, so it's easy to refactor associated code (I see newer MySQL supports this to a very limited extent, though not with joins)? How about column-based storage, so joins are cheaper (why should joining 10 tables be so much more expensive than 1-2)? How about making it possible to specify KEYS separately from INDEXES? How about relation-valued attributes, or array-valued attributes, or distributed foreign keys, or temporal support, or any of the stuff that I read about in theory books and then wish I had when I'm coding?

This post brought to you by whiny database dude.

Re:10,000 customers?

[Anonymous Coward]

I happen to know that Slashdor is powered by my submissions. It says so on the front page. My submissions. Not yours. Mine.

Re:10,000 customers?

[Anonymous Coward]

oh yeah, like I want to use InnoDB and have the transaction log eat up my entire disk

Re:10,000 customers?

[QuantumG]

Allow me to put this another way...

If you think Postgres is so much better than mySQL (and I'm honestly not debating with you that it isn't) then go start your own open source company around Postgres to provide support services. You should make a fortune and put mySQL to shame.

We'll wait here till you get back.

Re:10,000 customers?

[Anonymous Coward]

Ask Jamie McCarthy what kind of contortions they go through to make MySQL usable. Or CmdrTaco, who has stated his 2 biggest regrets are using mysql over postgesql and calling the iPod lame.

Re:10,000 customers?

[tedu_again]

http://www.enterprisedb.com/

Re:10,000 customers?

[mynameismonkey]

I pay for it. I have several installs where MySQL is the right tool for the right job and it needs to be up, we need the security of vendor support and so does my client (a large US state government).

I have many more installs where I don't pay for it.

You said "So the odds are 1 in 1,000 that a user of your product would actually pay you anything? Those are TERRIBLE numbers...."

I say "So you get 1 person in a thousand to pay you for your product, even though it's totally free? Tell me more".

Re:10,000 customers?

[Dan Farina]

Someone has tried this before...Pervasive Software.

They failed, but not for the reason you might expect:

What have we learned? While we always knew that PostgreSQL is a solid product with advanced database capabilities and that it has a very real opportunity to shake up the high-end database market, we underestimated the high level of quality support and expertise already available within the PostgreSQL community. In this environment, we found that the opportunity for Pervasive Software to meaningfully increase adoption of PostgreSQL by providing an alternative source for support and services was quite limited. Accordingly, we have made the decision today to substantially curtail our focus on our PostgreSQL initiative.

Re:10,000 customers?

[suv4x4]

Ask Jamie McCarthy what kind of contortions they go through to make MySQL usable. Or CmdrTaco, who has stated his 2 biggest regrets are using mysql over postgesql and calling the iPod lame.

Can you provide links to provide context and support of your statements?

Re:10,000 customers?

[Anonymous Coward]

"stored procedures/functions with the same name, but different # of parameters"
You don't even know the name for that?

Re:10,000 customers?

[mcrbids]

If you're tossing Wankr 2.1 [parm.net] together in your bedroom then MySQL free, pgsql, or even sqlite is more than enough to meet your needs. If you run a large business that relies on MySQL to actually make some $$, then purchasing support is a rational choice. Especially since TCO is still about an order of magnitude less than competition.

I make money with my PostgreSQL database. My small-but-growing business will pass the $1,000,000 gross income mark this year, with over 30% profit margins. My issue here is that PostgreSQL has "just worked" with zero significant support issues in almost 7 years.

24 x 7 x 365.

Over 70 school districts use our product, and while there are the inevitable wrinkles, uptime is *never* a problem, and never has a single support thread been tracked back to a fault in the database.

So, no. That's not an issue. At least, as far as I can determine.

My experience with "paid support" is that people who are essentially hired monkeys and who do not understand what they're talking about do google searches for you while you wait. If that fails, they talk to their managers who then do more google searches.

Typically, it's 4 or 5 days before you get ahold of somebody who does anything but google searches, and he/she then needs to spend a few hours determining that there is a real issue to be addressed. And when that happens, they'll implement one of the solutions found in a google search first.

But a support e-mail list usually has meaningful answers within a few hours if your question is reasonably well written, and doesn't cost jack.

Which do YOU prefer?

Re:10,000 customers?

[jadavis]

I think that MySQL could show a little more restraint in the configuration file and other options. It's too easy to have two apps -- both of which support MySQL -- which can't work together on the same instance of MySQL. I think it's better to make changes more slowly and send signals to the users about what the better practices are, even if they change over time.

Also, the fact that a table can behave differently depending on the storage engine is quite worrisome. If a storage engine doesn't support a constraint efficiently, support it inefficiently. Then people can evaluate for themselves how important that constraint is given the alternatives that exist.

In this respect MySQL is like another popular open source project, PHP. Both technologies -- if you'll forgive the overstatement -- have configuration files that look like a FAQ: "How do I get MySQL to do xyz?" - "It's a configurable option". PHP is starting to change the trend and be more forceful to its users about using "good" practices. Not everyone agrees on every detail about what's "good" and what's not, but steering towards accepted practices and away from configureware makes PHP apps more compatible, and have less surprising results and interactions. I think MySQL will need to do the same.

It's easier to provide a configurable option (or a new table type that silently ignores whatever is inconvenient) than to educate users about how to use a pre-exising, consistent system to accomplish their task. I think there are some instances where both PHP and MySQL are guilty of that.

This is a major (if not the major) difference in philosophy between MySQL and PostgreSQL.

Re:10,000 customers?

[jadavis]

Says here you get 24x7 web and phone support plus 30 minute emergency response time.

Sun Microsystems offers 24x7 PostgreSQL support.

Eat that, pgsql-bugs

I've always found the mailing lists to be great. I'm sorry you didn't have that experience. By the way, pgsql-bugs is not a typical support channel, you'd be better off in pgsql-general or #postgresql if you have support needs. Unless you have an actual bug, of course.

Re:10,000 customers?

[chromatic]

MyISAM is certainly not a joke.

I can't believe you're doing data warehousing without ACID compliance! Do you not care about the integrity of your data?!

Re:10,000 customers?

[headLITE]

That's not exactly the same. PostgreSQL vacuum frees up disk space used by deleted data; the data stays on disk until you do that and it's not overwritten by new inserts/updates. InnoDB in MySQL doesn't free up space for deleted data immediately either, but it does re-use that disk space for new inserts/updates. You can cause InnoDB to free up this space manually, and yes this is similar to vacuum in Postgres, but unlike in Postgres it's not necessary for the database to work.

In other words, if you have a database where you delete 100 rows and add 100 rows each day, the Postgres database will continually grow by the size of those 100 rows added until you vacuum. With MySQL, the database size will be the maximum size ever required for the database until you alter table engine=innodb, which will reorganize the table. But it will not grow by 100 rows every day and in fact not grow at all most of the time. And if you configure InnoDB to use one file per table, you can restrict this to tables that actually need it.

Re:10,000 customers?

[headLITE]

You can't completely turn of non-transactional storage engines, but you can configure a different default engine. If you configure InnoDB as default-storage-engine in my.cnf, then by default your tables will support transactions.

So in total, all you're complaining about is settings you can actually change easily, either in my.cnf or in the server command line. Maybe I'm missing your point, but what exactly is it?

After all, *some* setup has to be the default, and you can't just dismiss MySQL because its default is not identical to yours because there are likely millions of users with defaults different from yours.

Re:10,000 customers?

[headLITE]

Furthermore, the license behind PostgreSQL is MORE FREE than the one behind MySQL. You can build a commercial, shipping product with PostgreSQL and not be beholden to per-sale fees, as you'd see with MySQL. That's just the old BSD vs. GPL discussion, no point in trying again to establish which one is more free. But consider this: If PostgreSQL was LESS FREE in your terms of freedom, it would be the only Oracle compatible open source database by now. As it is, EnterpriseDB, which is PostgreSQL with added Oracle compatibility, is a closed-source, commercial product.

Re:10,000 customers?

[headLITE]

My experience with "paid support" is that people who are essentially hired monkeys and who do not understand what they're talking about do google searches for you while you wait. If that fails, they talk to their managers who then do more google searches.

See, that's exactly what MySQL support is not. I invite you to go over to mysql.com and have a look at their job offerings, where you can read the qualifications required for a support engineer.

Re:10,000 customers?

[Anonymous Coward]

That is a feature, for fucks sake. It's so you can have your databases reside on write-once mediums such as optical disks. This is basically a requirement if you're data is sacred, such as in banking. Yes, it may appear strange if all you ever do is use your database for toying, but I work with international banks, and they wouldn't want it any other way. Tamper-proof is really important, vaccum is a consequence of that.

Re:10,000 customers?

[Bronster]

Perhaps it has multiple different names (polymorphic indeed) and the grandparent wanted the damn kids to get off his lawn^W^W^W^W^W^W^W^Wto provide maximum clarity about the actual feature wanted and how it would work.

Re:10,000 customers?

[killjoe]

"An admin utility (no, phpmyadmin doesn't count for crap) that doesn't suck. Please, just take pgadmin and make it connect to mysql. PLEASE. MySQL Administrator and MySQL Query Browser work very poorly."

Try sqlyog. It's free and it kicks the ass of pgadmin any day.

Re:10,000 customers?

[kv9]

I find it curious that, on Slashdot, every armchair-software-engineer keeps slamming MySQL, yet in the real world it seems to be working just fine. and please, no stupid "everyone uses Windows too, so it must be GREEEEEAT" analogies. you know what I mean.

Re:10,000 customers?

[kv9]

1) An admin utility (no, phpmyadmin doesn't count for crap) that doesn't suck. Please, just take pgadmin and make it connect to mysql. PLEASE. MySQL Administrator and MySQL Query Browser work very poorly.

HeidiSQL. it's sexy.

Re:10,000 customers?

[killjoe]

I would not call it "more featureful" I would rather use "differently featureful". Mysql has lots of features postgres doesn't including a cluster table type, case insensitive collation, multi master replication (over a WAN!), etc. It takes an average person five minutes or less to set up mysql replication, compare that to slony for example.

Re:10,000 customers?

[headLITE]

I never data is sacred, whatever that means. But you can use logging in MySQL and log all transactions to some write-once media if you like.

Re:10,000 customers?

[TheRaven64]

Although I've not used these features, I believe PostgreSQL supports partitioning.

Re:10,000 customers?

[cheater512]

10,000 paying customers. 10 million installs. Seems pretty straight forward to me.
They do have a lot of non-paying customers you know.

The paying ones pay MySQL for support and a few goodies I think. Check their website for details.

Re:10,000 customers?

[TheRaven64]

yet in the real world it seems to be working just fine.

And let's pick a not-so-random name from that list; Wikipedia. Remember when they had some major hardware issues, which caused their database servers to crash. The admins announced proudly that almost half of their database servers had come back online without data corruption! They actually believed the fact that half of their MySQL installs had trashed their data due to not supporting ACID properly was a good thing just because not all of their machines had been in the middle of a sensitive operation at the time.

This is why people laugh at MySQL users. They honestly seem to believe that a product which doesn't even fulfil the basic requirements for a database is adequate for their needs. Try an experiment; get a real database and a MySQL server running, and keep firing transactions at them. Then pull the power cord out of the machine. Do this a few times. See which database has any data left at the end. A real database will lose the last transaction (there's no way it can't, unless you have redundant everything including power), but it won't lose the rest of the data.

Re:10,000 customers?

[TheRaven64]

Ignoring your double negative, the only place I have seen MySQL out-perform PostgreSQL is on simple SELECTs, and if that's all you are doing with your database (or even the majority of what you are doing) then you might be better off with SQLite (which takes a little bit of getting used to, but is a real joy to use). As to stable, I wonder if you are using concurrent connections? In my experience, MySQL falls over under heavy concurrent load, while PostgreSQL just slows down.

Re:10,000 customers?

[isorox]

See which database has any data left at the end. A real database will lose the last transaction (there's no way it can't, unless you have redundant everything including power), but it won't lose the rest of the data.

Decent servers have two PSUs, and any buisness that needs to run 24/7 will have 2 mains supplies, fed from two directions, with 2 seperate UPS systems

Having said that we have a crtitical piece of software runnign on mysql 4.0.16 (yup, that's right), with myisam tables, and the application that runs on them is flakey at best. About once a month it crashes leaving the table indexes mucked up, which means a 30 minute downtime while they are rebuilt if we're lucky (the worst time was 8 hours of downtime)

Troubel is, that piece of software is out of our control.

Re:10,000 customers?

[destiny_uk]

I subscribe to the top level support package - and I have to say it's the best tech support I've ever experienced bar none. You get direct phone/email access to the guys who develop it with 30mins response! They are not monkeys!

Re:10,000 customers?

[Anonymous Coward]

Furthermore, the license behind PostgreSQL is MORE FREE than the one behind MySQL.

No it's not. MySQL sucks, but that has nothing to do with the license.

There is only one license problem with MySQL -- they insist on using the GPL for the connection libraries (not even the LGPL, the full GPL). Which, like Trolltech and QT, is a blatant attempt to use an inappropriate license to scam money.

Re:10,000 customers?

[codepunk]

AUTO vacuum...yea keep telling yourself that. I have been running production postgresql databases
for about 7 years now. The large ones I got rid of because it just took too much baby sitting to
keep the things up and running. vacuum for one thing takes forever to run, often fails requiring me to dump and reload, locks everyone out while it is running etc. I love the feature set of it but
down time to vacuum a table to restore performance in a 24x7 operation just don't cut it.

Re:10,000 customers?

[rtaylor]

I can't even find the Postgres analogue of that to make fun of.

Your Google foo is pretty weak then. I found that the words PostgreSQL Support brought up a number of companies, some of which (Enterprise DB) seem to have some fairly large clients.

Re:10,000 customers?

[vasanth]

why is MyISAM a joke.. if you don't require all the fancy features like transactions etc... it provides really good performance benefits.. We use MyISAM and we deal with millions of records and the performance is really good considering that most queries are for read...

Re:10,000 customers?

[plopez]

1 customer may mean more than one install. E.g. Foomatic, Inc. could have 50 MySQL licenses.

Re:10,000 customers?

[Anonymous Coward]

And when your motherboard decides to blow a capacitor and die in the middle of a transaction? Then what genius? Real databases are made to stay intact no matter what happens to shut a machine down or kill a process.

Re:10,000 customers?

[LurkerXXX]

In total? Didn't you read the rest of his comments? Strict isn't 'strict' at all. You have to go in and twist a bunch of knobs to even come close. And real databases don't make you twist that first knob to set InnoDB as a default. They assume you actually care about your data and want data integrity. I can dismiss MySQL because by default it thinks data is garbage.

Re:10,000 customers?

[tehcyder]

If you're tossing Wankr 2.1 together in your bedroom

Absolute genius.

Re:10,000 customers?

[MikeBabcock]

For one, MySQL supports multiple storage engines. Simply write a WORM storage engine and use it with MySQL if you really think its that important to have as an option. Having it as the default storage engine seems moronic to me from an efficiency standpoint (and as a DBA, I tend to be paranoid about efficiency).

I use different table types in MySQL for different features and different situations. Website guestbooks and hit counters are perfectly well handled by MyISAM, most other data by InnoDB tables.

As another poster pointed out, MySQL also offers full (text and/or binary) logging, so you can always point those to a write-once device as well for nice and easy playback of transactions.

Re:10,000 customers?

[headLITE]

In total? Didn't you read the rest of his comments? Strict isn't 'strict' at all. You have to go in and twist a bunch of knobs to even come close. Yes, and you only need to do this once, outside any SQL session. It's just a matter of how you set up your server, and if you've seen how long it takes to install Oracle you can't really complain about setting a couple of switches in MySQL.

Re:10,000 customers?

[Dan Ost]

So, are you claiming that automatic vacuuming in postgresql doesn't work as advertised?

What are the limitations that you've experienced?

Re:10,000 customers?

[End Program]

So, again I ask.... Why would anybody use something with all the warts of MySQL?

Sometimes, it's out of your control. There are vast amounts of web hosting providers that pre-install MySQL on your hosted site. I am sure some people would like the chance to experiment with PostgreSQL, but to do this you would need to install that software yourself, assuming the provider allows you to do this.

Most of the people are worried about developing and launching their ideas and don't want to add more complexity than needed.

Re:10,000 customers?

[kilodelta]

The customers they speak of are those who bought support packages. And yes, I think they're in for a rude awakening. They dropped install packages for things like Debian, etc. to go with Novell's SUSE which IMO was a very bad move. And I'm a MySQL user who shrugs his shoulders as to why anybody would use Postgres.

Re:10,000 customers?

[Anonymous Coward]

1) MySQL-front

Re:10,000 customers?

[VGPowerlord]

How about making REFERENCES work in a column definition when InnoDB is in use? Right now, you have to go out of your way and create an index on the column, then specify the constraint after that.

Why isn't the parser smart enough to do that automatically when it runs into REFERENCES in a column definition?

Re:10,000 customers?

[mounthood]

Home > Support > Professional > North America
http://www.postgresql.org/support/professional_sup port_northamerica

Re:10,000 customers?

[isorox]

The OP was hung up on power. Obviously a bomb exploding in your apps room will knock out your database, that's where you use something like a ndb to spread your database over multiple locations

Re:10,000 customers?

[Anonymous Coward]

An admin utility (no, phpmyadmin doesn't count for crap) that doesn't suck. Please, just take pgadmin and make it connect to mysql. PLEASE. MySQL Administrator and MySQL Query Browser work very poorly.

The command line is the best way to administer a MySQL install. (along with direct edits of the ini/cnf file)

In fact I far prefer command line and configuration file edits to any GUI based tools. It is far more efficient and many GUIs are poorly designed and not intuitive at all.

4) Character truncation when inserting into char fields. (maybe this is fixed now? Last version I used was 5, just before it went GA)

Um, maybe just design your database properly and use the appropriate data types?

6) Get rid of myisam and make InnoDB the standard. MyISAM is a joke.

Very few database applications require an engine that supports transactions. InnoDB uses far more memory than MyISAM and is considerably slower. MyISAM is great and is all that most applications need. Try it, put a >500MB database on the InooBD engine and compare the speed and memory use to what it was on the MyISAM engine. I have and MyISAM is the clear winner.

Re:10,000 customers?

[Anonymous Coward]

And I thought that slashdot went over to an IBM database several years ago (and became more sluggish at the same time.)

Re:10,000 customers?

[ErikZ]

Because I can get it installed and working with almost no effort.

Would love to switch to Postgres, but I just don't get it. And short of buying a book on it and reading up (paying money) I really don't see postgres becoming as easy to install and use as MySQL.

Re:10,000 customers?

[Anonymous Coward]

It was from conversations in IRC and IRL at linuxcon 2004 (after hours). Drop into #slashnet sometime. If their not too busy with other shit, they're pretty accessible.

Re:10,000 customers?

[mcrbids]

Marten,

I'm probably not your demographic target market. Anyhow, technological inertia makes any switch at this point a non-starter.

However, I will say that based on my various comments herein, you've done a good job providing support for your clientelle.

I wish you the best of luck.

Re:10,000 customers?

[midom]

you probably miss the fact, that the servers handle many thousands of requests per second, and are functioning in somewhat async setup. though we'd be able to set up sync environment at every stack level, there's no need. mysql fully supports full data consistency, but you do not need in such distributed database environment, like we have one at wikipedia. Domas @ Wikipedia

Re:10,000 customers?

[Jay Pipes]

This is totally incorrect. Please don't spread FUD. MySQL has not stopped supporting any distro, or stopped providing binaries to the community, or anything of the sort. Please see Kaj Arnö's blog post here (http://www.planetmysql.org/kaj/?p=84) for *accurate* information, and feel free to contact the MySQL community team (community at mysql dot com) if you have any questions on this. Please, don't let the FUD continue to spread. Check the facts and ask us for information if you're unsure. Cheers, Jay Pipes Community Relations Manager, North America, MySQL jay at mysql dot com

Re:10,000 customers?

[kilodelta]

No but you DID stop the packages for other distros on your web site. Luckily others have been putting the packages together for products like Debian, etc. I'm very concerned with the direction that MySQL is taking.

Re:10,000 customers?

[Jamesday]

The database servers have two power supplies. The colo has dual power, UPS and generator.

Life gets tougher when you find that BOTH power branches are cut off at the same time, on the computer side of the UPS and generator, and you're prohibited from having a UPS in the racks. InnoDB recovers fine from this once the power comes back, subject to only one requirement: the disk drives must not lie about having put the data on the disk surface. If they do lie, life becomes more interesting, though I'll probably be able to get things up and running without losing more data than was thrown away by the drives.

The nasty Wikipedia power incident, while I was the active DBA there, happened when two different brands of caching disk controller with battery backup didn't bother to turn off the hard drive write buffers. Not a lot of use having a battery if you let the drives throw away things. The loss of power incidents since have generally been reasonably trouble free except for the boxes with this mis-feature.

Sorry to read that you can't get a more modern version of 4.0, at least. Can't stop the MyISAM indexes from being corrupted if there is a crash, though, that's a limitation of MyISAM, one not present with InnoDB.

Re:10,000 customers?

[Jamesday]

Lets take your favorite database and reproduce the scenario: 6 drives each throw away 8MB of different data randomly selected from the last 500MB of database writes. Repeat with two different battery backed up write caching controller brands with on servers with dual power supplies at a colo with UPS, generator and dual circuits. That's what happened at Wikipedia.

I'm very interested in knowing which database server, running on only a single computer, will recover all except the last transaction in this situation. Expect your database supplier of choice to laugh at you and say that it's an unreasonable test and of course it can't recover if you're throwing away data from 500MB ago.

For a story of another outage, where InnoDB was fine, read the story of the 2005 LiveJournal power outage, caused when someone hit the emergency power off button in their colo. EPO = required to turn off all power, including generator and UPS power. Say goodbye to power to both of the redundant power supplies at the same time.

Re:10,000 customers?

[Jamesday]

The MySQL Support team includes me, first DBA for Wikipedia, Domas, the active DBA for Wikipedia, among others with a fair amount of production experience on big, high-traffic, real-world systems. My own background includes a decade plus of application development work writing applications that worked with Sybase, Oracle and Microsoft SQL Server.

One thing MySQL Support isn't is a typical support operation. If it was, I wouldn't be working there: I'd be at Wikipedia or doing DBA or consulting work instead.

Glad to read that you've had good experience with PostgreSQL: competition helps to keep everyone on their toes.

Re:10,000 customers?

[LenZ]

No, MySQL AB did not stop providing binaries. The Debian folks were building their MySQL packages by themselves all the time before, MySQL AB has never provided DEBs for download. And as you can read on the MySQL 5.0 Download pages:

In contrast to the MySQL Enterprise Server, which receives both monthly rapid updates and quarterly service pack releases, there is no specific schedule for when a new version of the MySQL Community Server is released. While every bug fix that has been applied to the Enterprise Server will also be available in the subsequent Community Server release, there will be source-only releases in between full (source and binary) Community builds. So while the latest published community sources will always be available from the Source Downloads Section, the binaries listed on this page may be from a previous release. In any case, full binaries for all our supported operating systems are and will remain conveniently available from this page.

Oracle aquisition

[Anonymous Coward]

Yay, with MySQL public it will be easy for Oracle to acquire them in a hostile take-over and EOL their product. Go free market.

Re:Oracle aquisition

[timmarhy]

won't matter if they do, someone will fork the GPL version. ah the beauty of gpl. companys can totally fuck up a product and we will still get to use it as we please.

Re:Oracle aquisition

[Anonymous Coward]

leech.

customers won't pay unless they have to (don't give me crap about ensuring the success of the community if you're not paying anything).

Re:Oracle aquisition

[Anonymous Coward]

They could do the same if the code were BSD licensed. BSD forks just as easily as GPL.

For once

[G3ckoG33k]

I read all remarks before replying. I had planned to infer the same. But, there are other giants interested in databases too. IBM and Microsoft to names the largest. So, it wouldn't be cheap at all. 30 billion USD?

sorry...

[Anonymous Coward]

I see no point in using mysql except in preexisting software
. I use firebird for small app work, and postgres when I need a large application. No licensing costs for closed source, unlike MySQL's awful ripoff.

They claim the protocol falls under GPL because it uses structures in the protocol. If that is so, Gaim is violating the GPL over the aim, msn, and icq protocol.

Re:sorry...

[bill_mcgonigle]

. I use firebird for small app work, and postgres when I need a large application. No licensing costs for closed source, unlike MySQL's awful ripoff.

I'm just curious - why not postgresql for the small app work? I've been looking at firebird recently and wondering where it fits. TIA.

Re:sorry...

[Anonymous Coward]

firebird can run in "embedded mode" If the app is small then it doesn't require such a large database. Firebird can run in standalone or like postgres runs in a network mode. If the application is really small, then I use sqlite.

Re:sorry...

[headLITE]

I don't understand how MySQL is a rip-off - apparently you expect to get something for free that is created by 350 employees at MySQL? How is it a rip-off if MySQL wants to be able to pay those 350 salaries?

Not all public companies are worth billions

[Duncan3]

Wow, lots of rage filled comments so far.

Not all public companies are worth as much as GE or WalMart. Vast numbers of public companies exist, and many are only worth a few million. 10k customers paying for support (we all know they need it) is still millions in revenue a year, more then enough to go public without being bogus.

Public != Billions.

Invest or not invest?

[ClosedSource]

The dot-com craze proved that you can make a lot of money on stocks with a bad business model (at least for a while). So the question isn't whether a MYSQL company will be succesful in the long term, but whether there are enough people out there who believe it will to make it at least a good short-term investment.

Re:Invest or not invest?

[squizzar]

Might be completely wrong, but isn't there some kind of stock market scam you could play with this. Take an open source company with an established name. Put it on the stock market, make some money from the sale and let it go bankrupt. noone owns the IP so there's no assets to sell off (e.g. nothing to lose for going bankrupt) and when it's all over you pick up the latest CVS and start again.

Probably massively over simplified, but there's gotta be something along those lines that you can do.

Re:Invest or not invest?

[FallLine]

Might be completely wrong, but isn't there some kind of stock market scam you could play with this. Take an open source company with an established name. Put it on the stock market, make some money from the sale and let it go bankrupt. noone owns the IP so there's no assets to sell off (e.g. nothing to lose for going bankrupt) and when it's all over you pick up the latest CVS and start again.

Probably massively over simplified, but there's gotta be something along those lines that you can do.

Interesting thought, but most of the MySQL-employed developers and senior management would have non-compete and non-hire contracts in place that would legally prevent them from doing this for several years (presuming they're half-way decently managed). Besides, senior people would benefit far more from growing the company into a much larger and more successful company than trying to start over again (even with the code....especially with a now damaged reputation).

I hope my folks buy this time.

[mfh]

I told them to buy Google and they ignored me. Maybe this time they won't?

Re:I hope my folks buy this time.

[Anonymous Coward]

How is that paid account working out for you, big boy?

Re:I hope my folks buy this time.

[mfh]

Pretty damn spiffy.

Thanks for asking.

Go mySQL

[nnila]

Hope this proves to other companies that being Open Source and giving away your software for free can in the long run actually be profitable and make you many lovers along the way. Been using mySQL for years and love it. It set the way for free databases for using for projects on webhosts world wide. Good luck to it!!

Cost

[JoshJ]

Is there an estimated opening price for this? I'm not really familiar with how IPOs work, but if the shares are low enough in price to begin with, even someone who doesn't have a lot of spare cash could invest in MySQL- I would love to invest in Free/Open Source software but I also don't want to be pissing half the money I spend on the "investment" on a brokerage firm and related stuff.

Re:Cost

[nacturation]

Is there an estimated opening price for this? Maybe it will open at the same price that VA Linux did -- but I hope for MySQL's sake that their stock does just a tad better one year later.

Re:Cost

[JoshJ]

Sheesh. Google started out at approximately $100- I'd expect something like this to start out much lower. I was thinking something like $20-30, but if it's $200 like that VALinux, that seems like a bad deal.

Re:Cost

[KrisWithAK]

If it was just announced, I don't think a price has been set yet.

An IPO is an initial offer of stock for sale to the relatively general public -- primary market. Usually an equity syndicate team at one or more investment banks determine the best combination of price and quantity of shares to offer to maximize the capital raised for the company, while still making the value attractive to investors. Besides taking a cut of the capital raised, the banks might also buy some of the shares themselves before/after passing on the IPO offer to usually their most "valued" and "qualified" clients. If I remember correctly, the IPO for RedHat was $14 . If you don't get the "IPO", then you have to buy it once it opens on the open market -- the secondary market. If the stock is hot and has lots of hype, then it usually opens in the market much, much higher. It also technically shouldn't matter much if you buy 1 share for $100 or 10 shares at $10 each. It all depends on the perceived value per share. Other things to look at are earnings growth and/or dividends. I've enjoyed reading Jim Cramer's latest 2 books.

Disclaimer: This post is not an offer to buy or sell securities. Investing involves a lot of risk. You must determine for yourself your goals and risk tolerances... possibly with the help of a licensed financial services professional. I am not one, so please don't ask me!

Re:Cost

[daft_one]

Keep in mind how utterly arbitrary (and meaningless) the "price" of a stock is, especially for an IPO. If they want to raise $6 Million, they could theoretically sell 6 shares at $1M each or a million at $1 each--or anything in between/beyond either end. If the stock comes out at $100, it isn't necessarily more expensive than another stock that comes out at $20. (Aside from what follows, think about it this way... You have $500 to invest. Do you really care if you've bought 5 shares at $100, or 50 shares at $10, if its value increases by x%? You make the same amount of money either way.) What will really matter as far as the value of the stock, will be how much money they are earning per share, and what their growth rate has been recently/is estimated to be in the future--all relative to similar companies. Because I'm lazy and it's late, I'll just recommend you read something like this: http://investopedia.com/terms/p/price-earningsrati o.asp

Re:Cost

[maxume]

Redhat has split since the ipo, so to compare to today's price, the ipo was at $7.

A well managed ipo shouldn't move around much once it hits the market -- the company should get close to market value for itself. There is generally a target amount of capital to be raised; the investment banks figure out the best price and share issue to hit that target and start offering it for sale. They usually leave themselves a way out, should they decide interest is not high enough. You are right that the price is generally somewhat below market; this ensures, as you say, that there is sufficient interest in the stock and allows for favors to clients and such.

New Website

[Anonymous Coward]

They should think about hiring a decent graphics person for their website first. The new design is hideous. It's like they tried to copy suns website, but failed miserably. I think someone got carried away with the gradient tool.

Capital isn't the problem.

[Anonymous Coward]

PostgreSQL has done far more with far less capital.

Since you probably won't believe me, I invite you to compare the features of each. Visiting each project's web site is a good place to start. Once you see how much further ahead PostgreSQL is technologically than MySQL, consider how they managed to accomplish that with relatively little capital.

Re:Capital isn't the problem.

[Anonymous Coward]

Postgres (pre sql) was developed at UCB, so it was partially government funded. I agree that Postgresql is technologically superior, without the funding and fulltime employees that MySQL AB (and innobase) have.

Re:Capital isn't the problem.

[chromatic]

Once you see how much further ahead PostgreSQL is technologically than MySQL...

Are you familiar with MySQL's storage engine concept? It's powerful stuff.

Re:Capital isn't the problem.

[sunny256]

PostgreSQL has done far more with far less capital.

Since you probably won't believe me, I invite you to compare the features of each. Visiting each project's web site is a good place to start. Once you see how much further ahead PostgreSQL is technologically than MySQL, consider how they managed to accomplish that with relatively little capital.

One of my criterias for "software healthiness" is how much the community is involved in the development and management of the source code. It is interesting to see that the source code for MySQL is stored in a BitKeeper repository instead of something more community-friendly like CVS or Subversion. A real turn-off for me, at least.

Re:Capital isn't the problem.

[Anonymous Coward]

The POSTGRES project ended at Berkeley in 1993. At that time, the database system didn't even have SQL support. SQL support wasn't added until 1994, by Andrew Yu and Jolly Chen, leading to Postgres95. PostgreSQL was first released in 1997.

So as we can see, most of the SQL-related work happened after the project at Berkeley had ended. Yes, it did build on the foundation that they had built, but the past decade of work (including the support for SQL) did not happen until later.

Re:Capital isn't the problem.

[Raenex]

One of my criterias for "software healthiness" is how much the community is involved in the development and management of the source code.

The biggest limitation on community involvement in MySQL is that you have to transfer copyright to MySQL so that they can sell it for profit.

Geeks = Suckers

[Anonymous Coward]

Open Source company? Yes, to reap the benefits and take advantage of all geeks. But they read the GPL as the Devil would read the Bilble. Just using (not extending, changing or even looking at the source) their product means that all your source code must be GPL!!

For me this is the worst case of taking advantage of the good name of GPL and the open source movement to make some serious amounts of cash.

Geeks are just so easy to use by large corporations... *suckers* Why else do you think corporate America likes open source?

Re:Geeks = Suckers

[headLITE]

No, that's not actually true, you can use MySQL fine without having to use the GPL for your software.

However you may be using a GPL'd connector - so you're linking to a GPL library. Not actually new, it's the same for all GPL code.

But I don't understand anyway, why do you expect to get something for free while you don't seem to give away something in return?

Painful Irony

[bill_mcgonigle]

I've seen some guys at LinuxWorld who were offering commercial PostgreSQL support, and was going to post that link here, but I ran across this one from the official FAQ while googling for it and found it too rich to pass up posting.

Anyway, google for 'postgresql support' and you'll find several hits on the first few pages. A few have dorky requirements like buying a license to their rebranded 'distro' of postgresql.

Re:Painful Irony

[richlv]

i congratulate you on a joke that almost flew over my head :)
if modpoints would not expire exactly before i need them... yeah.

Traditional IPO?

[Falladir]

Would it be possible for this IPO to be run Google-style? I don't remember the details, but slashdotters surely remember the story. There was a big online auction instead of a traditional IPO, so that investors could deal (almost, I guess) directly with Google.

Is it that only a very high-profile company like Google can make that model work?

Just in time for Oracle

[Anonymous Coward]

Which software/media heavywight is going to buy them out.

I fell tbe bubble is rising, when will it pop?

Ramen

[remmelt]

The problem with your reasoning is that stockholders are very bad at long term projection. If they can turn a profit in a short time they'll jump at the chance, no matter what long term fall-out may be. This is true because stockholders don't care about the business itself but about the profit it makes. See, a typical stockholder doesn't care how good RedHat's maintenance subscription service level really is. If they can cut the service in half and still retain a number of clients, they will. This will ultimately be bad for business and it's immediately bad for customers and customer relations, but it will up the profit, so it's done.

If you would buy stock in a somewhat anonymous company, would you go and investigate what their business practices are like? Do you care about their customer service? I appreciate that there are exceptions, but most likely you won't. Yes, there are people who invest in companies that they know and care about (sports clubs come to mind,) but the majority of investors invest for a profit. If a company can turn a profit sooner rather than later, they will go for it. Most investors won't care about the database, the open sourciness, the service, the customers or anything else, but they'll care about the numbers on the yearly report. There is linkage, but if it's not apparent, if it's not 1-1 related, there won't be much interest.

Our guiding principle is "Do the Right Thing." This means doing what is best for our staff members, customers, business partners, and communities for the long term, and believing that "right" answers exist. It also means measuring our success, not merely in financial terms, but by how consistently we act according to this principle.

(From here: http://www.mathworks.com/company/aboutus/mission_v alues/)
I think that is very well said, and I think it's something that doesn't go over well for public companies. MathWorks is still privately held.

Re:Ramen

[malsdavis]

Although I agree somewhat with your point, I think you're greatly oversimplifying the issue. After all a privately owned company still has shareholders; the difference is just that not everyone can go and buy those shares as easily.

There are many different types of stockholder and although I agree going public inevitably opens up the company to the ones just looking for short term gain, its erroneous to think all stockholders think that way (many of the most savvy investors realise investing means potentially going in for the long haul). After all, unless your a hedge fund manager or such, then the chances of you making money from buying and selling stocks in the short term (i.e. less than a few years minimum) are actually pretty low.

Personally, I think the problem isn't as much about stockholders only being concerned about the short term - unless the company starts loosing money and/or avenues of revenue, when this happens stockholders do tend to panic and want random things changed - the problem is the possibility of stock holders seeing a good quality, well run, company (which happens to sell OSS) and buying a stake because they can see the company has potential but then attempting to shift the company towards a more conventional business model. This usually means less resources going towards the actual open source development and more emphasis on the revenue generating services on the side. I worry companies like Red Hat have been greatly affected from this stockholder driven change.

Re:Ramen

[Ash Vince]

The point the everyone seems to be missing with regard to stock holders is who actually gets to vote at the companies AGM. A large percentage of company chares in circulation are not owned by people, they are owned by investment companies (banks, etc) who get to block vote all the shares in one go.

For example:

Company A floats on the stock market and it share are purchased by Companies B (15%), C (10%), D(20%), E(6%) and a handful of smaller investors (49% total).

When at their AGM Company A wishes to appoint a new director they have to put it out to vote. But each person gets to vote according to the number of shares in Company A that they own. So if the directors of companies B,C,D and E get together in private and decide who they would rather put in charge, there is nothing all the smaller investors can do as even if they all voted the same way they would still only have 49%.

Now the numbers I quote above are a complete exageration but it usually amounts to the same thing in the real world. Its just that the other comanies would be made up of 10 - 20 investment houses (instead of B C D and E) and they would not initially all agree. So they would trade favours for voting the way another company would prefer in return for the same thing happening in reverse when a vote came up they veiwed as more important to their business. The have the opportunity to do this as they are still only 10 - 20 fundmanagers who probably drink at the same bar / club anyway.

Whereas the smaller investors are spread across a much wider geographical location and are much less likely to have the opportunity to meet. They are also less likely to trade favours the same way fundmanagers can as they probably dont own stock in such a wide range of companies so any favours on offer are less likely to be relevant.

This is usually the way things turn out because most of us do not own shares in a company directly, but our pensions and savings are invested on our behalf. In return for investing our money for us, the investment houses and banks get to use the vote that comes with the shares.

Re:Ramen

[dan the person]

If a company can turn a profit sooner rather than later, they will go for it.

Have you never looked at the track record of a fund before investing in it? I invest for the long term. Who cares if they are up 40% this year if they are down 30% next year?

Re:Ramen

[the_womble]

So why do you think analysts do fairly long term forecasts? Most investors in growth companies are looking at the long term, otherwise they would not pay hefty premiums for growth.

Trolling

[remmelt]

Are you familiar with stored procedures? Subqueries? It's basic stuff.

(Sorry. Yes, I know that MySQL has all those things now, my point is that pgsql had them ages ago. Also, I'm a happy MySQL user. Also also, phppgadmin is nothing like phpmyadmin. Etc, etc)

MySQL support

[shani]

I second this comment.

I've never had good IT support for software, except from MySQL. It was still a bit painful to get to the engineer who could actually make the code changes to fix the problems, but it did happen, and the problems were fixed.

Bad Idea

[plopez]

IMO.
The more I learn about how public companies are run, the less I want to work for one. Some reasons:

1) You become subject to the whims of the stock market, and the stock market is not the economy. If you ignore the market at best the shareholders hammer you, at worst the SEC shows up on yor doorstep.

2)What is good for the company may not be good for the markey value. E.g. long term R&D may be cut to meet quarterly earnings estimates.

3) The regulations you labor under may focre your decsion making. E.g from my reading it looks like at least some companies offshored because other companies were doing so and reporting huge savings. THe savings turned out to be an illusion, but failure to offshore could have been seen as mismanagement, and so the CEO's followed like lemmings.

4) Stock option back dating, and other such scams.

So I think the argument to stay private is much stronger than going public these days, IMO.

Re:Bad Idea

[kebes]

You make some good points. However let me just point out that private ownership has its own set of problems.

Beware: Anecdotal evidence follows!

A guy I know (now retired) worked as management in both private and public companies over the years. At one privately owned company (a medium-sized business owned by a couple of family members, sales in the millions), the place was severely mis-managed due to the fact that it was private. The problem was that the owner wanted to "do the right thing" which often came into conflict with the long-term viability of the business. He felt bad for the employees and was very reticent to fire people. At first this seems great: the company has a very 'family' atmosphere and everyone feels like they are 'part of the team.' However it was not sustainable, because many employees became complacent, even lazy, and were not performing up to the level of their salary.

Without being willing to make tough choices (including firing people, slicing salaries, etc.) a privately-owned company can become highly inefficient. In the long-term, this ruins the company, which isn't good for *any* of the employees. Again, you need to make some tough choices for the long-term benefit of the company.

Of course that's just one anecdote. There are lots of private companies with strong leadership that do very well. There are also plenty of public companies that do it properly, and don't compromise long-term business plans just to make quarterly earnings. But the 'pressure' of stock prices can also do good things for companies, helping them streamline and expand, which is good for the employees as well as the investors.

All I'm pointing out is that both public and private companies have their own pros and cons. (The 'better one to work for' probably varies from industry to industry...)

Why I think it's a bad idea

[C_Kode]

With Oracle's plans on stealing business looming, investors are going to question dropping money into a company like this. I think Oracle should just release the standard edition for free without support and forget its dreams of trying to hijack other company's products. Ellison is starting to grasp at straws.

MySQL AB is a proud partner with Scox

[walterbyrd]

So McBride influence is not out of the question.

You're starting to leave an impression with me :-)

[Qbertino]

Marten Mickos, this is the second time within a few days I've seen you posting usefull stuff on /. Thanks for you time and staying true to geekiness in general, allthough, I presume, you have quite a schedule to keep up to.

I'm an avid MySQL user. I too don't consider MySQL or any other classic RDBMS the cream of databases concepts, allthough the SAP DB/MaxDB thing and the attempt to make it compatible with MySQL SQL dialect did get my attention.
The prime reason why I'm using MySQL on a daily basis is that available object databases are to slow and/or exotic and for classic RDBMS - which in my book all suffer from SQL - I consider MySQL 5 at least nearly as good as any other. A big bonus is the available documentation and the amount of free tools that support MySQL, which is unmatched by any other product.

And here's my first wart that is bugging me:

When will we see MySQL Workbench finished?

Imho getting the DB-Designer Team on board was one of the smartest moves of recent in an impressive line of smart moves, and having a sophisticated free ERD Tool that fits MySQL 5's featureset is yet another killer criteria in favour of MySQL. Using Workbench would slash developement times yet again and give the entire MySQL faction yet another lead. ... When will we finally see it finished?

Aside from that: Keep up the good work.

A satisfied professional MySQL user

P.S.: What will the IPO per share be? I'm pondering the thought of hopping on board.

Let's think about this...

[Gorimek]

The problem with your reasoning is that stockholders are very bad at long term projection.

Assume there are two kinds of stockholders. Those who are bad at long term projections, and those who are good at it. Let both start out with equal amount of money. In the short term they do pretty much equally well. But in the long term (by definition) the second group does a lot better. Just to put some numbers to it, let's say that after 5 years, the second group has twice as much money as the first. It could be 10 or 3 years or whatever, it only changes the argument by degree.

Of course, the long term never ends. So after 10 years, the long term savvy investors have 4 times the money of their short sighted brethren. After 20 years 16 times more, and so on.

We have had stock markets now for hundreds of years. And this is why, pretty much by definition, the people sitting on all the old money is the very best at long term investing you can find. Because that's how they got that money. Sure there is always new money coming in, and it can be spent unwisely in the beginning. But it always ends up with long term term savvy investors over time.

Emperical evidence of why Open Source is best

[sideswipe76]

I can't speak for all project everywhere, but I write a lot of Java code and have a lot of recruiters calling me. And you know what, the technology they ask for is like a who's-who of Open Source in Java. Struts, Spring, Hibernate, Jboss, Ant, Tomcat, Lucene, MySql, Linux All open source. And if you have those skills, employers will beat a path to your door. I think the real reason these products are of such a good quality is that there is honest and very critical review of the source and design. There is also motivation to just throw it out when it makes sense. Let's face it, in corporate development there is no real forum to critique someone elses code. Infact, doing so usually makes you a new enemy and gets you a 1-on-1 with your manager for "interpersonal" skills. The Washington Post just posted an op-ed yesterday by Scott Rosenberg. The article is interesting in and of itself but look at the comments. Those are equally illuminating.

Open Source

[Ozgur Uksal]

Hope this proves to other companies that being Open Source and giving away your software for
free can in the long run actually be profitable and make you many lovers along the
way. Been using mySQL for years and love it. It set the way for free databases for using
for projects on webhosts world wide.

ozgur uksal
http://www.cbronline.com/article_news.asp?guid=26F 5F8B6-8CC6-4529-8DE7-65732FA84347/

There is more to "databases" than you realize

[crucini]

Not everyone using a "database" needs durability.

Durability is a requirement for some applications; a luxury in others, and completely irrelevant in a third group.

Durability comes with a steep price - syncing to disk after every atomic unit of data. This defeats the OS's carefully designed buffering. Most applications do not guarrantee durability. When you type a character into Microsoft Word, it is not synced to disk. When you make a TCP/IP connection, the kernel stores the connection state in RAM and never writes it to disk.

Just because an application uses a database for storage does not mean it wants guarrantted durability.