Does that matter? Those users who do care should get the support they can expect if management claims to be committed to Windows, regardless of how many they are. It's a simple matter of putting your money where your mouth is: if you care, you make sure you have people who can support Windows and like to take responsibility to bring it up to par with the other platforms, otherwise you make it clear that Windows is a second-class citizen and you should be prepared to get a poorer-quality product with less support.
From a short-term management perspective it seems better to have just the promise of support than to have nothing at all, but in the long term you're sending a signal that MySQL on Windows isn't to be taken seriously, and that you don't want people who are committed to Windows as customers. That's as fine a business decision as any, but then it's cheaper to cut out Windows development altogether than to go about it halfheartedly.
I can't think of one good reason why you would use MySQL on Windows. I know this seems trollish, but it's the honest to goodness truth.
Either way, it's not likely Windows is all that significant in terms of MySQL usage or you'd be seeing more activity and movement on that front. And Windows developers aren't really all that open source savvy yet either ...
It's what bensummers said above: it's not whether MySQL on Windows is a good idea at all, it's why MySQL management implies that it is by claiming to support it.
You're probably right that there's no technical reason to use MySQL on Windows, so the only reasons left are management reasons ("our code monkeys can install whatever they like as long as it's on our trusted Windows (tm) machines"). This explains why the technical support for MySQL on Windows sucks, while the management support is prominent...
I can't think of one good reason why you would use MySQL on Windows
Licenses can cost quite a bit when you have lots of systems. A former employer of mine saved several hundred thousand dollars by moving from MS SQL Server to MySQL and PostgreSQL.
I'm not sure what "a huge" is, but download numbers from the MySQL website would obviously be weighted towards platforms where the distribution providers for a given manufacturer's operating system don't ship a usable version. Since Microsoft doesn't provide MySQL as the OS manufacturer, nor do Windows distributors like Dell ship a packaged MySQL installation, then it makes sense that Windows would be well represented in download numbers of binaries from the mysql website.
There's also a trend, at least in performance centric circles, to start building from source with patches or using third-party binaries, like those provided by percona, and other stuff not provided for on the mysql website.
Downloads from any one website are ultimately not indicative of the actual usage of the software, especially when it comes to freely repackagable open-source.
I'm afraid I'm not a purist. You cannot deny the fact that windows is ubiquitous & for any software claiming to support it, it is 'huge', I repeat its 'huge'. Just check this report (http://dev.mysql.com/tech-resources/articles/mysql_on_window...) from MySQL itself on how 'huge' windows platform is.
raised left eyebrow Does anyone care about Windows users?
If you're doing any heavy development with a set of open source tools on Windows - I don't know it just seems like nonsense. Dual boot, vmware, virtual box, hackintosh, whatever means necessary, but it's always been my fundamental belief that I work in an environment as close as possible to what I'm deploying to. You're just creating headaches for yourself otherwise.
And if this is headed towards any kind of production/hosting thing, you - or someone above you - are making a series of pretty poor decisions. I'm not sure there is any other way to put it.
Please. Half the Windows developers in the world would be out of business if held to this standard. Almost certainly including Microsoft.
MySQL has bugs, but the basic functionality works on Windows. Or so I assume, considering that those bugs don't seem to have prevented tens of thousands of PHP developers from making good use of MySQL when developing their software. (I wouldn't know, of course -- I use Mac OS and Linux.)
I'm sure the bugs are annoying, though. Maybe the people who care about them should get out the GPL source code and fix them. Or maybe they should pay someone who knows how to fix them. Or maybe they should do what just about everybody else seems to do -- calculate the price of options one and two, compare them to the cost of running Linux, and deploy on Linux.
Or they can bang on the table and cry in the hopes that someone will fix their problems for free. Good luck with that.
I assume Windows developers and Microsoft care about supporting Windows, and develop on the platform. MySQL developers don't sound as if they do anything on Windows.
Given that MySQL put the Windows download link first in their list of downloadable binaries, you would have thought that it was a platform they might use for a fair percentage of their development work.
But you assume _everyone_ is deploying to a Linux and a server environment. What if you are writing a desktop app: a poker assistant, a contacts deduper, whatever. Some datasets strain the usual embedded dbs, and MySQL would be great... if it worked consistently on windows.
Download WAMP and you can use MySQL on Windows with no-problems. It can also run as a standalone Package. You can also download the Designer and it is ALL for FREE!
What problems? If there any, they are so minor the advantages by far outweigh these. I have been using it since probably version one and categorically without any hassle. (Even in the pre-WAMP days I never had any serious problems with it).
For me it has been the Golden Bridge allowing me to develop locally on a Windows machine and to deploy in the real world on LAMP!
Yannis, If you are using MySQL every day chances are very less that you haven't bumped into these nasty bugs. You can check the links provided by the author to the bugs.
There's a big problem with a product when you can't even get through the installer. I've tried 3-4 separate times to get MySQL installed on several different Windows boxes, and the installer has been broken. Eventually, I just gave up and used postgres, which has proved to be a great move in the long run.
25 comments
[ 3.6 ms ] story [ 83.6 ms ] threadFrom a short-term management perspective it seems better to have just the promise of support than to have nothing at all, but in the long term you're sending a signal that MySQL on Windows isn't to be taken seriously, and that you don't want people who are committed to Windows as customers. That's as fine a business decision as any, but then it's cheaper to cut out Windows development altogether than to go about it halfheartedly.
Either way, it's not likely Windows is all that significant in terms of MySQL usage or you'd be seeing more activity and movement on that front. And Windows developers aren't really all that open source savvy yet either ...
You're probably right that there's no technical reason to use MySQL on Windows, so the only reasons left are management reasons ("our code monkeys can install whatever they like as long as it's on our trusted Windows (tm) machines"). This explains why the technical support for MySQL on Windows sucks, while the management support is prominent...
Licenses can cost quite a bit when you have lots of systems. A former employer of mine saved several hundred thousand dollars by moving from MS SQL Server to MySQL and PostgreSQL.
There's also a trend, at least in performance centric circles, to start building from source with patches or using third-party binaries, like those provided by percona, and other stuff not provided for on the mysql website.
Downloads from any one website are ultimately not indicative of the actual usage of the software, especially when it comes to freely repackagable open-source.
If you're doing any heavy development with a set of open source tools on Windows - I don't know it just seems like nonsense. Dual boot, vmware, virtual box, hackintosh, whatever means necessary, but it's always been my fundamental belief that I work in an environment as close as possible to what I'm deploying to. You're just creating headaches for yourself otherwise.
And if this is headed towards any kind of production/hosting thing, you - or someone above you - are making a series of pretty poor decisions. I'm not sure there is any other way to put it.
MySQL has bugs, but the basic functionality works on Windows. Or so I assume, considering that those bugs don't seem to have prevented tens of thousands of PHP developers from making good use of MySQL when developing their software. (I wouldn't know, of course -- I use Mac OS and Linux.)
I'm sure the bugs are annoying, though. Maybe the people who care about them should get out the GPL source code and fix them. Or maybe they should pay someone who knows how to fix them. Or maybe they should do what just about everybody else seems to do -- calculate the price of options one and two, compare them to the cost of running Linux, and deploy on Linux.
Or they can bang on the table and cry in the hopes that someone will fix their problems for free. Good luck with that.
I assume Windows developers and Microsoft care about supporting Windows, and develop on the platform. MySQL developers don't sound as if they do anything on Windows.
Given that MySQL put the Windows download link first in their list of downloadable binaries, you would have thought that it was a platform they might use for a fair percentage of their development work.
Or better yet, use any of the FREE versions of SQL Server specifically slotted for the desktop.
For me it has been the Golden Bridge allowing me to develop locally on a Windows machine and to deploy in the real world on LAMP!
btw, the SQLyog product that Webyog makes is awesome. I use it all day, everyday.