Comments on: MySQL: Now and Then http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/ because technology is just another ecosystem Sun, 21 Mar 2010 23:17:05 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: The Open Database Alliance and the Future of MySQL | Tech-monkey.info Blogs http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-547540 The Open Database Alliance and the Future of MySQL | Tech-monkey.info Blogs Tue, 19 May 2009 01:17:27 +0000 http://redmonk.com/sogrady/?p=2613#comment-547540 [...] “Stated more simply: as long as MySQL remains committed to the dual licensing model, it will be unable to accept the same patch set that open source only versions of the code can, because they do not share the same licensing concerns. Which is why we’ve seen these spring up.” - MySQL: Now and Then [...] [...] “Stated more simply: as long as MySQL remains committed to the dual licensing model, it will be unable to accept the same patch set that open source only versions of the code can, because they do not share the same licensing concerns. Which is why we’ve seen these spring up.” – MySQL: Now and Then [...]

]]> By: tecosystems » The Open Database Alliance and the Future of MySQL http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-546930 tecosystems » The Open Database Alliance and the Future of MySQL Thu, 14 May 2009 19:03:36 +0000 http://redmonk.com/sogrady/?p=2613#comment-546930 [...] “Stated more simply: as long as MySQL remains committed to the dual licensing model, it will be unable to accept the same patch set that open source only versions of the code can, because they do not share the same licensing concerns. Which is why we’ve seen these spring up.” - MySQL: Now and Then [...] [...] “Stated more simply: as long as MySQL remains committed to the dual licensing model, it will be unable to accept the same patch set that open source only versions of the code can, because they do not share the same licensing concerns. Which is why we’ve seen these spring up.” – MySQL: Now and Then [...]

]]> By: tecosystems » The People of MySQL http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-518939 tecosystems » The People of MySQL Fri, 13 Feb 2009 16:58:08 +0000 http://redmonk.com/sogrady/?p=2613#comment-518939 [...] project in immediate jeopardy of being forked and coopted. First, because the risk of forking was present already, and needed no help from Mårten, Monty and co. Second, because with all due respect to the [...] [...] project in immediate jeopardy of being forked and coopted. First, because the risk of forking was present already, and needed no help from Mårten, Monty and co. Second, because with all due respect to the [...]

]]> By: 451 CAOS Theory » Commercial open source community strategies in 2009 and beyond http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-510146 451 CAOS Theory » Commercial open source community strategies in 2009 and beyond Mon, 12 Jan 2009 15:47:26 +0000 http://redmonk.com/sogrady/?p=2613#comment-510146 [...] also some discussion of this over the Christmas period, sparked by Stephen O’Grady’s Q&A on whether MySQL’s dual-licensing strategy impacts its ability to generate contributions from [...] [...] also some discussion of this over the Christmas period, sparked by Stephen O’Grady’s Q&A on whether MySQL’s dual-licensing strategy impacts its ability to generate contributions from [...]

]]> By: Log Buffer #129: a Carnival of the Vanities for DBAs http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-506743 Log Buffer #129: a Carnival of the Vanities for DBAs Fri, 02 Jan 2009 17:38:47 +0000 http://redmonk.com/sogrady/?p=2613#comment-506743 [...] O’Grady was keeping one eye on the road and one on the rear-view mirror, to get his view on MySQL: Now and Then. Stephen cites Jeremy Zawodny’s item The New MySQL Landscape, which I mentioned in [...] [...] O’Grady was keeping one eye on the road and one on the rear-view mirror, to get his view on MySQL: Now and Then. Stephen cites Jeremy Zawodny’s item The New MySQL Landscape, which I mentioned in [...]

]]> By: 451 CAOS Theory » 451 CAOS Links 2009.01.02 http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-506707 451 CAOS Theory » 451 CAOS Links 2009.01.02 Fri, 02 Jan 2009 12:28:32 +0000 http://redmonk.com/sogrady/?p=2613#comment-506707 [...] MySQL: Now and Then Stephen O’Grady, Redmonk [...] [...] MySQL: Now and Then Stephen O’Grady, Redmonk [...]

]]> By: MySQL: Now and Then… and Dual License Community Impact at Jeremy’s Blog http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-504438 MySQL: Now and Then… and Dual License Community Impact at Jeremy’s Blog Wed, 24 Dec 2008 16:09:53 +0000 http://redmonk.com/sogrady/?p=2613#comment-504438 [...] O’Grady recently posted a Q&A pertaining to the past, present, and future of MySQL. There’s been quite a bit [...] [...] O’Grady recently posted a Q&A pertaining to the past, present, and future of MySQL. There’s been quite a bit [...]

]]> By: ZUrlocker http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-504083 ZUrlocker Tue, 23 Dec 2008 16:10:59 +0000 http://redmonk.com/sogrady/?p=2613#comment-504083 Steve, good posting. We've never been great at handling external contributions. In the early days, the founders were too busy to put together a good contributors agreement and in more recent years we did not have the manpower to review all the contributions. The good news is, it's something that we will be staffing more in 2009. We've got MySQL 5.1 out the door, it appears to be working well, and we will assign some resources on this issue to make it easier. We'll also benefit from some of the experts at Sun who have done a good job in this area. Look for some news in the new year and thanks for all the good input. Merry Christmas, happy new year everyone... --Zack Steve,
good posting. We’ve never been great at handling external contributions. In the early days, the founders were too busy to put together a good contributors agreement and in more recent years we did not have the manpower to review all the contributions.

The good news is, it’s something that we will be staffing more in 2009. We’ve got MySQL 5.1 out the door, it appears to be working well, and we will assign some resources on this issue to make it easier. We’ll also benefit from some of the experts at Sun who have done a good job in this area. Look for some news in the new year and thanks for all the good input.

Merry Christmas, happy new year everyone…
–Zack

]]> By: Ken Crandall http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-503649 Ken Crandall Mon, 22 Dec 2008 22:40:40 +0000 http://redmonk.com/sogrady/?p=2613#comment-503649 Interesting that you mentioned "telemetry" as what ought to be MySQL's primary focus. I think that Sun is both "missing the boat" AND "banging the drum too hard" about telemetry, in general. As far as "missing the boat" goes, they have an EXCELLENT tool in DTRACE, that allows metrics to be collected in a way that is almost transparent to the application itself. As you mentioned in your comments on Amber Road, this is what really sets their use of [Open]Solaris apart from Linux-based and BSD-based NAS implementations. If Sun were so bold as to build a true MySQL appliance with DTRACE underneath (or to provide MySQL as an add-on to Amber Road, with the associated DTRACE goodies plugging-into the provided web interface) then they have some interesting ways to show some data. On the other hand, Sun is "banging the drum too hard" with respect to metrics/telemetry, because that seems to be pretty much all they have to offer above Linux right now. If it weren't for ZFS and DTRACE, I'd argue that Sun's relevance would be EVEN LOWER than it is now. As it stands OpenSolaris seems to be an interesting direction for the company. However, it isone that is playing catch-up with Linux distributions, in terms of features and usability. It's one redeeming aspect is the inclusion Sun's technological bits, like the aforementioned ZFS and DTRACE. What is troublesome to me is the still murky relationship of OpenSolaris to Solaris proper. It is still woefully undefined and (I feel purposefully) obfuscated. I think that if Sun REALLY wants to put its money where its mouth is, they should go with a release process similar to that of Ubuntu -- have only ONE OS, and that should be OpenSolaris, and choose to define certain releases for LTS, so that their traditional customers can still consume technology as they are accustomed to. I did not mean to make this into a rant on [Open]Solaris and Sun, but I think that MySQL is a case of a company/project that has been succeeding in spite of itself, and that the legacy Sun folks as well as the new MySQL Sun folks need to both take a step back and re-evaluate what will make them both a successful company as well as successful developers of OSS. Interesting that you mentioned “telemetry” as what ought to be MySQL’s primary focus. I think that Sun is both “missing the boat” AND “banging the drum too hard” about telemetry, in general.

As far as “missing the boat” goes, they have an EXCELLENT tool in DTRACE, that allows metrics to be collected in a way that is almost transparent to the application itself. As you mentioned in your comments on Amber Road, this is what really sets their use of [Open]Solaris apart from Linux-based and BSD-based NAS implementations. If Sun were so bold as to build a true MySQL appliance with DTRACE underneath (or to provide MySQL as an add-on to Amber Road, with the associated DTRACE goodies plugging-into the provided web interface) then they have some interesting ways to show some data.

On the other hand, Sun is “banging the drum too hard” with respect to metrics/telemetry, because that seems to be pretty much all they have to offer above Linux right now. If it weren’t for ZFS and DTRACE, I’d argue that Sun’s relevance would be EVEN LOWER than it is now. As it stands OpenSolaris seems to be an interesting direction for the company. However, it isone that is playing catch-up with Linux distributions, in terms of features and usability. It’s one redeeming aspect is the inclusion Sun’s technological bits, like the aforementioned ZFS and DTRACE. What is troublesome to me is the still murky relationship of OpenSolaris to Solaris proper. It is still woefully undefined and (I feel purposefully) obfuscated.

I think that if Sun REALLY wants to put its money where its mouth is, they should go with a release process similar to that of Ubuntu — have only ONE OS, and that should be OpenSolaris, and choose to define certain releases for LTS, so that their traditional customers can still consume technology as they are accustomed to.

I did not mean to make this into a rant on [Open]Solaris and Sun, but I think that MySQL is a case of a company/project that has been succeeding in spite of itself, and that the legacy Sun folks as well as the new MySQL Sun folks need to both take a step back and re-evaluate what will make them both a successful company as well as successful developers of OSS.

]]> By: Carlo Daffara http://redmonk.com/sogrady/2008/12/21/mysql-now-and-then/comment-page-1/#comment-503631 Carlo Daffara Mon, 22 Dec 2008 22:12:11 +0000 http://redmonk.com/sogrady/?p=2613#comment-503631 We found within our FLOSSMETRICS research that in general dual licensing tends to results in lower external contributions to the core source code, and more or less no difference in "peripheral" contributions (in terms of external packages or supporting tools as an example). This was found to be consistent with network-like models of contribution in OSS (for a small summary look http://guide.conecta.it/index.php/6._FLOSS-based_business_models#Assessment_of_FLOSS_business_models_usage) I believe that dual licensing can be a reasonable business models for companies that are not reusing a large number of external OSS libraries and for which external contributions are not relevant. As soon as the desire to reduce engineering costs arise, however, there is a strong pressure to move to alternative monetization schemes. We found within our FLOSSMETRICS research that in general dual licensing tends to results in lower external contributions to the core source code, and more or less no difference in “peripheral” contributions (in terms of external packages or supporting tools as an example). This was found to be consistent with network-like models of contribution in OSS (for a small summary look http://guide.conecta.it/index.php/6._FLOSS-based_business_models#Assessment_of_FLOSS_business_models_usage)
I believe that dual licensing can be a reasonable business models for companies that are not reusing a large number of external OSS libraries and for which external contributions are not relevant. As soon as the desire to reduce engineering costs arise, however, there is a strong pressure to move to alternative monetization schemes.

]]>