They’re just whining cuz it hasn’t been released yet. But once again the phusion guys have reposted the source on github.
I do think this should be remerged back into ruby core, but I’m glad that we don’t have to wait for that process to complete before getting access to the source.
Mark —
judofyr: The “Phusion way” is rapidly becoming the de-facto way in the Ruby (and Linux kernel) world due to Git. When you want to work on a Git project, from, say, Github, you can “fork” it, work on your own version, then anything the original maintainer likes can be merged up at some point. The Linux kernel has thrived on this model in recent years, in fact.
That said, there does seem to be a big PR / buzz angle on Phusion’s stuff which seems a little unsavory, but there’s nothing /wrong/ with it.. it’s just unusual. They do seem to pick their collaborators rather selectively.
PeterCooper —
@PeterCooper Yeah… I’d wouldn’t call the PR/buzz angle on Phusion’s stuff ‘unsavory’, I’d call it ‘refreshing’ or maybe ‘awesome.’ It’s sweet product with some creative marketing that builds some well deserved hype.
I agree w/ you though–their way of fork, develop and possibly merge is definitely becoming the de-facto–and as lifo pointed out, Matz has already asked about merging it into 1.9.
marcus —
PeterCooper: “forking the project”, not “forking the code”. They’ve forked the project and is promoting it as a different product. The only reason they would have done that, is to create hype around Phusion. (I’ve updated the post).
marcus: As a programmer, I prefer openess over “creative marketing”. If a product is awesome enough, they don’t need “creative marketing”.
lifo: Did you see any source of Passenger before 1.0? The COW-patch was originally a project by Hongli and completely open, but when they figure out they wanted brand it with Phusion they closed the doors…
judofyr —
My major beef with REE is the name and what it implies. When I first read the link on the mod rails site, before the REE site was really up, I was convinced it was a joke. Putting the term “Enterprise” onto anything suggests what kind of people/companies you want to sell this to - and I’m obviously not there. I’d rather not by something with an enterprise sticker on it. I do know quite a few people who would, though, and to them the Enterprise sticker is an assurance of quality. Marketing something as Enterprisey could mean spreading FUD, which shouldn’t be possible if you’ve actually created something of value.
Marius Mathiesen —
We had ALREADY sent this patch to the Ruby core mailing list. Search the archives if you don’t believe me. We’re providing Ruby Enterprise Edition because upstream is unclear about what they intent to do with the patch. But in the mean time, there exists clear demand for the functionality provided by this patch, and so we’ve chosen to temporarily maintain our own fork.
The eventual fate of Ruby Enterprise Edition is, and has been since the beginning, to be merged back upstream. This is explained on the Ruby Enterprise Edition website, which is currently being worked on. This is also why we have chosen to use git as version control system: so that any changes we make can be relatively easily merged back. This all has been planned since the very beginning.
The reason why we pulled it out temporarily is because we felt that there are improvements to be made. We don’t want people to use a half-baked product, because first experiences are very important.
We are a commercial company, and we’ve never made this a secret. We could have easily chosen not to write/publish any of this at all. And yet we have, and we’re even providing all this to you for free. I’m not sure what your point is. Regarding any kinds of profit that we MIGHT generate, should we apologize for trying to put bread on the table? That kind of defeats the point of being a company right?
@judofyr Good marketing doesn’t preclude openness.
It’s the whole idea that marketing isn’t necessary that relegates most OSS projects to obscurity. Rails, for example, succeeded because in addition to being a great framework, it was marketed really well.
Who knows how many great projects have never made it off the ground because the developers were content to let the project stand on technical merit alone.
marcus —
You can’t be enterprise with a logo like that!!!!
carllerche —
Have any of you started working on an open source project and NOT released the code from the get-go because maybe you wanted to perfect it first? Or maybe you got it working okay but the source needed serious refactoring.
Who cares that the source wasn’t available for Phusion’s projects … it is now! And the Phusion guys are very nice and open to contributions. Wanna collaborate? Fork, make your changes, & send them a pull request. I did and they accepted my contribution.
That said, I was very sad when the releases didn’t happen as scheduled - I think they didn’t realize how busy they would be at RailsConf and, before they knew it, they were on their way back to the Netherlands. It’s all good.
mod_rails is awesome and it’s something we all needed … adding Rack/WSGI support makes it friggin amazing.
Instead of bitching and moaning, we should be thanking Phusion and helping them make some money … maybe by donating money or by donating code contributions or by ‘purchasing’ licenses for mod_rails (aka donations) or … whatever.
Thanks Phusion guys. You rock.
remi —
remi: In response to your first question, no, at least not with the Ruby source code, because that’s against its license - http://www.ruby-lang.org/en/LICENSE.txt. You are not allowed to modify the Ruby source code without licensing it under the GPL or an alternative Ruby license that requires you make the source code publicly available.
However, this is a non-issue here since they have already demonstrated the patch publicly, although if they progress to doing any further work that isn’t publicly released, it has to satisfy the terms of the GPL. Passenger, however, as their own creation, can be licensed however they see fit.
PeterCooper —
They indeed did see if they could migrate it into core. I believe it made its way to 1.9, though it might not have. They were back porting it to 1.8.6, so we could benefit from it, which thing will never happen in core, since 1.8.6 is “stable and not due to larger changes” or what not.
Comments
Ill-informed and super lame article.
Well, super lame mainly because I hate stupid blogs without comments. So there.
And ill-informed because, oh well, take some time to read comments here at http://izumi.plan99.net/blog/index.php/2008/01/14/making-ruby%E2%80%99s-garbage-collector-copy-on-write-friendly-part-7/#comments
Once again, lame blog post.
They’re just whining cuz it hasn’t been released yet. But once again the phusion guys have reposted the source on github.
I do think this should be remerged back into ruby core, but I’m glad that we don’t have to wait for that process to complete before getting access to the source.
judofyr: The “Phusion way” is rapidly becoming the de-facto way in the Ruby (and Linux kernel) world due to Git. When you want to work on a Git project, from, say, Github, you can “fork” it, work on your own version, then anything the original maintainer likes can be merged up at some point. The Linux kernel has thrived on this model in recent years, in fact.
That said, there does seem to be a big PR / buzz angle on Phusion’s stuff which seems a little unsavory, but there’s nothing /wrong/ with it.. it’s just unusual. They do seem to pick their collaborators rather selectively.
@PeterCooper Yeah… I’d wouldn’t call the PR/buzz angle on Phusion’s stuff ‘unsavory’, I’d call it ‘refreshing’ or maybe ‘awesome.’ It’s sweet product with some creative marketing that builds some well deserved hype.
I agree w/ you though–their way of fork, develop and possibly merge is definitely becoming the de-facto–and as lifo pointed out, Matz has already asked about merging it into 1.9.
PeterCooper: “forking the project”, not “forking the code”. They’ve forked the project and is promoting it as a different product. The only reason they would have done that, is to create hype around Phusion. (I’ve updated the post).
marcus: As a programmer, I prefer openess over “creative marketing”. If a product is awesome enough, they don’t need “creative marketing”.
lifo: Did you see any source of Passenger before 1.0? The COW-patch was originally a project by Hongli and completely open, but when they figure out they wanted brand it with Phusion they closed the doors…
My major beef with REE is the name and what it implies. When I first read the link on the mod rails site, before the REE site was really up, I was convinced it was a joke. Putting the term “Enterprise” onto anything suggests what kind of people/companies you want to sell this to - and I’m obviously not there. I’d rather not by something with an enterprise sticker on it. I do know quite a few people who would, though, and to them the Enterprise sticker is an assurance of quality. Marketing something as Enterprisey could mean spreading FUD, which shouldn’t be possible if you’ve actually created something of value.
We had ALREADY sent this patch to the Ruby core mailing list. Search the archives if you don’t believe me. We’re providing Ruby Enterprise Edition because upstream is unclear about what they intent to do with the patch. But in the mean time, there exists clear demand for the functionality provided by this patch, and so we’ve chosen to temporarily maintain our own fork.
The eventual fate of Ruby Enterprise Edition is, and has been since the beginning, to be merged back upstream. This is explained on the Ruby Enterprise Edition website, which is currently being worked on. This is also why we have chosen to use git as version control system: so that any changes we make can be relatively easily merged back. This all has been planned since the very beginning.
The reason why we pulled it out temporarily is because we felt that there are improvements to be made. We don’t want people to use a half-baked product, because first experiences are very important.
We are a commercial company, and we’ve never made this a secret. We could have easily chosen not to write/publish any of this at all. And yet we have, and we’re even providing all this to you for free. I’m not sure what your point is. Regarding any kinds of profit that we MIGHT generate, should we apologize for trying to put bread on the table? That kind of defeats the point of being a company right?
Long reply here: http://izumi.plan99.net/blog/index.php/2008/06/05/re-ruby-fishy-edition/
@judofyr Good marketing doesn’t preclude openness.
It’s the whole idea that marketing isn’t necessary that relegates most OSS projects to obscurity. Rails, for example, succeeded because in addition to being a great framework, it was marketed really well.
Who knows how many great projects have never made it off the ground because the developers were content to let the project stand on technical merit alone.
You can’t be enterprise with a logo like that!!!!
Have any of you started working on an open source project and NOT released the code from the get-go because maybe you wanted to perfect it first? Or maybe you got it working okay but the source needed serious refactoring.
Who cares that the source wasn’t available for Phusion’s projects … it is now! And the Phusion guys are very nice and open to contributions. Wanna collaborate? Fork, make your changes, & send them a pull request. I did and they accepted my contribution.
That said, I was very sad when the releases didn’t happen as scheduled - I think they didn’t realize how busy they would be at RailsConf and, before they knew it, they were on their way back to the Netherlands. It’s all good.
mod_rails is awesome and it’s something we all needed … adding Rack/WSGI support makes it friggin amazing.
Instead of bitching and moaning, we should be thanking Phusion and helping them make some money … maybe by donating money or by donating code contributions or by ‘purchasing’ licenses for mod_rails (aka donations) or … whatever.
Thanks Phusion guys. You rock.
remi: In response to your first question, no, at least not with the Ruby source code, because that’s against its license - http://www.ruby-lang.org/en/LICENSE.txt. You are not allowed to modify the Ruby source code without licensing it under the GPL or an alternative Ruby license that requires you make the source code publicly available.
However, this is a non-issue here since they have already demonstrated the patch publicly, although if they progress to doing any further work that isn’t publicly released, it has to satisfy the terms of the GPL. Passenger, however, as their own creation, can be licensed however they see fit.
They indeed did see if they could migrate it into core. I believe it made its way to 1.9, though it might not have. They were back porting it to 1.8.6, so we could benefit from it, which thing will never happen in core, since 1.8.6 is “stable and not due to larger changes” or what not.
Post a comment