They owe the community a transparent and open development process. They are doing this with Lighthouse and GitHub. Anybody can submit a bug, patch, watch other tickets, see commits being made every single day.
They also owe the community responsible release management by releasing stable code. They are doing this by delaying the RC.
So in my books, the complainers really have NO angle here.
You missed one thing: They owe the community reasonable expectations of when new releases will drop. That doesn't mean they say a date and always hit it, it just means that they communicate that they are behind their original schedule, a new estimate, and a list of blockers. That way, the community knows what they can do to help.
Exactly, and while they certainly don't have to do any of that, when you have commercial developers counting on you, you will lose your community over time with a mantra of "Fuck you."
Also, while you could go digging through tickets to figure out what's holding it up, why have a release manager?
It seems as though the honeymoon has been over for some time with rails anyway. Between arrogent devs and the Rails 3 delay, I've moved on.
"We, too, are pretty annoyed that the release candidate has dragged on, but we want a candidate we'd feel comfortable deploying to production with our own apps. Thankfully, we've had tons of people testing the betas and bringing the polish up to release quality. That means a short RC cycle."
All told, I have to say that this makes me feel better about Rails 3 than I would otherwise. I'm perfectly happy building apps in Rails 2.3.x, and would rather switch to Rails 3 when the likelihood of it containing showstopper bugs that will make me get up at 4AM to fix production issues has decreased.
That's a very polite way to put it, imo. These same people demanding an immediate release would also be the first to weep prose all over the internets upon the first contact with a bug.
No one was demanding an immediate release. They were just asking for some form of communication. Their expectations were set for a release back in June. That date came and went without any information. A month and a half later, it takes someone asking about it just to get "real soon" as a reply.
Go take a look at how Django handles this: http://www.djangoproject.com/weblog/. For the 1.2 release, they posted every week with a detailed update of when they thought the RC would drop, what bugs were holding it back, and what areas needed the most help. Each week, you could see them getting closer, and when they got delayed, you understood why.
If I used Rails and had to put up with a release cycle that didn't meet my expectations, you can be sure I would fork it. The bunch of people in that thread can organize themselves and get intimate with the source tree (or switch to something else.)
Is it possible for both sides of an argument to be wrong? Yes, communication would have been nice. However, it's not like the workings of the Rails Core group is secret. You can look at the commit logs and get a feel for what's going on and how close they actually are.
Expecting open source developers to give accurate estimates is kind of silly. However, responding with "Fuck you" is unnecessarily rude.
It's not a matter of accurate estimates -- no one can do that -- it's a matter of managing expectations, and that is something a group of open source devs can do: http://news.ycombinator.com/item?id=1545892
Is it possible for both sides of an argument to be wrong?
Assuming that wasn't a rhetorical question: certainly. That's actually mostly the case, which is also the reason why debates can drag on endlessly, each one pointing out wrong parts in (usually irrelevant) parts of the arguments of the other, while drifting further and further from the original point. If people forget to focus on cooperation and reaching consensus, then constructive discussions are nearly impossible.
This Greg Donald guy is totally out of line, but he's just an irrelevant troll.
Edit: to the downvoters, what is your argument that he either a) isn't totally out of line or b) is relevant to the development of Rails or the Rails ecosystem despite his nearly complete lack of open source contributions (http://github.com/gdonald)?
The title here is "You stay classy, Rails," yet there is a single unnecessarily rude response from a Rails core member in response to criticism, followed by a bunch of inflammatory posts by a guy who apparently is just an immature troll who doesn't speak for anyone but himself.
> I was looking at Padrino recently and rather like it. It's a young
framework, nowhere near as mature, but has an interesting architectural
"value prop"... and at least so far I haven't gotten any STFUs from the
development team. :)
The client shouldn't care how the core team treats his/her developer. The client should only care about the product. While an insult from a core member may hurt feelings, it doesn't make ActiveRecord any less useful.
"The client should only care about the product. While an insult from a core member may hurt feelings, it doesn't make ActiveRecord any less useful."
Maybe. But you know that every lib and app has bugs, so when you start looking at tools you take into account how reliable it is now, and what happens down the line as bugs or issues are uncovered.
If the project members seem to have a shitty attitude about users' concerns, you should have second thoughts.
Realistically you want to look at the big picture and be a bit pragmatic about these things. People should by now know what to expect from various Ruby cliques, decide if it's worth the annoyance, and act accordingly. Overall, stuff gets done in Rails, even if not on the schedule people might want.
However, people should also look at the dozen or so other Ruby Web frameworks and see if their communities might not be more the kind they want to be a part of. The Nitro, and now Ramaze, community, was and is a big reason for my interest. The people and the code rocks.
Initially, I was going to flag this post. However, the link sets the context for what could be an interesting discussion.
What is the proper way to handle releasing open source projects? Does having a flippant, almost combative, attitude help project development? Granted, I would have preferred a blog post analysis rather with link to this flamewar as part of its evidence. Nevertheless, it could spawn a good conversation, as long as it doesn't turn to a flamewar on HN.
The right attitude to this sort of thing is simple, treat your open source users as you would your customers in a commercial setting.
The fact that the source is open does not mean the general rules of conduct change. A professional attitude coupled with quality stuff is what will get you contributions and respect.
For comparison, the recent Django 1.2 release estimates:
Update Tickets Blocking RC 1 Est Rel Est
Mar 09 120 -- Mar 22 Mar 29
Mar 16 84 -- Apr 05 Apr 12
Mar 23 69 -- Apr 05 Apr 12
Apr 01 54 23 Apr 19 Apr 26
Apr 07 46 14 Apr 19 Apr 26
Apr 14 49 14 Apr 26 May 03
Apr 22 51 8 May 03 May 10
Apr 28 48 2 May 03 May 10
May 05 --* 0 Released May 10
May 15 -- 0 Released May 17
May 17 -- 0 Released Released
These were taken from the Django blog: http://www.djangoproject.com/weblog/. About once a week, they had a comprehensive update with bug counts and revised estimates. If anyone wanted to help with the release, the blog told them exactly what parts needed the most work. They were nearly 2 months later than their initial estimate, but it wasn't a problem because everyone knew what to expect.
* Total ticket counts weren't given after RC1 dropped. All open tickets as of RC1 were translation or documentation related.
I'm perfectly fine with "it's ready when it's ready". But if you choose to give yourself a deadline by saying "RC in days" and a couple of months later there is still no RC, you should be able to empathize with people who are curious about the delay.
Why is this news? This doesn't belong on HN. One Rails community member got uppity and Jeremy linked the classic DHH FU image as a joke.
Come on people.
You may as well be asking where Ruby 2.0 is seeing as how we were supposed to have it years ago. Instead, most of us use flavors of 1.8.7, dabble with 1.9.x, and haven't heard any fish tales of 2.0 in a while.
As others have said, it's open source. Rails 3.0 will be out soon enough.
In the meantime, someone else mentioned Padrino. As a fan of things minimalist, I'll +1 people taking a moment to look at it while they wait on Rails 3.
25 comments
[ 2.8 ms ] story [ 54.8 ms ] threadThey owe the community a transparent and open development process. They are doing this with Lighthouse and GitHub. Anybody can submit a bug, patch, watch other tickets, see commits being made every single day.
They also owe the community responsible release management by releasing stable code. They are doing this by delaying the RC.
So in my books, the complainers really have NO angle here.
Also, while you could go digging through tickets to figure out what's holding it up, why have a release manager?
It seems as though the honeymoon has been over for some time with rails anyway. Between arrogent devs and the Rails 3 delay, I've moved on.
"We, too, are pretty annoyed that the release candidate has dragged on, but we want a candidate we'd feel comfortable deploying to production with our own apps. Thankfully, we've had tons of people testing the betas and bringing the polish up to release quality. That means a short RC cycle."
http://www.ruby-forum.com/topic/213412#927556
All told, I have to say that this makes me feel better about Rails 3 than I would otherwise. I'm perfectly happy building apps in Rails 2.3.x, and would rather switch to Rails 3 when the likelihood of it containing showstopper bugs that will make me get up at 4AM to fix production issues has decreased.
Go take a look at how Django handles this: http://www.djangoproject.com/weblog/. For the 1.2 release, they posted every week with a detailed update of when they thought the RC would drop, what bugs were holding it back, and what areas needed the most help. Each week, you could see them getting closer, and when they got delayed, you understood why.
Expecting open source developers to give accurate estimates is kind of silly. However, responding with "Fuck you" is unnecessarily rude.
Edit: to the downvoters, what is your argument that he either a) isn't totally out of line or b) is relevant to the development of Rails or the Rails ecosystem despite his nearly complete lack of open source contributions (http://github.com/gdonald)?
The title here is "You stay classy, Rails," yet there is a single unnecessarily rude response from a Rails core member in response to criticism, followed by a bunch of inflammatory posts by a guy who apparently is just an immature troll who doesn't speak for anyone but himself.
> I was looking at Padrino recently and rather like it. It's a young framework, nowhere near as mature, but has an interesting architectural "value prop"... and at least so far I haven't gotten any STFUs from the development team. :)
The client shouldn't care how the core team treats his/her developer. The client should only care about the product. While an insult from a core member may hurt feelings, it doesn't make ActiveRecord any less useful.
Maybe. But you know that every lib and app has bugs, so when you start looking at tools you take into account how reliable it is now, and what happens down the line as bugs or issues are uncovered.
If the project members seem to have a shitty attitude about users' concerns, you should have second thoughts.
Realistically you want to look at the big picture and be a bit pragmatic about these things. People should by now know what to expect from various Ruby cliques, decide if it's worth the annoyance, and act accordingly. Overall, stuff gets done in Rails, even if not on the schedule people might want.
However, people should also look at the dozen or so other Ruby Web frameworks and see if their communities might not be more the kind they want to be a part of. The Nitro, and now Ramaze, community, was and is a big reason for my interest. The people and the code rocks.
What is the proper way to handle releasing open source projects? Does having a flippant, almost combative, attitude help project development? Granted, I would have preferred a blog post analysis rather with link to this flamewar as part of its evidence. Nevertheless, it could spawn a good conversation, as long as it doesn't turn to a flamewar on HN.
The fact that the source is open does not mean the general rules of conduct change. A professional attitude coupled with quality stuff is what will get you contributions and respect.
* Total ticket counts weren't given after RC1 dropped. All open tickets as of RC1 were translation or documentation related.
Come on people.
You may as well be asking where Ruby 2.0 is seeing as how we were supposed to have it years ago. Instead, most of us use flavors of 1.8.7, dabble with 1.9.x, and haven't heard any fish tales of 2.0 in a while.
As others have said, it's open source. Rails 3.0 will be out soon enough.
In the meantime, someone else mentioned Padrino. As a fan of things minimalist, I'll +1 people taking a moment to look at it while they wait on Rails 3.