Roadmap and pending pull-requests

[Wild Nomad]

Hi!

I have observed that the development pace of Angularjs has slowed down considerably since last September. There are not many new commits other than minor docs/bug fixes.

According to this official blog entry, we should be already enjoying 1.2.0 for a month now!!!

But what makes me really sad is that I also counted a lot of opened pull-requests that meet the contributor guidelines, some of them that include very interesting new features. Sadly, they seem to be ignored by the main devs. This should be very frustrating for anyone that has invested some of their precious time in forking, making some changes to the code/documentation and implementing the required tests only to find their work lost in the bottom of the pending PR stack. I think that they deserve at least a reply by one the the dev team saying if their effort is going to be taken into account anytime soon.

C'mon! There are only 58 opened PRs and most of them are simple enough! Spending 5 minutes per PR, you could dispatch 6 of them in half an hour. If devs took just that half an hour per day, they could catch up in 10 days! This could dynamise contributions: no one will want to help if you are keeping this attitude for much longer.

Don't get me wrong. I only want Angularjs to get better and better, I just can't stand watching so much effort wasted in contributions and nice ideas that go nowhere without any explanation from the dev team.

Mike.

[Naomi Black]

Hi Mike,

Thanks for the nudge. As part of the dev team, I'm truly sorry that we've been moving more slowly recently. 

We took on more education/speaking gigs in the past few months, and got behind on our roadmap, as well as other things like reviewing PRs. I wish we could clear PRs in 5 min each :) Some are pretty quick to resolve, but many take more serious thought to review. Our goal isn't just to clear the PRs, but to ensure that we keep moving forward in a consistent and planned way. 

The good news is that most of our European tour is behind us and we're starting to catch up again -- you should see a lot more recent commits in the queue -- and we're working hard to get through the backlog and up to 1.2 very soon.

At any rate, we're sorry for the frustration and very glad to see the PRs that come in. We should have a roadmap update for you soon! And looking ahead, we're going to plan a little better around how we schedule talks around release timelines in the new year. 

-naomi

[Igor Minar]

On Sun, Nov 25, 2012 at 7:41 PM, Wild Nomad <wildn...@gmail.com> wrote:

Hi!

I have observed that the development pace of Angularjs has slowed down considerably since last September. There are not many new commits other than minor docs/bug fixes.

According to this official blog entry, we should be already enjoying 1.2.0 for a month now!!!

But what makes me really sad is that I also counted a lot of opened pull-requests that meet the contributor guidelines, some of them that include very interesting new features. Sadly, they seem to be ignored by the main devs. This should be very frustrating for anyone that has invested some of their precious time in forking, making some changes to the code/documentation and implementing the required tests only to find their work lost in the bottom of the pending PR stack. I think that they deserve at least a reply by one the the dev team saying if their effort is going to be taken into account anytime soon.

C'mon! There are only 58 opened PRs and most of them are simple enough! Spending 5 minutes per PR, you could dispatch 6 of them in half an hour. If devs took just that half an hour per day, they could catch up in 10 days! This could dynamise contributions: no one will want to help if you are keeping this attitude for much longer.

I really wish it was that easy. The quality of an average PR is unfortunately very low. Often there are tests missing, or tests don't actually test the code change, documentation missing, bugs in the code, etc. Don't get me wrong, we are very grateful for all these contributions and we are going to work through them all, but even for a simple changes (except for doc changes) it often takes 15min to an hour to get all the issues resolved and the single PR merged. And that's just for simple PRs...

When we merge something in, I want to be confident that it's a change that we can rely on. AngularJS codebase is a very complex machinery in which every piece relies on stability of other pieces and we use tests to verify that all contracts between these pieces are properly implemented. Making lots of changes to AngularJS is easy, but making them in a way that would allow us to be confident in the codebase and able to evolve the project further is a completely different beast.

But yes, we do need to figure out a way to speed things up. I'm happy to share with you that we are growing the core team at Google and we've also taken on external committers - Peter Bacon Darwin and Pawel Kozlowski (awesome guys you likely know from the mailing list already) who are already helping us with triaging issues and PRs.

As Naomi mentioned we are behind the planned schedule (due to one reason or another), and we also realize that the PR queue is frustrating, but I just want you to know that *we do care* and we are working on making things better.

cheers,

Igor

Don't get me wrong. I only want Angularjs to get better and better, I just can't stand watching so much effort wasted in contributions and nice ideas that go nowhere without any explanation from the dev team.

Mike.

--
You received this message because you are subscribed to the Google Groups "AngularJS" group.
To post to this group, send email to ang...@googlegroups.com.
To unsubscribe from this group, send email to angular+u...@googlegroups.com.
Visit this group at http://groups.google.com/group/angular?hl=en-US.
 
 

[wildn...@gmail.com]

Hi Igor,

As I said to Naomi, thank you very much for your update.

I recognise being a little moron about PR review time, sorry. I understand that landing a bad PR could have very nasty consequences and, therefore, you have to think twice (or thrice :-)

May be the only thing needed is just a little feedback on each new PR about if it looks good or not (i.e. not a good idea, needs more thinking, non-compliance with the guidelines, not well documented enough, non-existent or wrong tests, etc). This way the contrib can ammend his/her commits before the PR gets fully reviewed. This would also allow you have a deep look inside days later... and still everyone would be happy.

I really celebrate that Peter and Pawel got promoted. They spend a lot of time into the group and in GitHub. Congrats to both of you!

Did I say that I love working with Angularjs? :-) Thanks for sharing it with us.

Mike.

[Wild Nomad]

Hi Naomi,

Thanks for the update! I will be eagerly waiting for the new roadmap... And yes, I've been way too optimistic with the 5 mins per PR :) 

Cheers,

Mike.

2012/11/25 Naomi Black <nao...@google.com>

[Dan Denno]

If I see another simple angularjs implementation I'm going to puke.  It would be nice for the angularjs team to do more advanced applications talks.

[Paul Hammant]

At some level, I think that's for us in the community to deliver.

[Pawel Kozlowski]

On Mon, Nov 26, 2012 at 9:14 PM, Paul Hammant <pa...@hammant.org> wrote:
> At some level, I think that's for us in the community to deliver.

Couldn't agree more. I think that we are going to see more patterns
and best practices as AngularJS is used in more project. This is why
it is important to share community experience, be clear on what works
in practice and what is not. This is exactly why Peter and I started
to experiment with different approaches in a sample app.

Paul, are you, by any chance the same person who wrote Pico container?

Cheers,
Pawel

[Igor Minar]

On Mon, Nov 26, 2012 at 8:41 PM, <wildn...@gmail.com> wrote:

Hi Igor,

As I said to Naomi, thank you very much for your update.

I recognise being a little moron about PR review time, sorry. I understand that landing a bad PR could have very nasty consequences and, therefore, you have to think twice (or thrice :-)

May be the only thing needed is just a little feedback on each new PR about if it looks good or not (i.e. not a good idea, needs more thinking, non-compliance with the guidelines, not well documented enough, non-existent or wrong tests, etc). This way the contrib can ammend his/her commits before the PR gets fully reviewed. This would also allow you have a deep look inside days later... and still everyone would be happy.

right.. we are going to be better about this. but just by looking at the diff it's not always obvious that a test for certain corner case is missing. one has to check out the code, play with it a bit and only then things are obvious, but by then you are already 10min into the review.

Peter and Pawel have been doing an amazing job lately at screening PRs and issues and pointing out the obvious stuff before we even get to look at them. This seems to be working really well. 

/i

[l...@whitneyland.com]

Does this post still apply 1 1/2 months later, or am I missing an increased pace of development?  Are we allowed to know specifically what the core team is currently working on?  This is not complaining, this is just needing to know what's going on so that we can try to get involved and help.

On a more positive note, it seems the pace of interest in angular is actually doing well.  I made a quick utility to compare momentum on StackOverflow here:

http://data.stackexchange.com/stackoverflow/query/90306/tag-battle-mass-and-momentum

If you enter "knockout.js" and "angularjs" for the tags, you can see the activity vs. other MVC frameworks.

On Sunday, November 25, 2012 12:41:53 PM UTC-6, Wild Nomad wrote:

Hi!

I have observed that the development pace of Angularjs has slowed down considerably since last September. There are not many new commits other than minor docs/bug fixes.

According to this official blog entry, we should be already enjoying 1.2.0 for a month now!!!

But what makes me really sad is that I also counted a lot of opened pull-requests that meet the contributor guidelines, some of them that include very interesting new features. Sadly, they seem to be ignored by the main devs. This should be very frustrating for anyone that has invested some of their precious time in forking, making some changes to the code/documentation and implementing the required tests only to find their work lost in the bottom of the pending PR stack. I think that they deserve at least a reply by one the the dev team saying if their effort is going to be taken into account anytime soon.

C'mon! There are only 58 opened PRs and most of them are simple enough! ...<edit>... This could dynamise contributions: no one will want to help if you are keeping this attitude for much longer.