Meteor Community Coding


#1

There are lots of examples out there, lots of frameworks that grew because of the community. Not only posting issues, but actually coding for the open-source platform.

Now, I’m sure there are lots of people, who can invest a bit of time in the platform they love, and help these guys out. From my understanding we currently have 2-3 FTE on Meteor. This is sad. Imagine there are 100 people that want to help, and invest 10-15 hours a month.

The plan
We need a Meteor documentation. Not how to use it, but how it is architected. The code is documented, but it took me so much time in order to understand how they did oplog… maybe because I’m a bit slow, I don’t know, but I now I think I know by heart the code in: https://github.com/meteor/meteor/tree/master/packages/mongo because I wanted to move oplog to Redis. https://github.com/cult-of-coders/redis-oplog

We need a “community-help” label for issues. So guys like @benjamn, @abernix, @sashko, etc can simply say: this is not a super important thing to do right now, we need it but later, so let others try and do it “Community Help”, and leave a small guide-line on what to tackle, where to look. I would really like to do that in my spare time! Ofcourse, if writing the guide-line and time to explain it takes more than actually fixing it, maybe it’s not a good idea :smiley:

Let’s discuss other ideas here. Thanks!


#2

Meteor has a pull-requests-encouraged label with over 100 items, and there are many tasks in there which are reasonable to accomplish – even for first timers. These are issues which directly fall into the category of “let others try” and are labeled that way by MDG.

The Meteor community has pulled together some great stuff in the past, and a lot of amazing Meteor packages exist, but after six months of triaging issues I’m a bit disappointed with the amount of contributions that come in for the core tool from the community – despite the many threads which continue to pop up about contributing. That doesn’t stop me or discourage me, but I don’t get paid for this, and I don’t work for MDG.

As I said on this comment on the most-recently-popular thread about contributing, if developers would help triage issues (or even help advise/debug by commenting on them) it would be a huge help to the project and you could do it in an hour a week. I’m spending more than 10 hours a week right now just doing this because, like you say, I want to “invest a bit of time in the platform I love.” I’ve built products (and want to build more products) on Meteor and it’s completely free to me.

In the process of triaging issues, I’ve learned a lot about the tool and it’s enabled me to contribute more actual code. While some aspects are of Meteor are more difficult than others, it’s really not that difficult to contribute, people just aren’t doing it.

I’m not sure what the chances of an architecture guide getting written are, but if someone sees any of those issues they want to work on, all they need to do is run Meteor from a checkout and comment on the issue they’re interested in and ask for advice on how to proceed with submitting a pull-request.

Developers are standing by.


#3

@abernix

Now, what you say is true, I had no idea honestly about “pull-request-encouraged” label. Why not ? Because it needs to be BIG and on www.meteor.com website, or somewhere. And at the same time I should’ve looked better.

The reason I got involved lately so much is because Meteor finds itself in a crucial place. With MDG shifting focus to Apollo, people moving away from it for more flexible solutions like Webpack, Feathers, etc it looks like Meteor heads towards its inevitable death. Unless we resuscitate it somehow.

people just aren’t doing it.

Well, I don’t see any marketing put in this. Just asking for help is nothing. There are certain ways to convince people to contribute:

  • Give rewards
  • Gamify the experience
  • Meteor hackatons (Like they had on Drupal)
  • Make them feel guilt for not contributing (?) (barbaric but effective)

This is for attracting individuals. Now there’s another market, the parasite companies which work on Meteor. I call them parasites because they don’t use galaxy, they don’t open-source their packages, they don’t contribute to the community at all (they just take). However, not to underestimate those companies, they have good coders, we could convert them to a symbiotic state by doing something like:

  • Have a method of “rating” the difficulty of a task in points
  • There should be a top with companies (most points) publicly available on meteor.com. This way they gain marketshare, Meteor gains coding power.

However, the top, can be composed of individuals, why not ? The reason they would do this, is to have it on their Resume. It’ll look good :slight_smile: