I was awaiting the 1.3 release, I really wanted to use several technologies that seemed to be very well aligned with 1.3. Then I read blog posts about the future, in particular reactive graphql. I don’t have a problem with graphql. My question is about starting a new app on a platform planning on a major change. If I choose meteor to write my app, I want to know that there is going to be updates and improvements for some time to come without a need to re-write major portions of my stack.
GraphQL (Reactive GraphQL) doesn’t seem like a small adjustment, it sounds like a major change to how the app communicates, apis, etc. I think it’s a good idea and I’m impressed by the plans that Meteor has.
I wonder if I should be starting a new project on Meteor 1.3 when it sounds like the MDG already has plans to move away from it? Am I wrong?
Hi! I’m the tech lead on Apollo (the new name for the Reactive GraphQL project). That’s a great question.
Meteor is always going to be moving forward, and some of those will be big changes, for example the new GraphQL-based Apollo stuff. There are a few things to consider here:
Apollo is going to be beta-quality for a long while. App development with GraphQL isn’t yet as well understood as with Meteor’s current stack, so there could be major changes as we figure out how this stuff is supposed to work. So while it will be usable pretty soon, it’s not the place to go if you want stability and peace of mind.
When we start recommending Apollo as the data stack of choice for new apps, we’re going to put in a lot of effort to make it easy for people to transition off of the current thing. This is one great reason to follow recommendations in the Meteor Guide: http://guide.meteor.com/ If you’re doing writes via Methods, and using publish-composite to publish nested data, and splitting your presentational and data loading UI components, then you’ll be in a great position to migrate to Apollo if and when you want to.
Improvements are still being made to the Mongo/DDP/Livedata stack. Meteor 1.3 came out with several new ways to optimize query performance (blog post on that coming soon!) and we’re still working on updating the Mongo driver and other fun stuff for future releases. We’re also hard at work on making it easier to contribute to Meteor so that you can contribute the features you need.
There are a looooot of people with serious production applications on the current stack. It’s going to stick around, so it’s a top priority to provide support for that for a long time and also make the path to migrate to the new Apollo stuff as clear and optional as possible.
At the end of the day, you should use whatever will enable you to get your project done. Technology is always changing and improving, but that doesn’t mean the stuff before stops working.
