GraphQL
JakeHartnell opened this issue · 10 comments
@JakeHartnell Do you want to stick with polymer as the frontend?
For now. Unless it's lots of work...
Thoughts?
I guess it is okay to continue with polymer for now. But In my opinion react is the better FE framework.
I guess it is okay to continue with polymer for now. But In my opinion react is the better FE framework.
Polymer was good experiment, but I agree. As of now react is much better for building something we can put into production. I was noticing a decent sized list of problems in browsers that weren't chrome...
@JakeHartnell In my opinion polymer is good enough (even better in some respects) for small apps/prototypes. My biggest issue with polymer is it's limited programming model. ie helper elements, bounded properties and helper functions cant match the flexibility offered by react. I hate the way they designed observers(for example compared to $watch in vuejs) and behaviors( it is getting deprecated in 2.0, thankfully), the use of link imports (which is way inferior to es6 imports). One issue I had in the past with the state management and data flow can now be solved using polymer-redux or google/uniflow-polymer. I'm trying to solve the lack of build tools myself. I think these issues will get solved as the popularity increases.
Biggest plus with polymer is it's simplicity(there is nothing much to understand anybody can create a functioning front end). Learning curve is almost nonexistent. You'll get things done much faster in the initial phase.
@aruntk It would be cool if we used https://github.com/CommonGarden/graphql-api in this main Grow-IoT repo.
This is cool: https://github.com/APIs-guru/graphql-voyager
So, we're going to try an experiment and test out our new Open Collective: https://opencollective.com/grow-iot
If we can get Grow-IoT using GraphQL then we'll be a step closer to moving away from Meteor and MongoDB, freeing people up to connect different GraphQL endpoints (and databases such as Postgres, Dynamodb, timescaledb, etc).