Meteor meets GraphQL | Hacker News Reader

Meteor meets GraphQL

voice.kadira.io

112 pointspeterhunt11 comments

Hope some day Meteor meets Mysql.

By the way we have made air pollution monitor app using meteor. Will checkout possibilities of plugging GraphQL into it. http://oizom.com/

With GraphQL, it's possible to connect database or backend. That's the whole point here :)

Sure we will try out. If you get time check out Oizom app and suggest us can we get any benefit if use your kadira tool?

P.S I am still confused what exactly kadira does.

Okay. Kadira is a performance monitoring server for Meteor. With that, you can detect slowest parts in your app and place where it's uses a lot of server resources.

Then you can optimize accordingly. We've some guides. Start here: https://bulletproofmeteor.com/basics/what-is-performance

Thanks will check it out.

I agree, and I wish more JS projects in general would embrace SQL.

I think part of the problem is the Meteor team's reluctance to have Meteor thought of or even used as a traditional web framework.

Their vision is a framework for real-time applications that allows you to develop for web and mobile in parallel, but that puts them in a category where their intended use case is invariably losing to native every time. I think there are a lot of people who would love to use Meteor for RESTful mobile-first web application development, but are a little disappointed when they find Meteor is MongoDB-first and doesn't even have native support for routing.

For me, Meteor currently occupies a strange space where it's the web framework I'm the most excited about, yet it's the wrong tool for every project idea I have.

> For me, Meteor currently occupies a strange space where it's the web framework I'm the most excited about, yet it's the wrong tool for every project idea I have.

You've precisely captured a sentiment that I've been feeling but unable to express for some time. I think Meteor has tremendous potential, but there are a few very problematic issues, coupled with a severely walled ecosystem (Angular also has this problem, React I'm not so sure of but it seems to have it less) and unclear best practices (React has this in spades).

Meteor is working on it https://github.com/meteor/postgres-packages

I hope to see it soon.

You say that, but they've really slowed down in how much they commit - they were committing every couple of days but then it dropped off a couple months ago with only one rogue commit since. It doesn't get my hopes up.

What are the pros and cons of using this vs Relay?

Basically at the backend this is pure GraphQL. For Relay we need to have some Relay specific stuffs. That requires, more time.

In the client side too, we had to work a bit more.

That's why we go with Lokka[0]. It's just a client for any GraphQL backend. It has a simple cache.

It also supports Relay like declarative data fetching. Basically, you could build a complete app with it using Meteor's other tools like FlowRouter.

As a downside, you may need to fetch more data from the server.

As a benefit, you don't need to learn much to build a app. May be can reduce the code compared with Relay. (But that's subjective)

[0] - https://github.com/kadirahq/lokka

Reply on news.ycombinator.com