MDG’s plans and priorities

Geoff Schmidt (MDG CEO) here, with an update on MDG’s plans and priorities. For the most part this is a summary of the talk that Matt’s giving tonight at Meteor Night, so for more information check out the talk.

Our priorities

  • Make Meteor the “best of JS”: continue to pull in the best components from around the JavaScript ecosystem, replacing Meteor-specific technology with industry standards as they march forward (eg, ES2015 modules in Meteor 1.3)

  • Broaden Meteor core package development beyond MDG, increasing the number of outside contributors and finally broadening the core team beyond MDG employees (including commit bits)

  • Support all databases, not just MongoDB, through an optional new data API based on reactive GraphQL

  • Grow our commercial offering so that we have something valuable for every Meteor developer

Meteor

Meteor 1.3

Meteor 1.3 will be out in a few weeks. It’s a huge leap forward in our “best of JS” strategy, with initial support for importing npm packages on both the client and the server using the ES2015 module syntax. It’s also got a new official testing system (also based on ES2015 modules), an improved native mobile container (with both performance and hot code push improvements), build tool speed improvements, and more.

Meteor in npm

Once 1.3 is out we’re going investigate whether it’s time to start publishing Meteor core itself through npm. Our hope is that this will make it much easier to divide up the maintenance responsibility for Meteor (for example, splitting Blaze off as its own repository with its own maintainers, and not tying the version of every core package in lockstep to the Meteor release version).

New build tool

Though Meteor 1.3 makes significant improvements, the build tool has gotten slow with all of the fancy functionality that has been added to it over the years. We’re going to look into rebuilding it, possibly on top of an existing build engine (maybe Webpack?) The goal is faster rebuilds and support for the latest build technology such as code splitting.

Apollo

Our project to build a reactive GraphQL based data system now has a name: Apollo.

Apollo aims to be a complete data stack for modern apps, combining everything that’s good about Meteor with everything that’s good about GraphQL. It’s designed to be database agnostic so it can be used with SQL databases, MongoDB, existing REST services, or any other data source you may have. And it’s designed from the ground up to scale predictably to very large numbers of simultaneous clients.

Apollo will be released as a set of independent npm packages that you can add to any app, Meteor or otherwise. For Meteor apps built on Mongo and Livedata, you’ll be able to try out this new GraphQL-based approach incrementally, starting with just one or two parts of your app and porting more parts over if you like it.

We’re still in the early stages of Apollo and would love to have your feedback on our high level design, and soon, your code contributions.

Core team changes

Our company is now just about four years old and the Meteor core team has had very low turnover over the years. Going forward you can expect to see some changes, with some new faces joining and some familiar faces moving on to new things. In 2016 our biggest focus is Apollo and our team composition will reflect that.

The business

I’m proud to announce that MDG now has over 1000 customers. Given that the commercial side of MDG is only about half a year old I’m pretty pleased with that number. Thank you all so much for your support – we’re here because of you, and we can’t do it without you.

What’s next? The next stop is general availability of Galaxy Developer Edition, so everyone can take advantage of “zero devops” hosting for their Meteor app at a low pay-as-you-go price. Other forthcoming features include Let’s Encrypt (instant SSL certificates) and additional region support. Looking farther ahead, expect to see us offering a complete set of services and support around Apollo.

More information

As I type, Matt’s wrapping up an information packed talk at Meteor Night. For more details on the above, check it out!

90 Likes

I’m most excited about NPM support on client and server (no more going to the server for slug generation!) and testing as a first-class Meteor core package.

Thank you guys for all of the hard work and giving developers these great tools to build things with.

7 Likes

Great stuff. Really looking forward to the new data layer.

Personally, I think the thing I’m most excited about is being able to branch out of space-related themes, and into Greco-Roman mythology for my packages names ; )

18 Likes

That sounds good. Though will Meteor 1.3.x still be running on Node.js 0.10.x? Do you plan to upgrade to Node.js LTS (currently at version 4.3.1) any time soon?

8 Likes

Great Plan!!! All the best MDG!!!

1 Like

They mentioned during tonight’s talk that it’s next on the list of priorities to update the latest version of node, and also update to the latest MongoDB driver. But that comes after the current priorities listed on this OP.

7 Likes

+1 for Greco-Roman mythology names

3 Likes

Really glad to see the priorities are aligned with my needs.

2 Likes

We will be very happy here in Europe with additional region support for the Galaxy.

5 Likes

I feel that after all the talk of MDG > Community communication being lackluster, its really started to pick up and feel much better. I’m excited for the future.

10 Likes

Sounds great, but give us, please, some ETAs… When 1.3? When Apollo?
We all need some information for planning…

I think that Apollo is sill space related :slightly_smiling:

3 Likes

I think this is the most excited I’ve been about Meteor since mid-2013 (when I was very, very excited indeed). Wishing MDG all the very best with making this stuff happen.

5 Likes

Good enough for me… ETA’s on such big projects is really hard. And a lot of new strategies (like moving (core) to NPM) are still new, will take some time, and will be worth the investment in the end (at least I believe so). But how does one plan around such major changes? I have faith in MDG that they have a clear vision and work hard to achieve their goals. In the meantime I happily use what they offer right now.

@gschmidt, thanks for this update. Along with the recent Transmission podcast my faith in MDG has grown a lot. Keep up the good work (and communication!).

3 Likes

Apollo will be a game changer as Meteor was back in time.

5 Likes

Congrats on the 1000 customers, guys! I know you worked really hard to get there so it’s nice to see it paying off. Let’s see how long it takes to get to 10000 :slight_smile:

3 Likes

I’d really love to hear about your plans for supporting other regions in Galaxy. I’d switch over to it in a heartbeat if I could only host it in the EU. For a UK-based business, hosting in the US is just stupid, especially with a live-updating system where latency is crucial.

9 Likes

Native mobile container?!

What Geoff means by that is that in Meteor 1.3, we’ve rewritten the file serving and hot code push plugins for Cordova, moving most functionality to native code (you can find more design details here).

This rewrite improves performance and makes the process more reliable. One of the reasons for this is that faulty JavaScript will not prevent native code from running, so we can safely revert to a working version of the app and continue to download further updates.

11 Likes

Amazing stuff @gschmidt @debergalis and team! Thank you for continuing to invest in cutting edge innovation. Each major investment is a risk and it is amazing we have such a great team at MDG willing to take that risk and drive JS forward. I’m super stoked :sunglasses:

With so much innovation coming I get the same exciting feelings I had in the pre-1.0 days of Meteor… I feel like a kid again! :birthday:

7 Likes