I think I heard somewhere that most of that $20 million went into trying to get faster search performance out of Atmosphere. While Atmosphere performance issues still remain, I know that with enough community support, and maybe a dash of series C funding, MDG can make it happen!!
(EDIT: Just kidding/venting with attempted sarcasm; when I yell at Atmosphere it doesn’t seem to care, so I’ll vent with you guys …)
What!? $20m on one aspect of Atmosphere (which is surely going to be deprecated in favour of npm) ?
I agree with that point. We have being developing medium sized package based applications (~30 custom developed packages) and Blaze works very well. Our approach was to code view components that are initialized passing parameters to the template. From that point the template is in charge of subscriptions and handling internal state communicating using events with other components.
The main issue is that application grows very large and template naming for instance must follow conventions to avoid clashes. Incremental loading could be very interesting for performance optimizations.
Guy, this was a joke 
I’ve heard Atmosphere deprecation was a bluff - they are planning to disrupt NPM 
(sarcastic)
You may want to check out my Blaze Modules package, which addresses the template naming issue, and comment on the issue I’ve opened to get a similar feature added to Blaze proper.
I personally think my biggest pain point with blaze wasn’t even a blaze issue, but possibly a DDP issue. If DDP allowed for messages to be sent in large batches then blaze could re/render whole batches and not each time added/updated/removed is called. Virtual DOM would be nice, but I think batching the data changes would be a huge performance improvement alone.
Thank you for starting this. I’m totally new to Meteor, and Blaze resonated with me especially because of prototyping speed.
React seems significantly more laborious at first glance, so I was dejected about it eventually being the only option. Definitely hoping Blaze gets a 2nd wind.
Hey guys, FYI we can now put all Blaze discussion in it’s own brand new forum category!
Let’s rock… uh… let’s Blaze! 
Probably an attempt to keep disparaging remarks about Meteor hidden from the main forum!
I’m happy to give them the old Blaze version of Telescope to use as a community platform : p
Hey I created the blaze category after a request from @miro - I think it will make conversations about blaze easier, not harder!
I’m also interested in this. I have a long standing love affair with blaze.
He’ll come back saying it’s a huge risk. The reality is that it’s not.
AirBnB is using it. If Facebook was trying to use react as a way to sue and take over startups, they’d have taken over AirBnB and be reveling in the spoils…
and they’d have done the same with Asana, Atlassian, BBC, Chegg, CloudFlare, CNN.com, Codecademy, Coursera, Craftsy, Dailymotion, Dropbox, Expedia, Facebook, Feedly, Flipboard, HipChat, IMDb, Imgur, Instagram, Khan Academy, KISSmetrics, Mattermark, Minerva Project, Netflix, OkCupid, Rackspace, Rally Software, Ralph Lauren, Reddit, Redfin, Salesforce, Squarespace, The New York Times, Trunk Club, Twitter, Uber, University of Cincinnati, Venmo, WhatsApp, Wired, Wix, WordPress, Yahoo, Zendesk and everyone else using react.
But they haven’t.
So I think the 99.99999999% of developers using react on their revenue-less startups will most likely be okay… worrying about this is analysis paralysis at it’s best. analogous to fretting over unit tests for every single function in your MVP.
Beyond react’s utility to facebook and their type of app, most likely Facebook is putting react out there hoping it is widely adopted for other reasons. For one, it makes acquisitions 100X easier when the company you’re buying uses your entire stack/philosophy/architecture that you use.
Full Disclosure: I don’t even use react. I use Blaze and love it.
Man, I actually REALLY like blaze, I think it’s extremely well designed for usability. I find most other frameworks out there have a ridiculous amount of background before even making something as simple as a list from a query with a search box.
In blaze? Maybe 1-2 minutes of programming.
I’ve tried react and it is extremely… meh. Angular2 is okay, but a little wonky.
What is currently limiting blaze? Why do people prefer other frameworks?
Blaze is OK, but MDG was consuming too much resources on it. They decided to stop investing time and money on Blaze when huge companies like Facebook and Google are developing view layers as well. One great thing with React is React Native: hardware acceleration on mobile.
Personally I think React jsx code is a mess you can’t read, so I moved from Blaze to Vue.
yep! Here we are! 
Well, there’s a new repo, slack channel, and forum category now, as linked above! Sounds like people are pumped about getting a community together.
I am glad to see all this enthusiasm for contributing to Blaze. I think at this moment we are still stuck at waiting for MDG to allow external Blaze repository to work with current releases. And to move stuff to NPM, but once we get to there, I think we are then open to start evolving Blaze further.
Feel free to open issues at https://github.com/meteor/blaze with ideas what should be next things to work on.
Well, there’s a new repo, slack channel, and forum category now, as
linked above! Sounds like people are pumped about getting a community
together.
Somebody can make a pull request and add links to all that (and auto-invite Slack button for channel) to README at https://github.com/meteor/blaze.