Yes, that was the post I was referring to. It caused a lot of doubts and uncertainty back then, when most people were still using Blaze in production and React wasnāt even tightly integrated yet.
Itās quite ironic listening to the excitement for whatās coming, knowing that almost 4 years on we are pretty much still in the same spot
In the episode, Paul Dowman said he was expecting the community to grow at least 50% as a result of the move to NPM, Sashko said internally they expected even more than that.
This was right around the time they started on Apollo I think, Sashko was talking about āthe data storyā and how what they were going to build around that was going to be a āgame changerā.
So, yeah, if MDG would have delivered on the promises they made about Meteor back then instead of getting side tracked by Apollo, I think we could be in a different spot for sure. Although my experience reflects @captainnās, in that much of the reputation I encounter in practice does seem to be stooled by little actual knowledge of the framework. So it might also not have mattered.
It is frustrating though, how they suddenly completely changed course, leaving us stranded without really communicating about it. Iām exaggerating a bit because the framework is still great as is, but listening to this podcast, hearing Ben talking about HMR (again: feb 2016), I canāt help but feel a little sad for what could have been.
To clarify, I was referring more to the excitement about how the community and adoption was going to grow, and to the specific improvements talked about in the episode (contributing to core, testing, HMR). It could have been recorded last week in that regard. Genuinely not sour, just saying I understand the sentiment. And listening to this episode did remind me that at some point MDGās messaging has totally shifted.
But youāre right, it doesnāt help to focus on the negatives, weāre better off investing our energy in moving the framework forward.
I hope the next paragraphs are seen as some positive and productive input
I was reviewing some of the frameworks from the survey and all I see there is hours of wasted time, because of setting up architecture and basic features (that already come out of the box with Meteor) or just miss important functionality that is crucial for our use cases (reactivity). Letās be honest: If I really want this kind of pain to setup everything from the ground I would go for Java EEā¦
Besides that I agree that there is few knowledge about Meteorās current abilities outside of this community:
No one besides us knows that the survey is built using Vulcan = Meteor. I think no one complained about the quality of the survey website (it worked so well I was really impressed). So let them know about this! āDid you know the famous State of JS survey was built using the tool, which the most participants thought to be outdated?ā
No one besides us knows that Meteor has always been on the edge of the new ecma versions.
No one besides us knows that you have out-of-the-box support for the major frontends. I mean other frameworks either support excatly one or none (and you have to set it up all yourself again).
No one besides us knows that you can write using DDP/websocket and HTTP and that you can also use express as middleware by just using one line of code.
No one besides us knows that you have a full oauth2 based authentication workflow out of the box. This is still taking ages to implement on your own, even with packages like passport
Also some ideas for concrete marketing steps could be
a new major version, just to show we are far far away from 1.0 days
a new website with a fresh design, built with Meteor, hosted on Galaxy (shows that you eat your own food and it tastes good)
an SEO campaign that downranks all those old and deprecated articles
an improved atmosphere registry where abandoned packages are ranked down
a curated gallery, where companies and community can submit links to great Meteor websites and apps for promotional purposes
We are following everything that is happening very closely and working to revert as much as possible.
Itās also important to attract new people to Meteor. If you read the new Meteor roadmap with this perspective you will see many items related to documentation and tutorials. These items are very important to keep the community growing and to show that Meteor is very up-to-date and a great tool to start new projects.
Again, Iām confident that the future of Meteor is going to be amazing, we already have a great tool and it will be even better in the upcoming years. We need to learn from the past but move forward.
Iām a Meteor user for many years then I understand the frustration but I try to focus my efforts in the future and in what I can do now to help Meteor. Go go go!
@filipenevola I already read the roadmap and I am also confident about the next moves. This was rather a braindump after the survey because I canāt get in my head why devs are currently so hot for static side generators (back to the 90ās?) and serverless / JAM stack (GDPR nightmare) oO"
I will be making a ton of Meteor content in 2020. Iām not sure when to get started, but Iām excited to fill in the void of video tutorials. What kind of content do people think will help the most? Iām planning some basics content with React and some more full stack intense stuff with Apollo, TS & SSR.
In 0.5 days, Meteor was briefly advertised as āthe .NET of Javascriptā. Itās a good selling point that resonates with people, because itās needed. The problem is that in making that kind of claim is that it requires taking a stance on the UI layer.
Thatās already completely possible. There are actually multiple ways to integrate Apollo, some that leverage DDP, some that bypass it completely. React has been supported for ages, and Typescript is a core feature as of 1.8.2 - and thatās pretty dope.