What does the Meteor.js community thinks? — April 2019

What do the results of the survey say and some background.

BackgroundThe constant cycle of “Meteor is dead” posts has been haunting the Meteor community for the last few years. Yet here we are, Meteor just released version 1.8.1, 1.9 is already being sketched out and 1.8.2 is under active development.

Yet, it is being constantly pointed out that there is a lack of leadership which causes myriad of other problems that the community is facing. Some of that is justified, some of it is not.

From the MDG side, there has been a transfer of resources to focus more on Apollo, which addresses one of the early complaints on Meteor (that you can’t use other databases) and has found great success in the wider javascript community.

With the rise of many new libraries that gained a great hype, we saw a lot of top community contributors leaving or as is common in life, moving on to other things or getting busy with AFK life.

Besides the rumbling on the forums, there has been an activity to address these issues behind the scenes. Although initial talks did not pan out to everyone’s satisfaction, I can tell you that MDG is still committed to Meteor development (they rely on it themselves). Still, we probably won’t see as bigger engagement from MDG as we saw in the growth phase of Meteor.

Now then this is where the community comes in. Meteor is a mature framework now, but when it comes to community packages, things have been lacking lately. Most of this has to do with attrition on of the big name developers who have developed packages that have become the de facto standards and there are other issues that community can assist with.

In early March 2019 an open conversation was started between developers on what to do. In summary, first was the creation of a GitHub organization which will act as an umbrella for important community projects where there are people willing to maintain it. This is especially important if the original author does not have time to do so. We are currently in the process to determine which packages will be added here and figuring out an automated process for publication (suggestions and PRs encouraged).

There were many other ideas what to do with the community aspect, but eventually, it was decided to conduct a survey to determine primarily what people were looking for. As such a survey was conducted.

What follows is the summary of the results with my light personal commentary which is a mix of analysis and providing context.

Participants

First though a note on participants. As of the time of this writing, only 57 people have filled out the survey. Though very limited these people are very likely to be actively engaged with Meteor and want to see it succeed, as the survey was advertised on Meteor Forums and Meteor Spectrum channel.

Still, a long-running survey will be designed to get as many people’s feedback as possible. Maybe something akin to State of JS survey.

Survey results

First of all, let’s take a look at the easily measurable results (multiple choice answers).

For me, this is the main interest, as I have volunteered to do the newsletter. Overall there is an interest in a Meteor newsletter. Possibly since there isn’t a dedicated Meteor newsletter. One of the old ones have either ceased operating or have shifted focus towards general JavaScript.

Personally, I think the issue is that those newsletters were coming out on a regular basis which leads to burnout from the author and having to work super hard when there is less content. As such the new community newsletter will be occasional mainly centered around important news about packages, discussions and always to inform about Meteor releases. More on that sometime later.

distalx will be very happy that his as this was his idea and he was very excited about it. I think this follows the popularity of the Transmission podcasts that Sashko Stubailo used to run.

Transmission podcast thumbnail

Same numbers as for newsletter. Though this one might be a bit harder to setup. I guess that it is a human need to socialize and make connections, but at the same time as developers, we are more likely to be introverted types.

Probably the most contentious question. 💰

There has been a discussion to speed up development and maintenance and reward developers for doing amazing work for the community. The idea of supporting developers for doing FOSS is more accepted then it was about 5 years ago, but the hesitation is clearly visible and understandable. Hence it will be essential that any effort in this area is open and clear.


Now we are moving into an area that required a bit of interpretation of the results (especially towards the end of this section), so please check the data for yourself to get the whole picture.

Following from the previous there have been many suggestions on which way to support. Open Collective is clearly the winner here and seems to be the way for many FOSS projects. Second is Patreon, a popular support platform in general, though there is question of sustainability and probably not the best for org as a whole. That said, some individual developers already have Patreon. You can find more info in the related forums thread and would encourage everyone to support the developers and for developers to post where they can be supported into that thread, be it on Patreon, LibrePay, Crypto or Ko-Fi.

With a lot of overlap, we can see that Meteor Docs and Guide are the most popular place to go for information about Meteor, closely followed by the forums. This shows how important it is that we put our efforts into improving the docs and guide as much as possible. Later we will see that it has been often identified as lacking. That is also probably why GitHub and Forums are so much visited, besides it is the place to keep a tab on development and community news.

Finally, written articles are also sought after, but they will often be outdated as the bulk of them was written during the growth years of Meteor in the pre 1.0 or 1.3 eras and will often outrank in search newer articles.

Analyzing the question about Meteor misconceptions required a lot of picking and choosing on my side into which category I should place which comment. Not to mention there have been many answers which weren’t strictly answering the question.

The most common comment was about the hype train with other technologies and how Meteor is dying. Not much can be said about that what hasn’t been said already. “You are not Google” article by Oz Nova is a great response and summary to most common talking points in this regard. The second most common concern of scalability can also be answered by that. From my point of view through the only problem is just informing people on how to do it (which applies for the point of slowness as well, especially given the few latest releases just focused on that point).

The last point I would like to highlight is the point about Meteor being outdated. This could be merged with the hype answer as well. Common are the calls about getting everything to npm and Fibers and those are things that have been talked about for a long time and they are (with other points) on the roadmap, but there are other more pressing matters before we start moving onto Meteor 2.0. There is also an ongoing debate if that is a good direction given the benefits that tightly coupled framework offers.


Finally, you will find a long list of Meteor packages that have been highlighted as needing to be updated and that developers have taken onto themselves to update locally.

Quite a few are around Blaze and this will probably result in the creation of a Blaze centered working group. Beyond that is the legacy of Kadira and alanning:roles. Roles has been something I have been personally interested in as well, so I think that will become our first major project once we figure out the workflow on smaller packages.

What is next?

Newsletter

In the coming days, I will be putting up a welcome email to those who have submitted their email in the survey and preparing the first newsletter with other community members (feel free to contact me if you want to help).

Virtual Meetup

I think we will investigate some options on how to make this happen, but I wouldn’t expect anything major soon unless someone takes that as their priority.

Podcast

Looking forward where distalx is going to take this now that there has been a clear interest shown. I think we will see a forum thread about this soon.

Support for developers

We will have to figure out how to get some finances flowing to developers of popular packages, this will go a long way to ensure their long term prospects. This will take some time as we will have to figure out how to do it properly. Keep a lookout on our organization GitHub for an issue related to it (feedback is always welcomed).

Crossroad page

Finally, all of this will require some crossroad page as the overall efforts are not limited to just the forums and GitHub. This will be probably a GitHub hosted a static webpage that we will put up in the near future as a temporary solution until we have a better idea of what to expect from it.

Survey

As eluded another survey will likely be constructed to give voice to more developers in the community over a longer time period.

As you can see we are at the beginning of creating a new chapter for Meteor community. I would like to encourage anyone interested in helping or just sharing an idea to reach out to us, either on our GitHub organization repository or by creating a forum thread with your ideas, suggestions and/or to volunteer to do something. Now is the best time since we are starting from scratch and you can have a great impact.