Week 18 - Concatenating small things

#web #emberjs #open_source #github #help_wanted #good_first_issue #fork #contributing #ember_scroll_modifiers

The Ember Initiative needs more members so we can spend more time on it 🙏 Since my re-assignment, I could spend a bit more than 2 days working on the Initiative, and this time allowed me to do a few first things for the community. But to organize and tackle important topics like the new Route manager API, we need to invest a time that we don’t have just yet. Mainmatter new plans make it easier for small companies to get onboard.

In mid-July, after week 17, I had to stop the Ember Initiative because I was needed on a different project at Mainmatter. It was a bit sad for me to stop when we were so close to releasing Vite support for the inspector. But this is how it is, and now I am back on the Ember Initiative! 🎉

Unlocking Ember Future with the Ember Initiative #

Even though I was working on a different project, I continued to prepare the talk I presented with Chris Manson at EmberFest Brussels. It was about modern Ember, the recommended path to update to modern tooling, and of course, the Ember Initiative itself. We explained what it is, why the community needs it, and introduced a brand new plan that should make it easier for smaller companies to join the initiative. See the talk on YouTube.

Weeks 16 & 17 - Moving forward on the new plan

#web #emberjs #ember_inspector #virtual_files #embroider_compat

New backers for the Ember Initiative #

We’ve got new sponsorships for the Ember Initiative! 🎉 This means we have enough budget to work until September. This should be enough time to finish the Inspector’s compatibility piece and unblock Vite users. If possible, we will have started the next topic by this time.

Continuing Vite support for the Ember Inspector #

The Inspector’s compatibility piece gives the impression of moving forward slowly. It’s not that Chris and I are slow; it’s just that the project is stupidly hard.

Week 15 - Reworking the plan

#web #emberjs #ember_inspector #rfc #embroider_compat #thought #taking_breaks

The main event of the week is that the RFC we wrote for Ember Inspector won’t be accepted as is. The compatibility part of it is roughly ok, but what the Ember Inspector should be in the future should be redesigned more drastically. In other words, we designed a solution based on how to reuse existing pieces, but the Core Team is unhappy with the existing pieces in the first place.

Week 14 - Ember Inspector & preparing Ember Fest

#web #emberjs #ember_inspector #ember_fest #auto_reveal

There was not much coding this week. The primary goal was to draft the first version of the RFC so that the core team could challenge it. And to draft the RFC, I had to investigate more about the relevance of what the inspector currently imports.

Watching for rabbit holes #

This week’s investigations led me to a bug in the current Inspector: most services are marked as computed properties instead of proper services. This is because the way the inspector detects if an object is a service no longer fits the supported Ember versions. (A week later, this kind of problem will explain a downside in the approach we suggested for the new API.) The reason I found this bug is that the piece of code involved uses instanceof Service, where Service is imported from Ember.

Week 13 - Ember Inspector blog post & manual testing tips

#web #emberjs #blog_post #ember_inspector #vite #npx #pnpm_dlx

I promised it last week, here it is! I wrote the blog post The road to Vite support for the Ember Inspector. It will explain everything my team is doing with the Ember Inspector in the big lines: what’s the problem, what’s the plan, where we are… Everything I mentioned in the past weeks, but after taking a step back now that we are making proper progress.

Little break and come back

#web #emberjs #ember_inspector

For budget reasons, the Ember Initiative stopped on April 25th. After a one-month break working on another project and enjoying vacation, the Ember Initiative started again on May 20th.

That said, May is a weird month for French people: We have a lot of public holidays, and unfortunately, I had to take a few additional days off for health reasons. To sum it up, I didn’t work that much in May. Because of this, I spent all of my work time (a very short time) focusing on our current topic (the Ember Inspector) and didn’t communicate. I prefer to forget that May 2025 ever existed in my professional life and consider that the Ember Initiative truly restarted the first week of June.

Week 10 - Ember Inspector feat. Embroider

#web #emberjs #ember_inspector #vite #embroider #addons #blog_post

I am posting this summary a few days late since I couldn’t work on it last Friday, but here it is. Unfortunately, it might be my last Ember Initiative weekly update for a while, unless the Initiative receives more support. But let’s start with the good news and return to this later.

Ember Inspector #

Running the proof of concept #

My primary focus this week was to get the Ember Inspector working for a Vite app — in other words, getting Chris’ proof of concept to run on my side.

Week 9 - a plan for Ember Inspector

#web #emberjs #docs_writing #ember_inspector #vitejs #embroider #virtual_files

This week was a lonely week for me since both my coworkers and my manager were on vacation. I had to move forward with all the topics we had on the table, and I focused on three of them: improving ember-vite-codemod, preparing the communication about the Ember addons audit, and working on Ember Inspector.

First things first: rewrite the README #

Like every senior developer, I was a junior once. My most traumatising memories from this period relate to a particular part of error solving: guessing what other developers mean when they write stuff on the Internet. At some point, it seemed to me that they were unclear on purpose, I was suspecting some unconscious ego issue, like “If I add more detailed explanations about something, it means I think it’s not obvious. But if other developers think that I think it’s not obvious, but it’s obvious to them, then they will think I am incompetent”. Something like that. I was desperately seeking a logical answer to the questions: “Why are all your docs so BAD? Why don’t you want to EXPLAIN? You’ve written a lot of stuff here, but what’s the CONCLUSION? What do you MEAN? What’s the matter with YOU?”.

Week 8 - a new release for ember-test-selectors

#web #emberjs #embertestselectors #pnpm #workspaces #babel #viteplugins #rollupplugins #emberinspector

ember-test-selectors 7.1.0 is out, along with strip-test-selectors 1.0.0. After this topic was closed, I could also look into some issues on ember-vite-codemod and pair with Chris on our big next step: getting Ember Inspector work with Vite.

Tips from ember-test-selectors #

About workspace:* #

ember-test-selectors is now a monorepo with the following structure:

• ember-test-selectors: the classic addon that used to exit before
• strip-test-selectors: the transform functions you can configure in your Babel config directly when your app uses Embroider+Vite
• tests:
  • classic-app: tests the compatibility of the classic addon
  • node-tests: unit tests for the plugin that strip data-test-* from JS objects
  • vite-app: tests the configuration for strip-test-selectors

Working with monorepos requires familiarity with the concept of workspaces. To simplify the idea, it’s a protocol that allows you to declare what are the different packages contained in your monorepo.