After Radio Silence, Kbin App Artemis Shuts Down

What do you do when your developer disappears?

Artemis is a beautiful client designed for Kbin, built with the intent of prototyping a mobile app for the fledgling community groups platform. While Lemmy has enjoyed a plethora of client apps, Kbin has yet to implement a third-party API for developers to build on.

Unfortunately, it appears that the project may now be abandoned. Community moderators and contributors have released a statement:

Many people in the Artemis community have been patient and understanding of the lack of activity, ever since a prior period of radio silence was justified by Hariette saying that she was recovering from chronic pain.

However, this has gone on for long enough that the official Artemis Camp instance has failed over and never recovered. This is a problem for Artemis app users, since it is the only instance that is supported.

We are also unable to continue performing our duties as moderators in the Camp instance as a result. This effectively means that the entire project is dead.

Statement by Artemis Moderators

How did we get here?

Artemis was initially developed by Hariette, a software engineer who was inspired by the growing popularity of Lemmy and Kbin. She had seen a number of Lemmy apps emerge in the face of the Reddit Migration, but noticed that nobody had really written anything for the second platform yet.

The reddit blackout got me exclusively in the

Now, reading into ActivityPub to see how I can make an app for 👀

— 🖤🇵🇷 Hariette 🌺🖤:verified: (@hariette) 2023-06-13T14:11:37.750Z

In her own words, the app was hugely inspired by Apollo, a popular Reddit client that found itself at the center of Reddit’s API controversy. After spinning up a project and running a community poll, the app was named Artemis, as a clever nod to the project that inspired her.

@christianselig Hey you! I’m making an inspired app for and , codename . Huge fan of your work and used your app for years now.

Just wanted to let you know based on polls, this app will probably be named Artemis as tribute ✨

Def check my pin post for a glance.

— 🖤🇵🇷 Hariette 🌺🖤:verified: (@hariette) 2023-06-18T03:26:29.522Z

Over the first month of development, Hariette managed to create a feature-packed Kbin app with themes, swappable icons, microblogging functionality, card views, and conversation threading. It even featured a plugin system! Her efforts faced two particular hurdles:

  1. At the time, Kbin’s API was largely unfinished, meaning that technically, no Kbin instances supported apps of any kind.
  2. She didn’t want to release the code to something prematurely, with people mistaking pre-alpha development as something intended to be production-ready.
The Kbin API work is a series of smaller Pull Requests involving everything required for app support.. Over six months of development have elapsed since the initial creation of the ticket.

Dealing with the API Issue

Lack of a formal API to work with was tricky. There were discussions in Kbin’s development tracker about formally merging in support for apps, but it wasn’t ready for prime-time. Typically, adding a client API that’s well-rounded enough for mobile apps necessitates a lengthy checklist for requirements. Long story short, it’s a time-consuming process.

Somehow, it looked better than regular Kbin.

Hariette used the WIP code branch to launch her own Kbin server, Artemis Camp, for the express purpose of testing the app against it. Sure, the client only worked with her server, but this wasn’t necessarily a problem. Testers enthusiastically joined, a moderation team was put together, and the app continued to make progress.

To handle the second problem, the Artemis project first announced a closed beta to a small group of testers. After enough useful feedback had been collected, the project released an open beta to the public via iOS Testflight and Google Play. With the launch of Artemis Camp, the feedback collection process made a significant amount of headway, and an ambitious plan was put together for a 1.0 launch in September.

A conversion of the document’s timeline goals. Late June to Early September is an intensely aggressive development timeline.

To Hariette’s credit, all of these features appear to have been implemented in Artemis. However, as time grew closer to the September 1st release date, things began to break down. Occasionally, the server would hiccup, clients would freeze on login, and people would have to find workarounds to log back in. These are relatively normal growing pains of early development; server failures are not at all unheard of.

Screenshot from an Archive.org copy of Artemis Camp

As the deadline began to draw near, Hariette stated that she needed to slow down in development due to personal matters. She became less active in both the community group as well as the chat, and seemed to have also taken a break from Fediverse communities in general.

Screenshot from an Archive.org copy of Artemis Camp

Nearly a month later, Hariette posted in her communities that she had come back, fully recharged and ready to work. Worryingly, she hasn’t been heard from since. According to the Artemis community members that we talked to, the developer has not been heard from or seen since October 27th. Multiple attempts to reach out were made over the past few months, but no responses have been made, to anyone’s knowledge.

The Aftermath

On November 14th, the Artemis Camp server went down for good. With no way to maintain the server or work on the app, the community moderators resigned. For now, anyone who has the app installed is unable to use it.

The only thing I can try to log into is a dead server.

In an ironic twist of fate, Artemis was stopped in its tracks because the source code hasn’t been released, and the app is only set to work with Artemis Camp. The two workarounds used to get the project going have taken a turn for the worse, because all of Artemis development depends on one person, and one server.

To be clear: the biggest, most important detail to highlight is that nobody really knows what happened. Some people might blame Harriette for abandoning the project, but it’s clear that she was passionate about it, and put in a tremendous amount of work. The development turnaround was incredibly aggressive, limitations in the Kbin platform drove her to host and maintain a fork, and the project relied on one person with good intentions.

While it would be wonderful to have the source code out and community development humming along with volunteers, the bigger concern at this point is the developer’s health. Maybe someday, the project will start up again, and we’ll all be able to use Artemis across the Threadiverse.

From We Distribute, we have to say: Hariette, if you’re out there, we hope you’re doing okay, and that you feel better! You did amazing work, and deserve to take all the time you need to recover. If Artemis ever gets up and running again, a lot of people would love it.

Sean Tilley

Sean Tilley has been a part of the federated social web for over 15+ years, starting with his experiences with Identi.ca back in 2008. Sean was involved with the Diaspora project as a Community Manager from 2011 to 2013, and helped the project move to a self-governed model. Since then, Sean has continued to study, discuss, and document the evolution of the space and the new platforms that have risen within it.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button

Discover more from We Distribute

Subscribe now to keep reading and get access to the full archive.

Continue reading