Page MenuHomePhabricator

Potential RedWarn 17 fork
Closed, DeclinedPublic

Description

I don't want to see RedWarn, a fundamentally different tool that focuses on being an on-wiki tool, similar to Twinkle, be abandoned. In fact, RedWarn's growth should ideally be hand-in-hand with Teyora, despite the two projects being disconnected other than by common ancestry. This means that the two major areas for counter vandalism tools, on-wiki tools (RedWarn) and standalone (Teyora) are both catered for in a user friendly way by actively developed community tools.

RedWarn 16.1 has been the current version for nearly over a year. If the RedWarn team does not increase development speed or seek financial or volunteering support, my concern is that the risk of the tool being abandoned is substantial. Therefore, forking RedWarn's TypeScript rewrite may be a consideration if resources allow. This will not effect any license agreements currently in place.

Event Timeline

Ed6767 triaged this task as Lowest priority.Jan 11 2022, 4:43 AM
Ed6767 created this task.

Hi there, Ed. RedWarn project lead here. Just a few points I want to make here about the task, and the idea of it in general.

RedWarn is currently dealing with a common problem in the software developer world: "If one can make an alternative (for example, a free Wikimedia-like alternative to YouTube or TikTok), what's stopping them?" There's a simple answer to that question: taking the time to build alternatives from the ground up is extremely difficult. RedWarn was originally built as an alternative to Twinkle, and we're still under that expectation. Likewise, RedWarn 17 (or RedWarn TypeScript rewrite, whatever you want to call it) is a far cry from RedWarn 16.1 — nearly everything has changed and old code has been gutted beyond recognition.

We've removed most of the iffy components which was previously in RedWarn 16.1 which have proved themselves inefficient, difficult to maintain, and worst of all, insecure. In its place, we've been adding in out-of-the-box internationalization and per-wiki localization, not just for Wikipedia but for any MediaWiki wiki who wants to run RedWarn. Fact of the matter is: RedWarn's TypeScript rewrite is quite literally a rewrite. In other words, in the context of the previous, it's like an alternative to RedWarn 16.1 which is still under development.

It requires a lot of human effort to build in order to reach the functionalities of the existing version. Not to mention the amount of restructuring and organization we've had to do to enforce a more consensus-based method of running the team, rather than a "one-man-leads-all" style of leadership (which encompassed even user-important data, such as the unilaterally-written-and-imposed-and-now-deprecated Legal Policy).

With our main developers currently under a lot of stress, some under their academics and some in isolation due to COVID-19, we've been taking our time in keeping our on-wiki lives and real lives balanced. Not all of us can dedicate all of our time on maintaining onwiki tools. That said, we've been keeping the train going behind the scenes and is still actively in development. The fact that RedWarn 16.1 is still the active version on the English Wikipedia is not something to be proud of, but this isn't any incentive for someone to think that it's "abandoned". Major security patches are still being pushed out whenever they can, albeit as subtle as possible to avoid drawing more attention.

The idea that RedWarn "should ideally be hand-in-hand with Teyora" should have been considered when a split between RedWarn and Teyora was first proposed. Now looking back at it, I can say with certainty that I and the rest of the incumbent RedWarn team lost some morale due to certain issues with how the split was executed. I can't help but take offense at the fact that you're assuming that we haven't been performing any work on RedWarn for the month that you've been gone, and that none of us were notified about the idea of this task beforehand. And it's not just me, all other RedWarn team members have expressed their disdain in internal channels after reading about this.

Nevertheless, we'll take this task as a challenge to get things done faster rather than take it as a threat, even if this task stays open. The RedWarn team will continue to be a group of volunteers dedicated in creating a free and open-source anti-vandalism tool for Wikimedians. The way I see it, if this were to push through, the past month would have been the most roundabout way of kicking everyone off the RedWarn team and taking control of code worked on mostly by me and @Sportzpikachu all for yourself. Whether or not that's in the spirit of open source and the Wikimedia movement, I think the answer is pretty obvious.

Adding in the rest of the RedWarn team with Wikitech accounts, since they deserve to know this exists as well.

I wrote a longer and detailed rebuttal but I think I can boil it down to a tldr:

I should clarify that this is not a well developed plan or "threat" to steal or do anything not cool regarding RedWarn. Instead, it is an idea to ensure it remains maintained and support existing teams if any future Teyora team has the facilities to do so (hence "lowest" priority). It is not and never was an immediate or cunning plan to double-deal the RedWarn team, I never implied ties in terms of growth, rather this was an optimistic and idealistic view of both projects being released around the same time, I'm sorry if that wasn't clear. I will and have always wanted good for both projects.

All tasks here are transferred from Google Keep notes so aren't fully considered or committed ideas, I'm sorry if this freaked you out. Your commitment to the project is a driving force, and this idea was simply to ensure the codebase remained maintained in case team members became overwhelmed. It's excellent if this task has proved as motivation, as then it should hopefully be redundant.

I'll reconsider and reopen if I think it may be appropriate in the distant future given this clarification.

Whether it was a concrete idea or not, the fact that this idea had been publicized without any forethought comes off as extremely tone deaf. Until now, insinuating that we're incapable of handling this project by saying that this will be reconsidered still irks me. Either way, I hope this serves as a reminder for you to double-check possible consequences of posting your unfiltered ideas, especially those involving teams no longer under your active control, in public spaces.

EDIT: Rereading this, I find it hard to believe that this task was made for the "future", when there's specific emphasis on RedWarn 16.1 being the current version for over a year in the original task description, including what was essentially an ultimatum for us if we didn't act within the very near future.

Rereading this, I find it hard to believe that this task was made for the "future", when there's specific emphasis on RedWarn 16.1 being the current version for over a year in the original task description, including what was essentially an ultimatum for us if we didn't act within the very near future.

My focus is Teyora, not RedWarn, this had lowest priority. I have no reason to lie or backpedal, and I'd rather be transparent during my planning processes. I have already said this was never an ultimatum of a cunning plan, please stop assuming that I'm just going to act in bad faith and not stay true to my word. Even if I did decide to fork it, there is absolutely nothing stopping you from forking all the work back, because this is free software.

I've voiced my concerns about the risk of the tool being abandoned, and you have said about how busy everyone is in the last message, and maintaining something like RedWarn is a lot of work. I didn't label it as abandoned right now, because you guys are working hard on it. The last major release being a year ago is a concern, this is not indicating that no work is being done. There were no personal comments, and any actions boiled down to that it may be a consideration [if the tool were to become abandoned]. I apologise again if this wasn't clear in my original wording.

Either way, I'll ensure plans are more focused, thoughtful and transparent in future to ensure this doesn't happen again.

Qwerfjkl closed this task as Declined.
Qwerfjkl subscribed.

Oops