When i first started using the forums i quite swiftly reached TL3. The level itself isn’t that important but i liked making use of a few TL3 features such as updating post-tags.
After a short period of less activity i dropped out of TL3, since this trust level is time dependent. Now the interesting thing is that once you drop out, it becomes less likely than initially that you will reach it back.
I think there are a few reasons for this, but the main one may be this:
When you first sign up, all topics are new so every thread you open is counted as a new read. After you’ve been around for a while you’ve read most of the long-running threads and you become ‘dependent’ on actual new threads to reach TL3. There are only about 150 new threads a week in total on the subforums combined (slightly more after new releases), of which maybe 20% are relevant to you since most threads involve specific hard/software and use cases.
So with ~30 new threads that you may read weekly, which is basically everything that is posted that is interesting to you, you may be right around the threshold of reaching TL3. However, this implies that you have to keep doing this more or less daily, or you won’t reach the threshold.
In this scenario the trust level becomes mostly dependent on how much you scroll. It doesn’t reflect how much you otherwise have interacted on the forums or if those interactions were useful.
So my suggestion would be to lower the reading requirements, or perhaps as an alternative make TL3 sticky so users that are trusted at some point don’t have to keep scrolling in order to retain that level (in contrast to other trusted levels like TL4 and moderator that don’t have that requirement).
Indeed, it would be also possible to assign TL3 permanently: so we could add “permanent TL3” as possibility in between “normal” TL3 and TL4.
I am not so reluctant about this as we had already experienced in the past that giving TL4 too early can cause issues, while I understand that it can be demotivating to switch between TL2 and TL3 regularly → I presume the majority cannot keep fulfilling the TL3 requirement, at least as they are at the moment, throughout years. So to keep people motivated, TL4 is inevitable at some point.
However, another possibility: reduce only two requirements of TL3 → topics viewed (486) and posts read (2500) seem to be the major/sole issue that “breaks” the TL3 of people who are otherwise regulars who contribute a lot in Discourse. Also, I see the possibility that these two are the major reason why we have nearly no TL3 from SIGs/WGs other than people who are also active in ask.fedora: without ask, there is nearly no chance to open enough topics and posts to get TL3 → thinking about that case might be useful since Discourse has become a major point of interaction in Fedora - and the TL3+/moderation is yet mostly limited to people active in ask.fp. I would like to see the mod area becoming more heterogeneous with people also from other parts of our community who organize themselves in Discourse.
So we might consider to reduce these two requirements rather than considering a “permanent TL3”. Also, these two requirements can be easily achieved by cheating, so just randomly open topics and then skim quickly through. So I am not sure if they should be major determinants for becoming TL3 anyway.
So two further alternative possibilities here that could be discussed: “permanent TL3” between TL3 & TL4, or decrease two requirements (read posts / viewed topics)) to some extent (my favorite would be the latter, that would be also easier to tailor to without changing too much about how we organize) → of course there might be further alternatives, so I mean no restriction or so (including to keep it as it is)
Hi all, I’d like to add a related suggestion regarding tag editing permissions.
I’ve noticed that while helping users — especially in categories I’m familiar with — tagging is often inaccurate or missing altogether. As we know, proper tagging helps both in identifying issues and in increasing the visibility of threads for those browsing by topic.
Currently, tag editing requires TL3, but as a TL2 user who’s actively contributing in those categories, I often wish I could help correct the tags to improve discoverability and context.
Would it make sense to consider allowing trusted TL2 users to edit tags — perhaps with a certain level of category-specific activity or based on a moderator’s discretion?
You raise a good point about the need to increase appropriate tagging. I share this point indeed: this is one of the reasons why we always tried to increase the amount of TL3s, and also get more TL3s in the Project Discussions. The problem is that TL2 is quickly achieved, and especially in ask.fedora many users who get TL2 are not ready at the time to get this privilege. This could thus increase the problem / cause confusion. But if we decrease the need to “view topics” and “read posts”, many people who are persistently active in the Project Discussion would get TL3: these are the major requirements not fulfilled by passionate long term contributors like you @hankuoffroad - the reason is often simply that there ain’t sufficient topics in the Project Discussion to achieve that, and so TL3 remains limited to Ask.Fedora contributors (supplement: in some cases, Project Discussion contributors, depending on the teams they are active in, also lack a few likes - only 20 are necessary to get and 10 to give, but some teams seem to not use that at all, although this might be a minor thing we can give incentives about - the general issue remains the two requirements I mentioned earlier)
The idea to mitigate the TL2 issue with moderators would introduce two issues: first, the ticket numbers would strongly increase that way, and this would cause much work - unfortunately, moderators are a scarce resource, which I would not overuse (also, tags are most important in the beginning of a topic, and it can take hours if not a day until that would be processed). The other issue: I think we do not have a function that would allow us to link tagging to moderator actions, which means we would need to forward this upstream for implementation. I am not sure if that would be a function prioritized.
About category-specific, this is also not intended for TL, which are always Discourse-wide. Afaik, such a function exists only for moderators, but we discussed this long ago, not sure if something was implemented since then? (maybe someone knows more?) I expect not as this seems not interesting for most Discourse use cases (ours, ask+projectdiscussion combination, is a little special compared to others), but I don’t know for sure
I understand there are some reservations about. changing the principles behind TL3 promotion and default thresholds for TL3 in site settings. If there’s a need for countermeasures when relaxing TL3 criteria, perhaps we could consider some alternative eligibility paths.
In my case, my main activity on Discourse revolves around the Ask Fedora space and team workflows. I’ve been actively documenting recurring issues and contributing to Quick Docs to reduce repeated support efforts (for example, sound problems).
Could we explore the possibility of exceptional TL3 promotion for contributors actively working to reduce the load through support and documentation?
Definitely worth discussing, but a few implications about that, which a solution should try to mitigate:
Discourse is intended to promote intuitively, to allow people having a trust level that fits their Discourse and community experience. So far, many teams do not use the possibilities of Discourse, which often is one reason why they do not get TL3 (such as getting and distributing likes): exceptional promotions could foster that people and teams do not integrate with what Discourse delivers and the possibilities it creates to further bring different teams together.
→ Potential mitigation: Only bypass requirements that do not break the need to integrate with a Discourse organization. With regards to what I suggested about reducing needs for “posts read” and “topics viewed” → a bypass might be ok in the exceptional cases if these two requirements are bypassed when they are fulfilled, let’s say, 50% or so. But I would prefer to not bypass requirements such as the getting and distributing likes → loosing TL3 or not getting it at all for such reasons (likes) is a useful experience for people to get in the value of likes as core element in promotions (no individual promotes, but the community as a whole does → the result of likes of many different users). Of course the “exceptional” then needs to introduce alternative requirements.
A major foundation of TL3 is that the requirements are reproducible, and do not contain an individuals or groups opinion: it has to be a “mathematical” “distinct” fact that makes the bypass applicable, so if “100 likes received” then promote the very user, not something like “if the user shows to understand how Discourse works”. TL3 are an important institution for transparency: moderators have to discuss problem cases outside the public for obvious reasons. But TL3 ensure that the community determines itself people to add their opinions and review the moderator area without us moderators deciding who is allowed to review what we do → unbiased voices, feedback and reviews, that is also very valuable for us moderators to get evolving perspectives of an evolving community. The community needs to know that TL3 is not determined by individuals (we ain’t able to differentiate between different TL3) → We need to not break this institution
Moderators are a scarce resource, whatever is implemented here, it should not need much time investment: Whatever has to be checked by moderators, it needs to be a value that can be verified in a few seconds.
It has to be considered that some teams organize themselves in a completely different manner, not comparable to how Discourse works: not every valuable contribution qualifies for TL3, so we have to be careful in this respect as well. This might also lead to issues when people then can get TL4 quickly, which is then something else we need to consider (so adjusting TL4 requierements *¹ ) → a major element for a bypass imho should be that whatever property we use to bypass TL3, it should be something that happens in Discourse, and not from somewhere else, which also helps moderators to verify quickly as Fedora teams’ organizations are so diverse that it is hard to get an overview of who organizes how. I don’t think it would work if moderators would need to leave the “realm of our Discourse” given that this by definition needs time to get into something potentially new.
A manual promotion to TL3 is likely to become permanent and automatically bypass automated promotions/demotions → the automated demotion if no time is spend on / no action done on Discourse will not occur I think (maybe someone knows better? A related function or so?). Therefore, this is permanent and would quickly end up in the eligibility for TL4 without any further action to be done. With this in mind, exceptional cases need to be very exceptional, and contain only people with a long and trusted reputation that can be verified in a reproducible manner, far beyond likes and contributing to a team.
*¹ I discussed TL4 requirements with Ankur at Flock, and I wanted to open a topic about that already given issues with TL4 promotions in the past, something with making a minimum time of TL3 before TL4 promotion is possible (e.g., 6 months TL3 of the last 9 months → problem: how to measure/verify?), but I had not yet the time to open something dedicated about this. Of course many if not most variants of manual TL3 promotions would also bypass major requirements for TL4 if TL4 requirements ain’t adjusted.
I completely understand the value of natural TL3 progression and maintaining fairness within the community.
Since it would be hard to measure and would require creating a new policy, I understand that more input from the community is needed. For now, I’ll focus on following the natural progression toward TL3 through continued engagement.
As a note, during the upgrade period 41 to 42, the workload of reading all post really went through the roof, and made it very difficult to keep up.
Maybe a small reduction of reading required number of posts would be good. All the rest of the criteria still stand - they represent the acceptance of ones constructive involvement with the forum.
(supplement: in some cases, Project Discussion contributors, depending on the teams they are active in, also lack a few likes - only 20 are necessary to get and 10 to give, but some teams seem to not use that at all, although this might be a minor thing we can give incentives about - the general issue remains the two requirements I mentioned earlier)