> Not a project/lib announcement. Not an event. Not a blog post/news article. How do folks feel about this in the #announcements channel? Personally I see a podcast as conceptually very close to a news/article (strictly speaking a blog post may not be categorized as "news or an article" either). #news-and-articles desc may be expanded. There's also the possibility of renaming it to #content although it may not be particularly clarifying --- As for occasional non-lib posts with an near-universal appeal, honestly they are so occasional that they don't bother me, but they also have the problem of where to draw the line. (is posting $random_thing OK only if you are a well-known personality? Would seem a harsh thing to state) #off-topic has more members than #announcements so people shouldn't be afraid of posting things there. Sometimes it's inactive several days in a row so these kind of items in a grey zone will likely have decent reach.
btw #clojure is more focused than #off-topic and has even more members. Similarly to #off-topic , they have high reach but also normally will be muted, which IMO makes it a good candidate channel for grey-zone stuff (of course this is only an opinion/suggestion)
Personally, I like when there are frequent updates in #announcements . It shows me there are people engaged in making Clojure programming easier, more enjoyable, and more effective. Wether it is smithing tools, trying to improve Clojure itself, building or improving libraries, writing articles, publishing podcasts or videos, it all brings me happiness. If anything, I think the restriction to not crosspost is limiting, Somethinng could fit in #news-and-articles and also be generally announcable. I can also see how too frequent updates from the same source can delute the interest level. If I would write some posting guidelines for the announcement channel it would go something like so: Please help keep #announcements alive and vibrant by posting updates from your Clojure related work there. It could be about making libraries and tools, wringing articles, publishing podcasts, broadcasting videos, or just about anything aimed at enabling the Clojure community to thrive. Also help keep the channel interesting and relevant by filtering your updates some and make it about those major and important steps in your project(s). To further improve the quality of the channel, spend some effort on how you announce things. Craft it like you craft Clojure code, succinct and with clarity. After reading your update, it should be clear to the reader what the project is about, what the update is about and why anyone should take a deeper look at what is being announced. Also let people know where you want follow up questions and discussions to happen.
Thereโs another non-lib announcement posted today. I donโt think you can blame people for that mistake given the general name.
Seems people at least agree on a few things - announcements should be globally relevant to the community, i.e. no regional meetups or small niche topics - they should be significant, so major release yes, fix a single bug: no - no single actor should hog the channel The hard part is that all of these are fuzzy. What is too niche? When is something "big enough"? When is someone posting too often? But I think spelling these things out and letting people self-police would already go a long way.
you could also argue about novelty, I think a big function of #announcements is simply making people aware something exists. New conference? New clojure podcast? New ambitious library? I would want that in my face. Conference announces three more speakers? That can go into #events.
> But I think spelling these things out and letting people self-police would already go a long way.
I think this point is big - many communities (and individuals!) self-police if the rules are well-stated. HN is a well-known example, where jokes or flamebait are frowned upon
I also participate in another Slack community with a very strong emphasis on placing things in the right channel. People end up appreciating it, given attention is such a scarce/precious thing :)
...which brings up again the point of where to draw the line.
Maybe a possible rule of thumb would be 1 "novelty" announcement per party and year, assuming potential global interest
. Although a stronger categorization (i.e. make #announcements only for libs/projects) would be easier to understand and enforce
--- > Please help keep #announcements alive and vibrant by posting updates from your Clojure related work there. It could be about making libraries and tools, wringing articles, publishing podcasts, broadcasting videos, or just about anything aimed at enabling the Clojure community to thrive This line of suggestions would futher blur the line between #announcements and #news-and-articles . #news-and-articles has plenty of subscribers and a clear focus. Why make #news-and-articles less valuable (by suggesting #announcements is OK as well)? > Personally, I like when there are frequent updates in #announcements . I like what it signals (an active community), but as a consumer it's reasonable to have the practical problem of seeing far too much content, particularly as time passses. Personally I have to mentally apply an "ignore filter" to multiple lines of things. Seems far from ideal, and in the end people like me who really wanted to keep up with things end up quite exhausted.
I've heard several people say that they liked the more frequent updates as well. I think there has to be a place serving both users: those who want to filter only the new and important stuff: #announcements. And those who enjoy seeing ongoing activity about projects in Clojure land. I propose the latter to be #project-updates
It would sound good to me Maybe it'd generate a new blurry area (should my major revamped lib go to #announcements or #project-updates?), but it seems easy enough to tackle with some rule of thumb
I made a description in that channel that new + major should go into #announcements
> Share updates for Clojure projects without restrictions. This is more informal and chatty than #announcements where you can share new projects and major new releases.
Re fuzzy rules. Rules are really hard to make so that they are both precise and correct. What needs to be communicated is the intent, and also the problems that occur when the intent is not followed. The problems weโre seeing in #announcements is basically a variant of https://en.wikipedia.org/wiki/Tragedy_of_the_commons
Well, a better name (less general than announcements so it's clear that it's not a place for events) + a clear rule (new + major) + more options for people who want something different should solve most of this I think.
maybe #major-updates would be a better name
#major-project-updates even
As I've said before, if we change the name to make it seem lib/project specific then important near-global "announcements" will be discouraged and the vast majority of people want just one channel to follow. We already auto-join new members to #announcements #events and #news-and-articles (as well as #beginners #clojure #off-topic and #slack-help ).
By creating #project-updates we now force the admin team to decide whether that is yet another channel that new members should be auto-joined to on signup and, frankly, I think that's too confusing at this point.
Only one person so far has complained about the frequency and content of #announcements
Even if others may have concerns... they haven't expressed them.
And so far I've seen several conflicting opinions here about the "right" way to solve things so it's clear there's really no consensus (and also this is a small channel compared to the overall population here -- and I've not even seen a consensus among just the admin team on this: mostly the admins have expressed a preference for "live and let live" with occasional guidance given to folks in various channels where it seems needed).
I love the announcements channels, the more activity the better. Would be awesome if some Twitter bot could be setup to tweet them.
> Only one person so far has complained about the frequency and content of #announcements
I can count 3-4 people who have agreed with me.
Complaining takes effort, and the risk of offending people (which is why I generally take quite a lot of caution in how I word things). Generally, it's more normal to simply mute #announcements, or pay less overall attention to it. Which is the precise thing I wanted to avoid. Tragedy of the commons
seems a very precise characterisation of the problem.
So, sadly it's hard to measure these, but observing simple things like emoji, or how many github stars a new project gets, I'd say that overall engagement has decreased (to be clear: getting parrots or stars has a very relative importance. I care far more about being part of a healthy, active community of makers).
For lack of objetive measurement, common sense also applies: how many people can actively pay attention to barely moderated content for say 2 years? At some point, consciously or not, people get tired. It's the reason e.g. every social network loses its shine at some point. HN has not.
I made it quite clear in advance that I didn't want my suggested rules to be literally established. I'd be happy if any other form was applied. Tossing a coin, or following an individual admin's interpretation would be also fine, and better than the current status quo.
Btw I see more agreement than disagreement so I'd be optimistic about what can be achieved :)
btw at this point the amount of effort I'm spending on the topic starts to be ironic. I think I've said all the possibly relevant things from my side already, hope something useful can be extracted from those!
Several of the admin team are following this discussion @vemv so if there's consensus for proposed changes we'll figure out a way to try to make more people happy ๐
Part of the problem for this community, from a mechanical point of view, is that we have no good way to onboard people that takes them through "the rules". Slack is designed for companies that have well-defined ways to draw up policy and communicate it to their employees. A 15K+ OSS community doesn't have that...
Would it be worth collecting the number of people dissatisfied with the announcements channel? Perhaps an admin can post there asking something like, "Please indicate your feelings towards your current experience with the announcements channel with a reaction of โ or โ"? Every other reaction is not counted. 3 - 4 people who are varying degrees of dissatisfied seems like a small sample for sweeping changes.
From my perspective (as a member of Clojurians, not speaking in any official or admin-ish way) I agree with @vemv that #announcements has recently gotten slightly noiser than Iโd prefer. I think that with this conversation weโve gotten at least some increased clarity on what weโd like to see there. In my opinion, if those of us reading this discussion make a point of encouraging good behavior in #announcements, thatโs likely to be sufficient. In other words, if you see a clearly inappropriate post in #announcements, respond in a thread, nudging the author to delete and repost in the more appropriate channel. That may be more than enough to tone down the noise level, and I suggest that we try that for a while before considering an official policy. Policies have real costs, and theyโre a lot easier to add than to take away, so Iโd personally prefer to avoid adding more if we can. It also seems clear that thereโs not full consensus on this issue, which makes a policy change even less appealing.
There's the nuance that if you like an all-in-one #announcements channel, presumably you'd be happy as well with subscribing to 3 channels (#announcements + #news-and-articles + #events) and get the same exact content. i.e. a strictly-categorized approach can make two kinds of people happy, but a unified approach, only one Acccomodating diverse preferences can be more important than a strict majority
Yup, I'd be fine with that @eggsyntax
There are always going to be some announcements that don't fit neatly in "lib/project" vs "news/articles" vs "events" but still need near-global visibility and we need the flexibility to allow those (such as the upcoming State of Clojure Survey announcement). So "strictly-categorized" is probably going too far.
I don't think it makes sense to presume that. I do enjoy scanning the announcements to get a sense of what's going on code\meetup wise, I would prefer not having to divide time between three separate channels. Despite some of the noisiness, I do find borkdude's frequency inspiring providing a good example of what a good maintainer does. My design background has taught me the importance of collecting feedback to validate that there is a legit problem in the large and that the proposed solution will be well suited to fixing it. Such a poll would give us a sense of how many active members there are in the channel and a general consensus about the state of it. I agree that shouldn't entirely be a deciding factor but I would feel better about it than relying purely on speculation and correlations.
More importantly, I want to know that the community is changing for the benefit of itself as a whole versus trying to accommodate the most vocal and critical individuals as no solution is perfect and will make everyone happy.