I don’t think there’s any need to make a statement about leaving[1]. Like I said before: X has been hemorrhaging users for multiple reasons: it suppresses tweets that link out to a website[2], it promotes tweets from paid users over others[3], and they’ve got a huge spam/bot problem now. It’s universally considered a disaster except by people who have ideological reasons to ignore that fact.
It makes sense to simply move on.
Ok, then. If the consensus is to leave quietly, so be it. Some, though, may ask why.
Some quick points:
The account was unused, so it mostly makes zero difference. ↩︎
I did create an account on X, in case it would make a difference in what I see, but it didn’t. Still no “discover more”. I expect @notatallshaw’s speculation is right: that many big sites wait to “suck you in” before enabling their more obnoxious 'features" on your account.
The question to me is more whether they like it or not 
If a person doesn’t want to see random garbage, don’t use an option designed to show them random garbage. “Following” hasn’t yet shown me anything I didn’t explicitly follow. “For you” is all over the map, including things I found repulsive, and things I admired. That’s not usually what I want, but sometimes it will be. For example, I have extremely strong opinions and feelings about one of the wars currently going on, but seek out opposing views, and “For you” did show me pieces from both sides of that. That had value to me. I hadn’t seen those sources before. Nothing I saw changed my mind, but did better inform it.
Wholly agreed. Multiple sources would remain a good idea even if “almost everyone” was delighted with X.
What is your estimate of the number of users currently on X and on its competitors?
I don’t have my own estimate because why would I? Anecdotally, the majority of the community I followed has all left for other platforms, including a large proportion of the scientific community.
But if I wanted to have a pointless debate about this, I would still be on X.
My apologies; I supposed that when you said:
that you were suggesting that the number of users on X is (or soon will be) sufficiently small that the PSF needn’t be interested in reaching them. But since you don’t have an idea of what the numbers are, I now realize that you must have been suggesting something else.
was set up a bit over 2 years ago, and has 11.6K followers now.
was set up about 15 years ago, and has 677K followers (close to 60 times Fosstodon’s count). The number of views of a post is typically much less than the total number of followers regardless of platform, but the PSF’s post there from 3 days ago already got 6.5K views, over half the total followers on Fosstodon. A post from Jan 6 is up to 13K views.
So a Fosstodon alternative has been available for some time already, but, to my eyes, there’s no evidence that PSF followers are abandoning X to flock to it.
OTOH, who can know? I know for a fact that at least one PSF follower on X stopped using X entirely, but didn’t bother to close their account. So there are at most 677K-1 human followers
The PSF isn’t running discussion threads on these platforms anyway. It looks like there are almost never replies. So whether people like a platform for discussions seems largely irrelevant to me: the PSF uses them to announce info of general interest.
Maybe Bluesky will do better, but for now it’s attracted close to no Python-related content.
It’s quite possible for a platform to be a disaster for civil - or even informed - discussion, yet also be an effective way to disseminate information.
Within this discussion, multiple valid and quite understandable reasons have been cited for the PSF to depart X/Twitter as a means of disseminating information. The case has been made reasonably well. However, it has also been suggested that this departure be made without an explanation being offered to the community. Making the change quietly seems rather risky to me in terms of public relations. Possibly, as some of the previous posts seem to suggest, the rationale for departure will be universally understood without its being made explicit. Maybe it is true that everyone out there already agrees that X/Twitter is seriously problematic. However, just in case people begin to ask why the change was made, it might be good for us to consider what’s been said in favor of leaving, and to form in advance some consensus on which reason or reasons to cite as an explanation in response to possible questions.
I find there’s quite a lot of Python activity on both Mastodon and Bluesky, I’ve not missed leaving Twitter at all. Here are Mastodon “starter packs” that let you follow a group of people easily: Python community, Django community and Python core team. And on Bluesky: Python community, PyLadies and Python core team. See also the #python and #python hashtags.
I don’t personally have a settled opinion on the issue, so I’ve been interested to hear the thoughts shared here, and the only contribution I can add is a couple data points in case they help inform the conversation:
PS - Just to be clear, I’m on PSF staff & involved in comms, but I’m not a decider on this issue, just an interested party:)
I assume this is ongoing work, so removing a channel would save some effort?
I don’t know how one would measure the impact of a given channel, but I do wonder how useful the channels are compared to the announcements from specific people.
I’m told the startup cost of adding a new channel is larger than the ongoing cost, but yes removing a channel would save some work (for one example, managing DMs/tags/replies)
I’m not a social media expert, but my working understanding is that the impact of a given channel is roughly how many people in your “target audience” you reach and how many of those take up the “calls to action,” e.g. donating to a fundraising campaign or reading a linked blog. And yeah, I’d say that individuals sharing PSF messages can definitely be higher impact than the PSF sharing directly, both in terms of eyeballs and persuasiveness (anyone reading can please feel free to take this as an invitation to boost PSF announcements;)) But I’m not sure what the implication of that is for what channels to use.
I don’t think that, if PSF will leave Twitter, the fact will have any consequences on current Twitter policies. I feel sympathetic with the proposal anyway (following the evergreen paradox of tolerance)
Which is my primary concern: the PSF posts announcements hoping to advance specific ends (which vary by annoucement), but most of us (like me) have no access to measures of success (like, indeed, money contributed or number of hits to linked pages, or even, on many platforms, how many views the original announcement gets).
The original post didn’t mention those primary goals, and a number of replies just echo that they find X’s current general political leanings revolting. So I’m “don’t throw the baby out with the bathwater?” here. The primary goals should matter too.
Which you’re aware of, so I’ll shut up now 
Which has 138K followers, quite a healthy amount.
So there’s more out there than I was aware of. Now I’m concerned that you may spread yourselves too thin. There is no single platform everyone will like [1], the cost/benefit ratio will vary across each, and the cost is never 0.
I generally don’t “do” social media, so it doesn’t really matter to me directly, but I am keen to see the PSF continue thriving. Effective comms is a big part of that. “Not too many platforms, not too few” - there are tough tradeoffs to make.
for example, I closed a LinkedIn account years ago, tired of excessive email spam pointing me at self-promoting posts from “influencers” I neither knew nor cared about ↩︎
I try not to lean into anything that would lend to a “guilt by association” idea, but I also don’t use or reshare things people share that are on X/Twitter. There are certain lines that remove any possibility for me to view it as “just a platform in spite of who owns it”. It’s worse than what an unmoderated site ends up, as the owner has explicitly interfered with the site’s algorithm for hateful reasons, unbanned those who were previously banned for use of hatespeech and has declared things such as “cis is a slur” including so far as how the site is to be managed.
He has even lashed out at those who are “politically aligned” with him over small things like pointing out that he isn’t a “top gamer”, removing their blue check mark in retaliation to that, and use of the site and what people see on it is largely on a whim to what last set him off. Even if his politics were not abhorrent to me, and even if he separated his political leanings from any sitewide policies, neither of which are things he actually does, his behavior lashing out at those on his platform who disagree with him is reason enough to go elsewhere.
(Commenting as an interested party/observer.)
To tack onto what others have said, I’ll offer a strictly practical argument instead of a moral one[1]: when a service (any service) becomes unreliable or untrustworthy, minimizing its potential impact is sound strategy. This is especially true in the case of large presences like the PSF’s, where harm could be done to the PSF’s reputation[2] or ability to communicate should the PSF lose access to an account in an uncontrolled manner.
In the concrete case of Twitter/X: I think there’s ample evidence that the service itself is unreliable (from a strictly uptime/availability perspective), as well ample evidence that the service’s leadership behaves erratically/capriciously and lacks sufficient controls/internal policies around manipulating individual accounts/site content.
All told I think a controlled reduction in activity, amply signaled on other official channels, would pre-empt any potential harm to the PSF’s reputation and/or ability to communicate with the larger Python community. That would probably have to include leaving the current Twitter/X account in a “frozen but not deleted” state to prevent malicious takeovers.
I think the moral arguments for leaving Twitter/X are strong, but it’s good to have a sufficient argument from practical grounds as well! ↩︎
Concrete examples would include things covered under the moral arguments above, but also practically: Twitter/X’s staffing problems have been publicly documented, and suggest that the site has only limited staff available for security/incident response. This suggests that Twitter/X may struggle to respond to an account takeover of the kind that have happened to other prominent accounts (whether for misinformation, crypto scams, or anything else). ↩︎
I think you meant to say:
Some platforms have a keen interest to inflate their user counts by a varying range of tricks, from mass emails telling users that „someone tried to log into your account, come check that everything is ok” (account was unused for like 5 years) to cloning legit traffic x-fold to pulling user counts completely out of their arse, given the private company status none is actually verifiable.
My 2c: post in own channel, like here, and post only links on all the platforms we care about.
LinkedIn is very common for folks who attend conferences, both old-timers and fresh grads. It’s slower pace though. Source: PyCon APAC.
Ad hoc (ephemeral WhatsApp groups) is common for organising stuff live and it seems especially so for the younger, mobile generation. Source: PyCon APAC.
Mastodon exists, it seems to be more about individuals and connections between them, backed up by robust post history. Source: a few of my friends who are Pythonistas.
Matrix is another option, but I didn’t see much traffic so far.
I would not be surprised if TikTok and Instagram were best avenues to reach younger students, those who are not going to universities yet. I don’t have any data for that one.
Matrix is another option, but I didn’t see much traffic so far.
Matrix is a chat platform (similar to IRC, Signal, Gitter, Slack, Discord, WhatsApp, et cetera). It’s not a microblogging platform, so doesn’t really make sense to compare it to X, Mastodon, LinkedIn, BlyeSky, and the like.
Granted, there are social media platforms like Weixin/WeChat and Facebook which tend to integrate chat functionality, and so blur the lines in the mind of many, even though they’re really just aggregating different features and protocols behind the veil of a single interface/name.