-
Notifications
You must be signed in to change notification settings - Fork 71
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Close/de-list Jupyter Twitter account #170
Comments
Just for context: Mastodon accont is inactive and decision about instance has not been reached (@fosstodon has not changed its language rules, #146). Bluesky is not an alternative as of now as it is invite-only. |
Yes, as @krassowski there isn't yet a clear broadcast channel to replace our use of Twitter. Other OSS orgs like the PSF and the Apache Software Foundation are still using Twitter at the moment as well. The EC are working on standing up a Media Strategy working group to put more attention into the issue. |
Twitter isn't really a valid broadcast channel anyway, anymore, as it's completely lost its status as 'where people are', so I think there is little point in using it. I'm not sure keeping it is better than having nothing. We have the blog and forum, both of which are better at this point. |
I'm not sure that is completely true, the announcement from last week still got 3 Replies 11 Retweets and 26 Likes. I can't vouch for activity on bluesky as this is still private beta AFAICT. I'm happy to leave twitter in the long run, but I would prefer to have a proper media team before, otherwise we are just going to make our communication worse than it is currently is. |
It is so deeply disappointing and frustrating to spend so many months actively supporting a company that officially endorses, enables, and publicizes hatred and violence by continuing to promote the platform as our broadcast channel. Having nothing is better than what we're doing now. I really don't think we lose anything by just closing the account. |
I think the conflict may come from the multiple things your original message ask, and maybe is missing concrete action plan. If you want to stop listing Twitter on multiple places, I'm all for it, and will be happy to review merge your PRs. I do not have access to tw credentials in the vault, so can't change the I likely still have delegation access via tweetdeck, so I can schedule a "We are now at please unfollow us here once you followed is there" and we can look at how many of the current ~90k people we have unfollow us. And yes I would have love to have a social media team sooner. But I'm actually more embarrassed by still having a Facebook account at that point. |
I think that delisting links from official websites is ok but the other part:
Is IMO not something Jupyter should do as such an action can be interpreted as antagonising audience and those who are still on that platform. For what it's worth, there was a lot of engagements with the tweets on the JupyterLab 4.0 blog post (e.g. https://twitter.com/realpython/status/1667925184806899712 >1000 hearts, >200 retweets; and more here https://twitter.com/pycoders/status/1668258604984545281 and elsewhere) which highlights that it is a pity that instead of a tweet to blog post the Jupyter account retweeted a post from a private account which lead to such a low reach as mentioned previously. |
On the broader question of Twitter: I am very unhappy about the state of Twitter and its alternatives, but as of today, I get information from Twitter that I can't get anywhere else, and I am able to reach people that I can't reach any other way. I also think that @krassowski makes a good point about how the JupyterLab 4.0 Tweet was handled. @JasonWeill drafted a tweet for JupyterLab 4.0 with the intent that it would go out on the official Jupyter account, but was told to Tweet it himself (he has only 500 followers). As one would expect, it didn't go very far. Also, pinging @Ruv7 and @blink1073 who lead the emerging social media team for the EC. |
I'd like for us to adopt two social media habits to improve engagement:
|
Fair. Right now, all I'd propose is removing links, and stopping posts to the twitter account. I think it's already unmonitored. We could update the bio to just point to the blog or something.
What I'd like is to immediately delete the twitter account entirely, as I don't think anything would be lost. Second choice would be to make it private with a link to somewhere else (e.g. blog for now) in the bio. |
For me the big no on deleting is because I believe Medium (and Figma?) are login-via-twitter, and the second reason is if we delete there are risk of Someone picking it up to impersonate. |
Just to be clear what "delete" means, I think we should keep the name, but if it it decided to delete all the old content I would not oppose. We do have a history of keeping placeholder names like https://twitter.com/JupyterDev which we do have access to. |
Trying to step aside from my own thoughts about social media and Twitter in particular... This is a complex issue that involves a rapidly changing social media landscape and impacts multiple aspects of Jupyter (community, marketing, outreach, announcements, JupyterCon participation, fundraising, etc.). These decisions need to be made by the Executive Council (EC) with input from the community. Please keep the feedback and questions coming on this thread and the EC will begin to work on this and come back with a plan/questions/etc. One other thing that is may not be evident is that the EC has a large backlog of urgent and major items and "fires" that require significant work, discussion, and decision making. We are making progress and are doing everything we can to scale the work by involving others - please be patient though :-) |
To provide a brief update and to clarify our current structure for this decision, @blink1073 and I are working on a charter for a Media Strategy working group. Our first meeting was last week and our second meeting is tomorrow Thursday. Once this is ready we'll present it for a vote from the EC + SSC. In the meantime a response to this issue will be taken on the EC which as Brian mentions is a facing a large backlog. Please know it's on our radar and I agree a decision/strategy/next steps for this channel and the others is overdue. |
We should definitely work on a strategy. Promoting and using other platforms (such as Mastodon) may be the way to go. Unfortunately, we have not built much of an audience elsewhere yet, though I would not discount Linkedin where Jupyter already has 12k subscribers (I realised that when we worked on promoting JupyterCon). |
rOpenSci had a good post about this just yesterday. I hope we can follow their example. |
Hi everyone! This subject came up at the most recent Jupyter EC call (on Tuesday). Since the media relations working group is still ramping up and writing its charter, it's a very good time to get a read of what people think and to build a consensus around what our strategy looks like. Since this is a subject where people may want to have their say privately, we decided to create a form that can be filled out anonymously or not to ask people for their views on Twitter and social media in general. If you, the Subprojects you are part of, or someone you know has thoughts about these topics, please share the link to the poll: |
Is the stop (cross-) posting to twitter push due to an anti-twitter agenda or due to engagement numbers? Please be transparent about the social media metrics that you are using to guide the marketing efforts of this project. There are various services for cross-posting to various social media services. How many followers are there on each service, |
@afshin what do you think of :
I don't think this is metrics based, As for the metrics I think they are public. Facebook 5.8K followers |
This is about assessing the different sentiments expressed by our community in person and in issues like this. It might be informed by metrics at some point but that's not the primary angle we've been talking about so far.
I don't know if we even have these. But if we do, we haven't been discussing them. If we have them, we can try to find a way to share them. But as Matthias says above, this is not really a metrics-based conversation. It is more about values, given the sentiments being expressed.
|
I can post on Twitter with either/both of the twitter accounts. |
How very meta (no pun intended). Do you think it'd be beneficial on Twitter or will we just get flooded? Either way, I'm happy to defer to your judgment! I don't have strong intuitions about what the best way to gather this sentiment actually is. |
Well I guess that's all the questions, which channel and how are we communicating. I'm guessing I can tweet:
|
What do you think about this? We send this message to:
They can choose how they propagate this with their individual communities in whichever channel they wish. We are working on our communication strategy on social media, and in particular whether we should use Twitter. If you would like to share your thoughts with the Jupyter Executive Council, please consider one of these three channels:
|
And of course the GitHub issue will be available for the public to comment on as well, there would be no obligation to be a member of any of those bodies. The same applies to our office hours call. |
No objections, I'll do whatever the council asks. Note that [We are working on ... channel] is too long to fit in a tweet. |
Perhaps it would be best to pick a free service that will copy the shortest
message out to the other networks unless a Jupyter marketing team member
decides to program them independently.
What would be a reasonable quantitative or qualitative ROI metric to
compare the value of social media channels for open source software
projects (without conversion tracking)?
IIRC JupyterCon had their own marketing team?
…On Sat, Jul 1, 2023, 5:55 AM Matthias Bussonnier ***@***.***> wrote:
No objections, I'll do whatever the council asks.
Note that [We are working on ... channel] is too long to fit in a tweet.
—
Reply to this email directly, view it on GitHub
<#170 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAAMNS7S74YFCB3HHXFACRDXN7X2RANCNFSM6AAAAAAYZX2TQM>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
I think we should share this on Discourse too- I'm happy to post there if you want. Are you still working on the text? |
@manics I think you probably just received an email from me about this. Please feel free to share as you would like! |
Hi all, thanks for everyone involved in the conversation here and for those who have responded to the form. We have closed the form to further input, but the overall consensus in the form and in the EC office hours was that we should only use Twitter/X as a secondary "broadcast" channel to announce blog posts or Jupyter events, and look to join a Mastodon server as our "home" social media channel. The EC have been working on a draft charter for the Jupyter Media Strategy Working Group, which we are hoping to post as a PR for review in the next week or two. Establishing a "home" on Mastodon will be one of the first orders of business for the new Working Group. |
ipython/ipython-website#166 remove Twitter links and google+ (sic) links from IPython.org. Review appreciated. |
It seems long past due for Jupyter to close its Twitter account, or at least stop posting to it, promoting it, or linking to it. We have replacements at mastodon and blue sky in place, and at this point it feels rather embarrassing to still have an official presence on the platform that does anything other than point to somewhere else. How should we go about this?
The text was updated successfully, but these errors were encountered: