Mastodawn

Guy Van den Broeck Jan 4, 2023

On #scienceTwitter, my favourite thing was reading and, on rare occasions, writing "paper/preprint threads". From the researchers I followed and through the #Twitter algorithm, this became my most important source for new #research.

While people start to do the same on #Mastodon, I have the feeling that I miss important work bc no algorithm "saves" it for me if I don't watch my timeline constantly.

Two simple solutions would be: a commonly accepted hashtag that everybody uses when writing "paper threads" or an [...] @ a.gup.pe group with a similar adaptation rate.

#Question 1: is there already a mechanism for this that I missed?
#Question 2: What hashtag or group name?
I saw #TootPrint before. Maybe #PaperInAToot? #PaperInAPost? #PaperInAThread? #PaperPost? #MastoPrint?

Suggestions and boosts, please, we need reach for this! 🤓

#TwitterMigration #Science #Scientist @phdstudents @academicchatter @neuroscience @cognition

MWT Jan 4, 2023

@LeonDLotter I don't know the answers to your questions, but wanted to suggest also adding #Academia to your hashtag list.

I've also bookmarked your post so I can come back later to see what people said (and hopefully follow the consensus answer).

Leon Lotter Jan 4, 2023

@mwt
I think I’ll do a poll with the popular options after this post stopped going around 👍

Jascha Achterberg Jan 4, 2023

@LeonDLotter The ML community started using #PaperThread - @NicoleCRust set up #oldneuropapers to highlight fundamental neuro work and one could do #newneuropapers for new publications to keep it consistent. I feel it’s generally been a bit quite here over the holidays but I am sure the #neuroscience community will start sharing their new work again more actively soon!

Luiz Pessoa Jan 4, 2023

@achterbrain @LeonDLotter @NicoleCRust

It's been very quite... 😪

Todd Horowitz Jan 4, 2023

@LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

I agree, this is one of the main things I found useful about "Science Twitter". I like #MastoPrint, and also suggest #PaperThread.

Ken Miller Jan 7, 2023

For neuro paper threads: sigmoid.social/about/more has already claimed #PaperThread and #NewPaper (the latter announcing a paper without a thread) for the AI community. I enjoy seeing their papers too, but we need a distinct tag for neuro papers. For a thread, somebody suggested a #TootSuite, and I had suggested a #MastoPiece, but maybe we want something simple like #NeuroPaperThread and #NeuroNewPaper? or mix and match, like #NeuroSuite for a thread and #NeuroPaper for a paper w/o a thread?? Decisions, decisions.

I nominate @NicoleCRust to be the hashtag czar. Decide for us Nicole! (And anyone else, weigh in to help inform our czar!) And from your lips, or at least typing fingers, to the Mastodon's ears ...

@toddhorowitz @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition
@NicoleCRust
#neuroscience

El Duvelle Jan 7, 2023

@kendmiller @NicoleCRust
Since we can combine hashtags in searches, pinned hashtags etc. I vote for #NeuroPaper whenever we’re referring to… a paper in neuroscience, and simply add the #Thread if it’s also a thread?
Also if it’s a preprint thread it could be #MastoPrint but preprints and papers are really not that different nowadays…

Teixi Jan 7, 2023

@kendmiller @NicoleCRust @toddhorowitz @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition
We are living nowadays speedy overwhelming era:
New day, new neuroscience paper/s.
New every other day, new theory/ml free extra included.

Time to found a startup innovating and disrupting the market of daily neuroscience papers digest, with weekly top ten selection, monthly top five in-deep reviews, quarterly dual opposing columns.

ps: neuro millionares, vc, funds, bankrobbers: ping me!

Nicole Rust Jan 8, 2023

@kendmiller @toddhorowitz @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition
Done! Here:
https://neuromatch.social/@NicoleCRust/109654320540106174

Nicole Rust (@[email protected])

A big thanks to @[email protected] + for helping the #neuroscience community get its hashtag act together! https://neuromatch.social/@[email protected]/109645950754401219 > For neuro paper threads: sigmoid.social/about/more has already claimed #PaperThread and #NewPaper (the latter announcing a paper without a thread) for the AI community. I enjoy seeing their papers too, but we need a distinct tag for neuro papers. For a thread, ... maybe we want something simple like #NeuroPaperThread and #NeuroNewPaper? Great idea Ken - let's do this. ➡️EVERYONE: BOOKMARK THIS! ⬅️ and follow those hashtags. And then don't hold back. After all, it's what we all show up here for: to hear what you've figured out and learn from you. #neuroscience #neuroAI #psychology

Neuromatch Social

Ulrike Hahn Jan 9, 2023

@kendmiller @NicoleCRust @toddhorowitz @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

I'm still going to suggest using #PaperThread for all paper threads but just adding the discipline in a second hash tag- at least until that wonderful day where there are *so many* paper threads on here that just quickly discarding the ones I'm not interested in becomes really onerous... and more so than scrolling past other types of content I'm not particularly interested in

El Duvelle Jan 4, 2023

@LeonDLotter Also if people remember to add the appropriate themed hashtags (not just preprint thread, but the research themes), that would help (as you can follow hashtags or combinations of hashtags)

Brendan Jones Jan 4, 2023

Agree - as a non scientist those threads were often a great way to get the key info about new research, concisely, direct from the source, and not in academic jargon (unlike the paper itself). A great way to learn. Would love if there was an agreed upon hashtag for new research. #researchpaper might fit the bill. #breakingresearch is less obvious but could be good for denoting that it’s new research, or maybe just #preprint .

Giorgos Diakoulakis

@Brendanjones I am in favor of #preprint for unpublished work and #newpaper for published ones

Eric Lawton Jan 6, 2023

@gdiak @Brendanjones Yes, it's redundant to put "tooth or "thread" in the tag, but PLEASE use camelCase for screen readers.

#ResearchPapers is already in use.

Eric Lawton Jan 6, 2023

@gdiak @Brendanjones It seems I haven't been on Mastodon long enough for autocorrect to learn I'm as likely to mean toot as tooth. 😀

Brendan Jones Jan 8, 2023

@EricLawton @gdiak and autocorrecting to ‘tooth’ is obviously wrong here, it should autocorrect to ‘tusk’ 😋

Dr Annetta Mallon Jan 4, 2023

I am the same in missing the knowledge and information elements of Twitter.

Please share if you find any hashtags or other protocols here on Mastodon.

Thank you and have a great day.

Wolfgang Cramer Jan 4, 2023

@LeonDLotter @phdstudents @academicchatter @neuroscience @cognition Not sure I understand the problem. On twitter, this worked well because some people did those excellent threads and you followed them. Why would that not be the same here?
In my opinion, we just need to bring this activity here in a more coordinated matter, and we need to boost systematically the paper threads we see.

Thiemo Fetzer Jan 4, 2023

@wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

The problem is these threads are never seen unless people boost extremely generously as the timelines are chronological. And most users on Mastodon try to use it like Twitter without the sophisticated features.

I do genuinely think the next evolution of Mastodon should involve opt-in algo's that are transparent (e.g. only verified genuine human user boosts or favorites feature etc. in an algo curated timeline)

Ulrike Hahn Jan 5, 2023

@fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

the suggested consistent use of a hash tag such as “paper thread” would do that though, and given that you can follow hashtags you can collect those into an RSS feed which means you would never miss any, no?

Leon Lotter Jan 5, 2023

@UlrikeHahn @fetzert @wolfgangcramer @phdstudents @academicchatter @neuroscience @cognition

Yes, a widely accepted hashtag, or one general hashtag and associated field-specific hashtags (e.g., like #newPaper and #newNeuroPaper) would do the job. Each user can then decide on their own how they want to track the hashtag. Maybe a bot account boosting such posts would be interesting?
I nevertheless would be in favor of a transparent opt-in algorithm-curated timeline.

Russell Neches Jan 5, 2023

@fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

No need for an evolution. You can roll your own algorithm right now using the API.

Wolfgang Cramer Jan 5, 2023

@ryneches @fetzert @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition Can you explain a little more what that means, "roll your own algorithm"?

Russell Neches Jan 5, 2023

@wolfgangcramer @fetzert @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition Sure! The Mastodon protocol, and the API for the main implementation are open source. There are some intrinsic constrains because of the fact that Mastodon is federated, so there is no "Firehose" feed like there is on Twitter. But, within those constrains, you can hook software into an instance that does almost anything you want. https://docs.joinmastodon.org/client/intro/

Getting started with the API - Mastodon documentation

A primer on REST APIs, HTTP requests and responses, and parameters.

Russell Neches Jan 5, 2023

@wolfgangcramer @fetzert @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition There isn't really any technical barrier that prevents someone from essentially re-implementing the Twitter algorithm on Mastodon. The API would allow you to do that. Rather, the problem is social. You can't impose your algorithm on people; you have to convince server admins to give you API keys, and users look at your algorithm-derived feed.

Thiemo Fetzer Jan 5, 2023

@ryneches @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition
Yes I know, the issue is that most people that stick around on Twitter are not technically skilled to "solve that problem" for themselves. So they just remain there. And the sophisticated users dont need those "tools". So we are in an equilibrium where Mastodon may remain small, which again, is fair enough.

Its a classical coordination failure.

Russell Neches Jan 5, 2023

@fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition I wouldn't be so pessimistic. Lots of people are really interested in making algorithmic feeds for Mastodon. For example, there was a discussion about how and why to do it on episode #390 of @talkpython : https://talkpython.fm/episodes/show/390/mastodon-for-python-devs

Mastodon for Python Devs

Wondering what Mastodon is all about? More importantly, what does it offer Python developers and other open source folks compared to Twitter? There is a huge amount of interest in the tech community about what's happening at Twitter and whether they should expand to or even move to a new location. So I decided to put together a set of experienced Python developers who have been Mastodon inhabitants for a long time to discuss what this unexpected shift means for one of our important online watering holes.

Felix Reutter Jan 5, 2023

@ryneches @fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition @talkpython
Also, have a look at this #algorithm project - still under development:

https://mastodon.social/@MattHodges/109451111674479285

https://github.com/hodgesmr/mastodon_digest

Next step should be an easy way to use it for the masses.

GitHub - hodgesmr/mastodon_digest: A Python script that aggregates recent popular posts from your Mastodon timeline

A Python script that aggregates recent popular posts from your Mastodon timeline - hodgesmr/mastodon_digest

GitHub

Russell Neches Jan 5, 2023

@x1l3f @fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition @talkpython Well, the next steps will be tons of experimentation. Do you want the algorithmic feed to replace the social feed, as threshold triggered notifications, a separate feed on the side, or mixed in? Does it run in the client? Plug into the instance server? Its own service provided to instance users? A third-party cloud thing?

And that's just implementation, not, like, an actual algorithm.

Mauricio Foronda Jan 5, 2023

@ryneches @x1l3f @fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition @talkpython

Yes! So far I have

- The twice-daily digest page https://mauforonda.github.io/mastodon_digest/
- The morning email https://github.com/mauforonda/mastodon_email_digest
- The link collector https://mauforonda.github.io/mastodon_links/
- And the rss reader for everything else https://mauforonda.github.io/miniminiflux/

This is a problem that should probably be tackled by clients: @elk @apps #pinafore , and the like.

Mastodon Digest

Zacharias Steinmetz Jan 6, 2023

@mauforonda @ryneches @x1l3f @fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition @talkpython @elk @apps these tools are just awesome 😍 Didn't know about them before. Thank you!

Richard Van Noorden Jan 6, 2023

@mauforonda I was very excited to see this so I tried to fork your digest page and get it running (I am a beginner, cannot program, but can follow instructions), but alas I don't seem to have a gh-pages branch in the last step under 'to run your own'. Only 'main'. Are there other things I need to do to make it work? Thankyou!

Mauricio Foronda Jan 6, 2023

Hi!

I see that you didn't run the "My Mastodon Digest" workflow. First, enable github actions by going to your repository "Settings" -> "Actions/General". Then go to your repository "Actions" tab, select the "My Mastodon Digest" workflow and "Run workflow". That should run the workflow for the first time. Then you'd see a "gh-pages" branch you can publish from "Settings" -> "Pages".

Let me know how it goes!

Richard Van Noorden Jan 6, 2023

@mauforonda Success!
I have very little idea what just ran behind the scenes but I now have my own digest. That's terrific, thanks!

Mauricio Foronda Jan 6, 2023

@richvn
Awesome! Glad to help

Richard Van Noorden Jan 6, 2023

@mauforonda Your 'links' project looks great for research communities. As you say, set up a list of people-to-follow, & find their links or popular posts.

Mauricio Foronda Jan 6, 2023

@richvn
Right, creating an account that follows all users from a well-defined community and creates feeds for things they look, like papers, for sounds valuable and easy enough. For example an account for all repec-registered economists https://ideas.repec.org/i/emastodon.html Though maybe most "well-defined" communities already have more formal aggregation channels (?).

RePEc-registered Economists on Mastodon

JB Emmons Jan 6, 2023

@ryneches
@x1l3f @fetzert @wolfgangcramer @LeonDLotter @talkpython
I'd also encourage y'all to look at what @Colarusso has done with @icymi_law, a bot that effectively works as a well-designed algorithm serving up posts from lawyers & legal-tech types.

Don't sleep on it just because it's not directly in the sciences; I'd say LawFedi is as strong a community as many others here, in no small part due to that bot, & other communities would benefit from a similar tool.

https://icymilaw.org/about/#bot

About In Case You Missed It (Law)

@ICYMI_Law is a bot, originally created to help members of the legal community discover content on Mastodon. At first all it did was boost posts, but it quikly grew as I added new features. This page explains in plain language how each of these work so readers/listeners can decide how to fit them into their media diet.

David Colarusso Jan 6, 2023

Thx @IntlLawGnome

FWIW, @icymi_law is explicitly billed as an opt-in algorithmic feed for non-tech folks designed to address the discovery problem. Also, its workflow/algo choices are made clear on the associated site: https://icymilaw.org/about/

It has grown to include a digest on that site, part of which is a collection of preprints (made easy by the fact law folks all use one preprint server). It also collects hashtags.

@ryneches @x1l3f @fetzert @wolfgangcramer @LeonDLotter @talkpython

About In Case You Missed It (Law)

@ICYMI_Law is a bot, originally created to help members of the legal community discover content on Mastodon. At first all it did was boost posts, but it quikly grew as I added new features. This page explains in plain language how each of these work so readers/listeners can decide how to fit them into their media diet.

Thiemo Fetzer Jan 6, 2023

@Colarusso @IntlLawGnome @icymi_law @ryneches @x1l3f @wolfgangcramer @LeonDLotter @talkpython

This is really good. I will see how one can deploy this for #econtwitter #econtooter - or maybe @paulgp are you aware of any bots that may do something similar?

Is this open source?

David Colarusso Jan 6, 2023

@fetzert thank you. It's been a lot of fun to build. I'm still trying to decide if it's stable enough for a refactor and share. However, @MattHodges has a similar project up on GitHub: https://github.com/hodgesmr/mastodon_digest

@IntlLawGnome @icymi_law @ryneches @x1l3f @wolfgangcramer @LeonDLotter @talkpython @paulgp

GitHub - hodgesmr/mastodon_digest: A Python script that aggregates recent popular posts from your Mastodon timeline

A Python script that aggregates recent popular posts from your Mastodon timeline - hodgesmr/mastodon_digest

GitHub

Matt Mascarenhas Jan 6, 2023

I'm loving this and that it was amongst the first stuff I saw back "on grid" after the hols!

It's a response to chronological timelines, motivated to improve discoverability, right? I have notes (https://miblo.net/ppic_notes.html) coming at this from another angle: basically "most recent ≠ most relevant / informative". So its main presentation shouldn't be primarily chronological!

@fetzert @MattHodges @IntlLawGnome @icymi_law @ryneches @x1l3f @wolfgangcramer @LeonDLotter @talkpython @paulgp

Notes • Practising Programming in C • The Life and Times of Miblo del Carpio

David Colarusso Jan 6, 2023

@miblo yeah, for me, discovery was the motivator, but only in conjunction with the understanding that discovery was an issue folks in the community saw as an issue (e.g., not enough or too much content). So I wanted to do what I could to help the community grow. I really like the idea of decentralized community-driven social media, and want it to succeed.

@fetzert @MattHodges @IntlLawGnome @icymi_law @ryneches @x1l3f @wolfgangcramer @LeonDLotter @talkpython @paulgp

Matt Mascarenhas Jan 6, 2023

Cool, cheers! And yeah, same here. Maybe I should try following your lead and build on our already-existing Mastodon, but even if I do go for something new I'd want it to be compatible with the Fediverse.

@fetzert @MattHodges @IntlLawGnome @icymi_law @ryneches @x1l3f @wolfgangcramer @LeonDLotter @talkpython @paulgp

Ingo Lantschner Jan 6, 2023

This is a really useful hint but needs some background in software-development.

Just tried to fetch some #hashtags from my own instance to prove if it works – of course it did 👍.

So to explain it for non-IT people: Anyone can also access all the posts and data we see in our browsers in a *machine*-readable form. This in turn allows anyone to filter and recombine this data and present this filtered data again in a human-readable form. For example, one can create a web page that displays all posts from a particular group of users that contain either "academic" OR "science" as a hashtag. This can be extended much further: Get the data not just from one #mastodon instance, but from a whole range of such instances, exclude troll accounts, but also include #peertube instances. Full text analysis should also be possible in this way.

@wolfgangcramer @fetzert @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

ArneBab Jan 5, 2023

@fetzert Comparing what I saw on twitter and what I see here, I have to say that I see a lot more interesting things here.

The simplest way to make sure that you see the threads is to suggest to the authors you know to include a hashtag you propose.

That way you build your own algorithm based purely on information you control. @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition

Ed Cashin Jan 5, 2023

@fetzert @wolfgangcramer @LeonDLotter @phdstudents @academicchatter @neuroscience @cognition Yes, opt-in algorithms would be very helpful!
The thing is, the way I interact with my stream has tons of information that an algorithm could use, so I think that it's not Mastodon that needs to integrate opt-in algorithms but apps, like Tusky.

Nick Jan 4, 2023

@wolfgangcramer Depending on how many people one follows, and one's schedule, it might be the case that someone might only scroll through an hour's worth of toots per day let's say. But what happens if someone posts their paper thread in the other 23 hours? With twitter's algorithm based on what it thinks you are most likely to engage with, it would likely show you that post when you arrive, but with mastodon's chronological algorithm, you've missed it.

Nick Jan 4, 2023

@wolfgangcramer Hashtag for searchability and boosts for spreading beyond an existing group are very likely to be part of the solution, agreed. So settling on a particular hashtag may help.

Bryan S. Weber Jan 4, 2023

@nyates314 @wolfgangcramer
I do like the idea of hashtagging all the research under #paperthread. I have been thrilled to get such rapid access to papers - it is why I am here.

Wolfgang Cramer Jan 5, 2023

@nyates314 OK, I see the point, thank you.
What about paper announcements that are not actual threads but merely say, in one toot, "here is this paper you might be interested in, and it is about this topic"?

ArneBab Jan 5, 2023

@wolfgangcramer there boosting works the same — we could have a hashtag, too, but keep in mind that hashtags are open for all, not only for the accounts you follow.
@nyates314

ArneBab Jan 5, 2023

@wolfgangcramer the hashtag NewPaper seems to be used by some for that. @nyates314

Verena Jan 5, 2023

@nyates314 @wolfgangcramer
I was thinking about that too. Doesn't that mean that one more or less always misses the toots from other time zones, when one sleeps? Unless they are boosted a lot also later.

Nick Jan 5, 2023

@verysummer @wolfgangcramer Yep, unless you follow few people and can scroll back through 24 hours worth of toots at a time (or 8 hours if checking immediately after waking up). Boosts, following hashtags, notifications for important people or hashtags, and lists, can all help alleviate this issue, but it is an inherent problem in a chronological firehose of posts.

katie Jan 4, 2023

@LeonDLotter #science doesn't work?

Leon Lotter Jan 5, 2023

@rizzo
#science or #research might be to general if Mastodon keeps on getting traction? I’d opt for sth specific to the idea of explaining new research in a toot/thread. Maybe even specific to broad fields, e.g., someone suggested #newNeuroPapers for the neuroscience field.

Dr. Ruchira S. Datta Jan 5, 2023

@LeonDLotter @phdstudents @academicchatter @neuroscience @cognition Would also like to know.

Anne Carpenter Jan 5, 2023

@cognition @academicchatter @LeonDLotter @neuroscience @phdstudents
Tootorial is my favorite term so far for what was called Tweetorials over there.

Firoozye Jan 5, 2023

@LeonDLotter @phdstudents @academicchatter @neuroscience @cognition would be great to have a threadreader app on mastodon just to compile threads. And, on the bird-site, IFTTT gives at least some functionality to save based on specific triggers (eg, If i star then tweet ⇀ pocket). Even this basic stuff is hard to recreate on the fediverse. Your “trigger” of a thread is a bit more sophisticated.