3 ms·

> There is a difference between tweets and feeds compared to peer-reviewed papers in magazine-like journals. > Federation might be a part of a solution to the

> There is a difference between tweets and feeds compared to peer-reviewed papers in magazine-like journals. > Federation might be a part of a solution to the problem, but no where near the whole solution. To be clear, federation and even the ActivityPub protocol have nothing to do with the tweet-like interface of Mastodon (and ActivityPub isn't the only way to federate, either. Email is federated), although TBF the article itself fails to make this distinction clear. There are a number of ActivityPub applications which use a different presentation: * Pixelfed: Basically federated Instagram * WriteFreely: federated super-minimalist long-form blogging * Wordpress: Ubiquitous (maximalist?) long-form blogging platform that has recently added ActivityPub support That said, I'll admit I have no idea what the obstacles are since I never got into that world. Are there technical requirements which nothing else fulfills? Does it mainly require providing a means of searching for research documents in different stages of review and validating that an individual is a legitimate reviewer/contributor (something like the authorization around git, come to think of it)?

noirscape 3y ago
To be more precise, ActivityPub is pretty much just a JSON/HTTP version of email. There's a few fancy things to have stuff like user profiles and stuff to make it "look nice", but if you look "under the hood", the data structures you find ends up and the methods of communication end up looking more or less like a modernized version of email. --- Mastodon is a sub-implementation of AP that follows some of the components and then weirdly tweaks some of them in weird ways (because Mastodons internal representation doesn't actually match up with these data structures and Gargron wanted some features like Content Warnings without bothering with the entire process of protocol modifications, so he ended up cannibalizing a different field for it, resulting in an implementation that... mostly results in people yelling at each other for not using CWs "appropriately"), resulting in strange compatibility issues when writing your own AP instance. Most AP instances typically end up trying to then match Mastodons weird implementation (basically look at the social feed of any AP implementer for long enough and you'll find complaints about how Mastodon doesn't follow the spec), simply because of Mastodons excessive popularity.
zoogeny 3y ago
I'm not an expert on ActivityPub nor academic publishing. What I'm trying to say is that what is built on top of the protocol has more to do with its suitability for a particular purpose than the foundational technology. The first quote is an observation that It is unlikely that Mastodon will be a suitable drop-in-replacement for the current process. The second quote is an observation that perhaps there is a way to use the technological pieces that are currently available for federation and apply them to the organization of scientific/academic societies. But that whole system is going to be significantly more than the federated aspect. Part of this is I remember Google Wave and the unreasonable hype around it. It would replace email, newsgroups, instant messaging … basically all forms of text based communication! Some of the hype for these more contemporary niche technologies is even wilder, this article for example claiming it should force a complete re-think of academic publishing.

noirscape 3y ago

To be more precise, ActivityPub is pretty much just a JSON/HTTP version of email. There's a few fancy things to have stuff like user profiles and stuff to make it "look nice", but if you look "under the hood", the data structures you find ends up and the methods of communication end up looking more or less like a modernized version of email. --- Mastodon is a sub-implementation of AP that follows some of the components and then weirdly tweaks some of them in weird ways (because Mastodons internal representation doesn't actually match up with these data structures and Gargron wanted some features like Content Warnings without bothering with the entire process of protocol modifications, so he ended up cannibalizing a different field for it, resulting in an implementation that... mostly results in people yelling at each other for not using CWs "appropriately"), resulting in strange compatibility issues when writing your own AP instance. Most AP instances typically end up trying to then match Mastodons weird implementation (basically look at the social feed of any AP implementer for long enough and you'll find complaints about how Mastodon doesn't follow the spec), simply because of Mastodons excessive popularity.

zoogeny 3y ago

I'm not an expert on ActivityPub nor academic publishing. What I'm trying to say is that what is built on top of the protocol has more to do with its suitability for a particular purpose than the foundational technology. The first quote is an observation that It is unlikely that Mastodon will be a suitable drop-in-replacement for the current process. The second quote is an observation that perhaps there is a way to use the technological pieces that are currently available for federation and apply them to the organization of scientific/academic societies. But that whole system is going to be significantly more than the federated aspect. Part of this is I remember Google Wave and the unreasonable hype around it. It would replace email, newsgroups, instant messaging … basically all forms of text based communication! Some of the hype for these more contemporary niche technologies is even wilder, this article for example claiming it should force a complete re-think of academic publishing.