Synthesizing Social Media/First Breakout Session: Difference between revisions

From Synthesis Infrastructures
No edit summary
 
(18 intermediate revisions by 3 users not shown)
Line 1: Line 1:
Back up to [[Part of::Synthesizing Social Media]]
== Possible Topics ==
== Possible Topics ==


Line 4: Line 6:
* Using [[Braid]] - https://braid.org/point
* Using [[Braid]] - https://braid.org/point
* [[A System for Interleaving Discussion and Summarization in Online Collaboration]]
* [[A System for Interleaving Discussion and Summarization in Online Collaboration]]
== Patching on top of masto ==
=== First pass ===
* Using the profile tags to be able to assign prefixes to a digital object
** So say i have <code>paper:paper url</code> in my profile
* Then being able to refer to that object using @jonny:paper in a post
* And then posts beneath that use that as the subject when resolving [ [Semantic::wikilinks] ]
=== Second pass ===
* Metadata tag in masto profile that links to a wiki page
** <code>wiki:https://synthesis-infrastuctures.wiki/Jonny_Saunders</code>
* Page has a series of [[Property:Tag]] annotatations that link a short tag to a further page or URL. The linked page could have additional metadata like DOIs, authors, etc.
** <nowiki>[[Tag::shortname::Longer Page Name]]</nowiki>
* Posting a thread with the first post using a syntax like <code>@jonny:wiki:shortname</code> then treats posts underneath that as having that as the subject, and a bot then scoops those up and puts them in the page enabling the use of semantic wikilinks, etc.
* A bot then is monitoring opt-in accounts for these kinds of threads and exporting them to the relevant wiki page as subpages containing the threads.
** Like the [[WikiBot]], additional [[Wikilinks]] within the thread also embed a link to the thread (probably want to have the thread be its own page then, and have some syntax for declaring a title)
* The bot could also have interoperable export formats out to [[Braid]], etc.
==== TODO during the workshop ====
* Deploy a glitch-soc instance in the synthesis-infrastructures domain to test
* Write the above bot and make relevant wiki components :)
==== Further integration into client ====
* UX elements to be able to select and resolve tags for a given author within the instance that has enabled this syntax.
* An overlay could search for linked wikis and work backwards - so given a DOI, find the page with the <nowiki>[[Has DOI::...]]</nowiki> page and then overlay the threads underneath it.
== Use cases ==
* Shorthand identifiers using profile metadata (as above)
* Using semantic links in streams headed by shorthand identifiers
* [[Attention]] data on reading: taking eg. zotero and converting that to a post stream or something
=== Mastodon/Twitter Paper announcement use case ===
* One of the common norms that have developed on academic social media (Twitter/Mastodon) is announcement/summary threads of one’s paper. e.g.: https://hci.social/web/@Mor/109309396530932422 (others of interest: job market announcements, taking PhD announcements)
* Imagine if you could create for yourself a unique, short, readable identifier for a paper, and add that in your post thread about your paper announcement.
** Possible syntax:
*** <code>@mor:crisis-informatics</code> user namespace
**** downsides: there are multiple authors, so an author may feel left out if another author's id gets used more often or authors could coalesce around one author (ie, 1st author)
*** <code><nowiki>#</nowiki>cscw22:crisis-informatics</code> conference namespace
**** downsides: possibility for name conflicts, you could use the conference ID but it's not very readable
* Things you could do as a result:
** Presentation: put all your different paper threads on your academic webpage and associated with each paper, for example
** Synthesis: collect all the paper announcements for a conference/topic and organize/synthesize it using other tools
** Discussion: people could reference your identifier in their conversations, and a bot could post the paper announcement thread as a reply (https://hci.social/web/@lewischuang/109331737391363720)
** Discussion: you could collect all discussion around your paper in a separate place, or a bot could post pingbacks on the original paper announcement thread
=== Sharing attention data on ActivityPub through Zotero (?) extension ===
* Attention data (~digital pheromone trails) is an important step in the knowledge synthesis pipeline- what am I looking at and my assessment of it
* Sharing attention data using linked data over ActivityPub-
** Pros- catching Mastodon wave, good expressivity but
** Cons- writing it manually creates friction and so reduces usage by researchers
* An extension can create automatic linked data posts based on existing user activity in Zotero
** Adding a paper to a reading list
** Tagging
** Annotating inline, connecting pieces to other papers
** Reviewing/assessing/endorsing content
* Question - what are interesting downstream applications for attention data?
** Content recommendation
** Layering paper viewers with endorsements/reviews, popular tags & annotations
*References
**[https://dl.acm.org/doi/10.1145/2818048.2819949 Opportunities and Challenges Around a Tool for Social and Public Web Activity Tracking]
**[https://dl.acm.org/doi/abs/10.1145/3511095.3536361 From Users to (Sense)Makers: On the Pivotal Role of Stigmergic Social Annotation in the Quest for Collective Sensemaking]


[[Category:Conversation]]
[[Category:Conversation]]

Latest revision as of 17:35, 12 November 2022

Back up to Synthesizing Social Media

Possible Topics

Patching on top of masto

First pass

  • Using the profile tags to be able to assign prefixes to a digital object
    • So say i have paper:paper url in my profile
  • Then being able to refer to that object using @jonny:paper in a post
  • And then posts beneath that use that as the subject when resolving [ [Semantic::wikilinks] ]

Second pass

  • Metadata tag in masto profile that links to a wiki page
  • Page has a series of Property:Tag annotatations that link a short tag to a further page or URL. The linked page could have additional metadata like DOIs, authors, etc.
    • [[Tag::shortname::Longer Page Name]]
  • Posting a thread with the first post using a syntax like @jonny:wiki:shortname then treats posts underneath that as having that as the subject, and a bot then scoops those up and puts them in the page enabling the use of semantic wikilinks, etc.
  • A bot then is monitoring opt-in accounts for these kinds of threads and exporting them to the relevant wiki page as subpages containing the threads.
    • Like the WikiBot, additional Wikilinks within the thread also embed a link to the thread (probably want to have the thread be its own page then, and have some syntax for declaring a title)
  • The bot could also have interoperable export formats out to Braid, etc.

TODO during the workshop

  • Deploy a glitch-soc instance in the synthesis-infrastructures domain to test
  • Write the above bot and make relevant wiki components :)

Further integration into client

  • UX elements to be able to select and resolve tags for a given author within the instance that has enabled this syntax.
  • An overlay could search for linked wikis and work backwards - so given a DOI, find the page with the [[Has DOI::...]] page and then overlay the threads underneath it.


Use cases

  • Shorthand identifiers using profile metadata (as above)
  • Using semantic links in streams headed by shorthand identifiers
  • Attention data on reading: taking eg. zotero and converting that to a post stream or something

Mastodon/Twitter Paper announcement use case

  • One of the common norms that have developed on academic social media (Twitter/Mastodon) is announcement/summary threads of one’s paper. e.g.: https://hci.social/web/@Mor/109309396530932422 (others of interest: job market announcements, taking PhD announcements)
  • Imagine if you could create for yourself a unique, short, readable identifier for a paper, and add that in your post thread about your paper announcement.
    • Possible syntax:
      • @mor:crisis-informatics user namespace
        • downsides: there are multiple authors, so an author may feel left out if another author's id gets used more often or authors could coalesce around one author (ie, 1st author)
      • #cscw22:crisis-informatics conference namespace
        • downsides: possibility for name conflicts, you could use the conference ID but it's not very readable
  • Things you could do as a result:
    • Presentation: put all your different paper threads on your academic webpage and associated with each paper, for example
    • Synthesis: collect all the paper announcements for a conference/topic and organize/synthesize it using other tools
    • Discussion: people could reference your identifier in their conversations, and a bot could post the paper announcement thread as a reply (https://hci.social/web/@lewischuang/109331737391363720)
    • Discussion: you could collect all discussion around your paper in a separate place, or a bot could post pingbacks on the original paper announcement thread

Sharing attention data on ActivityPub through Zotero (?) extension

Retrieved from "https://synthesis.jon-e.net/index.php?title=Synthesizing_Social_Media/First_Breakout_Session&oldid=1186"