Property:Contains text
- Text
- schema:text (schema | Schema.org, V 14.0)
D
That said, the more abstract idea of defining a data model plus execution semantics that any programming language can plug into looks very promising. That aspect of WikiLambda was in fact one of my inspirations for developing [[Digital Scientific Notations]]. +
I'll try to turn this thread into [[Project Ideas#Federated knowledge synthesis]]: identify protocols, data models, tools, practices, etc. that can support the process of synthesizing and formalizing scientific knowledge, then build on these ingredients. One dimension is going from narratives via discourse graphs to knowledge graphs. Another dimension is going from conceptual ideas to formal systems. +
Thanks <@322545403876868096> ! Added to https://synthesis-infrastructures.wiki/Discourse_Modeling. I guess I could have used [[Wikibot]] for that, but it was easier to do it by hand than figuring out the intricacies of Wikibot. +
Just added a "proposal" tag to our discussion. In scientific discourse, that would be a category used in opinion papers etc. Is this already part of the [[Discourse Graph]] repertory? +
Note to <@305044217393053697> about [[Wikibot]]: it doesn't pick up edits on messages that it has already added to the WIki. The version in the Wiki ends up being obsolete. Could be important when someone edits to add "not", for example. Discord users are used to having this possibility. +
Yes, that's a prominent use case for [[Glamorous Toolkit]]. +
Note that [[Glamorous Toolkit]] is not (yet) a development environment for Python. What is described here is "data science" on a Python codebase. You analyze the code, but you cannot change it. For Pharo Smalltalk, there is excellent code refactoring support in addition to analysis features. +
those brackets cue the [[WikiBot]] to link the message to the wiki page containing the mentioned terms +
From the Gutenberg city of Mainz, the [[CLAIM]] home of modern intellectual synthesis and dissemination - thank you for your participation! I've enjoyed our discussions and look forward to their continuation! +
[[claim]] claims and questions dominate in natural conversation; the imbalance of sources & evidence is quite stark. This aligns with my mental model of *conversational charity*, where we assume our interlocutors *could* ground their statements in evidence if pressed, but skip this step in the interest of time. +
I've also recently been using logseq. I like how it just writes to markdown. I've been wanting to parse that markdown, look for we--known #hashtags and [[wikitags]], and build an rdf dataset. It looks like SBML is kinda like XML, so maybe something similar is possible there. Have you done anything more with logseq since this post in November? +
Konrad, are you familiar with [[Chemical Markup Language]] (CML)? I stumbled across it on Twitter a few weeks ago via discussions about open publishing, and was surprised at the longevity of the project. I don’t love XML, but it seems to have gained some traction in its day, though I am not sure how active it is these days. https://en.m.wikipedia.org/wiki/Chemical_Markup_Language +
ah, that is both informative and sad to hear. i think ahead of its time is a reasonable diagnosis.
[[ScholOnto]] I think was also ahead of its time: had a working prototype integration into a Word processor for directly authoring discourse-graph like things while drafting a manuscript (described here: https://onlinelibrary.wiley.com/doi/abs/10.1002/int.20188) +
the idea [[DiscourseGraphs]] is rooted in a bunch of models like [[SEPIO]] (h/t <@602622661125996545>) and [[ScholOnto]] that have been around for various amounts of time, though not yet with (to my knowledge) serious widespread adoption. +
we think the problem now is user-friendly tools and workfows that can create discourse graph structures, and have seen some exciting progress across a bunch of new user-facing "personal wikis". but bridging from personal to communal is still a challenge, partially bc of tooling.
this is why i'm excited about the [[Discourse Modeling]] idea, which i sort of understand as a way to try to instantiate something like [[Discourse Graphs]] into a wiki (bc wikis have a lot more in-built affordances for collaboration, such as edit histories, talk pages, etc.), which may hopefully lead to a lower barrier to entry for collaborative discourse graphing.
a high hope is that we can develop a process that is easy enough to understand and implement that can then be applied to discourse graphing the IPCC or similarly large body of research on a focused, contentious, interdisciplinary topic.
other examples include:
- effects of masks on community transmission (can't do decisive RCTs, need to synthesize)
- effects of social media on political (dys)function: (existing crowdsourced lit review here, in traditional narrative form: https://docs.google.com/document/d/1vVAtMCQnz8WVxtSNQev_e1cGmY9rnY96ecYuAj6C548/edit#) +
in human-computer interaction we have a similar problem of trying to think about and synthesize across many genres of contributions/research. one map (adapted for information studies) breaks things out into "empirical" contributions (these most often follow the standard intro/methods/results/discussion format), "conceptual" contributions (which are often more amorphous theory papers), and "constructive" contributions (making a new system/method)
from here: HCI Research as Problem-Solving +
I think of all of these tools as "personal hypertext notebooks" - basically taking what is possible in wikis (organizing by means of linking, hypertext) and lowering the barrier to entry (no need to spin up a server, can just download an app and go).
The common thread across these notebooks then is allowing for organizing and exploring by means of bidirectional hyperlinks between "notes":
- In [[Obsidian]] each linkable note is a markdown file and can be as short or long as you like
- in [[Logseq]]/[[Roam]] and other outliner-style notebooks, you can link "pages", and also individual bullets in the outlines on each page.
In this way, the core functionality of these tools is similar to a wiki, but they do leave out a lot of the collaborative functionality that makes wikis work well (granular versioning and edit histories, talk pages, etc.). So for folks like <@305044217393053697> who are comfortable with wikis already, they add marginal value IMO.
Their technical predecessors in the "personal (vs. collaborative) wiki" space include [[TiddlyWiki]] and [[emacs org-mode]] (and inherit their technical extensibility: many users create their own extensions of the notebooks' functionality. an example is the [[Roam Discourse Graph extension]] that <@824740026575355906> is using).
These tools also tend to trace their idea lineage back to vannevar bush's [[Memex]] and ted nelson's [[Xanadu]]. +
These tools are still not entirely mainstream compared to tools like [[Notion]], which is related to your experience trying to learn more about the tools - so they tend to have a steep learning curve!
IMO the best way to get a feel for what they are is to see some examples/videos.
I like this video for an overview of [[Logseq]]: https://www.youtube.com/watch?v=ZtRozP8hfEY&t=6s
I describe [[Roam]] and the [[Roam Discourse Graph extension]] in this portion of a talk I recently gave: https://youtu.be/jH-QF7rVSeo?t=1417 +
i agree it's not universal! my feeling is that [[Claim]]: a statement (claim or evidence) might be the more universal element:
- empirical work also consists of statements about the world (this is less controversial)
- design/technological innovation rests in part on claims about a) what is needed in the world, what is hard to do, constraints, and b) what is needed to succeed: examples here: https://deepscienceventures.com/content/the-outcomes-graph-2 (h/t <@559775193242009610>)
- theories often consist of systems of core claims (e.g., in models like what <@824740026575355906> and <@734802666441408532> are working with, where we can think of the claims as subgraphs of the overall knowledge graph)
see, e.g., [[Evidence]] from this review of models of scientific knowledge https://publish.obsidian.md/joelchan-notes/discourse-graph/evidence/EVD+-+Four+positivist+epistemological+models+from+philosophy+of+science%2C+including+Popper%2C+emphasiz...+statements+as+a+core+component+of+scientific+knowledge+-+%40harsDesigningScientificKnowledge2001
and [[Evidence]] convergence/contrasts across users of the [[Roam Discourse Graph extension]] in terms of building blocks: common thread across all was Evidence +
let's dump into a page! [[Discourse graphs within survey reading course]] +