Volunteers Needed: Abstracts & Tags for old Advance Readings

An important part of how/why a design workshop can generate so many collaborative works in just a few F2F days is having people to submit “Topics & Advance Readings” and reading those of by others before the event begins. This allows ideas that have critical mass of interest to percolate and rise to the top.

One of the changes at #RWOT8 that I’m most pleased with asking everyone to add abstracts and tags to their “Topic & Advance Reading” submissions into the /topics-and-advance-readings/README.md. Given that we had 85 topics submitted for #RWOT8, this helped everyone before the event prioritize their reading, which has become particularly challenging.

For future #RWOT events, at minimum we need a PR template in GitHub that makes sure that people remember to add the abstract and tags to the README.md. But we may want to have a more automated process/script for doing topic submissions.

We also have a large number of old topic papers, some of which are still quite good. I’d love to find a volunteer to go through the old papers and add abstracts and tags to them (I think there are over 400!).

We also need to find some way to make these discoverable on our website by tag, which may be another volunteer.

– Christopher Allen

I’ve been in touch with Shannon and Eric after the conference and it seems they are making slow but steady progress on this project. Contact them directly!

1 Like

Slow is a matter of taste…

Yes - very much so. I am moving one existing RWoT database to Airtable this week. The next step is to create a ‘form’ for tagging readings. The final step is to recreate the ‘cluster maps’ we already have , but using the crowdsourced tagging.

From airtable we’ll pushinto a repo-based snapshot. I have a graphql wrapper around a sqlite file (that would live in the RWoT repos) if that is of interest?

Anyone out there w/ Airtable experience? Anyone want to jump in to do ‘semantic tagging’?

Plz contact eric@korsimoro.com if this of interest?

sheesh Juan… slow… what’s your hurry!!! :wink: :wink: :wink:



1 Like

Sorry, that was very imprecise and rude of me! You guys rock for working on this, I will be in touch directly.

Absolutely no worries - but we are making real progress. I am also working a similar thread w/ infominer - the goal is simply this: provide a “repo-distributed, active data model” - likely a sqlite file - and look at how to integrate that with other ambient systems.

I’m looking at Airtable this week, specifically because of two features a) it allows data input in a forms-like fashion with a few bells and whistles missing from google forms, and b) it has integration w/ github pages so you can use query data rendering a page. It can be shadowed in a sqlite file, in RWoT repos as well - meaning that RWoT can publish the “official state of the db” from time to time and does not bind itself to Airtable IAM or accounting - Airtable (like google) just becomes a front-end option.

In the context of RWoT I think that git-repos give a pragmatic level of decentralization and distribution, and are a reasonable option for authoritative records - systems like Airtable and Google Sheets/Forms, on the other hand - provide a lot of out-of-the-box value, but must be treated with caution - the risk of that slippery slope is that we’ll end up with all our collaboration on the RWoT Facebook page :wink:

1 Like

I hear that we could target potential RWOT attendees using FB! We should look into it! We could very carefully pre-curate the topics using high-precision segmentation! :slight_smile:
(just kidding!)

1 Like

woo hoo - nice progress today.

we’re working w/ a test repo here : https://www.korsimoro.com/pubtest/

it pulls in data from a bunch of sources, normalizes it as Sqlite when possible, and does other transformations (such as TOML to JSON) as required.

one section of the site actually provides full text search of the topics papers from multiple events.

it’s a work in progress, but… it is also a deployed PoC (or pool of PoC’s!)

@korsimoro This looks interesting. We would have fork of this with our own sources? What is the format for the input of the citations?

– Christopher Allen

The goal would be move it right into the rwot repo - it’s actually just tooling around github pages. Right now we’re using “pubtest” as a proving ground for ideas - getting everyone on the same page technically and exploring some different integrations for user input. Ultimately, any changes collected from input source would generate a pull request for someone in an editorial capacity to merge (and thus trigger the gh-pages rebuild).

Right now I’m linking to the RWoT repositories using git-submodules, so we’re actually drawing data out of those repositories - but the whole thing should wrap up into a pair of directories and a continuous-integration config (we’re using circleci right now - so it is .circleci/config.yml). Adding it to a repository should involve copying those directories and twiddling the github pages settings.

Of course, setting up a “library” repo that uses the submodule strategy is also an option.

Right now, we’re looking at

  • the minimalmistakes package for jekyll github pages integration
  • using mkdocs for certain subsections - it is good at structuring otherwise free-form markdown data, complete with embedded text search.
  • generating profile pages out of sqlite databases (e.g. people/organizations links are that effort)
  • setting up a keyword search ability based on active tagging by authors
  • syncing that sqlite data w/ external sources which can provide input, specifically
    … editable google sheets
    … AirTable embedded forms (which might be able to sit inside the gh-pages)

But, ultimately, it would go in any repo of interest and be integrated with the gh-pages rendering of that repository, drawing on data sources within that repository. Thanks to sqlite gives us the ability to run full-on SQL queries against repo-resident files - and use that to generate static pages - so we get much of the benefit of an active data system, but without incurring the overhead of maintaining a separate data server, API, etc.

@JuanSC and I have also been working on an iteration of this project, which is now live at

didecentral/rwot-dir@github (1 link limit for new forum users)

That site and it’s repository are a good way to contribute to this effort, and being a live PoC \ Demo is nice, because you will be able to see results from your work in a relatively “instant” fashion, while the big vision is still being built out.

“abstracts” could be stored in the “excerpt” frontmatter variable, making it quite portable (or easily have their own abstract variable). Storing this type of information as structured data is extensible, and aims to flow along w @korsimoro’s vision, until our cumulative efforts combine into something super awesome.

For example, we are already using frontmatter variables to link back to the source of every page back to its home repository, which is displayed at the bottom of the page, so that the website can draw visitors back to the source for each page.

Storing all the metadata as variables (title description abstract images links to the source at different stages from draft to final), allows us to generate different pages to display that information in a variety of fashions, using a simple scripting language.

For now, the first RWoT is mostly “complete” for demonstration purposes, the rest just need a little tlc to make it all work together nicely, and is easy to collaborate if you have the time or inclination to help.

Please check out our blog at didecentral dot com where there are plenty of guides and how-tos that can bring you up to speed with GitHub, GitHub Pages, Jekyll, Minimal Mistakes, and the operation of the majority of sites in DIDecentrals network of micro-publications