Note: OGM Forum is no longer active. This is a static snapshot as of February 2022. You can browse the site to see posts, but the functional features of the site will not work. You can search or download a zip archive of the files from the site at github/OpenGlobalMind/

Invitation: Design + build a collection and feed of project invitations/requests

This prototype project is supposed to collect project proposals from many sources, and provide them as a feed, so people can “subscribe” and review the offers/invitations and get in touch for working on the projects they’re interested in.

This has a data gathering/curation component, also needs some design (ideally some federation/categorization/tagging too eventually, what kind of service is it - app, mailing list, other?), and then the software development to set it up and get it running.

In terms of content, surely the result should be universal and work for many types of data, but for our OGM purposes, the kind of service we would like to set up and run in this experiment is one to cross-promote small-scale, prototyping, experimental projects on a scale of 1 day to ~3 weeks (which then can branch off into a larger effort), and preferably also pretty inclusive, meaning that participants in different roles would be helpful, not just all-tech or all-conceptualizing or all-content/-curation. Emphasis also on an open invitation to anybody with no prior knowledge/experience, to peer-learn skills and try things in the process.


So there are a few “projects” with visibility as far as I can see. Cicolabs Education effort (one for kids, one for adults), the structuring of OGM work, possibly a food one emerging. And then yours which is on the harvest side of things. Are things called out via tags? A reporting place?


This is great! You’re already starting a list, and I therefore declare this project-invitation/exploration started :slight_smile:

Tags and a reporting place, and other ideas as well if proposed, feasible, useful. I don’t have some potentially super-cool tech to power it ready just yet, that’s some separate, somewhat slower and more difficult basic research which needs to go on in parallel, but for the time being, other makeshift interim alternatives might do just as well.

Recently I started some early code that’s generating a “Progressive Web App”, very simple, quick, cheap way for app programming which is a Website but also is supposed to work on/for/with mobile. That could be pre-loaded with the list of projects, or update them from a feed, and maybe suggest a “project of the day” or notify about new ones. Maybe us investigating projects and review/discuss them with the people who do them, and then offer a summary/intro to a wider audience. Little bit like our own producthunt, just curated and picked, from our context and our purposes :slight_smile:

So the next steps for the next week could be:

  • Gather/compile the easy, “near” projects and proposals and describe them a little, so there’s some examples/data to look at and play with
  • Turn the entries into a somewhat presentable form
  • Publish this early prototype result at a place so it becomes easy to invite others to have a look at it

Very rough/broad, please suggest anything more or different if you want (also in terms of degree of detail, if the steps are too vague in their description). If this is more or less fine, feel free to implement some or all of these yourself, and I’ll try to do the remaining ones or all of them, in latter case the result then for you to review/reflect :slight_smile:

One note regarding the content, I won’t put any of my own stuff on there except if it’s specifically for OGM or related tooling for this activity. I believe that neutrality should be maintained and potential conflict of interest avoided.

“Richard Feynman was fond of giving the following advice on how to be a genius. You have to keep a dozen of your favorite problems constantly present in your mind, although by and large they will lay in a dormant state. Every time you hear or read a new trick or a new result, test it against each of your twelve problems to see whether it helps. Every once in a while there will be a hit, and people will say, 'How did he do it? He must be a genius!”
― Gian-Carlo Rota, Indiscrete Thoughts


I won’t be on the call this week - work calls - but maybe you or someone can raise visibility of this and invite submissions of project-lets. If we had the minimum specs of a description such as:

  • Name of Project/Problem/Thing
  • Who is playing
  • Purpose: The purpose of this thing is to (do what/make/enable/ect) so that (who) can (do what)… and we will know we are making progress on this thing/project/problem when…
  • Time Horizon
  • How to join

Thanks a lot for this outline! If I have to come up with one myself and also build tooling for it, I’m always uncomfortable worrying that I just look at it from my own perspective for myself, but now knowing that this is the first version of scope you would be interested in, let’s go for it and figure out in the process how things work out.

Content-wise, I can try to mine project proposals/invitations/requests from the recordings. Didn’t do that too much as a practice yet, and it’s not that much fun to re-watch, but going forward, sure, why not have an ear for it? In parallel, entering/capturing such data could be made easy with some tooling, as well as the presentation and subsequent feed/notification. Hopefully make some progress in that regard over the coming weekend.

In terms of content, let me very quickly just drop a few links/projects mined from GameB:

Like this, such proposals could be collected from various groups and individuals, for then be regularly reviewed if they received some attention/traction, if they got started/abandoned, what they each want/need/request and offer, what they learned, grouping similar activities, cross-community promotion for synergies, etc. etc. To the degree possible/appreciated, some substantial portion in the long-term could become to establish contact to project owners/maintainers/coordinators, and especially so in regular intervals or around expiration/end-of-time-interval, or similar.

1 Like

Couldn’t resist to apply some immediate action: from some other context, I happen to have an early, simple form generator around recently (a little bit like our own Google Forms potentially to capture/apply semantics on data entry - not an editor yet, not saving to a repository yet). So I put your structure into it as a template (literally so, half-joking, but likely already works just fine!) that produced a corresponding form (this link is only temporary, may disappear at any time - also, isn’t saving online, therefore after clicking “save”, continue with a “Save as” in the browser to keep a local copy which then can be submitted elsewhere, but doesn’t need to be). Comes with no style/layout of itself, so the appearance is everybody’s own preference as configured in their browsers :slight_smile:

The result files might easily end up in an app, repository, digest newsletter, other.

Two practices come to mind. DURING the calls, anyone who catches a WHIFF of a project types PROJECT in the chat. It serves as a simple timestamped flag. If they catch it, include NAME OF PERSON who mentions it/Any of the fields above. Then we can at least have a weekly list of potential/real projects and ideally, a name of a person who might be willing to fill in the form/invite in others.


I and for that matter others who aren’t attending the call don’t have access to the chat, and I’m not aware if it’s shared/published somewhere. Therefore, me listening to the recording afterwards, and for some other people maybe way later catching up, or of course independently/asynchronously can extract/mine such project proposals made during the calls. But sure, if the chat log is available somewhere, it’s well imaginable to automatically scan it for such tags/indicators together with the timestamp, for aiding with the discovery of project proposals/invitations quite a lot.

If video recording and corresponding chat transcript would be filed/referenced in a proper index, all kinds of tags could be automatically extracted if people like to put these in there in real-time during the call.

Chat for this week is here:

I see, so that’s not only the copy of the chat, but also the provisional video index (apart from the YouTube playlist, both not being useful, clean data sources). :slight_smile:

The downside is that this Website is cluttered with all kinds of stuff that belongs to the user interface and visual rendering, and I think TheBrain doesn’t really want to offer the graph nodes nor the notes as a clean data source. Therefore, why manually paste/publish the chat in there for then having to manually copy it off from there and clean it up, when the chat started out as simple plain text (I assume) to begin with?

There’s the option of trying to somehow discover/extract the note data endpoint like this or this one, and from these, HTML can be read (which happens to be not well-formed, unfortunately), which potentially doesn’t matter that much of semantics (except per convention maybe) so it could be thrown out in order to arrive at plain text again for parsing. Maybe the table(s) are useful to separate a writer from the chat message, but there’s other clutter and messiness in this source, so wonder if it’s worth bothering. Any thoughts/suggestions/solutions?

zoom generates a text file too, of the chat. I’m in over my head on the tech stuff.

From a practice standpoint, it is the noteTAKING practices that are the starting point. Calling out ACTIONs, noting shifts in the conversation with things like +++++ to separate topics, noting resources or questions. So we need a few simple conventions to make the raw data potentially more useful.

1 Like

Is Zoom really doing much there? It’s simply a regular chat like any other, and people typing text into it, isn’t it? OK, there might be some additional time synchronization, and that could be as simple as absolute clock date + time stamps which sync the video recording with the text messages. For the things I found out about how TheBrain goes about storing the notes, maybe some software developers or data/information architects have some suggestions about how to go about it, to get the chat messages out of it again. Also seems that the Markdown table or whatever it is is broken (didn’t successfully converted to a HTML table in its entirety) and the table headers are messages too, not headers/captions.

Are there particular troubles with tech? In general, there’s tech everywhere, be it using a computer, or when a car is involved, or a pencil or language. Probably having a brief look at it what it is, how it works, and not everybody needs to understand all of the details of how, for example, electricity works, for navigating the complexity the world inherently is by its nature.

The poor technical quality that’s dominant everywhere makes it unnecessarily difficult, costly or specific to manage data well and do useful things with it, so it’s already a great improvement to do less of the unnecessary degrading of data quality.

If you or others want to start experimenting with such textual indicators/commands, either in the live chat during a OGM call or otherwise in plain text, if I can somehow get to the text (either by manual copy&paste, but ideally from a decent source that’s not cluttered where the chat log got published somewhere), I think it would be best to simply start inventing such special terms. There are a few design rules, which help a great deal for their automatic recognition and processing and potentially making use of some supporting structure around it, but not too much to worry about in advance, it can be as simple as a regular text search to get started :slight_smile:

I wonder if the question here is about real-time processing/mining/curation of a chat, because if something else is going on like audio/video at the same time, the writing is always behind the live-feed of voice/conversation, and people tend to be busy with other things than doing curation work. Surely there can be supporting roles for facilitation, but then the linearity of synchronous real-time is rarely a useful, constructive mode for knowledge work anyway, as there’s fight over airtime, topics branching into all kinds of directions, hard to pick up and expand earlier themes, no way to rewind/repeat/pause, etc. And a whole bunch of people aren’t attending all these calls anyway, including me, but I and others would be perfectly capable of looking at the chat log later, and sometimes what’s in there isn’t even that useful (and instead the so-called “second channel” to correct, expand or branch what the linear medium didn’t support/cover), but that’s usually where links can be provided/shared in a textual, clickable way.

This weekend, I managed to duct-tape together a simple form that can take project descriptions, and the semantic data collected this way can automatically go into a first, early “Progressive Web App” prototype for presentation.

Such PWAs are basically just a regular Website, but can be added to the mobile homescreen like other apps. Maybe the version of this prototype doesn’t work completely yet. We could even have a little OGM app store of ours :slight_smile: For this project list, there could be updates/notifications for new submissions, or a random project description presented per day, or a project of the day following a certain order. And lots of other things to add, like reviews, or of course improving layout, and so on. Also have to clean up and publish the software which generates the app, but that’s just for the prototype, could be a fully automated online service as well.

Here’s a technical repository on GitHub to “server-less” collect the raw data of our project list/feed.

If people create new project entries, I would assume that I or we as maintainers of that repository/feed add them to the repository, so it can serve as the single master source for such data, and the place to insert + subsequently replicate corrections and updates, for the time there isn’t a better way to allow this kind of forking, revision control, etc.

Might push to and/or pull from Murmurations eventually, but they might not care that much, as our project list isn’t co-ops.

Here’s a workflow which takes the project description files in the repository and automatically generates an early Progressive Web App prototype to navigate through them. This is something software developers might occasionally do, and also work on to improve the result.

Update: the PWA doesn’t work properly on Google Chrome (that’s Android) in terms of A2HS (app to home screen) as Google continuously is either too incompetent or deliberately sabotaging to get the most simple technical basics right. I want to produce a single, standalone, self-containing Website file which can also be passed around and shared (offline, sideloading, distribution) without relying on a particular server (too bad that PWA’s rely already too much on a domain), and Google is apparently unable to support an inline Webmanifest, and also unable or unwilling to accept a serviceWorker which isn’t a file on some server (bad enough that they insist on a service worker at all, as an “indication” despite/besides the Webmanifest that a Webpage file is an PWA, even if the serviceWorker.js is empty or not relevant or of no use for the PWA anyway, because all data is already inline/embedded/offline-enabled in the Website/PWA with no need to interact with any server/domain). So OK, simply use it as an ordinary Website or local file/app on your computer or mobile device. If Google can’t or doesn’t want to get its very own stuff right as late as 2020, why care too much about their segment of mobile. Much more in favor of the open Web, not Google’s crap.

I’m sorry I’ve been MIA. My schedule is crazy and I can’t make the Thursday calls for the next 3-4 weeks so I’m out of commission.

No hurry, we don’t have any kind of deadline or similar, only pretending that we do :slight_smile: Also, attendance of the weekly OGM calls is not much relevant for the practical project of this thread.

i like this idea a lot!! would it be more helpful to use #hashtags ? anyhow — direct relevance to #hashbins as well