How do we organise, archive, and communicate our collective knowledge?

Hello everyone,

Per the title of this post, I’ve been thinking about - given the incredible collective knowledge we have in GOSH - how we organise, archive, and communicate that knowledge?

Lots of discussions happen in this forum, while others take place in Whatsapp groups; Things happen in the GOSH GitLab repositories, and of course there are the valuable resources on the GOSH website such as the policy briefs and the super-comprehensive events framework. I also know that much work is being done via numerous Google Docs documents that have been shared in this forum.

My concern is that, other that what’s published on the GOSH website, much of the knowledge we’ve produced will not be retained very well and eventually be lost to time.

I don’t have a good solution in mind, but would love for some way to organise our collective knowledge so that it’s not only collectively remembered, but (relatively) easy to pass on as well.

Anyone has thoughts or experience with this? Or is anyone interested in perhaps forming a working group on this problem???

3 Likes

I’m biased here because I use it all the time, but we have a general Open Hardware library in Zotero, we could also have one for GOSH.

I like it because all your links are in one place, you can easily cite, the magic tool for adding resources makes it super easy. I use it not only for docs, also for repos, so it could work with the gitlab ones.

5 Likes

Hi hpy,

This discussion reminds me the warnings from one of the Internet’s pioneers: Vint Cerf
https://en.wikipedia.org/wiki/Vint_Cerf

And his warnings about the risks of a digital dark age:
https://en.wikipedia.org/wiki/Digital_dark_age

Google’s Vint Cerf warns of ‘digital Dark Age’:
https://www.bbc.com/news/science-environment-31450389

Best Regards,
Markos

1 Like

I worry too.
I’m worried that we get to dependent on proprietary platforms, and that to produce open science we are going to use and navigate on proprietary networks and platforms (more about it here ). That’s bad because we will be subjected to their “laws” and organizations.
Yes, is easier, faster, user-friendly, but not free as in free software.
(and sometimes is the only option . . . :confused: )
However, to use open/free platforms we need engagement on documenting, patience on using and commitment. And that’s not easy, demands a lot of energy. We do have people to volunteer on infrastructure maintenance, but we still need general engagement.
I’m not saying that’s easy, but i thought i should point out.

Juli suggestion is also good. A wiki is also interesting (less user-friendly and don’t look modern). Git is not the right tool, and unnecessary hard, but also an option.
,

3 Likes

Building on top of what @marinappdf says, it would be nice to pair this effort with Self-hosted apps for GOSH - #5 by Diapason

Personally, I’m a big fan of Low-tech Magazine’s solar powered website

2 Likes

Ni! Hi @hpy . I feel that for this to be a useful one you must be more explicit about what “knowledge” you are talking about. You say only the website is good, but I don’t see how “knowledge” on Gitlab and on this forum - and perhaps even on Google docs - can so generically be said to be, in any meaningful sense, “lost to time”. Cheers,

1 Like

This is tangentially related and I’m not suggesting that it addresses all the issues raised here, but there is a FOSS solution for creating copies of common knowledge stores and making them available offline. I have only spent a little time experimenting with Kiwix, but it worked fine in my trials.

This eliminates reliance on Internet connectivity and server infrastructure, and they have instructions for a server appliance that can be set up to serve a school, research group, event, etc on a wireless LAN. Maybe a config could be created to generate an offline archive of open science materials and related sources?

1 Like

Hi all, thanks for the responses! :pray:

@jarancio: Thanks! I completely forgot about shared Zotero libraries, even though I use Zotero for myself all the time! :sweat_smile: I suppose we can create a shared library there as an “index” to keep track of the disparate pieces of knowledge that builds up all over the place. IIRC Zotero has types for forum threads, multimedia resources, software, reports, etc. in addition to the usual suspects like academic papers. I am also a fan of Low-Tech Magazine.

I echo @Markos’s sentiment of the digital dark age. I feel like recorded knowledge has actually become more ephermeral now, because in the past if things are on paper - like books - they can last a loooong time but now not so much… This is why I love the Internet Archive. In fact, we could archive some GOSH outputs there, too. I vaguely remember @hikinghack saying something about storing things there but I might be wrong.

@marinappdf: Thanks for linking to that paper, it articulates so well many of my worries that I haven’t been able to focus and communicate. It’s very true that the problem isn’t just about which tools we use to organise and archive our collective knowledge, but also that we need people to do the task. I don’t have a good solution in mind, but that’s why I wanted to at least begin a reflection.

I think wikis were discussed in the other thread about self-hosted/open source apps for GOSH, and the problem we identified was that it takes a lot of effort and dedicated people to maintain/organise…

@solstag: Good point about defining what is “knowledge”. I think what I’m trying to get at is that there is a lot of informal “know-how” that comes from the conversations in GOSH, such as discussions in this forum or elsewhere. We learn a lot from this but it’s easy for this kind of knowledge to get lost. For example (and I know this is not an open science hardware-specific example), we had a long forum thread about election voting methods. We collectively learned something from that discussion, but eventually that thread will be buried deep in this forum. Most of us would have forgotten where it is, and eventually no one will even remember that the discussion took place or that we’ve investigated and learned about the different solutions. In other words, even if a forum thread or GitLab/GitHub issue technically exists, after enough time no one remembers that it’s there or what we learned from it. Basically, that knowledge is effectively “lost”. Having good organisation and archiving of more “formal” knowledge, such as papers or the publications on the GOSH website are already hard, but for informal, ad-hoc knowledge production it is even more difficult!


My summary

From what I can tell, looks like:

  1. A shared Zotero library (I think they’re called Zotero Groups?) might serve as an index to track various “pieces” of knowledge that are all over the place.

  2. But in addition to having good tools, a bigger challenge might be having the people to collect, organise, and archive that knowledge.

I suppose something we could try is to simply start experimenting with different solution. Try starting a Zotero group/library to track our knowledge and see how hard it is. Maybe play around with a wiki. And once we’ve organised something, we could try to make a separate archive with Kiwis as @jmwright suggested. I don’t really know… :woman_shrugging: :sweat_smile:

I am also really curious how other organisations have tried (and failed?) to track their learnings/knowledge???

2 Likes

Hi everyone! Just wanted to share here that a few of us will be meeting next week to discuss Zotero as a knowledge management tool. You can read more about the next meeting on this thread: Self-hosted apps for GOSH - #27 by briannaljohns

Feel free to join!

2 Likes

Ni!

Not actually, it won’t… The forum is searchable. If people ever discuss this again they can either remember there had been a discussion or simply be smart and start by searching past exchanges. I recently wanted to try to list all GOSH exchanges about the roadmaps, and to find where the issue of policy emerged, and using the search led me straightforwardly to the bulk of the info I needed (including links to specific GoogleDocs to examine). Information that is searchable is not lost, so I think that, for such cases, one may see a problem where there isn’t one.

The fact that forum is public, centrally searchable and managed by us is important for that. While information shared on Whatsapp or other places do not fall into this case, while some things like Google Docs might not be as bad since people tend to share the links here with some context, that makes them indirectly searchable. I think a more general practice of “use the forum wherever possible and, if not possible, leave a trace there” would solve a lot of the issues you raise, with little or no extra work.

2 Likes

Thank you for the input @solstag.

The fact that forum is public, centrally searchable and managed by us is important for that.

I agree this greatly helps keep information in one place. And Discourse’s search and typing suggestion features help find that information. This is all very good!

However, I probably wasn’t clear enough about the problem I am concerned with. I was using this forum as an example of a higher level challenge, which is how institutional/organisational knowledge is synthesized and effectively passed on.

To use the forum example, sure conversations are recorded, but what was learned from them are not collectively remembered, and (mostly) only people who were there would ever think of to search for that information in that thread. Not to mention when you search, you don’t know what things you are missing that’s not shown in the search results.

Another example from this forum is that discussions around knowledge management also just came up in this other thread about self-hosted apps. @dusjagr provided a great summary of a chat we just had about the topic (thanks @dusjagr!), but it is not obvious at all that a substantial comment about knowledge management happened in a thread titled “Self-hosted apps for GOSH”.

@briannaljohns’s newsletters got me to think more about this, and brought me to this excellent overview of digital gardening by Maggie Appleton. Digital gardening involves a specific methodology for synthesising, organising, and recording learnings, but for this post I’ll share this illustration from Appleton’s article:

On the left are what I think of as “raw data”, bits of conversations and messages some of which are potentially valuable. On the right are highly synthesised outputs like books and research papers. For GOSH, things like the Events Framework, policy documents, or the GOSH Manifesto would lie on the right. Forum threads would be more on the left, though not as far as Twitter threads.

The digital garden is in the middle, where there is continuous “cultivation” where things we are learning from the raw data are summarised. Something similar to this are @briannaljohns’s newsletters. The thing with newsletters is that they are organised by time, rather than ideas or concepts in a digital garden.

@dusjagr mentioned they tried “wiki gardening” for Hackteria, but it was difficult to pull off. I suspect this is true if there’s no sustained effort from dedicated people. Maybe this is also true for GOSH, but as a learning experience and experiment, I’m happy to give it a try.

And there would be value in this if digital gardening is successful. For example, someone completely new to GOSH could look at the garden and get a better overview of the conversations going on, the main topics/ideas that we GOSH has collectively learned about, and perhaps even who to talk to about what. This is less intimidating than being thrown into a forum where you might be overwhelmed with a ton of past threads you didn’t participate in.

I don’t know how far this idea can get, but would be happy to explore it with some of you if there’s interest. We can try different tools and methodologies and just see what happens! (?)

2 Likes

@briannaljohns: Is there an archive of past newsletters? I’ve been thinking one (relatively) simple thing we can do is to maintain such an archive, but also organise it by theme/topic/tags rather than just by time. This way, the synthesis you do every month can be added to a growing semi-knowledge repository?

Hey @hpy! You can see the past newsletters under the newsletter tag on the forum, each thread has a URL preview of the monthly newsletter (except for October and November 2021 - there were issues with these ones for some reason). However, I’m open to suggestions as to where we could store them!

1 Like

I’ve got other notes coming to this thread, but a quick random thought first:

Rather than trying to gather and organise all of our collective knowledge in one go, there’s this thread where we’ve been discussing the numerous open source tools used for GOSH. Maybe we can do something in a more structured way to capture that knowledge and put it somewhere, maybe as a page on the GOSH website that explains our tech stack and tools we use, plus where we’d like to be…

1 Like

You might also want to contribute to my GraphCommons “Open Hardware Landscape” (up 500 nodes allowed in the free plan). So far, there are no weights behind nodes and edges, so please don’t klick “Play/Simulate”, but feel free to add your knowledge about stakeholder of Open Hardware!

Supplement: GraphCommons Website and GitHub

2 Likes

Thanks for sharing GraphCommons, @Paul! At first glance, it looks like an implementation of mind-mapping. Is it a lot more than that? Also, too bad that it’s closed source (their GitHub repositories seem to only have demo use cases and client-side code), but that’s beyond our direct control. :sweat_smile:

On a separate note, I recently stumbled upon Diátaxis, a “systematic framework for technical documentation authoring”. The framework breaks technical documentation into four categories:

I only had time for a skim, but found the underlying philosophy to be intriguing. I wonder if it’s applicable to open science hardware and open source hardware documentation? Has anyone used this documentation framework before?

BTW, the text of the framework itself is open source under the CC BY-SA 4.0 license and the source is here.

P.S. I stumbled upon a thread here that links to some other discussions about Diátaxis.

1 Like

My thoughts are slowly navigating towards making a GOSH wiki.

At this year’s GOSH I realized that at least a few sessions subjects had already been discussed in the forum. However previous discussions are not really organized by subject.

To me, as it stands, this means that un-conference sessions, great as they are for many things, cannot be used as a tool to grow collective knowledge.

A wiki would also provide a persistent place where those sessions could be documented, and slowly digested into proper subject articles.

So here is a request: can we start a Wiki?

Some caveats have been mentioned by @julianstirling and @dusjagr here and here. I’m requesting this with those in mind.

1 Like

I understand what you are saying here. but back to my earlier argument…
We CAN use this discourse as wiki pages and collaboratively build knowledge together, sorted by topics and following discussions.

Imho it’s more about “how to motivate and value” the archivers who structure the knowledge, and not “what technical tools are out there”. None of the tools will do the work for us.

As example the notes from the unconferencing from gosh 2022 panama, sofar are group as they happened in the schedule of the days:

we just have to sit down and unknit all the topics individually, so it’s easier to find and contribute. Also discourse even suggests, when something should fit into an existing topic. please use that function!

2 Likes

I’m glad you mentioned wiki posts, it got me thinking about that feature again. :slight_smile:

I have set up a few to try out the feature, and proposed something similar here.

While I agree, wiki posts are still posts in a forum, not pages/articles in a wiki. There are many differences to consider, not just the collaborative editing feature.

For example, a readability issue is evident in the first post of that thread: GOSH 2022 SESSIONS Documentation - THURSDAY Block # 3 (01:00 pm). Longer posts with more varied content will also be harder to edit here.

My main argument against the wiki is splitting content and participation into different websites. A massive amount work went into putting the Roadmap into GitLab issues, and this ended with only 2-4 users participating. Using Discourse wiki posts would not generate this problem.

I also imagine that re-purposing Discourse as a wiki could work. But I would be careful with planning how, and consider the amount of work that it would actually take.

While I agree with this idea, I believe that this matter it is both about motivation and tools.

I will make a few comments on this at the next Roadmap Issues meeting!

PS: I just found an example from OSMC.tv on how Discourse wiki-posts can be set up to auto-publish as articles on a static page. An “edit”/“discuss” button could be added to each one, pointing to the forum.

2 Likes

I am following this thread with great interest and would love to be a part of a working group on this, with more attention to spend on it in the new year. Having worked on distributed open publishing models during my career as an academic research librarian, I am keen to help out, keeping things values-focused, such as is outlined in the recent release from the Next Generation Library Publishing Team in the FOREST Framework

2 Likes