Loomio
Fri 29 Jul 2022 9:44AM

Project: How to Marry the Needs and Offers

I Ivana Public Seen by 159

This is a continuation of conversation started in another thread but which deserves its own Project.

It seems to me that there are a lot of people out there who wonder what they can do. My interpretation is that this doesn't only mean to sort our the recycle-bins and have short rather than long shower and switch the light off when not in the room, but they also may mean: 'how can I help'?

At the same time there are many ideas for projects that are struggling to start - because they need help. Sometimes that is financial help, but many times they would benefit from a good brainstorming session, or from bouncing the ideas around with other people, or maybe they need an expert help for a few hours every month and maybe somebody could sponsor this expert to actually help them? There are so many possibilities...

But where can one state one's needs, what infrastructure exists, what could (easily) exist if it doesn't already etc...

One thing to try and understand is the Murmurations protocol (mentioned in the thread ...) https://murmurations.network/2022/07/05/what-can-you-build-on-the-murmurations-protocol/

some other links that could be useful

https://open.coop/2021/06/16/a-new-model-for-grant-funding-to-finance-the-collaborative-commons/

https://www.postgrowth.org/store/products/the-offers-and-needs-market-a-90-minute-process-to-reveal-and-mobilize-community-wealth

So... what does already exist and serve the purpose or marrying the needs and offers? What could easily exist and what is needed for it to come into being?

Then how to spread it, scale it, grow it so that we can say to everybody who wonders how they could help: ok, put your profile here, tell us what you can do/offer, and see who needs that in your locality or elsewhere? And equally, we can say to struggling projects: tell us what you need, put your profile here, and see who has what you need in your locality or elsewhere?

Then some kind of matching algoritm does its thing, et voila! we get the ball rolling a bit more...

Thoughts? Ideas? Reflections?

Thank you.

I

Ivana Sun 1 Jan 2023 9:21AM

Here is a an application that offers one solution to the 'needs and offers' . Please use it and share it in your social, professional or any other circles. https://www.sharebay.org/

Hapy New Year to all.

OS

Oli SB Thu 27 Oct 2022 9:24AM

this is a great idea - and should be pretty easy to develop, based on the Murmurations API, so that people can 'post' the the index straight from their browser plugin and searches the index (for similar Offers local to you) when you search via the usual search bar... although we may need to find a way to limit it so that it's not searching the Index every time you search for anything...

Let me think about this a bit more...

GR

Grace (Rebecca Rachmany) Thu 27 Oct 2022 2:35PM

Awesome because the person I am talking to is in the GEN-EU network and I know they are working on some stuff with the Murmurations team. I think she said the application for funds wouldn't be until August, so that's a pretty long time, but maybe if we have a team and a budget, we can explore other sources of funding as well.

MS

Matthew Slater Thu 27 Oct 2022 12:42AM

The beauty of the REST service approach is that yes, you can just build a browser plugin if you want. While you are seeking funding for future work, do put in a word for those who did past work without waiting for funding! @Bob the offer_want schema currently deployed is aimed at local community groups more than products. We could have broken it into two levels to better support future innovation but I recall there was a preference to keep it simple so we just defined one to get going. Some technical debt there perhaps.

BH

Bob Haugen Wed 26 Oct 2022 7:41PM

d a Browser Add-in which would pop up whenever you were shopping and offer you local goods instead of the online goods?

That's a great idea, but different from at least one version of offers and needs (which is if course not the only version, and yours is also doable). The version I have seen is often not commercial, but offers and needs among friends in a community. Could also be local businesses, but also barter or just stuff to give away.

GR

Grace (Rebecca Rachmany) Wed 26 Oct 2022 6:45PM

A couple of us were thinking this through and we thought: what if instead of needs/offers, we had a Browser Add-in which would pop up whenever you were shopping and offer you local goods instead of the online goods? The local businesses could sign up and post updates to what they have to offer, and then everyone in the network installs the browser plug-in which just simply searches for whatever term you are looking for within a certain radius, or with the attributes you want (ecological, accepts the community currency, etc.) Does anyone know of such an add-in or have thoughts about how to create this? We were thinking we might be able to apply for the Erasmus funding for that kind of a project

JD

Jennifer Damashek Fri 28 Oct 2022 1:30AM

Thank you for the invitation, @Renata Ballesteros. I’ve learned that my personal experience of facilitating events with more than a small number of participants that I don’t know well is stressful and quite unpleasant for me. It’s one of the reasons I’m working on co-creating small communities of people who have decided they want to care for each other. I hope you find someone to help you facilitate an OANM.

RB

Renata Ballesteros Fri 28 Oct 2022 12:42AM

@Jennifer Damashek and @RobertD would you be interested in co-facilitating a market event?

SG

Simon Grant Wed 26 Oct 2022 5:48PM

I don't yet have any takers for a collective conversation around the nature of needs and offers, but would be happy to facilitate one if anyone else shows an interest. If there were an offers and needs market of interest to me, I would be happy to test it out and give constructive feedback...

RB

Renata Ballesteros Wed 26 Oct 2022 3:39PM

Hello everyone, I have been following this conversation superficially, but have been too busy to get involved. However, I took the PGI's Offer's and Needs Market facilitator basic training this Spring and am gearing up to take the advanced one this Fall. As a requirement to enter this course, I have to host a market.

@Simon Grant, perhaps experiencing one of these along with the collective conversation you are hosting could help folks gain some clarity around the concepts?

SG

Simon Grant Fri 21 Oct 2022 8:26PM

I'd like to offer to host a collective conversation around the question: what is needs and offers anyway? I'm not at all convinced that people have the same idea, and having compatible ideas about what it is is vital to any constructive dialogue.

In essence: let's try collaborative / participatory / collective ontology design around this, so that we can actually communicate meaningfully

JD

Jennifer Damashek Tue 18 Oct 2022 7:18PM

Hi @Oli SB thanks for your feedback and for your questions!

“is this just a demo and you are thinking / hoping people will deploy local versions for their communities, is that right?”

Yes, the site I linked to is just a demonstration. The idea is that each community sets up their own version with a separate database.

The 'distance' field didn't have any units... I presume this would be miles from the location posted?”

You can enter a number and the units of your choice. For example, if you are only offering the item to people who live within 1 mile of your address, that is what you would enter, 1 mile. The fields are optional, so they can also be left blank and details like that can be worked out between individuals.

“Is there an Admin interface for Managers, where you can configure things like the currency etc?”

Yes, there is an admin interface where currency and language can be configured. It’s in the github repository.

Is there any matching and notification functionality built in? e.g. if someone posts a 'need' for window shutter would they see a match for my 'offer' and would I be notified?”

No, there is no matching or notification functionality right now. We are in the very beginning phase of using the app with small communities. It is meant to support personal meetings with as few as two people. In our small groups, we have regular meetings dedicated to sharing offers and needs. In my experience, this is where the magic happens. It truly is amazing how much each person has to offer and how many offers and needs can be matched within a very small group. There is so much abundance all around us, it’s simply invisible.

The app is designed to be a place to save offers and needs within a community that are shared during personal interactions. It may also happen that people simply enter offers and needs into the app without having the personal sharing, but my feeling is that this will be a secondary use.

In between the live meetings, people can use the app to find their own matches to offers or needs and save them for future reference.

This isn’t possible right now, but our plan includes enabling people to determine how broadly to share their offers and needs. They will be able to choose to keep them to one small community that they belong to, or have them posted in more than one community, or they will be able to post them publicly. When the app has that capability, we intend to offer a way to intelligently match offers and needs. That function won’t be part of the app but a separate service invoked from the app. We are working on that service for intelligent knowledge searching, organizing and sharing. This service will be used to enable matching of offers and needs when the amount of data becomes too large for individuals to do it themselves quickly and easily.

This app is just one part of a long-term community building project. We are in the very beginning stages, although my husband and I have been thinking about it for decades. We’re offering the app to anyone who might find it useful on its own. We would be thrilled if anyone wants to use it for other projects and/or help us improve it.

If we could add functionality to post the Offers and Needs as .json files which are registered with the Murmurations Index this would provide the basis of a fully decentralised, federated Offers and Needs marketplace ;)”

That doesn’t sound difficult, and we would be interested in that experiment to see if people would actually use it. I suspect more would need to be put into the project than just adding that functionality for people to start visiting and using such a marketplace. But I’m really curious to see what would happen.

We will need a JSON generator to feed the searching service that we’re planning. So adding that capability will actually help us with our project. If you want to talk about it more, let me know.

OS

Oli SB Tue 18 Oct 2022 1:22PM

Hi @Jennifer Damashek

Thanks for posting this - the app looks great! I just posted some window shutters as a test... I presume https://nbcoc.ca/ is just a demo and you are thinking / hoping people will deploy local versions for their communities, is that right?

It works brilliantly.

A few comments and questions:

  1. The 'distance' field didn't have any units... I presume this would be miles from the location posted?

  2. Is there any matching and notification functionality built in? e.g. if someone posts a 'need' for window shutter would they see a match for my 'offer' and would I be notified?

  3. Is there an Admin interface for Managers, where you can configure things like the currency etc?

If we could add functionality to post the Offers and Needs as .json files which are registered with the Murmurations Index this would provide the basis of a fully decentralised, federated Offers and Needs marketplace ;)

JD

Jennifer Damashek Tue 18 Oct 2022 10:58AM

Hi @Larry Reader we just now have made the repository for the offers and needs app public on github. So your question had perfect timing! Here is the demonstration app. https://nbcoc.ca/

Here is the repository on github. https://github.com/rdamashek/Community-Through-Caring

The app design is based on a prototype offers and needs app developed using Glide: https://oanm-prototype.glideapp.io/dl/6471c6. We didn’t want to use that app because it is based on google sheets.

We wanted an app that was very simple, open source, and designed for one purpose: to support ongoing exchanges of offers and needs in a small community where people know and trust each other. We couldn’t find anything like that, so we had it developed and offer it as open source so anyone can use it and improve it.

Let me know if you have any questions or feedback. You can post here or email me at jenniferdamashek@protonmail.com.

MS

Matthew Slater Tue 18 Oct 2022 5:20AM

Hi Larry there is a new an aggregator up at https://murmagg.communityforge.net but no items have been indexed yet, because I have other priorities, and no other platforms have implemented yet. I can give you pointers if you want to implement.

LR

Larry Reader Tue 18 Oct 2022 2:31AM

Hello, Forgive me for jumping in from no where but here I am! Noticed @Jennifer Damashek post and am curious about the status of your "Offers and Needs" app. It does seem that offers and needs outside of a small local trusted group or tribe is not very successful as far as I see. You seem to be more focused on a local group or project which makes sense to me as I ponder leveraging existing software for Offers and Needs or resorting to rolling my own rudimentary implementation since I'm not finding a fit out there.

JD

Jennifer Damashek Mon 5 Sep 2022 6:24PM

The sector I’m working on right now is my personal relationship to nature and especially the environment where I live. I’m studying permaculture through Earth Activist Training and plan to put into practice what I’m learning on the land where I live in Eastern Canada. I also am engaged in community building. We are starting with people we know and trust to build community, and we’re designing simple technological tools to be used in this context. One of them is an Offers and Needs app. I think as enough people engage in similar activities we will naturally begin to cooperate and eventually we will be amazed at what emerges. At least this is my hope.

SG

Simon Grant Mon 5 Sep 2022 2:07PM

Looks like I agree with you @Jennifer Damashek — of course I meant too complex for human rationality alone to solve. What would be your preferred sector to start on?

JD

Jennifer Damashek Mon 5 Sep 2022 1:56PM

So, rather than trying to solve the global problem of offers and needs (which I would say is absolutely insoluble, being too complex) how about identifying one particular sector?

If the global problem of offers and needs were too complex to be solved, why was nature able to do it long before humans were on the planet?

No one solution or group of people could hope to solve the problem. However, if we look to nature herself to find clues about how to solve it, I believe we can do so.

I agree a great place to start is by identifying one particular sector, or one project simple enough for our human minds and tools to grasp and solve.

Item removed

SG

Simon Grant Mon 5 Sep 2022 10:36AM

Some great issues are coming up for me here. One is the dimension of complexity / knowledge-intensity/richness. It's pretty clear that in simple, closed-world domains matching is relatively easy. That depends on the simplicity of the domain leading to shared representations or indeed ontologies. Also I hope it is obvious that one cannot generalise from simple domains to complex domains. One's whole approach needs to shift (see Dave Snowden's Cynefin model...) On the other hand it may be possible, and worth investigation, to break down highly complex domains into simpler sub-domains.

Where there is a well-defined ontology — which could be helped, e.g. by international standards — the complexity is reduced and matching becomes possible for supply chains, for instance. However, it seems too easy for international trade to simply ignore the vital issues of fair trade, environmental impact, and other ethics. Those issues introduce added complexity: no wonder businesses tend to be shy of them. So we often make do with small actions that make sense to us (switching off the lights etc.)

There are many existing localised initiatives like freecycle/freegle and charity/second-hand shops and sites. They seem to work for items that are simple to recognise and in constant demand. The physical shops do take time, though; and the fact that there are competing sites and shops is a real pain if you have an unusual need or offer.

At the other end of the scale, I've been taking an interest for many years in two very popular web-based services: recruitment and dating (“How to Marry…” :-)). And again, the more we reduce ourselves to commodities, and the more we simplify ourselves to fit in with social norms, the easier these things seem to work. A perverse incentive.

So, rather than trying to solve the global problem of offers and needs (which I would say is absolutely insoluble, being too complex) how about identifying one particular sector? I'd be happy to help...

If it's passing on/selling used things for reuse, I would say again that this needs domain specialists, together with storage space (shed, garage, barn, etc.)

MS

Matthew Slater Sat 1 Oct 2022 9:39AM

Hi Simon, and everybody,

My murmurations aggregator ( https://gitlab.com/community-forge/murmurations-aggregator ) (php) is now basically working, and am about to deploy it.

It is designed to aggregate any schema, by extending a base class.

(work in progress)

It's quite crude, it filters geographically using a square instead of a circle, and there is currently no ordering of search results.

I will imminently put this aggregator at https://murmagg.communityforge.net, where anyone can query it.

Soon my Drupal smallads ( https://www.drupal.org/project/smallads/ ) module will also be compatible, and I will set it up for:

  • HUMANS, a Mutual Aid Network in USA.

  • Selidaire, the French organisation that manages insurance for the LETS nationwide.

  • Gradually more of the Francophone LETS

However as far as I'm aware, nobody else is yet submitting offers and wants to the index.

The aggregator currently does nothing to order/prioritise the search results.

It would benefit from a basic elasticsearch or there is now scope for more adventurous AI - whatever you want to do with it.

SG

Simon Grant Sat 1 Oct 2022 7:52AM

I'm inclined to agree that fully automated matching is overall detrimental, maybe raising as many issues as it solves. But what about something that, from the automation point of view, looks semi-automated; or from the human point of view, looks like assisted human choice? I'd be looking at using the technology to do what the tech is good at: looking through large swathes of data looking for patterns, and then let the humans do what we are good at: using all of our senses and intuition, even more strongly in collectivity, to home in on what is emerging; what needs to be done; what the next minimal step (or match) might be. Etc. Talking of personal relationships, I'd love a personal catch up with any of you good folks here, to continue the endless quest for more and more fruitful collaboration.

I

Ivana Sun 21 Aug 2022 10:42AM

I don't think there will be going back either - evolution doesn't have the reverse gear. But what will happen is impossible to say. We can model things based on limited number of variables, and those models could be right or completely wrong. As a layperson - what I think will happen is that we are likely to receive some kind of genetic update from nature which will make us able to act more in tune with it - if it wants to preserve us as a species. (Which I don't think will be necessarily a pleasant or elegant process). Why I think this is a long story and probably not very relevant here.

But wanted to say that all these different ideas of needs and offers exist and it is not a choice of either-or but as-well-as. There are needs and offers that must be satisfied locally if at all, where real people will meet and talk to real people, and decide if to trust them or not, the way they always did. Digital solution for these types of needs could act as a bit fancier notice board, with some notifications, maps and such like? And sure, data sharing protocol where that makes sense.

And there are still the multi continent supply chains - sadly, which will continue to exist for awhile at least. I hope they decrease in importance by a lot though (because of their incredible cost to the evnironment)- but whether they will, I don't know. Digital solutions for those kinds of things would obvisouly be much more complex. People who know more about those kinds of supply chains than I do can comment on what the needs of those supply chains are and therefore what the responses could or should contain.

So it seems that we have different types of needs here, and different types of ideas for responding to them. Fair enough. What could be the next steps?

Could streamlining activities and setting up separate discussion threads for worldwide solutions and for hyperlocal ones help?

VG

Vincenzo Giorgino Sun 21 Aug 2022 10:42AM

It's a very interesting discussion. Gary raises an open problem but locks it up in a "small village" perspective as Rebecca points out. And I would add that he maintains those social intermediation positions - with ambiguous terms (facilitators...) that are not clarified but can be translated as (more or less professional) mediators, politicians, certifiers of various kinds - that technological development, especially post 2009 (the birth of bitcoin blockchain), wants to supplant. This is the real advance that technologies can allow: an answer to the problems generated by the past cooperation efforts, the hierarchical structures of left parties, trade unions, cooperatives etc.. I fully agree with Rebecca's observations and I would like to know more about what has already been implemented on this level.

GR

Grace (Rebecca Rachmany) Sat 20 Aug 2022 3:02PM

This can work at very small scale. In other words, we can go back to small village societies and then we don't need automation at all. The only problem is that our problems are planetary. We need planetary trust in place to run transportation and energy. We need regional trust to share goods along a group. We need regional and planetary sense-and-response systems to remain within planetary boundaries. I wish that I believed that a small-scale approach will somehow work. But IMNVHO this is naive and somewhat luddite. We live in a global system, with global economies and global supply chains. If you believe that the collapse is inevitable and we will end up with small villages again, I think you are correct. Only these small-scale reputation and trust systems are the answer. We don't need automation and we can go back to how things were in the past. If enough people die out because we've exceeded planetary boundaries, this should work swimmingly. (And the smart money is on the die-out, so hey, go for it.)

Another way this approach would work is if we take all of our junk-piles and upcycle them, so we never have to use new stuff again and we can live in our little villages. Not unlikely but then we would still need to scale our knowledge bases and understand who is trustworthy in terms of how to upcycle telephones or re-use silicon chips or create better irrigation systems out of old tyres. In any case, you will still need larger scale trust than people can reliably keep in their heads.

In my case, I think that there is no going backwards to the village model, and that we could prevent some of the collapse if we get this right, and that it is our obligation to care for all the living beings on the planet. And, even if collapse happens, we are still a global species. It is also naive to think we can salvage the planet, so I'm not claiming a superior position here. It's just my position that we have no choice but to create something that is truly going to counteract the existing systems and provide a viable, large-scale alternative for stewarding the planet.

GA

Gary Alexander Sat 20 Aug 2022 1:24PM

Why "bots to monitor our searches..."? Why not people who are keeping an eye out on community transport noticing that people have both asked and offered a second hand bicycle?

MS

Matthew Slater Sat 20 Aug 2022 12:26PM

Thanks for saying so Gary, I have to say as well I'm less excited about the kind of semantic automated matching that would serve larger, more anonymous marketplaces, and which I've noticed a lot of people enthusing about in all the years I'm trying to get a services running. A nice map interface with points and images of items in the locality fulfils most of the needs I have in mind. Maps are irrelevant beyond the local area coz things are done by post and online.

There is a role however for some kind of notification service e.g. notify me when someone within 10 miles has a second hand bicycle available. This might require bots to monitor our searches and present us with notification options rather than hardcore AI

GA

Gary Alexander Sat 20 Aug 2022 12:10PM

Hi folks, I'm sorry I'm coming in late to this discussion, but I would like to express a dissenting view. Those who know me know that I have a very long standing interest in online tools for building a collaborative economy, but I am not convinced that automated matching of wants and needs is desirable. The social vision of the commons-based collaborative society is based on personal connection, totally unlike a free market approach where you are usually dealing with strangers with whom there is no ongoing relationship. I think it is vital that we build personal relationships at all scales of our new economy, and would like to see local moderators/facilitators monitoring the offers and wants and helping with the matching on a basis on knowing the people concerned and building trust. The social vision rests on small scale networks with a basis of trust, linked together at larger scales, still maintaining the trust. You are mostly dealing with people you trust, or people who are trusted by someone you trust. Sometimes at larger scales, there may be several hops of trusting connection, but always connection of some sort. The monitoring and moderating is a significant overhead for a truly collaborative society that is bottom up, and needs backing up with good protocols for handling conflict and a core of peacemakers who can intervene when needed. Keeping the society stable and peaceful is a primary activity, not an add-on, so lots of people helping with the matching of wants and needs is an important and desirable activity. Gary

R

RobertD Mon 22 Aug 2022 5:21PM

Sorry I’ve been watching the thread but a bit engaged in a work effort. So I’m going to try to contribute a little now to this and the one other related thread I saw emerging over the weekend. First, I very much appreciate this healthy discussion on the attitude of caring and love we need to foster via personal relationships and small communities. In my mind and heart, this clearly is true. Yet I agree with @Oli SB that we also need matching capabilities to share valuable knowledge across community boundaries, and even across language barriers in order to care about and care for others who don’t yet have access to it. I’ve come to firmly believe that our approaches to and ideas in life are always failing, and that’s a norm. But as long as we ourselves don’t believe we are failing, we keep on learning from those failures and move on to new approaches. After all, it took Edison over a thousand tries to get a lightbulb that actually worked, and we all benefited immensely.

So, in terms of AI, I’ve attached two slide decks that a friend and colleague, Chuck Rehberg, and I put together to try to explain to my son, a recent Carnegie-Mellon School of Computer Science 4.0 average graduate, that a different way of thinking and approach the problem of AI-driven matching might actually work. We had presented something similar to DARPA as part of a conference exploring Third-Wave AI (Note: DARPA says that 2nd wave AI is what is widely in use today using neural-network based machine and deep learning). We made a little headway with DARPA and a little with my son, and I think it might help with this conversation as well.

The approach is a type of third wave AI that is core to what’s known in the field as Digital Twin technology. Digital Twins evolved out of the need to address issues in large-scale manufacturing and maintenance (think airplanes, vehicles, etc.), and the twin is effectively a digital, dynamic model of the object of interest. The model learns by observing both changes in the state of objects and their behavior in real-time, and then correlating those so that knowledge can be used to predict behavior, and prescribe actions that lead to desired behavior (Note: matching is one application that can use both capabilities - that’s for a subsequent post if there’s interest).

One interesting fact is that, though these AI models learn like ML models, they don’t need to learn a lot of data to be useful, since the data they learn is in the context of narrow events (e.g., state changes). Since very few people have been exposed to these kinds of large scale problem domains, it’s understandable that few know about the value and promise of this new wave of AI. Also, a lot of these solutions and applications are being kept proprietary or under wraps for commercial or security purposes (e.g., for use in driverless vehicles).

That said, we are specifically looking to make the technology accessible and available to address the needs of people and bolster equity in society. One specific application we are exploring is for more effectively matching offers and needs across communities, and sharing needed knowledge more widely and on a large scale.

Enough talk - here are links to the two documents I mentioned - the first presents the challenges and the second addresses how the technology helps address those. Thanks in advance for spending the time to read my post and review these, and I hope you are able to approach these with an open heart and mind!

Robert

I

Ivana Mon 22 Aug 2022 10:21AM

Many projects already exist, as we have seen in this thread, so maybe they could drive this effort according to their own needs, both met and unmet? And tell us what needs of their users they already do meet, and which ones they don't - to see what would be useful. I can see that people are keen to buld something so let that something be built...

M

mike_hales Mon 22 Aug 2022 10:17AM

A slight side track here, on 'AI'. I'd personally be gald if we managed to stop boosting the hype that now is associated with the 'intelligence' thing. 'Pattern recognition' is a good plain term? 'Mirror analytics' on the'wild' scraped data of our own widely distributed and emergent actions is good (as a term and an ambition, not as an existing practice in the hands of the GAFAM crew!).. Robotics and plain simple bots and software agents and open-app algorithm ecosystems can be good. Even hashchains have some good to be elicitied in them somewhere :-p Commoned infrastructures of distributed interoperable tools and media under commoned protocols are good. Etc.

Let's be as specific as we can, about the current and desired repertoire of software tools and digital ecosystem-infrastructures? And regard eveything as being constructed in commons, under joint producer/user stewardship, rather than cropping up in some academically or corporately or professionally defined field lime 'AI'?

OS

Oli SB Mon 22 Aug 2022 9:29AM

I think, if we are aiming to develop online tools (decentrlaised tools too) to assit with matching offers and needs we have to assume the 'problem we are trying to solve' is to match any offer or want anywhere... so, locally, that may be best done offline, as with your hardware shop example... but all other scenarios will need smart, computer-driven matching to help cut through thee noise and to deliver you relevant data... so imho some sort of matching and mapping technology will be necessary... although it should not need to be scary! Personally I think the term "AI" has been a bit over-used and miss-used and I'm sure it's even accurate to call a matching algorithm "artificial intelligence" ... but whatever, if it helps us link offers and wants to increase efficiency and reduce waste and cost/s I'm in!

I

Ivana Sat 20 Aug 2022 8:08AM

Cool, thanks for your reply.

Here is an example of the matching needs and offers in real life. I went to a local hardware shop to see if there were any cards on their community board left by builders - as need a builder. And sure, there are. Needs and offers matched. So it all depends what kinds of needs we are talking about and how they could be matched with offers. It seems to me before the implementation starts, we need to define what problem we are trying to solve - not that I discovered the world in saying this. It is pretty interesting, though, to see all the different approaches and all the different ideas driving them towards different kinds of aims that can all fall within the 'needs and offers'.

GR

Grace (Rebecca Rachmany) Sat 20 Aug 2022 7:19AM

First of all, there is no "small scale deep learning". Without a decent data set, you simply are unable to match needs and offers in any meaningful way. Each person would have to manually go through lists and hope that they found the best option. Even a search engine uses large-scale AI to understand how to meet people's requests. Otherwise we are basically going back to a phone-book approach where the matching is done manually. My assumption is based on things like Numundo.org which started out with this idea but had to pivot, or the needs and offers engines that were made during the pandemic but never really took off because there was too much manual work involved in finding what you wanted to find. Today, every needs/offers matching engine from Kayak to ebay, is reliant on AI. Conceptually, the needs/offers idea is ancient, but the technology for matching, mapping, and understanding relationships has evolved so much, it feels to me that creating directories is obsolete. Here's a great recording of 6 hours on different "Tools for Thinking" which help people and groups categorize information. IMNVHO, anyone developing matching and mapping technology should at least have decent knowledge of these types of technologies and leverage what's already been built. https://player.vimeo.com/video/739652984?h=ecdefd58ba

I

Ivana Fri 19 Aug 2022 2:43PM

Perhaps a stupid question but here it is anyway: why do you think the 'large scale deep learning AI' is necessary? In other words - what can be achieved by it than cannot be achieved without it? What kinds of needs exist, or what kind do you have in mind, that can only be addressed by AI?

GR

Grace (Rebecca Rachmany) Fri 19 Aug 2022 5:48AM

Needs/offers matching has reached an incredible level of sophistication but it is quite difficult to achieve. You need a large, active marketplace, a non-biased search engine, and a large-scale deep learning AI. All of this has been developed outside of our groups. I see very few AI experts in the opensource movement who have a feasible approach to this problem. As a result, we've seen numerous failed attempts. How do we get in the proper AI professionals who can approach this from a modern perspective?

I

Ivana Mon 15 Aug 2022 6:29AM

There is a bit of a mixture of all sorts of comments in this thread - but here is one more - Open Food Nework is known around here, I believe. I voluteered for them for awhile and know that some of their staff in France have left the OFN to dedicate themselves to creating a data sharing protocol between OFN and other food networks.

The reason for this was that food producers who sold via OFN were also selling via other networks and to reconcile the stocks and receipts from different networks was too time consuming for the producers. Hence the protocol, as I understood it at the time, would allow producers to post their items for sale just once, they would be automatically visible on all the networks that adopt the protocol (not sure if producer got to choose where they want to post?), and any updates, such as amount sold etc, would automatically feed back to the producer.

I know this work was underway, but I lost touch with them and I don't know how far the project got, how widely it is adopted, etc.

If anybody has connections with the OFN France, maybe they would be willing to do some knowledge and experience sharing in a call - which may be useful for Murmurations?

I

Ivana Sat 13 Aug 2022 6:14AM

Here are some people down under who appear to want to do similar things https://senvic.org.au/about-senvic/

OS

Oli SB Thu 11 Aug 2022 1:57PM

I just deployed a new aggregator, using the Murmurations Wordperess Aggregator Plugin - which shows a mix of Organisations from Karte Von Morgan data and @Matthew Slater 's Community Currency projects data: https://wpagg.murmurations.network/ - this is a useful viewing the results of test Murmurations Profiles which are published to the Test Index... You can make a Test profile by using the Test Profile Generator (selecting test_schema-v1.0.0 at the bottom of the list, making a profile and publilshing it to the Test Index directly via the Profile Generator site) but it might not show up on the map right away as I have not set the map to auto-update yet... but will do soon.

OS

Oli SB Wed 10 Aug 2022 3:42PM

Hi All,

Good to see some of you yesterday - and thanks for all the follow up messages ;)

To facilitate those of you that would like to experiment with sharing Offers and Wants via Murmurations I have prepared this short guide. Let me know if you have any questions!?

Oli

I

Ivana Sat 1 Oct 2022 8:55AM

I am interested but not working on data sharing protocols or two-way search engines so probably not the right audience for the 'in great detail' session. Although, if one were to be organised, I'd probably want to listen in...

SG

Simon Grant Sat 1 Oct 2022 7:57AM

This is nice. May I add one thing, though, that seems vital to me? Some projects will accept any volunteer, but others need or want specific kinds of people, maybe with specific skills, or attitudes, or values, or principles, or whatever other quality is relevant to the task. So I don't want to see all the projects that fit my criteria: I want to see all the projects that fit my criteria AT THE SAME TIME AS me fitting their criteria. This is the principle of a two-way simultaneous search (which I can go into great detail about if anyone is interested.)

OS

Oli SB Thu 18 Aug 2022 2:16PM

Yes, I guess anyone who shares data would want to know those things! For Murmurations it's pretty simple: any data shared will be in the public domain, on servers hosted with Digital Ocean, which are paid for out of the tiny BTC donation Murmurations received a while ago... There isn't actually a legal entity behind Murmurations... it's stewarded by The Open Co-op... and although we're not paid we're very happy to provide support, which can be accessed via the telegram channel or the contact form on the website. Does this help?

OS

Oli SB Thu 18 Aug 2022 2:10PM

These are good questions... but I'm not sure I have the answers... my hunch is that in order to encourage people and Orgs and Networks to share their Offers and Needs we first need to demonstrate we have some working technology which will help them...

...and since Offers and Wants are harder to make work than simply mapping Organisations my first objective is to build up a map and directory of Orgs (and hopefully soon Projects too) which is now up and running at https://regen.murmurations.network/ to which we can import large data sets if Networks can supply .csv or have an API, plus anyone can manually add Orgs to the map by creating profiles via https://profiles.murmurations.network/ and selecting the 'organizations_schema-v1.0.0' schema

OS

Oli SB Thu 18 Aug 2022 1:57PM

Hi Robert,

Thanks for posting this. Can you share any inks or info about the 'microservices API for predictive semantics'...? It sounds really interesting and might fit well with Murmurations (as it uses JSON) to provide a matching service... it would be good to see a demo of this and of the matching service provided by https://github.com/researchstudio-sat/webofneeds in a future session.

@Oli SB In terms of the Murmuration protocol, what would it take to extend this to support individual people as well as organizations? That’s clearly a core concept and both people and organizations will have offers, needs and projects to explore.

We already have a draft schema for Individuals... (which includes a range of standard fields, but also the somewhat experimental 'member_of', and 'knows' (think 'follows' or 'likes' ... this field could be the start of a 'Web Of Trust' / FOAF aspect of Murmurations) ...) it just needs merging in to the test-environment so we can experiment with it...

ATM any entity can make posts using the Offers and Wants schema within Murmurations (so these could be from Individuals or Organisations) but clearly it will make more sense to link Offers and Wants to existing Murmurations profiles for Orgs (or Individuals) so we can drill down a level and see e.g. All the Offer/Wants from X. How exactly we do this is yet to be defined...

I

Ivana Sun 14 Aug 2022 7:14AM

If I were to knowingly share my data with somebody, I would want to know where my data is, exactly what data, where are the servers that handle it, who is behind it - ie what legal enetity runs the operation? I would want to know how they make their money, what they pay themsevles etc... I would also want to know is any tech support is available, what kind of support, how do I know that I can trust them etc.

I

Ivana Sun 14 Aug 2022 6:53AM

There are many mapping efforts around and most exist in isolation. To my mind, if I work on some of those isolated mapping projects, most probably voluteering or with limited time and resource, and I am asked to do additional work, I would want to know what is in it for me. I don't mean that in a bad way, but how would my addiitonal work benefit the world. We would somehow need to know what need/want for sharing their data all these individual effort have, why would they want to share it and who with. On the other hand we would want to know that there are people out there who are ready tolend a hand to any worthy project. But do we know that? I mean can we demonstrate it? My insticts say that this is true, but my instincts are hardly the most sellable thing of all.

I am wondering then if anybody has a very active social media account where they could run a survey that would give us some numbers that we could use to support our claims that a lot of people are out there ready tolend a hand - so they need to know who has the needs for their skills, time etc. That could be potential motivation for isolated people/projects/ initiatives to share their data?

Or maybe somebody among us has another way to get us some useful numbers? How can we know what needs exist in a particular location/place? How could we invite people to post what they think is a genuine need in their location on a map? If they did, then that would give us something to work with - we could see what opportunities exist in a particular location for starting a new business or activity or a gaden, and then it would be possible to identify what that new venture needs to be able to come into being. Then that means real concrete needs and real concrete ways to help things along by offering time, skills, space, furniture, material etc. So this invitation to place needs onto the map (and from there it could be shared with other interested platforms) ... where can that be done from? What is a good place to shout: you're all invited? And who knows somebody who stand sor sits at that place and wouldn't mind shouting free of charge or for a modest compensation which we could crodfund)?

Then if we have a 100 locations, for example, we can say to other paltforms - look what we've got.. all these people want to start all these projects, come and share what you have with us and with others that you want to collaborate with... In this case it would be easier for the individual projects to see what they could share, and why they could do it and how they would benefit or how they could contribute to somebody else benefitting.. Any thoguths? Leads?

R

RobertD Sat 13 Aug 2022 2:04PM

@Oli SB @Vincent Arena @Tibor Katelbach We have something similar to “Opportunities” internally in our schema, which I’ll share soon in JSON. We’re using “Goals” instead, since I’ve had a vision for quite a while that people should just be able to post their goals, and just like @Ivana said, they would be able to see/review matches for those. So Offers and Wants (or in our case Needs) are just types of goals. I agree @Oli SB that it would be good to start with the existing Murmurations portion, since accurately matching semantically is a much bigger challenge.

As I mentioned on the BBB, we have a partner in our endeavor who has technologies and patents in both the semantic NLP space as well as in observation-driven, semantic predictive and prescriptive analytics (this is also known in the AI field as “Digital Twin” technology). I think this is worthy of a follow-on discussion, but that need not hinder the work on the existing Murmurations expansion.

Just to whet your appetite a little - you can apply predictive semantics to decide if there’s a good fit between the offer and need if you can turn the underlying data and related profile knowledge into time-stamped facts about the needs of the observed object of the profile, and their (or other’s) observed experiences (e.g., events or significant state changes) with the offers. Note that all these can be expressed in JSON and there’s a microservices API for this. Similarly, you can even apply prescriptive semantics to identify what changes need to be made to the object’s profile to achieve a goal by expressing that goal (or intent - in JSON as well), and adding that to the observational analysis. I would be glad to point to a couple of artifacts and setup a demo.

@Oli SB In terms of the Murmuration protocol, what would it take to extend this to support individual people as well as organizations? That’s clearly a core concept and both people and organizations will have offers, needs and projects to explore.

OS

Oli SB Thu 11 Aug 2022 1:45PM

I don't have any specific communications materials, other than what's online. Normally I email, linking to the main Murmurations website, and explaining in brief what we're trying to do and asking if they have a data set they can share, or an open API. Here's a list of some of the old mapping projects I have contacted to ask... There's a useful link in row 32 to an old list of co-op directories which I have also used to try and find more source of relevant regenerative economy data... if anyone fancies contacting any of these and encouraging them to share data please feel free - and add notes on what you've done ;)

Thanks!

OS

Oli SB Thu 11 Aug 2022 11:19AM

yes - nothing to stop us trying to get some Offers and Wants data into the test-index right now... but displaying it anywhere will take more work than displaying Orgs

MS

Matthew Slater Thu 11 Aug 2022 10:54AM

It seems clear to me that the matching is not part of murmurations, but either

a) some kind of application that reads a murmurations index and which is queried separately

b) an extension to a murmurations index specific to the offers & wants, perhaps offering an additional REST endpoint.

Thus it makes sense to get the offers and wants working and populated first before addressing that challenge.

OS

Oli SB Thu 11 Aug 2022 10:20AM

That's great Tibor :)

Thinking a little more about this tho... I think it makes more sense to start by sharing data on Organisations and Projects. Offers and Wants are a key component of the new economy, but they're also more tricky to make useful (since really they require a matching service and notifications etc) ... so perhaps we should aim to walk before we run, by seeing if we can share and aggregate data on Orgs and Projects first, then move on to Offers and Wants. This will also allow a bit more time to make sure we have developed the best O&W schema possible... I liked the way @Vincent Arena called both Offers and Wants "Opportunities" which seemed to have some semantic as well as programmatical simplicity...

If you agree that walking before running sounds like a good plan and would like to share data about regenerative economy Organisations over Murmurations, the process is pretty much the same as I outlined for Offers and Wants: Create an endpoint (or .json files) for the data you want to share, but which validates to the Murmurations map schema then make sure your Org profiles validate with the JSON validator, then post them to the Index.

Let me know what you think and if you need any help...?

Oli

TK

Tibor Katelbach Thu 11 Aug 2022 5:12AM

Hi Oli

if we can also get data from Communecter and Catalist Network we will be off to a flying start ;)

with pleasure , just give me a bit of time to look into your doc and produce a murmuration feed for you to play with

I

Ivana Wed 10 Aug 2022 5:32PM

@Oli SB do you have any communication material, or you explain it to everyone verbally from scratch every time? And well done on making a start. I wonder what other uses cases people can see?

OS

Oli SB Wed 10 Aug 2022 3:55PM

This is a really helpful example of a (the?) use case which Murmurations aims to cater for... but, as you rightly note, there is a lot of work to do to get there... because for any map / directory like this to be useful it will need to hold A LOT of data. So far we have schemas for Organisations and the experimental Offers and Wants schema but it will not be hard to develop a schema for Projects and one for Tasks, and for whatever other entities we think we need. The harder part will be in convincing like-minded networks to share their data with us (via csv, so we can import it), open up their APIs, or make their data available as compatible JSON. I'm slowly contacting other networks to explain our objectives and encourage them to open up their data... but it's time-consuming and I don't always get a response, even from co-ops who you would imagine would be keen on sharing... so, if anyone else has time to spare which they could use to assist this mission, in order to help develop the type of map which Ivana describes, it would be really useful ;)

We already have some great data from Karte Von Morgan, and some old data from Co-ops UK (they're not responding to my attempts to get access to their latest data set) and some from the Open Food Network, so by the time that is all imported and presented on a map, we should have ~50k nodes, which will be a good starting point... if we can also get data from Communecter and Catalist Network we will be off to a flying start ;)

I

Ivana Tue 9 Aug 2022 6:34PM

As a future user, what I am hoping to have from this sharing exercise is something like this: I go to on website/platfrom called, for example, wtfcanido, and I type in my search criteria: I want to share certian skills with projects of such and such type, in such and such location 5 hours a week in September. And when I hit search, what I want to see is a map with pins of all the projects that could interest me based on my criteria. Zoomable map, and when I click on the pin it takes me to the project page. There I see tasks (like in Tibor's presentaiton) and if I judge that I am up to any of those, I apply etc... This can include projects or timebanks and all of the platforms so far mentioned and many more. For me, I'd like to be able to go to the wtf website, wherever I am on the planet, at any time, and see who does what at that location, so I can lend a hand.

Or I want to see if there is a permaculture project where I am, or garden produce exchange, or community tree planting effort, mutual credit facility, or peer lending.. whatever.

As a person who may be in need of help on a project I'd like to be able to search for people who offer certain skills, or to post a project and invite people who have certian skills to lend a hand (paid or unpaid), with start date, end date etc.. all the stuff that people already talked about.

The main point being, I want to see all organisations, projects, people who do anything vaguely of interest to me, in one place - so they would all need to plug into the Murmurations. Obvisouly, there is work involved in getting them to do that, work in syncing all the profiles, some kind of stages approach would probably be well advised, etc but ... if my instincts are right and there really are a lot of people out there who want to DO SOMETHING other than putting the paper in the blue container and plastics in the yellow one whilst looking at governments and waiting for Godot, then... I think it's possible. Any thoughts?

MS

Matthew Slater Tue 9 Aug 2022 7:20PM

Yep that's me being all over it

The follow up document, if we get to it will talk about murmurations and some other glocal approaches to open source software.

MS

Matthew Slater Tue 9 Aug 2022 7:18PM

Thanks mike

I'm all over it already.

But not talking about it much in public because there's a lot of politics involved!

M

mike_hales Tue 9 Aug 2022 4:53PM

Another review doc on timebanking softwares to date: https://credit-commons.gitlab.io/principles-for-timebanking-software/#introduction

M

mike_hales Tue 9 Aug 2022 4:47PM

Timebanking community is taking stock of their tootls and intentions? Relevant to Wants&Need?

https://fyodorovich.github.io/Timebank-Toolkit/Trove.html

"Timebanking organisations worldwide are considering choices for the next generation of software. Recently, the Aotearoa (New Zealand) Timebank software working group published a report digging in to many aspects of timebank software and the needs of timebanks."

OS

Oli SB Tue 9 Aug 2022 10:45AM

This looks highly relevant - although the demo at https://www.matchat.org/ does not seem to be working atm ;(

M

mike_hales Tue 9 Aug 2022 10:41AM

Haven't been following closely so pardon me if Valueflow.ws is already under discussion here. But look here https://www.valueflo.ws/appendix/usedfor/ for a summary of things that are being done and developed with the valueflo.ws mapping framework and tools ecosystem.

That page includes a reference to

Web of Needs, in progress. The goal is to do automated matching of intents with attached actors/resources on the one side and interaction patterns on the other, such that meaningful interactions can be automatically suggested to users. The code is on github.

But here I'm intending to emphasise the entire valueflo.ws relevance not just this specific branch. @Bob Haugen and @lynn foster have built great foundations here, and this set of projects listed is entirely relevant to this 'wants & offers' thread, I would say.

NS

Nick Sellen Fri 16 Sep 2022 10:38PM

thanks @Graham :) I didn't get a notification as far as I can tell, but just randomly stumbled across this thread...

we have rudimentary "offers" feature, which isn't used a lot, compared to the core feature in karrot of "activities" (who is doing what, where, when, etc..).

I would love to expand the offers/needs part of karrot, and we have some interesting groups using karrot who are quite into that... for us actually, the "needs" might often be expressed in "activities", e.g. "we need 3 people who can drive to come to this address at 10am to carry x to other address".

anyway, I could talk about the promised future for hours, but actually have to focus on more mundane realities of life most of the time :)

G

Graham Tue 9 Aug 2022 10:19AM

Not sure if it's mentioned here or whether @Nick Sellen is aware of this thread, but I thought that https://karrot.world might have something to offer in this offers/wants context.

OS

Oli SB Tue 9 Aug 2022 8:32AM

Hi All - since there will now be a few of us in the Offers and Requests meeting at 4pm today (UK time), in order to save time I wondered if I could ask anyone who has an Offers & Wants/Needs system in place already (or in design) to add their initiative - and its fields - to this field mapping spreadsheet, so that we can see where the overlaps are, and where the gaps are: https://docs.google.com/spreadsheets/d/1W9o5zW77UrDmwm90tetp6NQq0ZtFleJ-txNhQb_G5pM/edit?usp=sharing

OS

Oli SB Mon 8 Aug 2022 3:50PM

really interesting - looking forward to chatting tomorrow ;)

R

RobertD Mon 8 Aug 2022 2:51PM

@Oli SB @Jennifer Damashek Oli, Jennifer and I are so glad to see the work you’ve been exploring with Offers and Wants, and also the discussion you and Ivana had. Jennifer was in the Offers and Needs Market (OANM) circle with the PGI and we both took their facilitator training. For a year or so we’ve been exploring building a simple, very accessible, open source app/web app to facilitate Offers and Needs (or Wants) exchanges, with an emphasis on building community relationships, and then as a foundation for intelligent matching of the two. Early on, we communicated with URI NOY MEIR, who built a demo of such an app using Glideapp, based on the PGI OANM model.

We very much liked Uri’s design and its simplicity, but since we wanted an open source app accessible to any community, not based on a platform that came with licensing constraints, we decided to commission a developer to create one for us. We now have a GitHub repository with the first Alpha version of the software as well as a demo web site we’re going to use to explore it’s use with a community of friends. Our website is hosted on CanTrust.coop and we’re still working through a few issues with the developer such as the password reset email going straight to spam or not getting through to people because of Google spam filtering, and some issues in the mobile phone UX. So, the repo is currently private until we have those things worked out, but I would be open to inviting anyone from the Open Coop to take a look - just send me an email.

We definitely are interested in a range of enhancements focused on intelligent matching/searching with individual/community control of the knowledge shared. We would love to ensure we could do that with Mumerations profiles, and with its aggregated open datasets. Jennifer and I are going to join your call tomorrow to learn more!

All the best,

Robert

I

Ivana Fri 29 Jul 2022 5:03PM

I'd like to understand some things around Murmurations better... In its webpage (link above) it says quite a few things, so I'll take them in order. On the website it says:

"For example, an ‘Alternative Berlin’ map could feature a mix of co-ops, using data from German Cooperative and Raiffeisen Association (filtered by location), other green initiatives and social enterprises from the Karte von Morgen data set (filtered by topic tags), plus bespoke data from local research and organisations that have added their profiles to the Murmurations network directly"

To understand how tis could happen: how would the data from Cooperativ and Raiffeisen Asso end op on the 'murmurations agregator'? Would agregator then have itw own website (or app?) and it would 'pull' data from other websites without asking? Or would the entities/websites who wanted to be represented on the agregator need to somehow put data onto the agregator website?

If so, how much work would be involved in settings it up for all who wanted to be on the Murmurations ag?

Or could, for example, the Karte von Morgen become a Murmurations agregator ad add data from other entities to its already existing map?

It's a bit hard to imagine how it all funcions, who needs to do what for it to come alive.

Same questions about agregating the blogposts. Say we have hundreds of websites, each with its own blog posts, that we want to agregate on display somehow on one webpage. What would the blog owner have to do to make this happen?

Let's say I'd be interested to run an Open Projects Hub aggregator site. What would I need to do to set it all up? The article is obvisouly written by somebody who knows how it all hangs together, and so they didn't see a need to explain in detail.. .Or maybe those details are stated elsewhere?

And then we'd need to add a 'bit of matching' and bounties, et voila!

But ... what does it all mean for somebody who is not a developer on one of these projects? What does actually need to be done and how can it be done?

I hope somebody has patience to comb through all of the above and reply... @Oli SB do you? Or do you know who would?

Thank you for reading.

Take care.

HB

Hugh Barnard Mon 15 Aug 2022 8:08PM

Sorry, have to make this shorter than I'd like. But. With Basis/Mistral in the 1980s, we worked in Brussels in 5 core languages, also the OJ, Officlal Journal was already in every community language. So one thing we did was have table translate of all keywords to extend any search. No machine translation except for Systran: https://www.systran.net/en/translate/ not v. good then.

At some stage when CELEX which is now Eu-Lex had full text, Eurovoc was compiled for core languages in about 1984: https://en.wikipedia.org/wiki/EuroVoc

So both the multilingual aspect and the (controlled vocabulary) thesaurus aspect gave a lot more breadth to search. Ironically, the last project I was assigned to, before leaving Brussels was a 'simplified search interface', no boolean, natural language (only 2 languages though) parse.

Long story short, this is how this approach helped us find useful stuff, from day to day, in an ever growing quagmire of legal text.

As for now, if anyone would like to think about this, I only have Twitter as social media because it's the nearest thing to a social epistemology tool: https://plato.stanford.edu/entries/epistemology-social/ I use it to identify plants, share gardening tips, discuss politics, discuss music, mess around etc. But I've been thinking about a borough/town/village focused instance of mastodon with set of bot helpers (with some of the characteristics above), clippy but useful, not annoying. I believe (see above) the simplicity of interface is a great virtue.

Hope that helps.

M

mike_hales Sat 13 Aug 2022 10:44AM

could you say more about ways of approaching synonyms and thesaurus? linking with machine learning maybe.

HB

Hugh Barnard Sat 13 Aug 2022 10:39AM

Sorry, I actually lied about what I was up, maybe I should go into politics? I almost did: https://en.wikipedia.org/wiki/2018_Newham_London_Borough_Council_election#Forest_Gate_South That is, in addition to the SIC, I've added folksonomy to generate a simple word cloud as well. I agree that SICs are prescriptive but there's also prefigurative/hybrid cases where (for example) boroughs/councils decide to run their own currency. I live in hope.

I was involved in search in the 1980s using: https://en.wikipedia.org/wiki/Basis_database and Mistral and one thing that struck us was 'synonyms' and 'thesaurus'. People express their fondest desires in different ways and (in our case, then) if you restrict them to booleans, then you will need 'high priests of search'. It was that kind of thing that left me rather agnostic. So I think there's some interesting AI style, recommenders for example, to be done in this area, maybe: https://ai4good.org/what-we-do/#otherprojects might take an interest?

M

mike_hales Wed 10 Aug 2022 12:21PM

Nice bunch of links @Hugh Barnard thanks 🙂 I wonder about this . .

I've been moving to SIC codes as a starting point. They don't really do fine grain but are a 'start'.

problem with SIC is not so much that the categories are too coarse? Problem is that they are categories of exchange-value based commerce? When trying to map (and promote) a contribution economy aka commons, there’ s a different ontology underneath?

SIC is mapping the top-right quadrant of this schema below? What about mapping the other quadrants (including ‘public’ provision)? And the (transformative) hub?

Maybe this ontology issue is where the timebanking community has things to offer? As distinct from tools?

HB

Hugh Barnard Wed 10 Aug 2022 11:34AM

Hi folks

Only just seen this, I'm doing some paid work for Nottingham Uni at the moment and some local gardening projects. So I'll make a micro-contribution:

  • I stuck a category matching RSS feed into https://sourceforge.net/projects/cclite/ about 10 years ago, it was originally based on stuff from LETs groups

  • For Cclite2: https://sourceforge.net/projects/cclite2/ I've been moving to SIC codes as a starting point. They don't really do fine grain but are a 'start'.

  • I'm sticking a partial set of fields that 'should' enable murmurations into the registry (database, trading group) definition in Cclite2. It's probably been involving since then though.

  • After saying all the tech stuff, I'm not really a 'believer', it's interesting to watch what my non-tech friends use, basically: https://www.ilovefreegle.org/ and ebay for 'stuff'.

  • Lastly, about 10-odd years ago, I was one of the founders of HEN (Hackney Environmental Network) which is now: https://sustainablehackney.org.uk/ this used Elgg, a little known social media platform. But the 'idea' is that the groups are present and aware of each other, communicate but maintain their own autonomy. I'm now trying to get this done in another borough.

  • I've argued in this (sorry, not sorry, self-promotion): http://hughbarnard.org/index.php/2021/11/27/print-copy-of-the-algorithmic-anarchist/ that informal physical space and, to some extent, serendipity is also v. important. The deep tech stuff goes over most people's heads and projects come first.

  • Interestingly (or not) on the above, I've just started to read this: https://en.wikipedia.org/wiki/Proofs_and_Refutations categories come out of experiments and refutations, so all the tech, love it though I do, is highly iterative.

Enough already, later Hugh B

TK

Tibor Katelbach Wed 10 Aug 2022 4:41AM

Hi Ivana ,

It was great indeed to share with so many entities working around the subjects
meaning that when we all have the protocole running , it'll be multi sourced automatically

Wow... So many raised hands. How exciting. So nice to see you all. I am not a developer, but lots of experience in QA and some PMing, if anybody needs help in that way. I am basedin France, so especially interested in what is happening in France, but not only. My are of interest is more the Planet Earth...

would be a pleasure to exchange directly , most of our actions are in France ;)

My thoughts about possible next moves:

  • for each platform - what do you want to share with other platforms, now or in the future?

  • What other platforms do you need to share your things with?

  • To be able to share what you need to share, what kind of entities do you need defined inthe protocol. Tibor spoke about People, Organisations, Projects and Events. Anything else?

  • What does your platform have in common with other platfroms? Does the protocol cater for that?

  • What are your specificities and does the protocol need to cater for those?

  • Inventory of all your fields - and identification of what already exists, and what is missing. What is the must have for you, and what is a nice to have?

  • What are your concrete needs and wants?

  • Perhaps a collaborative doc could hold all these and serve as an input to future work? Any pointers where this could be?

Thanks again for taking up this little invitation and i hope the collaboration not only continues but leads to great things. Take care.

we could try out creating our own community project "Needs & Offers" in each of our tools and set up demos. This could create a cross contributive dynamic where we all contribute feed back to each others plateform ? to make them better openly.

for the pad , whenever a group or a project is created in CO (COmmunecter) , that community has direct access to a whole stack of open source tools , like rocket chat , wekan (kanban), codimd (pad), next cloud...
Great to see this wave starting

cheers

I

Ivana Tue 9 Aug 2022 4:32PM

Wow... So many raised hands. How exciting. So nice to see you all. I am not a developer, but lots of experience in QA and some PMing, if anybody needs help in that way. I am basedin France, so especially interested in what is happening in France, but not only. My are of interest is more the Planet Earth...

My take from this meeting is that some people are ready to write code here and now and some people are at other stages - trying to udnerstand concepts, think through their own needs etc.

My thoughts about possible next moves:

  • for each platform - what do you want to share with other platforms, now or in the future?

  • What other platforms do you need to share your things with?

  • To be able to share what you need to share, what kind of entities do you need defined inthe protocol. Tibor spoke about People, Organisations, Projects and Events. Anything else?

  • What does your platform have in common with other platfroms? Does the protocol cater for that?

  • What are your specificities and does the protocol need to cater for those?

  • Inventory of all your fields - and identification of what already exists, and what is missing. What is the must have for you, and what is a nice to have?

  • What are your concrete needs and wants?

  • Perhaps a collaborative doc could hold all these and serve as an input to future work? Any pointers where this could be?

    Thanks again for taking up this little invitation and i hope the collaboration not only continues but leads to great things. Take care.

    Ivana

TK

Tibor Katelbach Tue 9 Aug 2022 5:17AM

Hi all

Thanks for the great feed , I'm also very interested , we also have a needs and offers module in communecter,
https://www.communecter.org/#annonces, we just finished implementing open Badges protocol as well

and we use OCECO as a horizontal governance system for communities, based on proposals, evaluation, financing and project management with an integrated local credit system.

so I'll be happy to be with you for the meetup today.
cheers

I

Ivana Mon 8 Aug 2022 4:00PM

So happy to see potential collaboration shaping up... See you all tomorrow.

R

RobertD Mon 8 Aug 2022 3:04PM

@Oli SB As I posted above, @Jennifer Damashek and I plan to join the call tomorrow!

I

Ivana Mon 8 Aug 2022 12:36PM

Count me in. I probably won't be able to contribute god knows how much initially, but i'd love to listen in. Speak to you tomorrow. Have a lovely afternoon.

OS

Oli SB Mon 8 Aug 2022 12:05PM

No problem! I hope it helped.

And thanks for the offer of help, we need all the support we can get! @Matthew Slater and I are meeting tomorrow (Tues 9th Aug) at 4pm UK time to discuss the new Offers and Wants schema... and how we can move this forward to try to create a decentralised system for sharing details of Offers and Wants across networks. You would be very welcome to join - as would anyone else here with an interest in this. We'll be on BB here: https://ca.meet.coop/oli-1vp-srh

I

Ivana Mon 1 Aug 2022 7:47AM

Hi @Oli SB / Thank you so much for the time you took to answer my questions. The picture is somewhat clearer now and once I work through your links I'm sure it will be even clearer. Interesting work - very well done, congratulations to all involved. If I can help I will - lots of qa experience, some pm, etc. Can volunteer a few hours a week - if of interest mail me at earthlingib (at) proton dot me otherwise we can continue the communication here. Now to read all the blogs and things that you directed me towards.

OS

Oli SB Sun 31 Jul 2022 3:32PM

Hi @Ivana,

These are excellent questions. Let me try and answer one by one...

There's also this blog post which attempts to explain how Murmurations works:

https://murmurations.network/2022/07/12/using-murmurations/

how would the data from Cooperativ and Raiffeisen Asso end op on the 'murmurations agregator'?

Someone would need to create 'profiles' for the Raiffeisen and Cooperativ data (or provide the data to us to import) and then an Aggregator could find the records they want in the Index, and import it into an aggregator.

Would agregator then have itw own website (or app?)

Yes, an aggregator is its own website.

would 'pull' data from other websites without asking? Or would the entities/websites who wanted to be represented on the agregator need to somehow put data onto the agregator website?

For the entities / websites to be featured on the Aggregator they need to be in the Murmurations Index. Anyone can add data to the Murmurations Index (so this doesn't need to be the entities / websites / organisations themselves) via the Murmurations Profile Generator, which can also host node profiles) but if they do the URL of the profile will be something like this https://test-profiles.murmurations.network/profiles/cl5attgay000109jtc5mn5qdk

Obviously only people from the Organisation in question can save their profile data at the URL they are describing (for example The Open Co-op profile lives at https://www.open.coop/open.json and is therefore more 'authoritative' than a profile hosted at https://some-other-urll.com/open.json ) but, if they do so, the new profile data will supersede the data which was added by someone else because it is more 'authoritative' - as I attempt to explain and demonstrate in this blog post and short video.

If so, how much work would be involved in settings it up for all who wanted to be on the Murmurations ag?

Well, we can import large amounts of data (from any open data set) to the Murmurations Index, fairly simply, as we're doing now for roughly 25,000 Karte Von Morgan nodes. So, if you have other open data sets, or know people that do, please let us know ;)

Or could, for example, the Karte von Morgen become a Murmurations agregator ad add data from other entities to its already existing map?

KVM (or anyone else) could use the data in the Murmurations Index to find the nodes they want and present this on a map, or in a directory. If they want to make new data available to Murmurations, they (or someone else) would need to add records of the new nodes to the Murmurations Index.

Say we have hundreds of websites, each with its own blog posts, that we want to agregate on display somehow on one webpage. What would the blog owner have to do to make this happen?

If the blog is running Wordpress, they could use the Murmurations Wordpress Aggregator Plugin to do this, or build their own aggregator (which is obviously a lot more complex). But first they (or someone) would need to make Murmurations Profiles for all the websites they wanted to feature, including URLs for the RSS feeds for the sites in question. But then, once the website's profiles were registered with the Murmurations Index, anyone could access their Profiles, find their RSS feeds, and aggregate their bog posts to create an automatically updated 'news' site featuring the latest blog posts from the hundreds of websites.

Let's say I'd be interested to run an Open Projects Hub aggregator site. What would I need to do to set it all up?

Basically follow the process above (for the news aggregator), but to make an Open Projects Hub aggregator possible, we'd need to make an 'open projects schema' (i.e. decide what data we were going to collect about projects. I've already made a start at this, and it only needs to be done once) and then create some Profiles for the Projects we wanted to feature on the aggregator.

I realise this all sounds a bit complicated, and can seem a bit complex, especially to non-developers, but what we're attempting to build is quite sophisticated... so, to combat that, and to make Murmurations usable by anyone we're developing the Node Profile Plugin and the Aggregator Plugin for Wordpress, which powers ~40% of the websites on the internet... Plus, we've also developed some Documentation to try and explain how to use Mururations and answer some FAQs.

They are all great questions - and I hope my answers help explain some aspects of how Murmurations can be used. We'll be developing more how-tos and videos, so do ask if you have more questions because it will help us develop useful guides ;)

Oli