Notes and references: in English
At the end of the meeting, this will be replaced by the notes taken in the following pad:
InterMapping Notes EN
This pad is intended for note-taking at the Intermapping Session on 30 June 2020.
Notes and References in English
1. Qualification of the Commons and their Organization
Commons, standards without top-down organization.
Look at research into decentralized web. Also Semantic Web. Currently ActivityPub is a standard for federated/decentralized network building on Semantic Web.
2. Venues to Discuss Interoperability
What is interoperability?
[TODO ADD QUOTED DEFINITION]
In Free Software Mapping community interfaces are not well understood. Everybody is doing there own thing. We need to understand the interfaces.
Goal: no restriction on kind of software that can interact with data.
You need diversity. No single project will succeed. We need to account for all approaches for the commons.
We can not lead others, we can only inspire others.
There are internet bodies (IETF, W3C) leading the direction of the web by defining standards.
IETF has a very complicated process. Lot of reading/writing every day over years. A long-term process. Also defines names IANA (URIs etc.) - kind of the ISO of the web.
W3C defining standards such as Linked Data, ActivityPub, Solid.
Social Community groups (of the W3C) have been able to publish standards that have become web recomendations (e.g. ActivityPub).
Usually an individual comes up with an idea and it grows from there. After years - even if not going trough internet standardization communities - there is a benefit for standardization.
Many Mapping communities have their own venues (Virtual Assembly, [others mentioned]). INCOMMON comes in not as another venue but as a federation of discussion venues. Has been doing:
- Collecting ideas
- defining models for interoperability
Tools used by INCOMMON:
- Intermapping Matrix channel for chat
- public pads (like this one)
- Mumble server (voice conversation)
- Resources from different librehosters (BBB from fosshosts)
3. Practical Interoperability with Linked Data
OpenEngiadina is to acknowledge that social interaction is also meaningful interaction.
Descriptions posted as local knowledge are difficult to present into a relational model. They all have their own language. One solution would be to use a graph as a data model, there is established standard as to do this RDF is the one we look at.
By using this model we render data immediately interoperable, activity streams alllows us to add the social interaction, ie this person has seen this shared theis liked this etc…
Linked data is a model that allows us to capture an amazing amount of information and make them interoperable it is already used by many institutions for example museums
Wikidata allows to reference comments
Transiscope is a workaround to interoperate the data. Semantic bus is an ETL (extract transform load) hosted at grappe.io. Everyone can create an account it becomes payment when you load a large amount of data.
When the data is modified at the source it can update
It can also Transforming non semnatic data into semantic data
The Challenge of Naming Things
How to do this:
Importance of naming things, using URI and name them by the hash
Using containers wich content… append only logs (CRDTs, DAT, SecureScuttleButt)
The difficulty of naming: how to deal with the challenge of naming a single thing that has two names?
Yes, the process of naming is inherently social.
ActivityPub is useful to make the naming social, a conversation.
AV / Transiscope: one database?
Transiscope is a not a database its more of a buffer that is updated everyday at 20hrs
Updating Data Across Databases
In Transiscope : need to have access to the data source to update the data.
It’s a workaround, we’re working on a solution to update the data in Transiscope to update the source.
In ActivityPub, every source can have an Inbox, so that the original orginal author can receive updates automatically.
In the case of different views then translating to ActivityPub by different streams you follow each stream would be a view
Using visual tools enables non-technical people to share responsibility more fairly.
ESS: Economie Sociale et Solidaire / Social Solidary Economy
Beyond an ESSGlobal foundation to maintain a vocabulary, Yasu proposes to create a whole software ecosystem across the world.
Can we prevent wasting resources on reinventing the wheel?
We need more note taking ~60-65 minutes about global consortium/unions.
How to maintain ESSGlobal? Maybe take inspiration from existing processes (SRFI, PEP, XEP…) @pukkamustard is interested in studying how this could work for this and other vocabularies.
Data Food Consortium
Vocab: e.g., by design, existing databases could not cooperate on tomatoes for the lack of common definition of what a tomato is. So the OFC managed to act as a convergence point to make a common vocabulary on tomatoes.
Create Semantic Apps from Forms
Vocabularies define world views. With Linked Data you can practice logical inference that allows compatibility with existing vocabularies and data.