Proposal for converging Semeoz work and Transformap wiki of maps

Tags: #<Tag:0x00007fb71be380f8> #<Tag:0x00007fb71be38df0> #<Tag:0x00007fb71be3a600> #<Tag:0x00007fb71be3b4d8> #<Tag:0x00007fb71be37ea0>

(Jon Richter) #10

The discription of the background above appears a little vague for my likeliness. Please allow me to recur on some of the points made.
Already from the title can be depicted what the suggestion is here: To integrate two bodies of work and to equal out their difference for mutual rapprochement.

In that sense the understanding of the case given is determined by the objects we are looking at: Is it to merge two up until now conflicting visions of a mapping inventory, or do we rather infer our approach from that case and generalise it into a strategy for not only federated geodata, but data catalogues.

Dig into for ideas.

Synchronisation is resourceful and always comes at a price.

Experiences from distributed systems engineering confront us with a high unlikeliness of successful easing here.

The challenges surrounding immutable infrastructure involve not the pattern itself nor the implementation in the runtime, but rather the process, human organization, and tooling that needs to be in place.
What Are the Central Challenges with Immutable Infrastructure?

The question may then transpose into how to express a shared understanding about (1) what a TransforMap Atlas is and (2) which minimal agreements have to be made for interoperability of those?

If you have read

you will have witnessed a report on how processual premisses give direction to such a synchronisation effort. Up until now I am only aware of

to tackle the organisational challenge imposed by merge conflict resolution.
Then instead of CKAN related solutions, we may also think of

Where do those go and how can they be kept useful for the wider context?

This can be automated.

I suggest we move the pieces of the conversation above which entail Librepay into a new topic.

Use data from different Semantic Mediawikis and also Communecter's database
(Maïa Dereva) #11

Why conflicting ? Could you explain where there is a conflict ?

(Jon Richter) #12

Conflicting in the sense of different, non-aligned implementations. Coming from

we are interested more in raw, usable data rather than catchy interfaces, knowing we also need the latter to communicate to a wider public.

While the Semeoz XML export contains mainly WordPress related metadata and else only a link and a title of any of those maps, the transformaps implementation via Semantic Mediawiki provides structured data about the entries available.

But maybe the work of @gandhiano on the IISA food sovereignity indicators can inform this discussion about how to enrich WordPress’ Custom Data Types with standards compliant vocabularies and data formats?

(Maïa Dereva) #13

Exactly. For me there is no conflict. Just a lack of information and of time. I can create all Custom Data Types needed if I’m asked to in order to reach an “aligned implementation”. Maybe be i will need some technical help but I really think it’s important to think about “catchy interfaces” AND usable data. Data is not usefull to me if it doesn’t tell a story…

However, in my mind, the best way to align implementations is to develop a simple PHP request able to displaying data from wiki directly in wordpress…

(Jon Richter) #14

I am happy with all the curation, visualisation and publication related activities from this proposal, yet I have strong concerns about the proposed im- and export leading to a synchronisation of datasets.

I would propose @gandhiano to investigate with his PHP coder network how to employ

to bypass the technical difficulties I have pointed out above. I would then donate the neccessary thumbnail generation logic, however implemented, to this bucket. Because we also have a special interest in that. A decoupled webservice, shared by the ecosystem, comes to my mind.

In my aspiration this would be yet another conversation which involves structured data on the web and client side embed logic. This has been explored at the fringes of the P6Data Hack Day together with @mattw et al. and later turned into

Finally sheds a completely new light on this conversation. Have I ever been mentioning CKAN before?

(Amaury_van_Espen_van) #15

May we could purpose, more than PHP only, though the Semantic Form from Jean Marc Vanel
a semantization of content =>
In order to store properly data

Is there a Smeantic Media Wiki plugin implemented into Transofrmap Wiki ?

Thanks to everybody

please join @almereyda @alabaeye @johan @Maia =)

Open real-time collaboration protocols for open communities. Avoid Slack.
(Jon Richter) #16

First few quick remarks:

Just to mention, I have been warned in the last years of his understanding of how and why to use linked data. Having said this, do you know of a running demo instance of that code?

Yes. See

One may want to compare this listing to


Now that this bucket has been funded and as said above

Could someone carrying out the technical work please scope how

  1. the data pipeline would look like in the desired bidirectional scenario?
  2. this preliminary work could prepare for a broader federation of atlases?

The following links could then be of further interest to us:

@toka Can we implement a basic DCAT-AP for the Collection of mapping resources?
Unfortunately there doesn’t seem to be any prior art based on Semantic MediaWiki, despite the site documenting this fact runs on it.

(Maïa Dereva) #17


first I must say it’s very hard for me to understand the technical exchange… so don’t worry if I don’t answer sometimes.

If I read carefully the initial proposal, I just see a work of cleaning, completing, exporting and importing datas.
It’s not a technical project of real synchronization that could come after. It’s a first step to a future permanent link.
Am I wrong ?

Anyway, that work would be useless if we don’t walk the second step… :slight_smile:
If you think about a meta-solution (like a broader federation of atlases) that could be a Wordpress plugin able to display datas coming from the wiki in any site who wants to display them, with options that user can check (For example :“I want to display only maps from France about cycling”). What do you think ?

(Maïa Dereva) #18

About the project itself, who is going to work on it in Transformap team ?
@Amaury_Van_Espen, are you OK to be part of it as Semeoz team ?

I opened a kanban :
Tell me here if you want to join it.

(Thomas Kalka) #19

I would like to join.

(Jon Richter) #20

I feel a bit misunderstood here, but never mind. It is not my project and I only tried to help with expertise - maybe a little too abstract and technical. Yet I would feel a lot better if some person knowledgeable about data curation and associated infrastructures could at least try to reply to the points I made.

I believe this will include to introduce the notion of an Atlas next to Maps in the directory.

(Maïa Dereva) #21

Yes, I feel sorry @almereyda; it’s a pity but I can’t answer you as I don’t understand your questions… :frowning:
I was clear on that point at the beginning: there is no technical people in Semeoz team. The only thing I can do is to let you access freely to wordpress and to the data… And i can tell again that I am ready to input datas directly in the wiki if you help me to synchronize the wiki and the wordpress… I think this kind of work could be usefull for many people, not only us. For example, at the assemnly of commons, we are looking for that kind of semantic-wiki/WP connection to create a directory of commoners…

@toka: I go to holiday until the 28th of august. I’ll stay connected if you have questions but I won’t work a lot.

(Jon Richter) #22

I am currently preparing my stay with the

crew who also use Semantic MediaWiki. Maybe this could be a chance to discuss using it for DCAP (Atlases, Maps, ESSGlobal, …) storage and WordPress as a viewer for that?

(Maïa Dereva) #23

Yes @almereyda, it would be great to include wordpress as a viewer. WP is a very effective way to create nice sites for many people, and I think it’s important to think about displaying datas too (we are not all developpers).
I’m sure something is possible, even with existing plugins like this one and/or with the help of Virtual Assembly WPLDP plugin.

Also, as I am the principle french user of the P2P Foundation wiki, be sure I’m very interested in the event in Montreal…

(Adrien Labaeye) #24

@Maia @toka: you are lead on this bucket. My two cent: it is great to receive input, but be sure you implement this at a level of complexity that you two can manage. Or make sure concrete arrangements are made with third parties. Rather make it small, simple, and well; than too ambitious, and sketchy. At the end of the day, it’s only a small bucket. :slight_smile:

(Maïa Dereva) #25

Hi @toka, I won’t have a good internet connexion until the 5th of september. So I will start working only during that week.

(Jon Richter) #26

It is developed by our friend from earlier times @acorbi

8 - 10 nov 2016 ⯇ reappropriating transformaps ⯈
(Jon Richter) #27

2 posts were merged into an existing topic: - Semantic Mediawiki

(Gualter Barbas Baptista) #28

How is the current state of development of this project? Is there any place beyond this thread, where we can follow and interact with the plan and ongoing development?

(Maïa Dereva) #29

Hi @gandhiano : the project is starting slowly. I created a kanban here :
And I wait for @toka’s news to decide who work on what.