How to solidify Solid Project #783
Replies: 3 comments 1 reply
-
|
What do you think if the website would be hosted in a Solid pod with all the content, html,css, templates and the relevent data would be pulled in from RDF sources? Editing possible via a CMS. Access and editing would be enforced by access to the Solid POD. |
Beta Was this translation helpful? Give feedback.
-
|
There is this debate about serving HTML from a Pod: https://forum.solidproject.org/t/is-it-secure-for-pods-to-serve-html-files/6379/9 |
Beta Was this translation helpful? Give feedback.
-
|
#248 + https://github.com/linkeddata/dokieli + magic would be one way. |
Beta Was this translation helpful? Give feedback.
-
|
I beleive the predicates below are sufficient for describing pretty much all of the contents of solidproject.org. I propose this as the basis of a shape for organizations and specifically as the shape for Solid Project and Solid Team and as a basis for RDFa on the site itself. To be addressed later - predicates that Organizations share with Persons, relationships between organizations, and more on types of organization. Organization Description
Organization Structure
Organization Deliverables
Orgainzation Stateholders
Stakeholder Deliverables
Notepto: is the prefix for the Product Ontology `http://www.productontology.org/id/. This ontology is a collaboration between Wikipedia, Schema.org, and GoodReations and leverages Wikipedia's many languages. |
Beta Was this translation helpful? Give feedback.
-
I began a discussion on dogfooding solidproject.org a while ago, but it's time for a fresh look. My own vision of this looks something like the following model, implemented so that machines can follow it to the RDF and humans can access it as a website.
Here are some of the issues that need addressing :
How do we generate the website?
Currently it is done with Jekyll but as my recent assessment of creators shows, the creators (myself included) are somewhat lacking in knowledge of Jekyll. I would like to see this eventually be done by our own scripts which generate the site using Solid Technology and taking advantage of existing Solid RDF data such as lists of team members. I have created several previous versions of such a method, demoed them, got some discussion going. I've applied this to RDF generated HTML-RDFa for team-members, glossaries, repositories, specifications, and ontologies. My most up to date version of a library for componentized website/app design - solid-uix - uses SolidOS as an example rather than Solid Project, but it can be, I think, adapted. You might also want to look at an earlier demo of an RDF-generated solidproject.org.
How do we edit the website?
Currently this is done with Github PRs. Both @timea-solid and I have worked on parts of this problem. She has a POC on creating a backup/restore system and I have been developing a series of editing tools (WiP) which allow editors without RDF knowledge to edit the RDF sources and see in real time how the changes look in their eventual HTML representation using editor selected templates. This relates to how the Creators edit solidproject.org but also to the more general question of the relationship of front-end developers and content providers with RDF.
Where is the source of truth and how is it stored?
If we're going to base solidproject.org on RDF we need a plan to know where the RDF is. Currently it's spread over HTML-RDFa pages, RDF documents on github, RDF on pods and other sources. Take the Solid Team info - the only actually complete and up-to-date (ish) data is hard-coded in HTML on the website. There is an incomplete RDF version in one of timbl's address books. There is the github list of who controls what repo. You get the picture ... We need some level of standardization and a way that editing the RDF gets reflected on the website rather than the current disconnect between the website and the data it is based on.
Which Ontologies?
The question of ontologies used on solidproject.org is relevant, see the discussion at solid/team#15 and solid/team#16. We also need an ontology to represent applications and libraries and pod-providers, etc.
How should organizations employ RDF to present themselves?
@timbl has been approaching this question from several directions. This article touches on how organizations define themselves. He has introduced a new predicate
solid:communitywhich used to interconnect organizations and their members with triples like<#me> solid:community <someOrganization>and a plan to use this to, for example expose the contacts lists and meeting-trackers of the organization to its members. See also my article on the topic of the RDF modeling of organizations.The future
Recently @danielbakas has expressed an interest in helping to pursue this vision. @jaxoncreed has mentioned the possibility of future work on a UI for solidproject.org. I and the SolidOS team will be pursuing the use of solid-uix as a components/templating system and its use to create access to RDF for front-end developers and content-creators without them needing advanced RDF knowledge. The Solid Team admins are progressing on moving solidcommunity.net to CSS and solidproject.org to solidcommunity.net. And ____ [please fill in the blank].
Beta Was this translation helpful? Give feedback.
All reactions