Repository: https://github.com/solid/specification - Technical Reports: https://solidproject.org/TR/ - Code of Conduct: https://github.com/solid/specification#code-of-conduct
Thank you again @elf-pavlik , the 303 redirect short web id solution works very well: https://frederick.trinpod.us/@
@gibsonf1 in https://frederick.trinpod.us/profile/card you should have statements about WebID with
https://frederick.trinpod.us/@ as subject, with this approach https://frederick.trinpod.us/profile/card#me would not be used any more
i get they are probably internal, but why expose them publicly?
4 replies
@elf-pavlik So completely removing #me from the picture would still be spec compliant as long as profile/card is in use? (My worry was that devs would hard code /profile/card#me)
if devs hardcode something like that their application wouldn't comply with the spec, /profile/card#me is just convention specific to internal choice of one implementation, clients have to "follow their nose"
I like how you put the code on the URL and the databrowser just loads the code on the page @ /profile/card#me because the databrowser loads it up like this https://portal.inrupt.net/?uri=https%3A%2F%2Ffrederick.trinpod.us%2Fprofile%2Fcard%23me
@mikeadams1 How does the databrowser work, I just see this and can't get anything else from that link:
tried to load https://frederick.trinpod.us/profile/card, got this error
InvalidPropertyURIException: https://ontology.graphmetrix.com/node/2
RDF/XML writer cannot figure out out to split such URI
it attempts to use
https://ontology.graphmetrix.com/node/ as the base URI, but then 2 is not a valid XML local name (needs to start with a letter)
we have a simple solution in that all of our ontology, including that 2, has a conceptual tag that is compliant, so instead of 2, it would be m_tag
will add that translation in
We are also looking at releasing the neo lower ontology to the public as well, which captures states/attributes/events/process/function/result, in which case the turtle for https://ontology.graphmetrix.com/node/1bpk would be neo:i_attribute or https://neo.graphmetrix.net/node/i_attribute
not the code you have on /card#me but the actual page it should produce
It doesn't seem to work when I try
@gibsonf1 although the mashlib data browser was able to resolve https://frederick.trinpod.us/profile/card#me was there a reason that you designed it the way you did? was it to prevent anyone using another web browser from being able to resolve the page?
I am trying to follow your work but it is just a little bit over my head...
@mikeadams1 I just changed things such that webid remains @, the radix nodes are prefixed with t_ and the ontology exports with conceptual ids rather than radix to solve the parsing issues @namedgraph_twitter found
@mikeadams1 The goal was simply to let people have an easy to remember and type webid.
thats what i get on chrome
I get the same thing on my cell phone with safari
I had to add the explicit @base there as solid-client tried to resolve the turtle nodes with the pre-redirect @ node
So are you creating your own browser, or is it that one needs to be signed into trinpod before any of the pages will resolve to something other than code?
my only thought is...well not the only one... but, if mashlib cant resolve the page...why not?
@dmitrizagidulin how about we transfer solid/identity-panel#1 to solid/specification repo? (Need to unarchive first and then transfer the issue.) I'm not sure if there is going to be enough material or even resources at this time for a whole panel on identity, but that's for participants to decide. If you want to revive, lead it. If in the foreseeable future it is just about having the Solid ecosystem incorporate DID or WebID being compatible with DID, then maybe issue in solid/specification will suffice.