POC applications for identity management and RDF data

Bingo! So how easy do you think a visualization of this would be to create? Simplified is the key to the adoption. Endless number of areas this could be applied today.

4 Likes

It is still sinking in but I would imagine that a talented illustrator could easily do the job. Would make for a great short video clip.
Reading about it definitely did not make the penny drop quite as hard as it should.

4 Likes

I love the concept, but where does fine control come in? What if I want some of my posts to stay on reddit, and others to be on steemit, and some to be both?

What would be amazing is when I make a new post or something, if I could have a multi-dropdown of “where do you want to post this to” eg: “what service should have access to this?” Is this something I could change after the fact? Can I post on patter but not not-patter in subsequent posts?

Brings in a whole new dynamic. I still need to learn more. But if my reddit and steemit posts both get posted, but someone wants to reply on reddit, but not steemit, and someone else is a flip flop of that. Others don’t know steemit exists and might be pissed if their post shows up there without them knowing.

I’m really just trying to understand how these new data structures are going to work. This is such a huge change from my SQL backend silo way of thinking.

3 Likes

Agree. Perfect.,…

1 Like

You need to think differently. Your posts will not be filtered by the service, but by who you decide can access them.

If they are public, anyone can access them through any app (or service) they choose. Or putting that the other way around - you can choose to view any data you have access to through any app you choose - which is what Gabriel just demonstrated in letting people switch from using Patter to NotPatter.

It’s a different way of things operating which takes control away from the service / app and returns it to the publisher (you) and the consumer (people you choose to give access to what you publish).

Patter and NotPatter are just switching how public data is viewed / created. But next you add creaton of groups and decide which groups or individuals can see which of your posts.

So that is when you get to decide, not what appears where (reddit or steamit), but what is visible to whom (me, members of SAFE group, everyone etc).

17 Likes

Really cool.

One thing about the data you need to enter though. Does it have to include both full name and nickname at the same time? Most of the time I would prefer to have either my full name or a nickname, but not both connected to a webid.

1 Like

There is Webid,www,but webhosting software is asking for Safe Network Public Id.
Why so many ids?

You could have two separate webIds if you want truly separate inboxes, is one option.

Otherwise as noted by @happybeing, it’s less of where it goes, but which ID it’s associated with.

so:

Others don’t know steemit exists and might be pissed if their post shows up there without them knowing.

This is more just I’m messaging a user… irrespective of the platform.


I don’t see this as being a hard fast rule, but more perhaps something that apps could implement:

If they want to only display messages from their service, they could add something [ideally a standardised field], to a message RDF. And then filter by this to only show those messages.

The flip side is, any other app could happily ignore this field, so you can still see all you’re messages wherever you choose to.


Right now, we’re opted to enforce ‘name’ and ‘nickname’, tbh, it doesn’t have to be this way. The only one needed in any sense is the nickname as this is (currently) used for the dropdown in Peruse…

3 Likes

@Nigel @happybeing @anyoneElse

If you’re having issues with Peruse not showing WebIds in the list (not authing), please 1) , make sure that you auto-auth (the wee padlock) turned off (locked), this should at least let you see when the error is occurring. 2) You should only need to close/reopen the browser and this issue should (probably) be solved. The unlucky may see it twice in a row. BUT>…

When the error does happen, can you please grab the peruse logs (as soon as you see the failed auth). Grabbing the peruse logs, as opposed to safe network logs is described here: IPC error: StringError(Utf8("invalid utf-8: corrupt contents")) · Issue #96 · joshuef/peruse · GitHub

You can post them here, or there. Then hopefully we can get to the bottom of this.

If you’re able to reliably reproduce this error please let me know

2 Likes

@joshuef Not able to use atm but just a quick note to say that I had auto - auth set when the problem started and turned it off to see if that was causing the problem.

1 Like

I think auto-auth just masks the issue (you dont see the error message).

The most annoying thing about this bug is how randomly it occurs. I’m pretty confident you’ll see it again… at some point. :frowning:

I have an inkling about what causes this though and just testing something now… fingers crossed. If i don’t see the message locally for the rest of the day, I’ll update in here.


Update:

It’s not what I thought. Back to the drawing board on this one.

2 Likes

I’ll try again tonight but since yesterday on multiple attempts I can’t get Peruse to pop up for reauth. I’ll create a new account and start fresh to see if I can get the logs!

I have a feeling that web-id have potential to be a real game changer, “one account to rule them all” which gives users so much potential value in effectiveness when time comes to switch from todays web to SAFE.
@joshuef @hunterlester @bochaco

Will there be an option to lock web-id so you can open many tabs, so that the web-id you choose to lock will follow to every new tab you open and stay “signed in”?
To copy/past text (mouse) and scroll zoom with (ctrl+mouse scroll), in Peruse, will that work in the future, don’t seem to work for me in the moment.

6 Likes

Should do. We’ll be gong through some more stabilisation milestones and adding in basic missing features in the near future.

I could certainly see this being a useful feature. Setting a ‘default’ id. Or perhaps associating an ID with a url by default…

6 Likes

I have the same error as @Nigel.

imagen

Yesterday all was OK. This morning, in other computer, no problems, but this night the error persist even close/reopen the browser several times.

The Log:

T 18-07-31 00:16:12.285699 [ :147] Creating unregistered client.
T 18-07-31 00:16:12.287701 [crust::main::service service.rs:556] Network name: Some(“alpha_2”)
T 18-07-31 00:16:13.290368 [crust::main::service service.rs:82] Event loop started
T 18-07-31 00:16:13.290368 [crust::main::config_refresher config_refresher.rs:44] Entered state ConfigRefresher
T 18-07-31 00:16:13.290368 [ :1395] Waiting to get connected to the Network…
T 18-07-31 00:16:14.374088 [crust::main::active_connection active_connection.rs:63] Entered state ActiveConnection: PublicId(name: 34e3eb…) → PublicId(name: 3a602f…)
T 18-07-31 00:16:14.374088 [crust::main::active_connection active_connection.rs:110] Connection Map inserted: PublicId(name: 3a602f…) → Some(ConnectionId { active_connection: Some(Token(24)), currently_handshaking: 0 })
D 18-07-31 00:16:14.374088 [routing::states::bootstrapping bootstrapping.rs:266] Bootstrapping(34e3eb…) Received BootstrapConnect from 3a602f…
D 18-07-31 00:16:14.374088 [routing::states::bootstrapping bootstrapping.rs:332] Bootstrapping(34e3eb…) Sending BootstrapRequest to 3a602f…
D 18-07-31 00:16:14.412113 [routing::states::client client.rs:91] Client(34e3eb…) State changed to client.
T 18-07-31 00:16:14.412113 [ :1413] Connected to the Network.
D 18-07-31 00:16:54.196445 [ :44] ERRNO: -205 IpcError(StringError(Utf8(“invalid utf-8: corrupt contents”)))

1 Like

Perhaps it could work a bit like container tabs in Firefox? With color coding etc to easily tell which is the tab is using

4 Likes

I try in other computer and work OK.

Strange, the main difference between both is the windows version. Windows 8.1 (error) vs Windows 10 (OK).

A small update to the browser builds (i will update the OP too):

Fixes webId urls such that safe://myid.atjosh/#me works without rewriting to an invalid safe://myid.atjosh/#/me url.

Adds a link to the authenticator to the webId dropdown/

12 Likes

Do we have an up to date list of profiles?