oxfordhcc / solid-media Goto Github PK

When I run https://oxfordhcc.github.io/solid-media/
i find it gets a stream of 403 (unauthorized) errors from things like

https://jackson.solidcommunity.net/movies/

It should not guess the URL of where the movies might be, but instead lead the user's profile, and then if they had a public type index load that and look in that.

It should also add an entry to this user's type index if it doesn't have one already from say MediaKraken.

It seems like the recommendations can include movies that are already in the collection, it would be useful to exclude those given that users already know about them.

To continue the MPC experiments, we need new hardware.

We also need some new hardware to host public solid web apps.

Can we use Ghent's set-up as the starting point?

update the implementation from foaf to vcard

Hi there!

Saw that you are planning to build cross-app or platform integrations beyond just Netflix (eg: Trakt, LetterBoxd, or other streaming services.

We've just open sourced a single SDK that can help you integrate all of those simultaneously (https://github.com/comake/standard-sdk-js). You could even make or use a common model and use that a unified API for the various streaming platforms. We're happy to help you if you like. Just let me know.

As introduced by Nitin in our meeting on 23 May, a new architecture (/protocol?) is proposed for performing MPC on distributed settings (e.g. Solid-like), calling it PPC evaluating. We are doing benchmarks (based on https://github.com/data61/MP-SPDZ) to obtain empirical evidence to learn:

• How well does the new architecture scale with the increased number of participants?
• What is the bottleneck of the architecture?

Two main settings:

• Homogeneous nodes with same hardware and network performance -- directly by the library, network controlled by tc.
• Heterogeneous nodes with different hardware and network performance -- on different Docker containers, with different network (interfaces) controlled by tc.

Following our discussions on 24 May, we would like to identify what kind of features that are going to provide some new additions to systems like drivers’ seats.

• Can we have all the data from other related systems? If so, how?
• We should aim for something unique and different, such as normalising this data across different work sites, enabling the communication of workers across different sites. What should this unique feature be?

Suggested timeline:

• Mid June: identify all the datasets
• End June: identify prototype timeline for a 4-week cycle
• Mid July: have the project intern start

It is an art getting apps to display stuff quickly. Some ideas, probably missing a bunch.

Make sure you display useful stuff the moment you can, and don't block the thread waiting for other things.

So if my collection is loaded don't wait for others' collections to be loaded to display mine.

Maybe display each movie when it has come in, so the user has something to look at. But then when they have a screen full, then the urgency to load more is less.

Pull in data from more remote places, and larger lists of movies, in the background.

• Display blank card for a movie and fill it in asynchronously as the data comes in?
• Prioritize filling out the blanks which the user is looking at?
• Display progress bars for long loads?

Test it with friends who have a lot of movies and some which have a few.

Hey, I just watched last week's Solid World. Thanks for the shout out :D.

I think it's great to have more media applications in the Solid space! And I specially like that you guys are tackling different aspects as well. I don't think I'll ever implement recommendations for my apps (at least not in the short to medium term), so I think both apps complement each other pretty well :).

So, wouldn't it be great if users could use both our apps at the same time? I think that goes right in line with the Solid spirit. If that works, users can start experiencing the true power of Solid.

I've played a bit with the app, and I think this should be easy to achieve, because I see that we're both using the same schema. I did a couple of tests, and I was glad to find out that if I already had a collection of movies created with my app, they showed properly on Solidflix! However, this doesn't happen the other way around. If a user creates a collection of movies in Solidflix that isn't registered in the type index, it won't work in other apps. That is, unless the apps look explicitly at the /movies container looking for schema:Movie. But I don't think that's the best approach, that's why I haven't implemented it in my apps.

What do you think?

Let me know if something could be improved in my apps as well on this regard. I've been eyeing the Solid Application Interoperability spec for a while, but it's still in progress so I don't think we should use it yet. Type indexes aren't officially specified either, but as far as I can tell they're the best existing solution and the de-facto standard for interoperability in Solid.