Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Amritanshu jain
    @jainamritanshu
    @cristobalmedinalopez Also as mentioned earlier I had a project. It was a simple Location-Based Chat written from scratch. The sockets are written in javascript. I have included cache management and pagination too for optimization purposes. Latch, If time permits please go through the application once, the registration won't take more than 2 minutes. It makes interest based rooms based on your location. Also you can explore the application if you have more than one registration to see how the peer to peer interaction works in this application.
    Buddhika Jayawardhana
    @buddhikajay

    Hi All,
    I am interested to build a webrtc virtual room with following features

    1. Collaborative Whiteboard
    2. Slide Sharing (May be only pdf in initial stage)
    3. Desktop Sharing
    4. Video and Audio Streaming

    I like

    1. The design of https://github.com/Kurento/kurento-room and
    2. The functionality of https://simplewebrtc.com library
    3. The user friendliness of https://appear.in

    Also it should be easy to deploy. May be we can use docker-compose to package the entire application including a signalling service and turn server.

    I would like to know the alignment of my idea with this project.

    Rakshith G
    @hehaichi
    Hi there @cristobalmedinalopez and @vicente-gonzalez-ruiz , I'd like to work on this idea. I was wondering what all features of P2PSP should this project implement apart from the ones mentioned above?(I'm talking about the set of rules).
    Cristóbal Medina López
    @cristobalmedinalopez

    Please, keep in mind the essence of the idea:

    The P2PSP virtual room application.
    Explanation: A virtual room where friends share videos among them in real time directly over the web browser, with synchronized playback and a video chat at the same time. It is like to be in the same room watching a movie but in a virtual way. Browsers are connected each other over a P2P overlay (using WebRTC), this means that we have a private communication, without going through a server. It's a browser-to-browser connection.
    Expected deliverables: A website where an user can share a video with his friends in real time, with synchronized playback and they can have a video chat at the same time. All of this over a P2P overlay.

    New features are welcome, but the main aim is to achieve the expected deliverable proposed in the idea.
    @vbnlohani you could read about WebRTC (and do some experiments)
    @jainamritanshu it sounds good. I'll try to check it out.
    Cristóbal Medina López
    @cristobalmedinalopez
    @buddhikajay your idea is great, but it's not exactly the same that we need. See description in P2PSP ideas page: http://www.p2psp.org/en/news/77-gsoc-2017-ideas-page (number 11)
    Amritanshu jain
    @jainamritanshu
    @cristobalmedinalopez Thanks for replying back. I just wanted to know would it be good if I made a prototype of what is in my mind so that you can suggest what I am missing or any improvements?
    Cristóbal Medina López
    @cristobalmedinalopez
    @hehaichi It would be an adaptation of the DBS.
    @jainamritanshu of course, it would be interesting. Maybe a mockup with the elements and their behaviour.
    Amritanshu jain
    @jainamritanshu
    @cristobalmedinalopez Sure! I will get working on this :smile:
    Vaibhav Lohani
    @vbnlohani
    @cristobalmedinalopez OK I will also start implementing a prototype of the idea
    Vaibhav Lohani
    @vbnlohani
    @jainamritanshu maybe we can work together on it
    Amritanshu jain
    @jainamritanshu
    9160ab7f35d5418a96ee5846c0a2e845.png
    @cristobalmedinalopez this is just the initial mockup about what I had in my head
    Just to explain a little
    Amritanshu jain
    @jainamritanshu
    Those avatars represent the live videos of the people in particular room. Just like a hangout with live streaming video. A user could send invites to his friends, which gets in the notification tab of a user. After Semester fun! :D is the title of the virtual room made by the user inviting others. We can include things like personal chat pop up with a person when a user clicks on the avatar of the person in the virtual room itself. That is all I could think of for now. I would love some suggestion from your side, like things I maybe missing or some more things we could add.
    Cristóbal Medina López
    @cristobalmedinalopez
    Ok, it looks good. It could be a first approximation. Maybe more information such as number of users, playback status, etc would be interesting.
    Aseem Shrey
    @LuD1161
    I would like to work on this project , I did try something similar in a hackathon last year but didn't know about it untill now.
    We were planning to make virtual classrooms where while the teacher is teaching the students can draw on the video in real time and so the teacher understands where the doubt is. We were using canvas for rendering video and so we could also draw but there were some issues related to HTTPS webserver certificate and chrome didn't allow video conferencing or video chatting on insecure channels
    it was only working on localhost :(
    Would love to contribute to this project
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    Hi @LuD1161 . Welcome! Feel free to star working on this idea or a variation of it. Comments can be done here. And thanks!
    Cristóbal Medina López
    @cristobalmedinalopez
    Hi there! @hehaichi asked me about the stream synchronization in WebRTC.
    At first glance, I think that nothing special is required. We could just share the current playback time between browsers. But, What do you think (@/all) ?
    Amritanshu jain
    @jainamritanshu
    Hi @cristobalmedinalopez I guess we discussed about it previously, I saw a couple a synchronized playback examples and I thought sharing the time would be the most efficient way to do so.
    {"type": "pause", "time": 21:24, "UserID": 105623} //When a user changes the playback
    {"type": "play", "time": 21:24, "UserID": 105623} //When a user plays the video again
    This way we can also keep a track of playback with the time of play/pause and the user who did it, and that we could show in the history of a virtual room
    Amritanshu jain
    @jainamritanshu
    Hi everyone! My exams are over now, and I guess we can start working on the project by discussing about the structure of the project and suggested changes in timeline if any
    Cristóbal Medina López
    @cristobalmedinalopez
    Tushar Aggarwal
    @tushar-97
    Hello everyone!
    I am a prospective GSoC student and have been directed to this channel for working on this project - "WebRTC implementation of a NAT traversal procedure based on port prediction."
    @cristobalmedinalopez The slideshare link was helpful, but I have a question.
    Tushar Aggarwal
    @tushar-97
    "TURN servers does not follow the distributed communication scheme of P2P networks". This is from the project description. What does it mean?
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    Hi @tushar-97, that means that TURN is not an option for real (and possibly massive) P2P overlays.
    Tushar Aggarwal
    @tushar-97
    Please forgive me if I ask doubts which are trivial, but doesn't ICE work behind symmetric NATs. Why is the project asking us to develop a system with multiple STUN servers. Is it because it is more efficient than using a TURN server?
    wei-apan-v10.pdf
    ^Also I found this paper to be an interesting read for port prediction
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    No, ICE (by itself, without using some kind of ports-prediction) can't connect (directly, without using a TURN server) two peers that are behind a symmetric NAT (SN). In general, the ICE candidates generated by standard signaling servers do not consider that two SN-ed peers will use different ports (to the used for taking with the STUN server(s)) to communicate each other. For this reason, we want to extend ICE's functionality (that, if I'm not wrong, is implemented in WebRTC) to perform port-prediction and solve some SN combination traversing cases. Thank you for the paper!
    Tushar Aggarwal
    @tushar-97
    Ah, I get it know. This is extremely interesting and something I DEFINITELY want to work on. Is there any material you would recommend I read to form a strong proposal or am I better off exploring my own interpretations?
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    Sorry, but unfortunately I don't know any good (or at least reliable) resource that I would recommend. However, if you google webrtc ice you will find a lot of information that can help you to do your proposal. Good luck!
    This can help you.
    Tushar Aggarwal
    @tushar-97
    Thanks @vicente-gonzalez-ruiz for all your help and answering my (basic) queries. I look forward to giving back to the community
    Freakston
    @Suraj1006
    Hello everyone !
    I am a student looking forward to work on this project.
    Is the project being worked on by anyone else right now ? If no then I would like to work on it.
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    Hello @Suraj1006. There is no one else working on the project at this moment. Welcome to the P2PSP.org!
    Freakston
    @Suraj1006
    @vicente-gonzalez-ruiz Thanks for the reply and I hope that I will be able to implement the project. I have started working on the project and will contact you if I need some information or have queries .
    Vicente González Ruiz
    @vicente-gonzalez-ruiz
    :+1: