Re: public interfaces for web of sensors

On Saturday, November 16, 2013 at 3:24 PM, Paola Di Maio wrote:

> Thank you Marcos
>  
> > We are basing the design on the extensive knowledge/experience of the nodejs community.
>  
> uhmmmm. are the nodejs the main target user group for the api?

Absolutely not. Our main target is Web browsers (hopefully Firefox/OS, Chrome, IE, etc.). It’s just being informed by the Node community as they have a wealth of real-world experience to share that browser makers currently lack.    

> I'd say, when you are ready, we could help to profile the user base for this  
> project, and perhaps organise focus groups on selected aspects of the api architecture.


We are still at the use case and requirements gathering phase, so now is the best time to get involved before any implementation occurs. However, once we get a prototype implementation up and running, then it would be great to have some more external folks try it out.  
  
>  
> > > Is there anything we can do to make the sensor web more
> > > accessible/useful?
> >  
> >  
> > Well, the fact that the browsable web can’t access any sensors without having to go through some intermediary is a problem. This means that many synchronous (real time) applications can’t be created.
>  
> ok, let me understand the problem. you are saying that software that comes with the sensors does not have interfaces? I think I have played with browser based sensor interfaces before.

I’m saying that those APIs are not enough and too mobile phone specific. So yes, much of the work of the W3C's DAP working group is making its way into browsers: e.g., proximity sensor, battery status, etc. However, this is fairly limited when compared to something like what you can do with the Johnny 5 JavaScript library:  

https://github.com/rwaldron/johnny-five

> >  
> > So, helping us actually making it possible to interact with hardware that interfaces with sensors would be mutually beneficial to both groups: as, at the moment, any semantic sensor data you guys might be getting can’t actually come from the users of the Web (only for intermediaries or otherwise indirectly).
>  
> are you absiolutely sure about this?? If you are, then I eed to do more homework before working more on this

Yes. I’m sure :) It’s why we started this group. Happy to provide you with more informations if you need it.  
  
> if so, that's the missing link to the semantic web. then we can have the vision finally up.


That’s the dream :) Open up new opportunities for innovation - particularly over BlueTooth, USB, and any other standards that are able to expose a serial port interface.  

> >  
> > We are just getting started - most of the discussion is happening in the Github repo:
> > https://github.com/whatwg/serial/issues
>  
>  
> OK- not sure if github is the best place to discuss for us 'users' but we ca pick up from there and roll on here or something
Happy to discuss aspects here also. However, GitHub’s collaboration tools are generally more effective than mailings lists - we encourage you to sign up and simply subscribe to the repo to watch the discussion. If you need help getting started there, happy to provide guidance.  

> will continue to follow closely
  
Great to hear :)

Received on Saturday, 16 November 2013 18:05:43 UTC