RIVACmd Cross-Platform Standardization Discussion

[SamVimes2]

DA,

Late in the 4.1 beta process, Dean added a RIVACmd Action, that sends arbitrary information to a RIVA client.

The goal of this thread is to suggest some standardized forms for commands in the hope that various extant and future RIVA clients will process them in the same way for a consistent user experience.

I'll try and keep this first post updated based on any discussion here, but hopefully we'll get consensus pretty quickly around new commands and Dean can move them into the documentation.

The first one under discussion is the "web widget kludge" to let the RIVA client know to render a native web browser window, similar to how the IV renders an IE window.

Suggested form:
Command="OpenURL"
Value1=URL
Value2=x1,y1,x2,y2
Value3="SuppressDialogs=n,AutoRefreshInterval=m"

[bryanb]

This may not be a direct answer to the question "other RIVA Commands" but more of a request for functionallity. I use the current iOS RIVA client in a couple of ways. One is from my iPad as a local client in my house. I also use the iPhone in a couple of ways. One is remotely accessing the house and the other is in a commercial system I sell to manage AV equipment in sports bars. My customer has the ability to use their iPhones to control the equipment while talking to customers in the bar. I very pleased with the RIVA client and I would like to see it enhanced.

The two main functions I would like to see added are 1) the web widget thingie. In my opinion, any good home automation package has to support the iOS devices and have the ability to display security video both locally and remotely. If the web widget thingie is the best way to have that functionallity then that needs to be in the client.

The second functionallity that I would like to see is to implement the multi-touch capability of the iOS devices. The iOS client has limited ability to scroll the screen by dragging the finger or a flick. But I would like to see that added on a CQC overlay basis. That way, as an example, if you have a list of things in an overlay and you flick the screen, just the list scrolls - not the whole page.

Thanks for developing the iOS client. It's a great product.

Bryan

[brianmount]

Bryan,

We are working on a backhanded way to get security camera-type content. That's what Sam Vimes was talking about earlier. Stay tuned for details.

What you want with the list scrolling is reasonable but completely impossible using the RIVA protocol, because the client doesn't know anything about what's on your template. We are developing a new app which uses a completely different way of talking to CQC, and will be able to do that sort of thing.

[bryanb]

Brian,

Thanks for the quick feedback. On the multi-touch capability, I've asked Dean to look into adding that functionallity to the regular IV also. In my commercial system, I use a large HP TouchSmart All-in-One PC as the main user interface and having multi-touch capibility would be really nice. I think he is considering the request and looking into the possibility of supporting that. If he can come up with a solution for the regular IV maybe that functionallity could be added to the RIVA system. Multi-touch may only make since in browser type widgets - like the List Broswer widget. Maybe the solution is for the IVs to have a set of new browser widgets that support multi-touch. Actually, you may not need new browser widgets just a new command added to the browser. Something like ScrollList-MT and that command understands the drag or flicks. Anyway, thanks for your support.

[wuench]

RIVA is going to be a dog for any type of dynamic animation (which is basically what you are asking for) the way it is designed. You just can't make dynamic decisions over 500ms (half a second) of network latency. Any changes to the IV by Dean wouldn't facilitate it for RIVA, they are really two separate things.

What is needed is a real IV client that runs on the mobile device and makes a lot of the rendering decisions on it's own. And it sounds like that is what these guys are working on.

[SamVimes2]

Guys, couple things. I was hoping this thread would be for other RIVA developers and advanced users to weigh in on some kind of standardization for client-side processing of these new arbitrary RIVA commands. We're breaking the fourth wall so to speak for the first time in RIVA and I'd like to keep it consistent for users if possible. I guess that's just Dean and Fonceur for now, but also anybody else hiding out there working one for any other platform.

Second, there is a new CQC iOS app in the works (been in the works for a while but real life keeps getting in the way). It will NOT in the short term be a native template viewer, though that could come at some point built atop the same architecture. I'll post some more information about it in another thread.

[Dean Roddey]

I don't have any strong opinions about what functionality you would use it for, though I'll certainly speak up with any concerns about potential technical issues with any you guys come up with.

It could be used to invoke any local hand held functionality, some of which may well be completely specific to the platform. But, where it's possible to create common commands, that would be a good thing.

The OpenURL one seems reasonable to me. Of course you can always implement it and see how folks think it works in practice, and then adjust if that is deemed necessary. For the most part it's difficult to really get significant user feedback on any proposed functionality, until you actually put it out there and they decide it's good or bad.

[Dean Roddey]

For the next beta I allowed the new RIVACmd to be done in the OnPreload, as requested.