LG Thinq and smart appliances

Beside Motorola’s Atrix 4G and the many tablets, one of the very nice announcements of CES is LG’s Thinq, with significant press coverage.

Connecting home appliances sounds kind of obvious, and the ubiquitous availability of smartphones and tablets makes it even more obvious. I have many times left my clean laundry sit in the washer for hours, just because I turned the thing on and forgot about it; no doubt that a message on the phone in my pocket would be a welcome reminder.

The devices shown by LG seem to be rather high-end stuff, with rather large LCD screens for the interface (and no buttons, so these screens must be tactile). Apparently, these devices are running Android, which could mean that we will soon find Android applications targeted at washers and refrigerators. This is great news at least for one point: I have no doubt that independent developers can imagine applications on appliances that LG wouldn’t dream about, and opening this market could lead to real improvements.

Now, I am wondering: do we need a smartphone in every home appliance? Do we need a 7″ screen on every appliance? A few rich geeks may answer yes, but most people are quite likely to stick with much simpler interfaces and cheaper electronics. You may have guessed already where I am heading: Java Card 3.0 Connected is the way to go. It has many qualities required to do that:

  • Web-based application model. A Java Card 3.0 servlet can provide the required communication link between an appliance and the controlling device, using a fairly simple and well-known application model.
  • Dynamic management of applications. With the addition of GlobalPlatform specs, Java Card 3.0 supports a wide range of application management models, and it could in particular be used in an application store model. For those who doubt, we can remind that the Java Card/GlobalPlatform is already used on a few billion devices, so scaling up is not a major issue here.
  • Small footprint, possible single-chip integration. Adding a full Android device to a US$1,000 washer is not a big issue, but adding the same device to a US$100 coffee machine will have a significant impact on its price.

Of course, it lacks a few things, in particular regarding I/O’s. In order to control an appliance, Java Card 3.0 needs to be able to control a few I/O ports, including a small display. But these APIs already exist in other embedded Java dialects such as MIDP, so why not doing it?

The bottom line is here that we can achieve with Java Card 3.0 Connected results that are quite similar to what is possible with Android, while making it accessible to a much wider range of devices. The only question is: Who is going to define the few missing APIs and make a reference design for a network extension board for appliances? I would definitely like to be part of that, but I’m not the one who will do the electronics part.

No Comments

Leave a Reply

Your email is never shared.Required fields are marked *