Mythical Empires

There’s no idea like an old idea. 26 years ago Echelon set up to connect things in the home. It had demos showing the lights being switched on, curtains opened and drawn, gates and garage doors opened and shut – all remotely.

Echelon produced (what else?) a ‘platform’, which it called called LonWorks, to make these connections ubiquitous and designed ICs, called ‘Neurons’ to control the connections.

All these foundered, of course, on the reluctance of hardware manufacturers to install or support common interfaces.

Then last week, Samsung paid $200 million for a two year-old US start-up which has had $15.5 million venture money with a the slogan “every home a smart home.”

The company, called SmartThings has (what else?) a ‘platform’ that connects a range of consumer IoT devices for the house together into a smartphone app.

26 years on, the world hasn’t come a long way – or any way at all.

Of course Samsung has a chance to do what everyone else has failed to do because it’s product range is so wide. According to an AnalysisMason survey covering six European countries and the USA, 35% of respondents had a Samsung handset, 25% of respondents had a Samsung TV and 11% had a Samsung tablet.

The strange thing is that this seems to be driven by companies’ fears rather than consumer demand.

When Philips made both DTVs and PCs its CEO told me no one wanted a merged device because the use model was so different – one you sit up to, while the other you watch while slumped on the sofa with a beer.

No, the motivation for domestic connectivity is not consumer demand but fear – which is why all attempts at connectivity have failed – because no manufacturer wants to enable their competitors, while every manufacturer would like to connect a string of interactive devices which would lock out their competitors.

If your competitor looks like he’s building such an empire you tremble.

So when, earlier this year, Google paid $3.2 billion for Nest and Apple released HomeKit – both aimed at connecting things in homes – Seoul trembled.



  1. You’re right on both counts, dft-guy, still it was a good story.

  2. David,

    I guess your last post here is actually intended for the TMSC FD-SOI chapter, not this one.

    And regarding to Fairchild – this one has been posted before (I feel like an elephant now – heck, I’ve seen this one before)

  3. If that’s possible it knocks FD-SOI on the head. And since there aren’t so many FD-SOI designs around, it seems a trifle premature to be developing a technique to migrate FD-SOI designs to bulk designs. Bonkers even.

  4. Well one would expect some adoption somewhere of a technology which has been under development for 26 years, Mike. My simple point is, and I know my points are always rather simple, is that if widespread adoption hasn’t happened after 26 years then maybe, just maybe, folks don’t want it.

  5. I think there’s more than a little NIH involved.

    LonWorks is an open standard (ANSI I think) but maybe the licence fees aren’t conducive to the price points being aimed at. There are few connected homes but I think we can assume the Shard is very well connected indeed and I bet it used LonWorks somewhere.

  6. So Apple, Google and Samsung are barking up the wrong tree – all this connected home stuff was done a couple of decades ago and we’re all enjoying its benefits.

  7. I think you’ll find Echelon and LonWorks are doing quite nicely thank you and are standardised in numerous applications including building automation, household appliances, some parts of train networks and even I believe some types of semiconductor manufacturing equipment.

  8. Well, Yes, hopefully it will, Keith, unless those pesky manufacturers find a way to impede it for their own ends.

  9. Hi David,
    That’s why the open source (OS) approach will work, where the closed ‘manufacturers only’ route will not. With OS drivers being written by the community and used with an OS kernel which supports these objects ( as I have mentioned before) then locking out competitors will not happen.


    PS I have no financial or other interest in ‘thethingsystem’, just think it it is a good approach to solving the problem.

Leave a Reply

Your email address will not be published. Required fields are marked *