Blog

Mondays are fine. It’s your life that sucks.

Recently I’ve been watching a lot of Casey Neistat videos on YouTube. If you’re not familiar with him, Casey is filmmaker. It’s fair to say he’s one of the founding fathers of the “daily vlog,” and has brought a methodology and a style to the genre that has since become the standard.

As someone who seems to define the medium in which he works, Casey often gets questions about what hardware he uses to create his videos. He rarely answers them, instead saying that his ideas are important, the tools he uses to express them are not. Pretty much any camera would allow him to convey his ideas, and therefore the best camera is whichever one he happens to have available.

This makes sense. I’d be willing to bet that the last time you went to see a movie the question of which camera model they shot it on never once crossed your mind.

I’m not a filmmaker, I’m a project manager. My tools are Outlook, Word, Excel, PowerPoint… the list goes on but you see what I’m saying. You could argue then, possibly quite persuasively, that the same thinking applies here. We could swap out our office applications for alternative tools, and it would make no measurable difference in any important way.

But that’s not what I believe, and I think I have excellent reason to be excited about my workplace’s upcoming switch from Microsoft Office to Google G-Suite.

This is so accurate it’s terrifying.

While I was waiting for the electronic components to arrive for my DIY Smart Home project, I had the idea of adding some voice control to our house.

Plenty of people have done this with the Amazon Echo, but that’s not available in Canada yet. Google Home has now been announced of course I want one, but I don’t know when they’re going to be available in Canada either.

After a quick google search I found the article I’ve linked above about adding a custom google now command using Tasker on Android – an app I already know and love. I set it up to trigger a call to the HomeAssistant API.

It’s been a little while since I posted about my smart home journey because I’ve been pretty busy of late with other things (like my actual job), but rest assured that there has been progress – albeit slow.

As I mentioned before, I’ll be adding three microcontrollers to our smart home setup in various parts of the house. Each project is going to go through three stages: proof of concept, prototype, and final product.

Today is our first proof of concept, and it’s going to test whether or not I can hook our existing doorbell up to WiFi.

Getting a notification when someone rings the doorbell was the first thing Flo asked for when I mentioned I was going to start making our home smarter, and I quickly set about figuring out how to do it: happy wife, happy life.

Some initial ideas (adding a relay to the doorbell circuit) were fairly quickly dismissed, and so I came up with a piece of out of the box thinking that I’m actually very proud of.

If you’ve been keeping up with my #SmartHome series (and if you haven’t, why not?) you’ll already know that I have plans to make more of my home “smart” using Home Assistant as the software that ties everything together, and some DIY NodeMCU-based hardware that I’m going to build myself as a learning opportunity.

Another important piece of the puzzle, but one that I haven’t previously mentioned, is MQTT.

MQTT is “a publish-subscribe-based lightweight messaging protocol for use on top of the TCP/IP protocol,” at least according the slightly suspect grammar of the person that wrote the Wikipedia article about it.

I learned about MQTT at the same time I learned about Home Assistant, although I didn’t initially appreciate its power. I’ve been using it from the start to enable Home Assistant to know where we are: our phones run an app called OwnTracks which publishes location data to an MQTT “broker” (server). Home Assistant subscribes to these updates, which means it immediately knows about it when our location changes.

I love this whole solution, not least because it’s very easy to run my own MQTT broker (I’m running Mosquitto in a Docker container on my home server) and I’m therefore entirely in control of our location data – it’s not being shared with the developer of some app or service I have no insight into.

This publish/subscribe model and the lightweight nature of MQTT makes it perfect for “Internet of Things” (IoT) devices to communicate with each other, and when you add Home Assistant into the mix it gives me all the tools I need for any sensors I build to feed their data into my smart home ecosystem, and for my smart home controller to feed commands to any devices.

Indeed, I’ve already built my first little NodeMCU app that leverages the technology.

I have begun to learn how to program my NodeMCU.

Granted, all it’s doing here is running the example “Blink” program that comes with the Arduino IDE, but the important point is that the little light is turning on and off because I uploaded some code that tells it to.

Ever since I bought my own home nearly a year ago, I’ve become increasingly interested in making it smart.

Right off the bat, I feel like I should clarify what that means to me. The ability to turn some lights on or off with an app is not smart, in my opinion – the smart way of controlling lights is by flicking a switch conveniently located in the room you wish to illuminate.

A smart home needs to be much more intelligent. It’s about automation. It’s about the home being able to notify me if something is happening that I need to know about. It’s about being able to accomplish things with minimal difficulty, not adding complexity and more steps.

That’s where off the shelf “smart home” solutions really started to fall down for me. I could spend hundreds or maybe even thousands of dollars, for what? The ability to turn on my living room lights while I’m still at the office? Why would I ever need to do that?

Nevertheless, the lack (in my opinion) of a pre-packaged, useful, holistic solution that accomplished my vision of what a “smart home” should be didn’t deter me from tackling things bit by bit. It started with our burglar alarm. It has internet connectivity which sends me alerts in the event that something unexpected is happening, and lets me arm or disarm the system from my phone – which I actually do find useful.

Next up was our thermostat. The one that was installed when we bought the house was an old-fashioned one with a simple mercury switch inside. You set the temperature, and that was it. We replaced that about a month ago with something programmable (it doesn’t need to be as warm in here at night as it does during the day; it doesn’t need to be as warm if nobody’s home), and I took the opportunity to get one with WiFi so I can set the temperature remotely. That’s not useful in and of itself, but if you take that functionality and look at it in the context of my wider vision then the thermostat is certainly something I’d like to be able to programmatically control.

It was around this same time that I discovered home assistant, and now my dream is starting to come alive.

Home Assistant is an open-source project that runs on a variety of hardware (I was originally running it on a Raspberry Pi, and I’ve since switched to running it in a Docker container on our home server). It has a ton of plugins (“components”) that enable it to support a variety of products – including our existing alarm, thermostat, streaming media players, and others (including, somewhat ironically, the colour-changing lightbulbs we have in our family room). It includes the ability to create scripts and automations, it uses our cellphones to know our locations, and can send us push notifications.

My initial setup was all about notifications. If we both leave the house but the burglar alarm isn’t set then it tells us (and provides an easy way to fix the issue). If we leave one of the exterior doors open for more than five minutes, it notifies us (or just one of us, if the other is out). I also created a dashboard (that you may have seen in my last post) to display some of this stuff on a monitor in my office.

Since installing the thermostat I’ve added more automation. The time we go to bed isn’t always predictable, but when we do go to bed we set the alarm. So, if it’s after 7pm and the alarm goes from disarmed to armed, the thermostat gets put into night mode. If nobody is home then the temperature gets gradually turned down based on how far away we are.

If nobody is home at dusk then it turns on some lights and streams talk radio through the family room speakers to give the impression that someone is.

This stuff meets my definition of smart, and I’m barely scratching the surface. The open nature of the platform not only means that I’m not tied to a particular vendor or technology, but also means that I can add on to the system in a DIY way.

Which is exactly what I’m going to do. I’ve bought some NodeMCU microcontrollers which are WiFi enabled, Arduino IDE-compatible development boards designed to the basis for DIY electronics projects.

Recently, I have been all about turning our home into a smart home. That’s mostly because I’ve discovered Home Assistant: a little piece of software that runs (in my case) on a Raspberry Pi and pulls together data from all the sensors and smart devices you own.

#DongleJokes

Recently, for reasons I’m sure I’ll write about in the future, I needed to find a way to use JavaScript to test if either of two web-locations are accessible – my home intranet (which would mean the user is on my network), or the corporate intranet of the company for which I work (which would mean the user is on my organization’s network). The page doing this test is on the public web.

My solution for doing this test was simple. Since neither resource is accessible publicly I put a small JavaScript file on each, then I use AJAX and jQuery to try and fetch it. If that’s successful, I know the user has access to whichever intranet site served the request and my page can react accordingly.

If neither request is successful I don’t have to do anything, but the user doesn’t see any errors unless they choose to take a look in the browser console.

This all worked wonderfully until I enabled SSL on the page that needs to run these tests, then it immediately fell apart.

There’s a popular myth that says we only use 10% of our brains.

It’s simply not true. Studies (including the source of all scientific truth: an episode of MythBusters) have proven that all areas of the brain have a function, and while the percentage that we’re “using” at any given time varies by task it can certainly exceed 10%.

One thing that seems very obvious to me without needing to cite a study about it, however, is that I certainly have unused brain capacity, and it can do amazing things when you leave it to its own devices.

As an example of what I’m talking about, I refer you to a link I posted on this very blog some time ago: Why Great Ideas Always Come in the Shower (and How to Harness Them).

In the brief commentary I added, I mentioned that never in my life have I had a good idea in a meeting. Great ideas come to me while I’m doing other things. Specifically, other things that do not take much in the way of thought and offer little in the way of distraction: things where my brain gets left to it’s own devices and has an opportunity to wander – showering, certainly, but also commuting, trying to get to sleep at the very end of the day (infuriatingly), and when I’m at the gym.

Talking of the latter one, I haven’t been to the gym for quite some time.