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.
Back to Mr. Neistat. If his views on the importance of the tools we use are so opposed to mine you may wonder why I cited them, but actually they illustrate my point very well. Casey preaches that his purpose as a filmmaker is to take his ideas and deliver them to his audience. Anything in-between, no matter how crucial it is to the process, is little more than a barrier. This is a fascinating perspective to me: by this definition every tool we use is helping us move forwards while simultaneously holding us back. And when the ying and the yang are out of balance? When our tools start to hold us back from our purpose a little too much? It’s time to reassess.
It is time to reassess.
Here’s a brief story: Not too long ago I created a shared OneNote notebook for a small team to help us collaborate on a particular task. It took me 15 minutes to create it and publish to SharePoint with all the right permissions, and another 15 minutes to show people how to open it and synchronize the content. People were blown away with the ease and simplicity of it, and the power of collaborating in real time on the same piece of content; not to mention having a single source of truth and no emailed document preposterously named something like “Important Document v9 FINAL for final circulation FINAL v2.doc.” My 30 minute time investment paid off, then – people even congratulated me on it. But what if we lived in a world where our tools were built from the ground up with that kind of collaboration in mind? What if we never had 12 different versions of every document, spread across four email chains? What if we didn’t have to use the word “investment” when we talked about getting our tools to work for us, and allow get things done in a way that makes sense?
This, is why I’m so excited. The more I lean about G-Suite the more it becomes apparent that it’s been built with this sort of thing in mind: the sort of thing where teams can’t reasonably gather in front of the same whiteboard, can’t huddle in front of the same computer… can’t afford for their tools to hinder their purpose more than they help.
There are countless stories like mine out there in my organization. We’ve become blind to them because we’re so used to the status quo. That’s not OK, but it does mean that with the right approach we could uncover them, collect them together, share them, and nurture them to take on a life of their own. If only we had a platform for that sort thing, hey?
In a first for in-car convenience features, Jaguar is teaming up with Shell to launch a new payment feature that lets you use Apple Pay or PayPal (with Android Pay coming later on) to easily pay fo…
This is pretty sweet. Paying for parking might be an even better use for this kind of thing, but I’m looking forward to it rolling out in Canada nevertheless.
Google Now already has a ton of useful voice commands built in. Thanks to a recent update to Tasker plugin AutoVoice, though, you can now create your own commands that plug directly into Google Now to accomplish anything that Tasker can do with nothing but your voice.
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.
I’ll be using magnetic switches, also known as reed switches, elsewhere in the house. In fact, we already have them on many of the doors and windows as part of our burglar alarm, and you may well have some in your house too. They work by opening and closing when a magnet is brought into proximity with the switch. My flash of inspiration came when I was thinking about how a doorbell works: an electromagnet pulls a hammer into one chime when the doorbell is pushed, and then a spring pushes it into a second chime when the button is released and the electromagnet turns off. So, if I put a reed switch next to the electromagnet then the switch will flip when the magnet turns on. Genius!
Happily for me, our doorbell has two chimes: one in the main floor hallway, and one in the basement. I wouldn’t want to run cables up the wall to the chime on our main floor, but the basement one is the mechanical room so the aesthetic is really not a consideration.
Using a switch as an input for the NodeMCU is simple. Really the only thing to think about is that the digital input pins are “floating,” which means that if nothing is connected then the result you get from them is unpredictable. To account for this you use either a “pull up” resistor between the input pin and the 3.3v pin, or a “pull down” resistor between the input pin and the ground pin. I went for a pull down resistor between the input pin and ground, then put the switch between the input pin and 3.3v.
For the proof of concept I wrote some simple code to display the input pin state over the serial console, taped the reed switch place and, despite some protest from the dog, rang the doorbell a few times.
NodeMCU, MQTT, IoT & Other Letters I Find Exciting
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’ve ordered almost all the components I need for my upcoming hardware projects from China, and they’re only just starting to arrive. Happily the NodeMCUs themselves were amongst the first shipments to land on my doorstep, so even though the only thing I can really do with them right now is programmatically turn their internal LED on and off, I have still been able to use this to start learning: I’ve made it an internet controlled LED.
As promised, I’m going to be sharing both the hardware and the software as I take this DIY Smart Home journey. There is no DIY hardware here, just the NodeMCU itself, but I’ve published both the Arduino sketch and Home Assistant configuration I’ve used for the video.
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.
It’s not connected to a computer here. The USB cable is simply plugged into a power outlet.
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.
Watch this space, because over the coming months I’ll be connecting our doorbell, garage door and laundry appliances to Home Assistant. I’ll be learning as I go, and I’ll share the hardware and software.
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.
Apart from a new WiFi thermostat that I was going to buy anyway I haven’t yet spent any money at all in my quest for this - I’ve merely connected together the devices we already owned (Chromecasts; family room lights; our burglar alarm), added in some data available through various APIs (weather; internet speed; server status) and pulled location information from our phones using an app that publishes the information to my own server.
Anyway, now that I have all this information in one place I wanted an attractive way to display it in my office. Happily I’d already bought a monitor with a built-in powered USB hub from a surplus equipment sale a while back, with this kind of project in mind.
So, my Friday afternoon project today became hooking up another Raspberry Pi to this monitor and crafting a HTML dashboard for it to display. Not everything works yet, but I think it’s pretty good for an afternoon’s work.
(Since I didn’t remove the clock from the frame when I shot the video, I can’t pretend this thing is quick to boot. I may have manipulated the video speed at a couple of points. But anyway).
Oh, and watch this space for some home-brew smart home devices, coming soon!
Using JavaScript to Identify Whether a Server Exists
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.
Both requests fail, because a page served over HTTPS is blocked from asynchronously fetching content over an insecure connection. Which makes sense, but really throws a spanner into the works for me: neither my home nor corporate intranet sites are available outside the confines of their safe networks, so neither support HTTPS.
My first attempt at getting around this was to simply change the URL prefix for each from http:// to https:// and see what happened. Neither site supports that protocol, but is the error that comes back different for a site which exists but can’t respond, vs. a site which doesn’t exist? It appears so!
Sadly, my joy at having solved the problem was extremely short lived. The browser can tell the difference and reports as much in the console, but JavaScript doesn’t have access to the error reported in the console. As far as my code was concerned, both scenario was still identical with a HTTP response code of 0 and the status description worryingly generic “error.”
We are getting closer to the solution I landed on, however. The next thing I tried was specifying the port in the URL. I used the https:// prefix to avoid the “mixed content” error, but appended :80 after the hostname to specify a port that the server was actually listening on.
This was what I was looking for. Neither server is capable of responding to a HTTPS request on port 80, but the server that doesn’t exist immediately returns an error (with a status code of 0 and the generic “error” as the descriptive text), but the server that is accessible simply doesn’t respond. Eventually the request times out with a status code of 0 but a status description, crucially, of “timeout.”
From that, I built my imperfect but somewhat workable solution. I fire a request off to each address, both of which are going to fail. One fails immediately which indicates the server doesn’t exist, and the other times-out (which I can check for in my JavaScript), indicating that the server exists and I can react accordingly.
It’s not a perfect solution. I set the timeout limit in my code to five seconds, which means a “successful” result can’t possibly come back in less time than that. I’d like to reduce that time, but when I originally had it set at 2.5 seconds I was occasionally getting a false-positive on my corporate network caused by, y’know, an actual timeout from a request that took longer than that to return in an error state.
Nevertheless if you have a use-case like mine and you need to test whether a server exists from the client perspective (i.e. the response from doing the check server-side is irrelevant), I know of no other way. As for me, I’m still on the lookout for a more elegant design. I’m next going to try and figure out a reliable way to identify if the user is connected to my home or corporate network based on their IP address. That way I can do a quick server-side check and return an immediate result.
It’s good to have this to fall back on, though, and for now at least it appears to be working.
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.
When we lived in our apartment there was a gym in the building, and that was great. I could easily fit in a solid 45 minutes there at lunch. Any spare 30 minute window in my schedule could be turned into 20 minutes on the stationary bike.
I want to go back, but now that we’ve bought the house there is obviously not an on-site gym. There’s a gym at the office (20 minutes away) and a Goodlife Fitness close by (10 minutes away) where I’d get a discounted rate, but small though it is even that travel time is putting me off. I will most likely join Goodlife, since I rarely go to the office these days and installing a home gym just isn’t in the budget right now, but I’ve been missing the ability to easily take 30 minutes and get some exercise, and I’m sad that none of the solutions will offer me that. In the absence of a perfect solution, I haven’t done anything at all… until yesterday.
Since the weather here in Calgary is distinctly spring-like these days, I went for a walk before I started my work day. I didn’t go far – a little less than 2km, according to the Google Fit data from my phone and watch – just down the road a bit and then back along the pathways that run through our neighbourhood.
I liked it so much I did it again at lunch time, and then for a third time this morning.
The physical benefits of this, though I’m sure not huge by any means, are probably much needed at this point. Really though what I like about it so much are the mental benefits. I’ve never been much of a morning person and I would never consider going to the gym before work, but rolling out of bed and attempting to be productive more or less immediately is not a recipe for success either. Feeling like my day has already started by the time I sit down to get some work done definitely gives me a mental boost that I’ve been able to capitalize on. More significantly though, there’s a lot to be said for the kind of problem solving that can only come from not thinking about something too much and letting my subconscious guide me in ways that I’d never have come up with if I were sitting at my desk consciously trying to focus on something.
It’s amazing what you can do when you’re not trying to do anything.
tangentwithtanner
blakekathryn
titsay
kaledo-art