Tom Knight

I was hoping to write about my new nCube smart home hub today as it arrived last Friday, unfortunately three showstopping issues have occurred and the developers are looking at them and trying to fix them for me and the wider user group. Hopefully I will be able to give you a full run down on it next week.

In the meantime I wanted to talk about maximum heart rate tests.

I’ve had a little bit of a break from running since a disappointing performance in a marathon 3 months ago following an amazing gamified training experience I put together. But now I’m ready to get back to training again, and as someone who likes to run by heart rate (HR) as a guide to effort rather that pace it’s important my maximum HR is calibrated correctly so all my effort zones (usually percentages of your maximum HR) are accurate.

I’ve not done a maximum HR test for a year or so, and even then I think my lack of understanding about how a Garmin samples heart rate might have incorrectly reported* my max HR – so there is no harm in doing it again, and if you’ve never done one and always assumed that ‘220 – your age = maximum HR’ is a substitute measure I’d advise you to think again and if you have a HR monitor it is worth your time to do a proper test. It will only take 20-30 minutes, and only 1200m of that will be hard work!

1. Run 1.5 km as easily as you possibly can. 

2. Run 400m as hard as you possibly can.

3. Recover for 2 minutes, this means run very easily (I’d not recommend stopping).

4. Repeat steps 2 & 3 until you have run 400m three times.

5. Cool down with a low effort run for 1km.

When you synchronise your HR data you’ll see where your heart rate jumps during the 400m efforts, the highest heart rate value you see recorded is your maximum HR.

Bonus content: how does Garmin calculate my HR?

First off, I’m only talking about how it seems to work in my experience, which has been with a Forerunner 225 predominantly.

It seems that the Garmin polls it’s sensors every 5 seconds and then averages the result against the previous 60 seconds results to show you your HR. This is fine as perhaps wrist based HR monitoring is too error prone to give real time output results and over the course of a long run it’s not really a problem. 

Where it does become a problem though are interval sessions with short efforts, as you may find if you go from very low effort to very high effort you will get a lag in the HR reporting for 30 seconds until the earlier readings are pushed from the averaging for that previous 60 seconds.

As you can see in the data below where using the above method, the real time heart rate (in red) jumps as the interval begins but the average HR takes time to catch up as it is still averaging the HR shown on the device (in green).

...but I suppose first I should tell you what it is, it’s an automation (or workflow) tool for an individual, which sounds incredibly dull. 

What’s important is what it does, and the greatest examples I can give of that is does it annoy you that your Instagram pictures post to Twitter as links rather than photos? Yes, it does, and it also annoys all of your followers. IFTTT can fix that.

IFTTT stands for ‘If This Then That’ and whilst it’s widely popular amongst technology purveyors it doesn’t seem to have broken out of that audience...yet. 

As a user, you create ‘recipes’ (or sign up to existing ones) that plumb your social networks, cloud storage accounts, news feeds, IoT devices, fitness trackers, calendars etc. together.

I’ve attached my favourite five beneath, they let me know if it’s cold outside (so I can think to go out early and defrost the car), save tagged Facebook pictures to my Google Drive, share my Instagram photos as proper photos on Twitter and remind me in the evening if I tagged a YouTube video to watch later during the day. None of them are rocket science or even nerdy, but all of them really useful, they make the technology work for me.

Earlier in the year I wrote this (How I’m training for my next marathon : Gameification), explaining my approach to marathon training this year was going to involve game-ifying the process, as one of the big problems I have with training is when you just have one thing, many months from now, to work towards it’s really easy to lose focus and slack off.

So I wanted to check back in, just 3 days ahead of my marathon, and look at whether gameification of the process had allowed me to prepare any better this time than two years ago. 

Below you’ll see three charts, the pink line is 2016 and the purple is 2014 (apologies on the colours - the charting tool was not flexible!), they represent progress across the year for kilometres covered, number of activities undertaken and finally time spent training. It’s incredibly clear, I think if the point of gameifying my training was too increase volume, gameification has completely succeeded. In each chart I have nearly TRIPLED my training volume by marathon day.

2014 value = 256km , 2016 value = 697km 

2014 value = 25, 2016 value = 70

2014 value = 20hrs, 2016 value = 61hrs

Importantly gameifying the process has had the exact effect I hoped for, when it was frosty/icey/snowy I was still motivated to go out and run for 2 hours. Now as I finish my training and start to think about the race the level of self doubt I have is tiny. I look at those charts and feel very prepared, if I ran a 3 and a half hour marathon on a third of the training last time imagine what I could do this time (though that is in itself a scary way to look at things, because if I only improve by 1% or actually don’t improve at all the whole feel good of this method is likely to be severely tarnished I suppose).

In conclusion, unless things go very badly, I can’t see why I wouldn't do this again. I've enjoyed the technology aspect of tracking everything, I’ve enjoyed the gameifying process and feeling achievement along the way. Micro-rewards have made me want to do every run, when I’ve felt like missing a run it hasn't felt like ‘oh it’s just one run of many’ it’s felt important because that run doesn't just contribute 1% to my over training goal, it contributes 10/15% of my way to my next reward, and that’s hugely motivating.This is probably the best experience I’ve ever had of sticking to a workout or training plan.

I like this one;

I’m a huge fan of gamification. I’m a game player at home and I’ve always loved the idea of introducing it in the office, though disappointingly we never seem to quite have anything that really fits the model.

If you don’t know what gameification is, I’d really recommend this talk by Tom Chatfield at TED from 2010:

Or at least read how Wikipedia defines it: ‘Gamification is the application of game-design elements and game principles in non-game contexts.’

Back to running. The only bit about running races I like is finishing them. Which makes 16 weeks of training for one feeling at the end, very difficult. I know this about myself, I know that continued motivation throughout 16 weeks with the carrot of ‘you get to run a marathon at the end’ will not cut it. So I have to look at how I can create what Tom Chatfield calls a ‘Reward Schedule’.

Sixteen weeks of training represents 80 sessions to me, and as Tom said (is it confusing we are both called Tom?) ‘5000 pies? boring. 15 pies? interesting.’ - 80 sessions is boring, but what about 5? Five was my first goal, then 10, then twenty, thirty, forty, sixty and eighty...and the rewards get greater, the first three are just my favourite chocolate bar - a caramac, the most luxurious of chocolate bars but later on things ramp up, the reward at eighty is significantly ‘rarer’ to put it in video game terms.

The really good thing about Garmin Connect (that Strava doesn't do) is it allows me to create all these goals and it puts them next to my training calendar. This really feels about as close to augmenting real life with gameification as you can get, I have actual progress bars like a game character that relate to a physical action I’m completing.

So that’s what I’m doing, I’ve got eight targets each with a reward, some cheap, some frivolous and some luxuries that ordinarily I look at the price of and talk myself out of. 

To paraphrase Steve Redgrave, after I finished the London Marathon I pretty much declared ‘if you see me near my running shoes, shoot me!’.

That lasted a little while, I went to New York that summer and couldn’t resist running around Central Park but otherwise I pretty much gave up. Until as a company we entered the JP Morgan Corporate Challenge and a little friendly peer pressure had me on board for that, anyway it’s such a short race (5.6km) training for it doesn’t really interfere with anything socially so I gave it a good go (…and was a bit disappointed with my time in the end but oh well.), and since that race I’ve ran sparingly, probably just enough to keep my base fitness in reasonable shape.

I was talking to my Cousin’s husband (let’s call him John) later in the year and he told me he was running the Milton Keynes Marathon and asked if I wanted to also, which I didn’t. Before I ran the London Marathon (the only marathon I’ve ever run) a runner friend said ‘you won’t just do London, you might do two but you’ll not do one if it’s London. You’ll enjoy London as an experience, but you’ll do another to get the time you want. You won’t get a time you are happy with at London.’ He was right. London is an AMAZING experience, I will never forget running across Tower Bridge, Mo Farah running towards me where the 14 and 20 miles are parallel and running down the Mall at the end. My time was awful though, I was caught in traffic for the first 5k, then saw how slowly I’d ran and panicked for the next 10k running far too quickly, I was all over the place.

I got back to John, and told him I was in for the Milton Keynes Marathon.

So I started looking at training plans a couple of weeks ago as I was approaching 16 weeks until race day. My first principle was that I wasn’t going to train as much as I did last time, as for me it wasn’t socially viable and I often skipped sessions last time round for that reason. This resulted in me taking a pre-canned session from Garmin (developed by FullPotential) which is four running sessions a week (two easier long runs and two interval/fartlek speed sessions) and swapping a recovery run out for a nice long evening with the foam roller!

The big difference with this plan though is it is nothing to do with pace, it’s all heartrate (HR) based training. When I go out for an ‘easy’ run I run by HR zone rather than a pace I think I justify as ‘easy’. Running by HR feels really good by the way, it is a bit odd though, you are training the Aerobic system (which accounts for 98% of a marathon run) in your body as slowly as you can to avoid fatigue and injury, which inevitably feels really slow when you are doing it.

That’s the HR bit, and I’ll update you in a few weeks as to whether I feel this slow aerobic development training is going to get me where I would like to be.

Maybe this list is a reminder for me, next time I think a £250 phone will last me more than a year I should consult it - or perhaps it’s a list for someone about to buy a ‘low price high spec’ smart phone, and can serve not to dissuade but at least allow you to see the types of things that start to go wrong and you can decide if the trade off against price is worth it for you. For me, it’s not going to be. 

This OnePlus has been an awesome foray into the Android world. The low price allowed me to move from iPhone with little risk. It looks cool and am unlikely to own another phone that get’s as much attention as it does - most notably when my 13 year old cousin told me I must be pretty cool for owning it (with zero sarcasm in her voice) and when a teenager asked if he could see it, as he’d not seen one before, in a mall in Orange County (…he didn’t then leg it.).

But let’s cut to the chase, I’m planning on switching to an established brand with a stock OS in the hope that they do a more to stop a deafeningly loud amount of software problems destroy their device’s reputation.

For those interested, it’s a 64gb OPO, unrooted, running CM12S (YOG4PAS3JL).

Overheating

…and not because I’m playing games or watching movies, just random overheating, where the phoen is just in my pocket doing nothing and it just fires up its internal blast furnace to try and cook my leg.

OK Google Detection makes me a robot

This is a very annoying one. Me: ‘Hello’ Caller: ‘I can’t hear you, you sound like a robot’ Me: ‘Hold on…!’ Caller: ‘No I really can’t hear you, I’ll need you to ring’

furiously do the following Home screen, swipe right to Google Now, click hot dog menu, Settings, Voice, “OK Google” Detection, tap ‘From any screen’ slider

Me: ‘Can you hear me now?’ Caller: ‘ahh, yes perfect.’

Every single time I get a phone call this setting turns itself back on.

Bluetooth is not reliable

The bluetooth is perfect for the car, or speakers or headphone. Anything else and it will either fail multiple times or will just never work. Chipolo, who make a bluetooth tile for your keys, went as far as saying:

@t0mknight unfortunately the OnePlus has a bad Bluetooth signal strength. Please write on our support mail: [email protected]

— Chipolo™ (@ChipoloTM)

February 18, 2015

Phone signal when disconnecting from wifi

Often when I leave a wifi network (happens without fail from my home one most frustratingly) the phone won’t connect to the data services of my network carrier until I put the phone into airplane mode for a minute.

Camera not saving HDR photos

New one from this festive period, the camera has stopped saving photos taken in HDR mode…it just spins (beach balls/pinwheels/hour glasses whatever you call it) and never saves. 

Video goes odd when recording/playback 

From the end of June through to the end of August I've been capturing ‘one second a day’ of my life using a nifty app called, imaginatively, one second everyday.

It’s been a really interesting experience and as I watch back the video above I get a lot of the ‘promised’ nostalgia you’d expect to, my brother and girlfriend are both in there lounging around on summer weekday nights, my little cousins blasting each other at a big family get together is represented, I went to Wimbledon, and (a very very wet) Silverstone Classic with my parents, and Intel, and ran the JP Morgan challenge in Battersea...and my displaced best friend (he lives in Dubai) was also back for a bit and made sure he go plenty of seconds! However, whilst all that is great, I think that’s probably the end of one second videos for me...

There are two problems I experience, the app is perfect - the problems aren't technical, it’s to do with my work life and worrying about missing a moment.

The work stuff first, the large part of my conscious Monday to Friday life is in the office, which means I either take clips in the office or after work, and if there is one thing this has taught me when I'm scrabbling around trying to find something interesting to film on a week night is that my week nights don’t look massively interesting. 

The other week I wrote a blog about what I called ‘the Microsoft Android phone’, where I explored how Cyanogen could displace all the Google services with Microsoft ones and Microsoft could move away from turnkey hardware/OS offerings and more towards services, something their CEO seems keen on.

Firstly, to burst through the fourth wall, I was amazed at the traction is got on reddit. It received thousands of views from around the world in a 24hr period and completely eclipsed my expectations for my lonely little blog. 

Then, as is the way with reddit, people started challenging it..and each other. Which was fascinating and I can image a little what it’s like to write scientific journal articles. 

One thing someone mentioned that I omitted in the first article was I’d missed a keystone, the store, they argued it was the lynch-pin and the reason Google couldn't be usurped as the defacto Android. It’s a good point, but I think if you look at Amazon’s behaviour in this space, particularly recently, you see that 3rd party app stores may have potential. 

The Amazon app allows app downloads, many of which are in both Google and Amazon app stores, offering a real alternative to the Play store...it even manages updates of installed apps. There is no reason Microsoft couldn't adopt this model to unhook from the Play store.

A couple of interesting other things since the first post, firstly my phone just updated to the latest version of Cyanogen where the homepage in Chrome has been set to...dun dun DUN...Bing. Hard for me to take that as anything but confirmation that I've hit the strategy on the head in the last post.

I started running in earnest in 2012 and have considered myself a runner ever since, though I’ve taken hiatus in those years, most notably pretty much giving up running entirely following the London Marathon in April 2014 until the beginning of this year. 

Two things have motivated me this year, or rather the same thing twice, competitiveness - and that’s competitiveness with specific individuals. I have two PRs I want to beat (10k and Half Marathon) but no matter how massive and interesting to myself I make my plan I cannot motivate myself to run it for myself. However some (not so) friendly competition with a colleague ahead of the JP Morgan Corporate Challenge this year lit a fire underneath me.  

  Above, my current and largely ignored 10k training plan.

  Following that race I was goal-less again and set my sights on an arbitrary local 10k in October, only to ignore my training plan again for the first 3 weeks of the plan completing one very poor 10k training run. Then on a quiet Monday morning I got on Strava and saw a friend was running a 10k pace near the exceptionally poor one I’d sweat over the week before.

The fire underneath me was re-lit, I had to reprove myself, challenge my PR and start posting faster training runs on Strava. I needed my friend to eat my dust, in a race he didn’t even know he was in. 

So, I’m training four times a week, and as always when coming back from being off the boil for a bit shaving massive amount off my times every week, as I find my groove and fitness I’ve found 20 seconds per km. The two biggest contributors to this have been running negative split runs (where you run the second half of the run faster than the first - hugely satisfying) and tick-tock intervals (I have just made that name up, so consider it trademarked!).

Intel, the computer processor company, develop their chips on what they call ‘tick tock’. For example, if year one is ‘tick’ they produce a chip that is speed 1 and size 1, the next year, a ‘tock’, the produce a chip the same speed as the tick but half the size (so speed 1, size ½). The next year, another ‘tick’, they create a chip the same size as the previous year but twice as fast (so speed 2, size ½) and so on, the half the size or double the speed.

…back to the running, ‘tick tock’ intervals are the same concept but for running. I alternate between sets of 200m and 400m intervals, so my ‘tick’ week are the 200m intervals. I blast it over 200m, real Zone 6 stuff, everything my legs have. The next week, the ‘tock’ week, I’m looking to get as close to the previous weeks times over 400m. Then the following week, the next ‘tick’ week, I try to go even faster. I’m not saying it’s easy, but it’s something to work at, last week my faster 200m tick was 2.59mins per km and my fastest 400m tock was 3.23mins per km…still my slowest splits were much closer, 3.47 and 3.51 respectively.

  Above: My sourdough starter refreshed at 8pm and then filmed over night as it comes to life over 11 hours. (...it was a little chilly that night hence the moderate activity)

In my previous post (I’ve finally cracked a Sourdough table loaf.) I mentioned I'd share how my Sourdough starter got going, and how I look after it.

So a little about my starter, it's 23 weeks old, a veritable baby by some standards - The Boudin Bakery in San Francisco's starter is over 150 years old. In the week it never really exceeds 200g in volume, but often bulks up for the weekend's baking projects. I've had two holidays during it's ownership where my brother has diligently kept it alive (Thanks Sam!) and it is the 5th or 6th Sourdough starter I've started, and the only one that's survived past 4 weeks.

Oh and it's a 'stiff' starter, which is unusual and is due to the amount of water being significantly less than the amount of flour, in most starters there is about 10% less water than flour in the recipe, in mine it's most like 30-40% less water. Another great thing about this type of starter is it seems far more robust than a standard one, although not recommended, at times my starter has gone 'unfed' for 3 or so days and not long term problems have emerged (i.e. it dies or goes mouldy).

Before getting into the recipe, I have to credit a baker in New Zealand (apieceofbread's instagram) who introduced me to stiff starters and is well worth following on instagram, he is a pioneer of modern sourdough baking.

Recipe and Method

Day One 100g Wholegrain Flour 200g Water Mix, and leave in sealed container (or bowl with cling film over the top) for 24 hours.

Day Two 100g Wholegrain Flour 300g Previous Days Mixture Mix new flour in with yesterday's mixture, leave in sealed container for 24 hours.

Day Three 100g Wholegrain Flour 90g Water 20g Previous Days Mixture Mix new flour and add in 20g of yesterday's mixture, leave in sealed container for 24 hours.

Onwards (Daily) 100g Wholegrain Flour 65g Water 35-40g Previous Days Mixture Mix all ingredients as before, you'll notice the mixture is more a ball now, hence the name stiff starter!

What's cool about this particular start is not just that it seems quite hardy but also it's versatility, for example if you want a to start a white or rye starter you can take the last step and switch the flour out using the wholegrain starter and it seems to just cultivate. I've run white starters alongside my wholegrain one for specific loaves using the wholgrain starter to get it going, then once the bread is made I stop the white one knowing I can always fire up a new one if I need to.

You may have caught the news that Microsoft are realigning their mobility hardware vision the other week (Did Microsoft just give up on Windows Phone?), and the word on the street is they are going to take the *slightly* messy Lumia random number convention and replace it with three phones, something budget (emerging markets/smartphone newbies), something enterprise (hello blackberry market share) and premium (for fanboys!).

That's all fine but I've seen something else a bit more interesting happening over in the Android space, that no one seems to be talking about, and it's the emergence of Microsoft's Android Phone.

Before I get into the nuts and bolts, let's baseline some primary components of the modern smartphone that I'm going to address here:

    You have an OS, on which there is a User Interface and then some capabilities (usually delivered as apps), in this case we are focussing on the dialler (it just feels wrong to leave the phone call bit off a technical reference model for a phone), the voice assistant, productivity suite, cloud storage and messaging. Now we've baselined what we are looking at, here is how I see the Windows Phone version of these technologies today (Windows 8.1 today but the same for Windows 10 when that is available too).

    And to its right, the existing Android stack represented as a TRM with the Google applications, which are likely the most commonly used suite on Android.

With that all baselined we can start to see how Microsoft returning to its service and software roots (after a try at the Apple hardware/software turnkey solution game). In the past three months they've released (albeit in beta for some services) Android versions of their software that transform the Google Android device into an Android OS running various Microsoft services.

The big thing that's going to pull this all together for them is the launcher, Arrow is a really basic launcher at the moment but with the announcement that Cortana is being made available for Android you can be pretty sure that they will be working on a way to integrate it into Arrow, displacing Google Now and OK Google.

This coupled with the whole office suite, outlook and onedrive, all having decent Android apps starts to make the whole Microsoft suite on Android idea really stack up. I've blended the models from above to show where I think it might land from a services point of view.

    If you then consider all the rumours earlier in the year that Microsoft were looking to fund Cyanogen (Cyanogen Wants to Take Android From Google, and Microsoft May Help), probably the biggest custom Android OS in the world, who are trying to wrangle Android off of Google it starts to make sense. Cyanogen have already agreed to install Microsoft Office apps as part of their new OS release(Cyanogen Announces Strategic Partnership with Microsoft), if Cyanogen unhook Android from Google then they could start creating a full Microsoft Android OS. Which gives Microsoft all the benefits of a platform for their services without the headache of growing a hardware business...something Satya doesn't see as the future of the business anyway. Who knows though, looking at the state HTC are in at the moment perhaps they could jump into this partnership as a hardware provider.

Recipe (Interactive Recipe Sheet for helping with specific quantities/percentages) 700g White Flour 50g Spelt Flour 50g Dark Rye Flour 560g Water 192g Sourdough Starter 16g Salt

Short-form Method

Autolyse 60-90 mins, Bulk Fermentation 3-4 hours, Preshape 30 mins, Final Prove (in fridge) 8-10 hours, Bake @ 230'C 30 mins covered, 25 mins uncovered.

Long-form Method

I’m not going to go into how my sourdough starter is formed (…that might be my next bread blog!) but briefly, I feed it 11 hours before I want to use it as I find that seems to be long enough for it to have begun to get active, but not so long that it’s hit its peak and is losing its activity.

Following that, I autolyse the total flour and water quantity. Autolysing is where you mix the two ingredients ahead of adding salt or yeast/sourdough starter - this is particularly useful when dealing with high hydration recipes (large amounts of water compared to your total flour) and can make it much more manageable.

Then I mix in the starter, I tend to do this in the bowl with a little folding followed by some ‘pincing’…three times over and then leave it 20 minutes before starting the folds.

I do three folds (Ken Forkish shows you how here), once per 20/30 minutes, at the beginning of the bulk fermentation phase. I add the salt on the first fold, this seems to work well and giving that time between the initial mix and the first fold where the starter is alone with the flour and water seems to make a difference.

Leave it to bulk ferment after the third fold for 3 to 4 hours.

Then I pre-shape the loaf, this is where I divide the dough into whatever size loaves I’m aiming for (in this case I just divide the dough into two) and then make firm rounds of dough. Regardless of the target shape of the dough I always make them balls here as they get shaped again after this. Let them rest on the counter for 20-30 minutes covered with a tea towel.

While the pre-shape is happening I throw quarter handfuls of white flour into my bannetons and work the flour into the grooves to stop the bread sticking, though sticking seems to be less of an issue since I started proving the dough in the fridge. Once the pre-shape is up, I’ll reshape the dough and place it seam side down in the banneton. Before placing the banneton in a sealed plastic bag (this stops it drying out) and then in the fridge for 8 hours.

After 8 hours I’ll take the bannetons out, and in a swift motion upturn and smack them on the surface to try and dislodge the dough at the same time. The oven is up to temperature by this point (230’C) so it’s a case of scoring the dough (I use a wet razor blade but used to do fine with a sharp, wet, knife) and getting it in the oven.

For the actual bake I use an upturned iron Dutch oven (or casserole dish/slow cook dish/crock pot, see below) over the top of the bread for the first 30 minutes. This allows the bread to grow without cooking the crust too quickly limiting the rise of the bread, after 30 minutes I take the Dutch oven out and leave the bread to brown, usually a further 25 minutes.

Got a new GoPro Hero HD and took it karting in electric go karts...the video is a bit rough and ready, it was done the day after we went karting cut down from 90 minutes of footage, but if I didn’t do it quickly it would have never been edited! 

Easily the hardest bit to do, so I’m trying to get it out of the way quickly. I’m going to blog about technology, baking, running and motorsport mostly I think...I’ll try and tag things properly so if you come here for one thing in particular, you’ll find it easily. 

At the moment I’m a bit obsessed or focussed on a niche of those areas, this will change I'm sure but for the foreseeable future I’m absorbing as much about connected homes (or Internet of Things) as I can be - I have a house full of philips hue, wemo, raspberry pis and ifttt workflow. This extends to my work life also, where I’m currently trying to devise a way of getting some plants to tweet when they need watering in a meeting room!

On the baking front I’m on test subject #30(ish) trying to crack the perfect recipe for a weekend loaf of Sourdough bread, nothing fancy, jsut something you can have for lunch and toast and things...the problem being in a process that can involve two 24 hour prove periods there are a lot of variables you can muck about with.

I ran a marathon last year and the plan was to retire immediately from running and switch to another sport...well I took a break and after a fierce training plan earlier this year to get me a good 5k time in a corporate race I’ve decided to return to long distance running with a goal of beating my 10k and half marathon times (41:18 and 1:30:39 respectively).