Transcript for Episode 14 – A graveyard of side-projects

back to the episode

Our podcasts are best experienced by being listened to. These transcripts are generated using speech recognition and human editing. There may be errors.

Today’s speakers are, in order of appearance:

  • Ryan – Trainer at Instil. Host and Editor in Chief for the WeekNotes podcast
  • Matt McComb – Director of Business Development, award winning Dahlia grower and World Champion Coloroids player
  • Matty Wilson – Principal Software Engineer and Instil's Apple expert

[00:00:00]Ryan welcome back everybody. This is Weeknotes by Instil. I’m Ryan. I’m joined this week by Matt McComb and Matty Wilson. And we’re talking about side projects. Let’s get to it.

You should totally put a microphone on your hardware budget

[00:00:15]Matt McComb I totally should. And I also need to get a proper webcam too, because in my confined space, I have a laptop stand and then the laptops in the corner and then the screens in front of me. So I have to keep turning round to the laptop. If you want to actually look at people on video calls, which is it’s quite James Bond.

To pivot round to look at people every so often it’s very dramatic. So a webcam mounted to the monitor would be nice. I think it’s probably got to that point in lockdown now where you really just do have to invest in the right equipment.

[00:00:42]Ryan Yeah unless we open up, in which case we’ll be back to the office and you won’t need it anymore. So

[00:00:47]Matt McComb That’s true.

[00:00:48]Matty Wilson I think it’s too late to buy any new hardware with my own money, because I feel that we’re going to be out soon. That’s where I’m at. That’s some positivity for this, and people listening to this episode, it’s almost done don’t buy any new webcams. The ones that you have will do.

[00:01:01]Matt McComb so then do we expect to see , the share price for zoom drop dramatically now?


[00:01:06]Matty Wilson Yes, time to sell.

[00:01:08]Matt McComb Yeah time Stock recommendations from Matty

[00:01:10]Ryan You don’t notice how bad a thing is until someone comes along and says, that’s terrible. And then you go, actually, yeah, you’re right. It is. I need to fix that. So camera setups and microphone setups.

We’ve been surviving perfectly Well well with crappy webcams, And laptops for years. And then all of a sudden somebody comes along and says, you know that camera’s not very good. You realize, yeah, that’s not and I really should invest some money and get something decent,

[00:01:34]Matty Wilson I never noticed how bad the MacBook Pro cameras were until we went into lockdown. It’s actually an embarrassment considering the front-facing camera that’s on your iPhone is pretty good. Like just put that camera in the Macs. like, you have the components there, you have a supply chain, , , you’d almost think it would be cheaper to just use the same camera everywhere. .

[00:01:52]Ryan It’s thickness It’s Jony Ive and his thinness It’s all Jony Ive’s fault

[00:01:56]Matty Wilson I think that the era of Jony Ive is dead. I think there’s going to be new Mac’s out this year and they’re going to be thicker. And they’re going to have more ports and they’re going to have a better camera

[00:02:04]Matt McComb Tara got one of the M1 Macs and his camera is a lot better. If you’re on a call with him

[00:02:10]Matty Wilson They say that nothing changed only the image processor, because it’s using an image processing from the iPhone now. But I think the hardware’s nearly the same.

[00:02:18]Ryan Okay. So let’s actually talk about side projects.

On the Northern Ireland tech Slack , there’s been discussion about interviews and doing technical exercises as part of an interview. And I know it’s not quite the same thing, but it’s slightly related. And it comes to the question of, should we, as developers always be prepared to do work outside of our day jobs in order to keep ahead, should we always be doing homework? Is that acceptable,

[00:02:43]Matt McComb don’t think it should be an expectation. I think everybody obviously has different levels of commitment in

Their lives. , and as a dad of young kids, I would love to have an hour every evening to program. Realistically that’s highly unlikely, most evenings. So I people have different circumstances, so I don’t think there should be that expectation, but from a CV and an interview perspective.

It’s nice to have something to talk about because usually if it’s somebody’s side project they’ll be in a way passionate about it and interested about it just generally makes for a nice topic of conversation. At times, in my experience, most successful side projects are not, when you’re trying to learn a technology, they’re the ones that end up on the scrap yard.

It’s when you’re actually building something that you have a purpose for or a reason to do it. That’s where I’ve had success in the past. When I say success, it’s not success just that they’re finished and they’re not abandoned.

[00:03:37]Ryan So you’ve actually finished a side project.

[00:03:40]Matt McComb I have, I have finished two or three side projects, mostly mobile apps Way back in 2010,

I built a to do list app probably along with about 3 million other people. The main value proposition for mine was it was really simple because there was lots of other apps out there to do list apps that were overly complicated it did actually get a few positive reviews. I think there was a couple of thousand people use that at the time and it was quite a good experience.

in that, when you build a mobile app, you have to do aspects of design. You move into doing bits of marketing, bits of sales, analytics, even in some ways bits of product management. So , it gives you a bit more of a rounded experience as well, but that was probably the most boring of the side projects.

After that made a bit of a step up I thought I’ll build my own game engine. Yeah, don’t do that. That was a really stupid idea. I came up with a concept for a game, which I still believe is a valuable valid concept.

[00:04:36]Ryan do you want to keep that secret or do you want to share it?

[00:04:38]Matt McComb No, if somebody can build this I’m happy to share the profit or even just a reference or an acknowledgement will do me fine so everyone knows asteroids obviously seventies arcade games spaceship You’ve got a bit of velocity from a little rocket that spins round and you have to blow up asteroids So I came up with the concept of color roids So each asteroid was a different color So it might’ve been red green blue and you had the match the color of your bullet fired from the spaceship to the color of the asteroid to destroy it And if you mismatch the color the asteroid spawned another three asteroids So it created this really weird sort of game dynamic where it was really calm at the start There was only one asteroid But it ramped up in difficulty because as soon as there’s five asteroids if you’re bullet it was the wrong color and it hit another one you would end up with three more So it went from having like maybe the first minute or two three or four asteroids to having like 60 so it was very hard very quickly but for at least three years whilst the game was in the store I remained the world champion Which is pretty good claim to fame

[00:05:44]Ryan And that’s on the top of your CV

[00:05:46]Matt McComb That is on my CV Yeah It still As it should be It’s no longer in the store It was no longer compatible with the later versions of iOS It fell away

and then there’s Trip NI

Trip NI was a iOS only mobile app probably in around 2013 that was built it was built at a time that TransLink our local bus and rail company had an older version of their journey planner Which anybody who used it probably remembered that it didn’t work very well you probably couldn’t find your stop And if you could it didn’t list all of the bus or trains going past and there’s quite a few limitations to it And there was no mobile app either and it Definitely Wasn’t a responsive webpage at the time And if you couldn’t find your bus or your train then you had to go and download the PDF from the TransLink website It started as an exercise to try and scrape the website for the PDFs and then parse out the time tables into a SQL Lite database so that I could then publish that and other people could use it

once I’d done it and then wondered if it could actually do live search on a mobile app over the SQL Lite database with all of the parsed route information in it And that will work quite well So it builds a little mobile app around it which let you see all the departing buses from a stop or plan a journey from a departure to the destination point and then moved on to doing journeys with stops as well all using locally stored data So you didn’t even need a connection

The most tedious part of it was that I wanted to do location based search if you were nearby to stop that you could search to find the name of that stop So you didn’t have to know where you were on the bus or rail map but obviously there was no GPS coordinates because there was no data So I went through the database and find the 100 or 200 top mentioned stops and then went on to Google maps To try and find them on a map to get there that long coordinates to manually put it into the database that was dedication

[00:07:44]Ryan That is dedication And did you crowd source that is that now officially published open data

[00:07:49]Matt McComb I was going to give it back to TransLink and they didn’t want it

I think at that time there was a really long running dialogue between the development community and TransLink before all of this where they were pushing for open data and TransLink every single year had responded with a reason as to why it hadn’t been published yet So I don’t know if I’ve ever published the database I might have I might not

[00:08:09]Matty Wilson I think Open Data NI is pretty good now has a] lot of this information You could probably build your app with a lot more ease these days I was going to build an app for the glider just to sorta send me push notifications whenever the next glider was coming And at the time there was no glider information on open data NI but it was promised that it was coming yeah open data NI as a treasure trove for this kind of stuff

[00:08:30]Matt McComb It is unfortunately at the time It wasn’t but it didn’t exist

[00:08:36]Matty Wilson The question I was going to ask Matt is would you define a side project as being finished would that be one that’s been pushed to the app store

[00:08:42]Matt McComb I think if you can take any side project To the point where it’s a publicly usable product it’s finished I think to get to that point is incredibly hard with a side project last couple of percent makes it humanly consumable I think that’s the challenging part especially as a developer as well when you’re trying to do design and stuff although there is great services neither for crowd-sourcing design as well current side project I haven’t even discussed that one yet I’m currently using Fiverr to get some app icon designs So I was able to get an App Icon designed for 10 15 pounds which feels like a real treat to yourself Once you put in that effort building the mobile if you can get an icon for for it feels like you’ve really rewarded yourself You’ve got something nice to badge it with

If you can publish it Yeah I think that’s the measure of success

Matty Wilson: [00:09:31] so if you go onto my github profile you’ll see like a graveyard of side projects there’s a website called killed by where it shows you all of the services that Google has started and then killed I need to create my own one where’s it’s like you and it’s just this huge list of side projects that never got off the ground

[00:09:52]Matt McComb I think that’s what you’re aiming for if you’re looking to learn something or looking to experiment something I think that’s fine I think it’s going to be abandoned I don’t think any good idea or concept is ever brought to life off the back of wanting to learn A particular language or framework or technology So I think having a good graveyard of abandoned technical side projects is a positive thing

[00:10:15]Matty Wilson Yeah whenever people ask me about applying for jobs especially like grads and stuff and say any advice I say set up a github account and just fill it full of junk doesn’t even matter What’s in it People will just look at it and go Yeah there’s some stuff on here And then move on As long as you have something that looks like you’re busy then that’s what’s most important

[00:10:33]Matt McComb caveat that by saying probably shouldn’t be template projects

[00:10:40]Matty Wilson Yeah Just forks of a whole load of other people’s repos

[00:10:43]Matt McComb Yeah, fork this and change the readme

[00:10:45]Matty Wilson Also caveat if you’re applying for a job at instil I’ll not be fooled I’ll go and have a look at the actual repos My side project started somewhere to matt just mobile apps really for me I have this rule where I get paid to do certain things as a developer So I get paid to write tests And I get paid to do back end work but for me the parts of development that I find that aren’t work is iOS apps So we’ll say any of my side projects that have actually put effort into have all been app based because it doesn’t feel like work to me Everything else feels like work So me and a few friends had a few different projects One of them was called Gumtree spy andthis was before the gum tree app was on the Android app store I didn’t know if there is a gum tree app on the app store or not but at the time there wasn’t so we obviously saw the gap in the market to put app on the Android play store with the word Gumtree in it because whenever somebody searched for gum tree our app would be top of the list So we had one called Gumtree spy and it it was a scraper based project as well It’s just script Gumtree For products that you’re interested in So let’s say you’re going to buy a piano you would say alert me whenever a piano appears in the Belfast area below a certain make the money and It would just ping you whenever that appeared that’s all happened on the device as well because you know we were young and stupid and we didn’t think about doing it in a back end So it’s just like probably completely decimated people’s batteries whenever this thing was running and major painpoints with it was just keeping it up to date with Gumtree because they were forever changing their site and it actually it was a paid for app so it made it made a bit of money Like it made enough money where we were like here this is pretty good And then we got an email one day from eBay who owned gumtree And it is pretty much takedown notice it was actually we were just told by Google your app has been taken off the play store And the reason why was because we used copyrighted name or maybe a trademark name Gumtree in the name of the app and the app icon was very similar to the Gumtree icon as well So yeah that was like one of the things where it was just like yeah I got a takedown notice for something that I made

[00:12:53]Matt McComb was because they felt threatened

[00:12:54]Matty Wilson I think so And I also have to say I did very little work on it as well I was just there just because it was Android I was like I don’t really want to work on this We’ll build an iPhone version as well but the iphone one never got off the ground so

[00:13:07]Ryan quite ironic that you could almost do that today on iOS and get away with it seems like that’s an issue on the app store impersonating other apps

[00:13:17]Matty Wilson That seems to be the way to make money in App Store now is to take somebody else’s App and build a clone and just beat them in SEO or was it up search optimization or is annoying because at the other time it was just like he could make some money on the app store but now it’s just way too competitive You gotta build a service instead

[00:13:37]Ryan Yeah it’s kind of funny that on Google you got caught whereas on iOS you could probably get away with it So that was your first one. any more apps

[00:13:46]Matty Wilson That was the one that was the most successful I would say because it made money I made similar the trip and I as well it was called Traffic watch NI and it pulled the RSS feed from the alcohol is there a website is the website actually called Trafficwatch NI mine was called something like that Traffic NI Traffic or something like that it pulled the RSS feed and plotted that all on a map it was rejected from the app store for being too simple which really hurt and all I did was you could share traffic incidents with your friends it allowed you to push it to Twitter or Facebook yeah I added that that featurein and that got me past the app store reviews as well So there was another landmark moment where I was actually rejected for making an app that was too easy which was I know I took it very personally because I spent ages working on it and they just flat out reject it for being too simple

No it would’ve just been allowed straight through No nobody cares about it

current side project then there’s an app called quiet time Bible It’s a Bible reading App and I built it because I needed it And this is probably why it’s been the one that I’ve put the most effort in I was Doing stuff on my ipad and I needed a good Bible app and the biggest one isn’t very iPad friendly So I just built an iPod friendly version that did what I needed it to do And really the thing that I needed to do was being able to copy text either by dragging and dropping it or just copying it out so I built that up and Yeah it gets regular downloads then I’m pretty proud of it I use it all the time My dad uses it as well He loves it So it’s good motivation And that’s the one where I’ve built it Not really to learn anything but just built it because I needed it on that’s what keeps me putting regular updates So I’m currently working on changing everything to use Swift UI and I’ve actually as part of that I’ve found a nice pattern which I think would if I was to work on a iOS app now I would probably copy that where instead of using Swift UI for my whole app because it’s a bit immature right now I use Swift UI to build different parts of like low level things like settings views and things like that andI can just drop them into my current navigation stack that I have So it means that I keep the part that’s of Swift UI we have find complicated at Instil is just dealing with navigation so you can still use UI kit to do all of that but then host Swift UI views inside of your existing stack And it’s just been a really nice way to migrate my app which I built before Swift UI was even out and it just slowly add new things So I started with my settings view Which looks like any other iOS settings we use swift UI’s perfect for that And when I was able to just drop it in and as part of that you get things for free Like you get dark mode support andthings like adaptive text where if somebody how their text size turned way up on their iPhone my view now supports that without me having to actually do anything So it’s nice as you’re working on a side project to figure out patterns that you can then bring into your day job

[00:16:42]Ryan Yeah cool But you would say then it’s just because your quiet time APp is published that it’s not finished

[00:16:49]Matty Wilson No it’s not finished because one of the things I wanted to do was to build an app that is always adding features to it with iOS releases So if Apple adds something new in iOS 15 I would try and find one thing That I would add to my app because I like doing it it’s more like relaxing for me to play out with that stuff And then I get to learn about that thing So in iOS fourteen obviously widgets where the big thing So I added a widget to my app there was a nice you know couple of hours of work just So whatever happens in iOS 15 I’ll just pick one thing and add it in to quiet time now So I also released this year the Mac version as well which was fun. It’s using Catalyst just it’s not a native app but I’m going to write my own native Mac App as well because Catalyst is nice but it’s just not ready yet It still feels like you’re using an iOS app On your Mark. Again it’s just a nice way to slowly migrate onto a different platform And I find that on the Mac app store you’ll get a lot more downloads and than you will on the iOS app store because of less Apps so it’s sorta like the early days of iOS where you can get to the top of search results a lot easier

[00:17:53]Ryan Yeah but I think Mac users tend to be a bit more discerning than iOS users Is that a fair statement

[00:17:59]Matty Wilson To be honest I have I’ve had a lot of iOS feedback very little Mac feedback just waiting for that to start It took a while Whenever I first pushed the iOS version there was radio silence for a long time But and I have been getting regular emails about adding different Bible versions things like supporting adaptive texts and stuff like that I will get emails from people Somebody complained that my dark mode was too dark and they actually sent me an email saying that they would like it to be a bit lighter Which I just thought was funny I’m just using like the native darkness stuff I didn’t do anything fancy for it it’s you have to take it up with Apple even get an email Like that’s nice to know that there’s somebody out there using the app that you made and hopefully they’re finding it useful no matter what it So if you’ve written a piece of code and somebody is using that it just feels really good I guess like we get paid to build products with people and people use it all the time but for some reason It doesn’t feel as satisfying as something that you’ve built just feels like more of an accomplishment

[00:18:53]Matt McComb It’s nice to think One day you might see someone using one of your side projects in the wild as long as it’s not causing them harm or telling them the wrong bus

[00:19:05]Matty Wilson you got to find something that doesn’t feel like it’s work That feels that you’re just playing about and It’s hard to do that sometimes think you just gotta be really passionate about what you’re trying to build if it’s just some thing that you’re lile Oh well this is a good idea Even if you think this is an idea that’s gonna make me millions If you’re not passionate about it you’re probably not going to work that hard on it

Thanks to Amy for producing these.