Reality According to Matt

I spent too long this weekend getting a “hello world” react native application up and running just the way I wanted it. I just wanted to detail my experience and my gotchas.

Goal: Get a “hello world” react native v0.49 app running on android on Kitkat API 19 using the Genymotion 2.11.0′s android emulator on OSX 10.12.

What I did: The installation of react native is pretty straightforward- you need ruby 2.4, nodejs, watchman and in my case, I wanted to use yarn over npm.

I didn’t want to use android studio so I opted to install things via the android sdkmanager (which you get from the sdk tools package you can download here). What isn’t detailed in the articles I found is what android studio is actually downloading for you when you install an sdk. For use with Genymotion (personal version downloadable here) I found I needed the sdk platform, build-tools, platform-tools and optionally the sources. The installation of these via the sdkmanager looks like:

sdkmanager “build-tools;19.1.0“ “platform-tools“ “ platforms;android-19“ “sources;android-19“

You can either let the sdkmanager use its default location to install to, your you can specify your own. I’m pretty sure it’s a better pattern to pick the installation location so that you can run multiple versions of android SDKs, but for to get it up and running, I just went with the defaults.

With the sdk + tools installed, next it was Genymotion. I installed it and then configured it- under settings > ADB, select “use custom android SDK tools” and selected the top level folder that contains all of the packages I installed. The tools Genymotion needs are aapt and adb, so the folder you choose needs to contain these files in its subdirectories. If Genymotion can’t find everything it needs it’ll tell you, otherwise you’re good to go.

After selecting an appropriate VM in Genymotion and having it start successfully I was trying to do 

yarn start

and launch my android connection but was getting an error like:

"could not install *smartsocket* listener: Address already in use"

After bouncing around between articles and reading about adb there are apparently a bunch of reasons this can happen. From the article linked above:

When you start an adb client, the client first checks whether there is an adb server process already running. If there isn't, it starts the server process.

and from starting adb in nodaemon mode I found that `yarn start` was starting adb server, but Genymotion was also starting an adb server itself- but to do this it terminates the running server that yarn started somehow. For whatever reason when you try to rerun the android connection from the react native application to bind to the android VM’s adbd process, it tries to start the adb server but throws the above error because the default port for the adb is used by Genymotion. Why don’t they coordinate? Why is this happening?

What I found is that it seems to have something with where the adb executable in invoked from and how. Because installing the sdk from sdkmanager doesnt manipulate your PATH, I noticed the processes had varying paths to the adb executable in ps. I figured I’d try making a symlink to adb in /usr/local/bin and gave it try- that fixed it! There must be something in the executable or a PID file or something somewhere that facilitates this restarting coordination and it took the symlink to bring everything in line.

I've never been a believer in mantras. You hear about them used by characters in movies or TV shows occasionally- usually by someone who's been painted to be particularly desperate for some kind of self confidence. In my head I had always thought that a mantra was something you had to take time out of your day to consciously do or observe (on TV they often listen to mantra tapes while sleeping), even if it means turning on that track in your car on the way to work every morning.

Recently, I watched a great interview with Ben Milne of Dwolla on ThisWeekInStartups (if you aren't tuning into this show, you should be, it's amazing) so I started following him on twitter. He posted a tweet that I really liked:

Ship or Die.

— Ben Milne (@bpmilne) October 8, 2013

So simple and succinct- it rolls off the tongue and has real meaning to me. I found myself repeating it whenever I needed inspiration or just...whenever because it sounds so cool. I found the quote to be so resonant- I decided to make it my password at work! (I waited to blog about this until after I changed it, don't worry ;) In the beginning I remember feeling really empowered when I was typing it...I'd get a little jolt every time I'd punch in those letters because this is what I really believe. The interesting part of this story is that I really feel like this password, something so simple, had a profound effect on me. The various transformations I had to apply to the quote to make it fit our password standards at work made the password somewhat hard to type. Normally after a few days, muscle memory takes over and you don't even think about what your typing in anymore- but not this password. I had to think about the keys were and in turn think about what I was typing since I was flying all over the keyboard. So in my head- 10 to 20 times a day I would repeat: 'ship or die'.

I've heard stories about startups that do something similar to what I had just experienced. The guys at Airbnb put up a poster in the middle of their bathroom mirror of a hockey stick growth chart so they had to stare at it every time they walked by. Alexis Ohanian from reddit put a quote up in the founders apartment from a business person at Yahoo saying that their '...initial traffic was so small it could be considered a rounding error' as motivation. I think these passive motivational ideas are really powerful- just like advertising, they get in your face and you can't run away from them- only- these ad's are targeted directly at you by you.

Either way, this password really affected me. I still find myself stressing out over things I can't finish quickly enough and being overly hard on myself about comments I receive in code reviews. I'll submit some code or perform a task and I'll always say to myself afterward that I should have been better, or faster- I find myself picking apart every aspect my work and working really late to make up for what I believe is lost time. I've sacrificed a lot of sleep and personal time in service to 'ship or die'. I think reflecting on the work you've done is really important to growing yourself and your skills. But you need to be in a reasonable state of mind- but it's important to be good to yourself. It's the weekend while I write this so it's easy for me to admit I've been overly critical since I'm caught up on sleep, but I'm still trying to strike that balance for those Tuesday afternoons when things just wont work. In the heat of the moment when the deadlines are looming and you find yourself chanting 'ship or die' you might not be doing yourself any favors.

I think 'ship or die' has a lot of power- and it will take time to understand that power and use it create a healthy drive. I really suggest trying out something like this yourself to see what influence you can have- maybe you'll tap into some of the power that I found- just make sure you are able to take a step back from work to be kind to yourself, we deserve a little love now and then.

On Saturday, I attended the Hopper Storm Hackathon at Notman House. My friend Victor from work asked me to go and hack for a few hours to learn a bit more and find out some more about what Hopper was and maybe win a prize.

Hopper (whose website I cannot find) offers a product called Storm that follows a map/reduce sort of process. Instead of running in batch like hadoop, Storm runs from a feed- most notably a social media feed like twitters firehose or any other massive data feed. For this hackathon we had a few feeds available to us, Victor and I picked up a third guy (George) and together we decided to pick something we knew we could finish in a day. We decided to visualize geotagged tweets ontop of google maps with the option of following tweets with certain hashtags. Not exactly a difficult project, but we knew we could get something running end to end in the timeframe we had.

The code provided by the Hopper team was all java examples (the Hopper team were all writing Scala) so we did our processing on their cluster then saved our results to a riak database provided by hopper (to which I must add, wow riak is easy...incredibly simple key/value store to connect to and query). Our frontend was some javascript to make an AJAX call to a python REST endpoint we were running that queried the riak database and just dumps the results back to the frontend. You can check the code on my github (its a real mess, but it was a hackathon- cut me some slack), but the final result was pretty cool.

We were getting down to the end and trying to think of something to add in the last few minutes. There was a sales rep there from a company called Semantria that does sentiment analysis and natural language processing for big data. We thought a cool idea would have been to add sentiment analysis to tweets showing up in the same area and try to do some more classification and grouping to provide more info on the frontend than just tweets in the same location. Unfortunately we just didnt have time, but that would have been pretty damn cool.

Unfortunately we didn't win the hackathon (not that I'm surprised, we didn't exactly innovate). The winners did a pretty cool analysis of the twitter feed using finangle and redis to find youtube videos linked in tweets to find videos that are being shared by users and deciding whether they are trending. They managed to pick out some seemingly popular youtube videos (several popular k-pop songs) that had been linked a bunch. It would have been pretty cool as well to see some time analysis too- since anything linked 12 hours ahead or behind us will be big in Korea or Southeast Asia.

In late 2012, a friend of mine, Manish Singh who lives in NYC was telling me that he was gearing up to quit work and start his own company (qhojo now live!). He was telling me all about his idea, and was looking for criticism and feedback to try to build it up and close loopholes. Manish is the second friend of  mine to ditch their full time job to start their own gig. Even earlier in 2012 I had another friend and collegue, Lex Youssef quit his job to work full time with Jintronix a company he has begun with a few of his friends a year or two prior. They had just been accepted into a Microsoft Accelorator program for Kinect based startups through bizspark and were moving to Seattle for a few months to work with the Microsoft engineers to help build out their product.

Getting to talk about business ideas and starting a company reminded me of being back in school. Getting to talk with my friends about how awesome it would be to start a company, be our own bosses and change the world. It gave me a huge surge of energy- why did I, all of a sudden forget my dream of starting a company? It’s like I got behind a desk and got swamped with work and it really killed my innovative spirit. I decided after speaking with Manish that I needed to get back on the startup bandwagon and channel my efforts into innovation once more.

When you really delve into the world of entrepreneurship, you really come to find the huge community that exists. There are so many people who are motivated and itching to find the next big idea that they can turn into something amazing. There are tons of excellent mentors: people who have been there and understand the ins and out of startups and how to begin. I was introduced first to Startup Canada(powered by Victoria Lennox) then in turn to Nottman House in Montreal, and finally to Startup Weekend.

I love hackathons. I love the idea of coding really hard with a few people for a few hours and coming up with something (maybe) in the end. When I read about Startup Weekend Montreal I thought it would be a great chance to meet some new, motivated people and maybe have a cool experience. Needless to say, its worth writing a blog post about.

I had a blast. For a quick introduction, designers, developers and business people begin on friday night by both pitching ideas for a business as well as being pitched to. People join the projects they like, and teams are formed. You hack all weekend and on sunday night you hold a presentation and a demo in front of a panel of judges (CEOs, VCs, etc). The winners get recognition and some free swag to help get them off the ground if they choose to continue working on their idea past the weekend. The event I attended in Montreal was a ton of fun- one of the highlights was getting to run into a guy I had done an internship with in Ottawa a few years prior, Jean-Pierre Simard. JP had founded a mobile dev shed in Ottawa and had brought his team to Montreal for the weekend to hack on an idea they had. Dubbed Dashbook, these guys KILLED. They had the expertise, the drive, the team...I was really jealous of their cohesion and felt stupid for not joining them. They came second in the end, but after SW was over, they were approached by Founder Fuel, a Montreal based accelerator to join their summer cohort to try to develop Dashbook full time to see what they could create.

Brandon Chatreau was interviewed for a promo for the next SW Montreal, he sums it up pretty well.

I met up with their designer, Mo Mozafarian again when I attended my second SW in Ottawa. This time, my friend Cole and I pitched an idea for building a community around artisan beer, and finding great local beer in your city. We got a hold of some amazing people- and managed to crank out a pretty amazing demo and presentation. We didn’t win, but we all love the idea and are moving forward with it. I just need to stop with all these weekend production support issues and maybe I can get some time to write it!

  Honestly, these events are so much fun. If you have an idea and are wondering if it has legs, this is such a great venue to explore it. I can’t even begin to describe the energy- its unbelieveable. The hustle is incredible and the feeling you get of drinking way too much coffee and being surrounded by people who are excited and motivated is really addicting. At the end of the weekend you’re so sad to go home and leave behind your cohort, it makes going to your day job the hardest thing to do on monday morning.

I’ve been an on and off reader of Penny Arcade for probably 8 years now- when I was more of a hardcore gamer I occasionally understood their jokes whereas now I’m so far out of the loop, I’m beyond help. I recently came back to it however with their release of PATV, which consists of an inside look at their office life and how they do what they do. I honestly love the insight in their day to day, but their newest project has me completely enthralled.

After a kickstarter they posted in July of 2012 they posted a stretchgoal of turning their usual season of PATV into a reality show called Strip Search: Finding Americas Next top Webcomic. I was hesitant at first to trade the show I really enjoyed for a reality show but I could not have been more wrong.

In short: I am totally hooked on this show. I’ve always been jealous of people who are gifted when it comes to drawing and illustration, and they managed to assemble some extremely talented people for me to drool over. The contestants come from all over the US and are (tastefully I add) stuck them in a house together to compete for exposure, office space inside of penny arcade for a year, and $15 000: big money for starving artists. Penny Arcade opened up by accepting applications of any american webcomic artist to be judged and ultimately narrowed down to 12 finalists (6 men, 6 women) who were to be flown to a venue in Seattle to compete. The great part about the show is that it really isn’t about strategy or drama like your typical reality show; the producers are treating the house as more of a common ground to house the contestants between challenges to assess skill.

The show is nearing its end (I’m praying for a season 2) and I’ve been keeping a very close eye on it. I follow all of the contestants work now and have been introduced to some seriously cool new art. Just over the weekend, I had the good fortune to send a tweet to one of the contestants, Abby Howard who happens to live in Montreal (@AbbyHoward) and asked her to join my girlfriend and I for a beer on a patio. Somehow she saw my tweet and agreed to meet up with us- so we got to talk about the behind the scenes of the show, what it was like during filming and how its affected her career. I don’t know much about driving traffic to a website outside of youtube, so it was really interesting to hear about how her analytics changed based on what was happening on the show and how it has changed her life as well as her ability to make money.

It was super cool to meet Abby. I was starstruck the whole time but I think I managed to not make a fool out of myself. It was more an interrogation than a conversation while we were drinking, but she was super cool about it. Abby was really funny and very cool- I need to meet up with her again, especially after the show end. One thing that has been made very clear after speaking with Abby is I really need to take an art class, I need to see if I can do this too!

Recently at work I've been doing a bit of WPF development and been having fun with it. I was excited at the prospect of working with WPF since I've been getting into some C# development and I've been enjoying the language a lot. I figured it would be a nice addition to my repertoire since my last crack at GUI development was with the Qt framework in C++ a few years ago.

I was writing a new component to an already existing tool that has been implemented using the MVP (Model View Presenter) pattern. I was told by a senior developer to check out MVVM (Model View ViewModel) as in his opinion, that was a more correct pattern for WPF development. After a lot of reading and thinking, I definitely agree.

I read many articles on WPF, most notably Josh Smith's articles (specifically this one), found some great examples online, and got a pretty good understanding of the subject. I've really come to see the beauty and power of WPF, and I barely scratched the surface. Through databinding, we can create a truly thin view layer, that gives us the flexibility we need for a lookless ViewModel, much the same way we separate the "what" versus the "how" in html/css. Commands serve to enhance this decoupling, since in patterns like MVP or PM we see coupling by way of events. Commands free us almost completely from this coupling by use of the RelayCommand (also mentioned in Smith's article), so we have a much more generic way of dealing with the actions created by the GUI.

While my application utilized probably less than 5% of WPFs capabilities, I was still able to find a few things that really caused me problems, the main one being Scrollpanes.

Oh my god, what a mess. This was given to me as a suggestion near the end of my project because the user wanted to be able to shrink the window to have it always present on their screen as they were working. Stupidly, I agreed outright thinking it would be trivial to implement...what a mistake.

The use of Scrollpanes makes most WPF components go completely haywire. I had been using a GridLayout with min and max dimensions set, with a ListBox inside a panel inside the GridElement, but adding the scrollpanel to the main window causes the ListBox to stretch uncontrollably and prevent my GUI from looking anywhere near proper or useable. I played for a very long time to tame the controls but to no avail...alas I had to hack it. 

I tried lots of different types of Panels instead of a StackPanel in my grid to limit the Listbox, but they all yielded the same result. I decided that since my ListBox sits inside of a Grid, I needed to bind the MaxHeight of that grid location to some other value in the control to keep the ListBox from growing or shrinking too much. The problem with this is that there is no element that you can bind straight to and get the exact height your looking for.

Enter the hack:

My height was just a tiny bit too big creating a weird always offscreen ListBox (in fact 33 pixels too large, which was the height of the label sitting above the ListBox). So I implemented an IValueConverter:

class HeightToAdjustedHeightConverter : IValueConverter {     public object Convert(object value, Type targetType, object parameter, 

CultureInfo culture)     {         var height = (double) value - 33d;         return height < 360d ? 360d : height;         //360 being the minimum height allowed for the listbox     } public object ConvertBack(object value, Type targetType,

object parameter, 

CultureInfo culture) { return null; } }

All I did after that was include it as a converter for the binding on MaxHeight (Note you need to name your usercontrol and bind to its x:Name. NOT ideal):

<Grid Grid.Column="0"     Grid.Row="1"     VerticalAlignment="Top"     ClipToBounds="True"     MaxHeight="{Binding ElementName=AdHocUserControl,         Path=ActualHeight, 

Converter={StaticResource HeightToAdjustedHeightConverter}}">

The only other alternative I could think of is to extend one of the panels and try to play with its growth behavior.