Great discussion between Ryan Androsoff & Public Digital's Anna Hirschfeld & Alexandra MacEachern. I sadly missed #FWD50 this year, but it was great to hear about the work to bring the leadership behind #ARadicalHow to Canada.

https://thinkdigital.ca/podcast/the-radical-how/

The Radical How (with Anna Hirschfeld and Alex MacEachern)

Canada has fallen from 3rd in the 2010 UN e-gov rankings to 47 in 2024. We’re stuck when it comes to modernizing government for the digital era. But o...

@RyanAndrosoff I forgot to tag you here.
@mgifford so glad you enjoyed the episode!
@RyanAndrosoff I was thinking of writing something recently about the missed opportunity with open source and #ArriveCan and then this episode rolled up in my podcasts. Perfectly timed.
@mgifford please do - would love to read it!

@RyanAndrosoff It wasn't much more than what I wrote on LinkedIn. It should be fleshed out more, but really...

Aside from different content, ArriveCan seemed like a much easier app to build.

It's a form that you need to fill out on a plane. That just can't be that difficult.

@mgifford @RyanAndrosoff whoa there was a LOT going on in the background of ArriveCa and constant changes in the rules. And people without phones and parents doing it for kids and elders and lots of logic rules. Not saying it was designed well (so much hardcoding 🤪) but let’s not go down the build it Ina weekend path a weekend path.
@mgifford @RyanAndrosoff and it was a form for boats, and different requirements for driving and walking across and places without Internet etc.
@lisafast @RyanAndrosoff so pretty simple functional compared to the Covid alert app.
@mgifford @RyanAndrosoff You mean Covid Alert was simpler? Because yes Covid Alert did seem much simpler and was built on a detection code base produced by others.

@lisafast @RyanAndrosoff i don't know what the requirements actually were, but ArriveCan was ultimately a mobile app with local storage and some forms. So was the covid alert app, but it also needed to connect to two types of networks to detect and communicate. ArriveCan sure seems simpler.

ArriveCan chose not to build on an existing, tested, open siurce code base.

@lisafast @RyanAndrosoff sure. That makes sense. Still you build for the simple conditions first. Test with users. Iterate. Much like the covid alert app did. Learning along the way.
@mgifford @RyanAndrosoff So much to say that I won’t. But simple it was not. There were two massive organizations involved and btw the rule changes that had to be implemented were top secret until announced - no testing with users when only a few people can even know what massive change to the functionality is coming…

@lisafast @RyanAndrosoff no worries.

But ya, sometimes bureaucracy shoots itself in the foot from just starting in the wrong place.

But even the starting from an open source solution would have given them more flexibility.

If that kind of flexibility were actually needed, the developers could have spent their time developing for 3-4 different possible requirements , rather than rebuilding the wheel.