davidbanham

From an Australian perspective the main factor in us not getting as many hazard reduction burns as we’d like completed isn’t financial it’s the weather. There’s such a thin window of time when the bush is dry enough that we can actually light it but the weather conditions aren’t so dangerous that to do so would be foolhardy.

We could spend money to eke out a little around the margins like using more paid firefighters to do mid week burns. It’s not addressing the main issue, though. The climate is changing.

I heavily favour goretex gear even when the item doesn’t really need to be waterproof. More than the membrane, I respect gear with the goretex label because of the durability standards they enforce on manufacturers using their product. I know that this piece has been independently tested and it will last me well for years before needing replacement.

I’m a bush fire fighter in Australia. We do make use of technology and we’re bringing more in but reliability and dependability is _so critical_.

All of my brigades trucks have iPads which are useful for accessing maps, running collaborative fireground apps like collector, firemapper or rfsBuddy, etc. All of these things are “nice to have” though. We don’t rely on it for anything we must have, because it fails from time to time.

We use apps like Rover and Bart to manage call outs. They’re great because we can see who’s responding, how far away they are, etc. We also all carry ancient pagers because they’re bulletproof and mobile phones aren’t. They’re served by a hardened transmission network where every tower has backup power in case the grid goes down.

Tech is great but in extreme environments like a wildfire I want to have a paper map with me and I want to know how to use it. Just in case.

Try volunteering. I had this exact conversation yesterday on a firetruck on the way back from a hazard reduction burn, that one of the things we all valued most about our volunteering experience was that we were exposed to people we wouldn't normally hang out with in our professional or social circles.

That said, on this particular truck three people out of the crew of four were programmers. We thankfully pulled up the conversation about the merits of Jupyter Notebooks for BI tooling before the fourth crew member, who is studying forensics, got so bored her eyes rolled right out of her head.

Ah, yes that makes sense. Thanks for clearing it up.

It's not clear to me how the Native File System API relates to the existing File System API.

https://developer.mozilla.org/en-US/docs/Web/API/FileSystem

Is Google proposing this as a replacement? As a supplement?

I made heavy use of the FileSystem API in a cross platform educational application a few years ago, tripping over a really hairy memory leak bug in Webkit along the way. By the end of that project the API had been deprecated, but apparently now it isn't any more and instead it's experimental? I have no idea what that means.

I suspended my inReach plan during the lockdown but was about to re-enable it for a motorcycle camping trip next weekend. Hope the website is back up in time!

Because then all the robots (vacation responders) reply too. The separate address in the body of the email is a little like a captcha humanity test.

Arch for my development machine. It’s been the most trouble free distro I’ve ever run.

All my production stuff is dockerised and running on GKE these days. Most are FROM scratch with a statically compiled go binary in there. Some are on alpine and there’s the odd bit of Debian here and there.

Tim worked for Amazon up until last month, so it's front of mind for him. The gag is that this is a proposal in a hyperspecific internal format being written in the public domain by someone who is now an outsider.

Does it allow you to set a hard time limit on how long the candidate has to complete the challenge?

> Your application doesn't need logging until you have a problem that is solved by logging. It's possible that you never have such a problem. If that's so, then building a logging subsystem is a waste of time.

Your application doesn’t need logging until _right before_ you have a problem that is solved by logging.

I don’t use frameworks these days. I’m glad I was exposed to their ideas in the past. I’m glad that exposure shaped my current approach. A framework is one of the ways that experienced developers can pass their knowledge on to less experienced ones.

I use QWERTY but on an ergodox. Massively improved my experience at my usual workstation and it’s still seamless to jump on a normal keyboard if I’m using my laptop or someone else’s machine.

Good technique helps prevent injury. Rheumatoid arthritis and carpal tunnel are no fun. Adding “it’s going to hurt” to the list of reasons not to sit down and knock out that feature really isn’t helpful to your productivity.

I ended up tracing my pain to lazy technique in my right hand. Sorted that out and I’ve been pain free for years. It was _no fun _ before I figured it out.

That's the joke.

> I don't think it would save the university money, if done well.

Why is that the targeted benefit? I would think the main thing to gain from online courses is the ability to disseminate knowledge and learning to a broader swath of students.

Oh hi! There are very few Banhams in Australia for some reason. I always need to spell it out for people. On the rare chance a British or Irish person needs to write my name though they know how to do it straight away and I get unseasonably excited.

I also used to take a massive efficiency hit when I had to move away from my multi-monitor setup. I wanted to be more flexible so I invested some time in getting comfortable with a tiling window manager.

My weapon of choice is Xmonad on Linux but I've also used i3 and awesomewm and they've been good. I hear good things about Amethyst on OSX these days too.

I'm very excited by this!

I currently use https://materializecss.com/ over several products. It's been great, but I'm stuck on the 0.100.x line because the upgrade path was too difficult. It's also hasn't been updated since the first 1.0 release some time ago. Not complaining since I got what I paid for (it was free) but it's left me pretty anxious to find a replacement.

Salesforce's Lightning components are good but challenging to understand how to work with outside the walled garden. Zurb's Foundation UI is interesting but seems to want me to use whole page templates rather than components.

This looks like exactly what I've been dreaming of. I've put my money where my mouth is as I can't wait to have more components built out to a stage where I can start using it for real.

Holy hell I never knew the old.reddit.com domain existed.

Thank you!

Whenever I can and it makes sense, yes. When I write a test first the resulting code is always better.

Sometimes the change is so trivial or type safe that it's not worth it, so I don't.

Sometimes I don't understand the problem well enough, so I learn more about it by doing some exploratory coding and prototyping. I usually come back and write a test after the fact.

Sometimes the project is on fire and I'm just throwing mud a the wall to see what sticks.

I should really get around to writing some docs for this, but you may find this useful:

https://github.com/davidbanham/kewpie_go

I’ve been using it for years across multiple products.

I also built this for a client lately. It lets them interop their Haskell services with kewpie a bit easier without needing to write a native wrapper.

https://github.com/paidright/kewpie-http

I’d love to be able to ditch my pager and rely solely on smartphone apps. Unfortunately there’s no way to get our alerting apps to override silent mode on iOS and they’re patchy on Android. So I keep a pager next to my bed to make sure I get woken up if something’s on fire.

Works as expected on an iPad. Might be a browser inconsistency.

I wonder why he's getting rid of the vehicle.

There's a lot more going on with a Pi than an ESP chip. More bits, more failure modes is my reasoning.

I use it to run my automated watering system in my garden.

https://github.com/davidbanham/relay_runner

I used to use an esp8266 but replaced it with a Pi after a lightning strike destroyed it. The Pi is just so much easier to write software for and it's been no less reliable in practice.

Very true! I think it's actually an interesting constraint to place on business model ideation. When you restrict yourself to opportunities that you could reasonably pursue solo it forces you to be more creative. The upside is that your total addressable market doesn't need to be nearly as large.

I've done this with the older lightning CSS based toolkit and it was great.

The front end was React/Redux and we just wrapped the bits of markup we needed into React components.

I volunteer on the board of my ski club. We have three lodges across two resorts and have used a couple of different booking systems over the years. They've all been pretty bad, usually falling over under load and then making financial errors.

Clubman is well tested (TDDish) software written in Go and hosted on Kubernetes (GKE) with payments through Stripe. It's done 6 bookings per second in real world usage and been synthetically load tested to 10x that.

Ski clubs are a small niche but they have very specific needs. There are _lots_ of booking systems for hotels but they're all geared towards filling as many beds as possible with people willing to pay (of course!) Ski clubs usually have rules to ensure all members get equitable use of the facilities while also trying to maximise utilisation. These rules can get very complicated with multiple classes of membership, multiple rounds of booking with different entitlements and limits on beds and guests, different rules for high and low season, etc.

Clubman implements a system of Rules based on Triggers and Actions. It allows clubs to reflect arbitrary sets of rules without needing to commission custom modules from the developer as with other systems.

Notbad Software is a one person operation. I've built Clubman over the last 12 months alongside consulting work and my other products. I'm very proud to launch it to all clubs today.