This article skips a few important steps - how a faster CPU will have a demonstrable improvement on developer performance.
I would agree with the idea that faster compile times can have a significant improvement in performance. 30s is long enough for a developer to get distracted and go off and check their email, look at social media, etc. Basically turning 30s into 3s can keep a developer in flow.
The critical thing we’re missing here is how increasing the CPU speed will decrease the compile time. What if the compiler is IO bound? Or memory bound? Removing one bottleneck will get you to the next bottleneck, not necessarily get you all the performance gains you want
At no point did anything say when I was creating stacks that this info would be public for other users of the site! I was shocked to find what I’d been ranking (which luckily was just chip flavors) available for anyone to see. What if I’d tried putting clients to evaluate in there?
You can’t just collect personal data and share it like that!
One thing this article doesn’t mention is how this all falls apart when you spend time in more than one country.
My UK bank (Barclays) won’t let me install their app on my US iPhone (i.e. my phone that uses a US based iCloud account). Tesco won’t let me use their loyalty app. I can’t install an app that’ll let me order Starbucks or McDonalds in the UK (I only have access to the US versions of these apps). I can’t watch Star Trek because the US paramount plus app detects I’m in the UK and I can’t install the UK version.
I could switch to a UK iCloud account but then when I’m in the states everything falls apart the other way round.
I’m really curious about the number of requests you end up making. I’d imagine that this would get expensive fast in a web app - AWS charge $0.40 per million read requests, so if you’re seeing ten request to your app server a second and that ends up making a hundred requests to S3, that’s going to come in at at about $34 a day!
A lot of this article seems to be conflating two things:
1. The ability for the code to process the error and recover from it
2. The ability to sensibly log, trace, and later for a human to understand the error after the fact
The later of these is best handled by using a tracing library systematically throughout your code to spit out open telemetr and “logging” the error within your local span. You can then debug the problem with a tool like honeycomb which’ll give you a much clearer idea of what’s going on. Having tracing in place will also help you debug in cases where you don’t have errors but your system is running slowly, doing something unexpected, etc.
This greatly reduces the problem domain you have to solve in actual error handling to just having the computer recover when something is recoverable. And in these situations it turns out that most recoverable errors are best expressed as “non-errors” once they migrate any distance from the origin of the error (e.g. the record not found turns into a standard “user not known” status return value - it’s no longer an error, but an expected behavior of the system!) So for these “short lived” errors that are either handled locally or bubbled up all the way to the top and never handled, I’ve found that Go’s native basic error framework is fine.
Before I put down cash I have basic questions, like “how do I type a square bracket?” and “how do I type a backtick?”
The on-screen keyboard is pretty good, but makes coding and typing markdown hard. Will this solve this?
Also, this seems to be missing cursor keys. How do I move back and forward precisely? If the onscreen keyboard is hidden, I can’t even long press on space to move the cursor.
I would agree with the idea that faster compile times can have a significant improvement in performance. 30s is long enough for a developer to get distracted and go off and check their email, look at social media, etc. Basically turning 30s into 3s can keep a developer in flow.
The critical thing we’re missing here is how increasing the CPU speed will decrease the compile time. What if the compiler is IO bound? Or memory bound? Removing one bottleneck will get you to the next bottleneck, not necessarily get you all the performance gains you want