I don't fully understand all the networking protocols involved but who/what is responsible to manage this kind of failure in the network? How should this ideally be handled besides cloudflare engineers calling someone?
The problem is there's no clear way to get started with setting up a python package. Every other day someone ask a question on SO about it and the first google result is outdated distutils IIRC
> Then there's the contemporary modus operandi that gives the console its name. Every Monday, via WiFi, owners receive a new game, the notification light on top of the case blinking to announce its arrival. Whenever you have five spare minutes, you’ll be able to reach into your own shirt pocket, and make time for your Playdate.
After the device is released and is available for sale, a new game will be released for it and downloaded automatically each week for the first 12 weeks, when downloaded a little light will light up and you can have small play date with the Playdate.
That's IMO the only clever part of the device, the syncing up with users everywhere anticipating the minute a new game drops. It'll work great until a game fails to resonate with any particular user at which point negative sentiment and disappointment get attached to the device.
I'm sure it's a fun challenge for the game designers with the limited screen, and a crank mechanic.
But I'm really hoping they are also able to make some games with some depth and replay value to them for the players, a la the early mario, spyro, sonic or pokemon games.
I like the idea but it seems the author is just slicing the normal distribution at a point of interest and calling it bifurcated? e.g. with the phone example. "good enough android" or "iphone / flagship android" all the actual phone models are hidden under two higher level labels. Isn't every decision making process with buying essentially try to get the best for your money?
One of the big challenges with data ownership is policing it. If I grant one-time access to my data to some company, the company can store the data and use it many times over, potentially distribute it etc.
I can see this working for datasets like website interactions or search history which companies could pay a subscription for and receive periodic updates which would be of continued value.
Some practical things that have helped me, by no means an expert:
1. Practice writing new project's from scratch. not so much todo apps but challenge yourself e.g. I wrote my own function as a service engine
The main point of the exercise is to show:
- Best practice is contextual, so first you need context ;)
- Think about the big picture and all the pieces to get there
- Repetition. It will help you improve your development flow and process, and become better at your craft
You can borrow design/implementation ideas from current projects that you're familiar with or use a design pattern you found on a blog etc. it will help you recognize when certain practices are the "best".
2. Rewrite this code continually. Is there a better way? more concise? faster? easier to test?
It seems there's many ways to do the same thing. Which one is best depends on the situation.
3. Optimize last. No seriously.
If I could count, the amount of times I've thrown away code because I optimized to early...
tl;dr best practice needs context. This requires big picture thinking. Learn to recognize contexts and the best practices are alot clearer
