This week has been pretty wild compared to others. For one thing, I'm writing this on Sunday night rather than Saturday morning. Publishing it on Monday morning. That's not a good sign.
This last week has just been one big escalation of people panicking and I've been spending a considerable amount of time reading into COVID-19.
My impression is that people are freaking out to early numbers for the mortality rates. These early numbers are misleading and will continue to raise sharply as more testing becomes available. I believe testing still isn't widely available. Testing for COVID-19 is reserved for those that are likely to have it. This is because testing kits are scarce. It creates a bias in the data that puts the average mortality rate much higher than it probably is. It's like to be under 1% overall. However, these rates are likely to be relatively higher for older age groups.
Sadly, people being scared and panicky is plunging the economy into another recession. It's concerning stuff.
On a milder note, I'm coining a phrase this week:
Legut Effect: Given enough time every prediction or scenario seemingly becomes true.
This past week has been less productive than I wanted it to be. I realized something valuable by trying to setup my repeatable distribution processes. That publishing to the web was the way to go initially. It helps iterate on design much faster and there is less to it. This helps with getting the software into tester hands faster.
I also made a small piece of marketing material that I didn't expect to make to help get testers. It may or may not get used for guerilla marketing.
I really don't know why I spent so much time on this. It really was a distraction and I got like maybe 1 tester out of it so far. Maybe I can end up using it again in the future but we'll see what happens.
I need to fix a CORS issue for the web published versions I can deploy. This results in a XMLHTTPREQUEST error otherwise. It's an inconsistent error and I find it interesting how it doesn't just fail all the time. I can't explain why that would be the case and I find that interesting. I would expect it to fail every search.
This past week has been mostly learning codemagic.io and how to navigate and use it. I've also noticed how stackery.io has now started offering http APIs to reflect changes in AWS. That is great. I am now considering changing my RESTful API gateway to an HTTP API because it would reduce costs.
It costs 1 dollar per million requests for HTTP APIs compared to $3.50 per million requests for RESTful APIs. You can see that here.
I've also stumbled across AWS activate. Which is a program to get some credits towards an AWS bill. It is something I have to read into more.
A closed alpha launch target has been set for May 1. I have three main concerns I need to focus on: Login, Inventory, & Stripe billing. Besides that, dismissive swiping where the price and sell through chance needs to be implemented correctly. That's probably all that really matters for that first alpha version.
A small to-do item is that I have to switch over to using https rather than http for my calls to eBay. They are strictly enforcing this at the end of this month. Ebay has a post about it.
Lots of what I needed to focus on last week will be carrying over to this week and through the end of the month probably.
Something I read about payment processors for mobile apps is that each store has their respective terms that try to force you to use their respective payment processor. However, I am uncertain about how true this may be.