it's so fucking hot in here and I've got no work and still 3 hours left to go...just let me go hooome!

Cordova or Xamarin?

I need to build a mobile app again, mostly for Android, because an interior Official app's functionality is so interior (written but monkeys) that it has finally pissed me of enough to take action.... And possibly be profitable (ads banner).

Mostly targeting Android but hey why not hit them all or at least also UWP.. I ain't paying for the Apple license... (Or maybe dfox could help release it...)

Anyway, which is better? I kinda want to pick up React and maybe try Typescript? (currently work with MEA(v1)N at work)

But I also use C# WPF, but mostly for my own projects these days when I need a desktop GUI.

I don't really feel like relearning Android SDK, I learned the SDK back when it was for Ice Cream.

IntelliJ, IntelliJ I love you but please tell me why making a jsp file from scratch doesn't work in the servlet but your template for a jsp does. It makes no fugging sense to me. StackOverflow doesn't know and I went there for help.

I was debugging that shit for 5 hours today over an issue that was fixed in 3 minutes. I need ice cream.

I’m working on a new app I’m pretty excited about.

I’m taking a slightly novel (maybe 🥲) approach to an offline password manager. I’m not saying that online password managers are unreliable, I’m just saying the idea of giving a corporation all of my passwords gives me goosebumps.

Originally, I was going to make a simple “file encrypted via password” sort of thing just to get the job done. But I’ve decided to put some elbow grease into it, actually.

The elephant in the room is what happens if you forget your password? If you use the password as the encryption key, you’re boned. Nothing you can do except set up a brute-forcer and hope your CPU is stronger than your password was.

Not to mention, if you want to change your password, the entire data file will need to be re-encrypted. Not a bad thing in reality, but definitely kinda annoying.

So actually, I came up with a design that allows you to use security questions in addition to a password.

But as I was trying to come up with “good” security questions, I realized there is virtually no such thing. 99% of security question answers are one or two words long and come from data sets that have relatively small pools of answers. The name of your first crush? That’s easy, just try every common name in your country. Same thing with pet names. Ice cream flavors. Favorite fruits. Childhood cartoons. These all have data sets in the thousands at most. An old XP machine could run through all the permutations over lunch.

So instead I’ve come up with these ideas. In order from least good to most good:

1) [thinking to remove this] You can remove the question from the security question. It’s your responsibility to remember it and it displays only as “Question #1”. Maybe you can write it down or something.

2) there are 5 questions and you need to get 4 of them right. This does increase the possible permutations, but still does little against questions with simple answers. Plus, it could almost be easier to remember your password at this point.

All this made me think “why try to fix a broken system when you can improve a working system”

So instead,

3) I’ve branded my passwords as “passphrases” instead. This is because instead of a single, short, complex word, my program encourages entire sentences. Since the ability to brute force a password decreases exponentially as length increases, and it is easier to remember a phrase rather than a complicated amalgamation or letters number and symbols, a passphrase should be preferred. Sprinkling in the occasional symbol to prevent dictionary attacks will make them totally uncrackable.

In addition? You can have an unlimited number of passphrases. Forgot one? No biggie. Use your backup passphrases, then remind yourself what your original passphrase was after you log in.

All this accomplished on a system that runs entirely locally is, in my opinion, interesting. Probably it has been done before, and almost certainly it has been done better than what I will be able to make, but I’m happy I was able to think up a design I am proud of.