I’m a staff software engineer at Sunrun, the USA’s largest residential solar installer.

I mostly work with kotlin, but also java, python, ruby, javascript, typescript. My hobby is picking up new hobbies. Currently bird photography and camping.

  • 19 Posts
  • 914 Comments
Joined 3 years ago
cake
Cake day: June 6th, 2023

help-circle

  • I was doing research the other day for my job and it involved all those data broker removal tools. One of them is by incogni and it’s supposedly “better”. They claimed all your data stayed in America except if you checked their JavaScript was literally sending data straight to an analytics company based in Europe.

    A different data removal tool called easy opt out on the other hand loaded zero JavaScript at all (or at least third party JavaScript) and had actual security claims on their site. But they haven’t done any audits so don’t trust them either!!!






  • I’ll give my reasoning. Or at least one story from a compendium of many.

    There is an OSS library called ktlint for Kotlin. It is a “no bikeshedding linter” that was supposed to be based on the default rules from JetBrains. That is, if a rule existed already from JB then ktlint would enforce it and if the rule wasn’t specified it would make and enforce a new one that the community agreed on.

    That was all fine and dandy until I and some others pointed out that one of the rules was the exact opposite of what JB had specified. When called out the owner of the repository didn’t care, until too many people were against his choices that he couldn’t ignore it anymore. He made it configurable.

    So the rule was fixed. Until he sold the repository to Pinterest. Pinterest just wanted to own ktlint so they could control the rules. They immediately changed it back and made it non-configurable. When pressed on it, they just refused saying it wasn’t going to be that way anymore. They didn’t care what the community wanted they just wanted to control the rules for themselves.

    The repository is open source, but they took the most widely used Kotlin linter and essentially stole it from the community. Getting enough people to switch because of one rule will never happen.









  • the amount of people that do not understand how big of an improvement passkeys are is really saddening. They think that somehow these tech companies are utilizing this in some nefarious way, rather than the very very simple explanation that … tech companies don’t want to be responsible for more breaches.

    Passkeys are so simple and such a huge improvement that it’s literally all upsides and no downsides. Either you use their passkey managers like you would their password managers and it’s safer for them, or you use your own password manager with passkeys in it and it’s still safer for them.


  • You have a vast misunderstanding of why passkeys aren’t transferrable or usable outside of those providers. It had nothing to do with lock-in, but because every implementation was different. And no, you do not ‘depend on the site’ to let you use that option instead of a major provider. There’s a standard now and everyone is following it. If you can use a passkey you can use your password manager to manage that passkey.



  • These were exactly the things that drew me to Svelte as well. And yes, the author is right. If they had claimed that you should use Svelte for everything I would have immediately stopped trusting them.

    Svelte is very nice for backend engineers to use for a frontend framework. It makes way more sense from a mental model, and it’s incredibly lightweight, at least from what I’ve seen in other frameworks.

    It’s the exact same way I feel about Ruby. You definitely shouldn’t be using it for everything, even most things, but the stuff it’s good at it beats Python by a mile. And it’s so much more pleasant to use.