• 1 Post
  • 59 Comments
Joined 11 months ago
cake
Cake day: August 9th, 2023

help-circle



  • An American leaving the US is more comparable to a European leaving Europe rather than their home country.

    Hilariously enough by percentage, the number of Americans who have never left the US is similar to the number of Europeans who have never left their home country (40% vs 37%). That’s honestly insane given leaving the US is much more difficult than hopping on a train to go to an adjacent country in Europe. I’ve never left the US, but there’s so fucking much here that with the exception of a few culturally significant places in the world (mainly Thailand) I have no desire to really travel abroad.


  • I’ve noticed it’s less common in the city and more common in rural areas. I live in SF and people here don’t call them gas stations unless they have gas, but in the Central Valley this is extremely common.

    I grew up there and I always forget how much more “proper” I speak at home vs where I grew up. My partner sometimes struggles to understand what I’m trying to say a lot of the time when I slip back into it when speaking with my family. Gas station is just one of the many overly generic terms. Another one is “Vallarta” which doesn’t necessarily mean the chain grocery store Vallarta, but a Mexican grocery store usually selling produce and with a meat counter.







  • That was a pretty interesting read. However, I think it’s attributing correlation and causation a little too strongly. The overall vibe of the article was that developers who use Copilot are writing worse code across the board. I don’t necessarily think this is the case for a few reasons.

    The first is that Copilot is just a tool and just like any tool it can easily be misused. It definitely makes programming accessible to people who it would not have been accessible to before. We have to keep in mind that it is allowing a lot of people who are very new to programming to make massive programs that they otherwise would not have been able to make. It’s also going to be relied on more heavily by those who are newer because it’s a more useful tool to them, but it will also allow them to learn more quickly.

    The second is that they use a graph with an unlabeled y-axis to show an increase in reverts, and then never mention any indication of whether it is raw lines of code or percentage of lines of code. This is a problem because copilot allows people to write a fuck ton more code. Like it legitimately makes me write at least 40% more. Any increase in revisions are simply a function of writing more code. I actually feel like it leads to me reverting a lesser percentage of lines of code because it forces me to reread the code that the AI outputs multiple times to ensure its validity.

    This ultimately comes down to the developer who’s using the AI. It shouldn’t be writing massive complex functions. It’s just an advanced, context-aware autocomplete that happens to save a ton of typing. Sure, you can let it run off and write massive parts of your code base, but that’s akin to hitting the next word suggestion on your phone keyboard a few dozen times and expecting something coherent.

    I don’t see it much differently than when high level languages first became a thing. The introduction of Python allowed a lot of people who would never have written code in their life to immediately jump in and be productive. They both provide accessibility to more people than the tools before them, and I don’t think that’s a bad thing even if there are some negative side effects. Besides, in anything that really matters there should be thorough code reviews and strict standards. If janky AI generated code is getting into production that is a process issue, not a tooling issue.



  • I mean if you have access but are not using Copilot at work you’re just slowing yourself down. It works extremely well for boilerplate/repetitive declarations.

    I’ve been working with third party APIs recently and have written some wrappers around them. Generally by the 3rd method it’s correctly autosuggesting the entire method given only a name, and I can point out mistakes in English or quickly fix them myself. It also makes working in languages I’m not familiar with way easier.

    AI for assistance in programming is one of the most productive uses for it.


  • I’m forever grateful to have been on Kaiser my entire life, and that all my employers have had it as an option.

    It’s expensive up front (~$5k per year, my employer covers it thankfully) but the most I’ll ever pay per year out of pocket is $1500. Office visit/urgent care is $10, ER is $100 and waived if you’re admitted, prescriptions are $20, and the most expensive surgery I could get is $150 which includes the hospital stay if needed. My partner got sterilized for like $35. The biggest thing for me is my therapy is free so long as it is virtual (my therapist is 4 hours away by car anyway), and $10 for an in person visit if I make the hike.

    It’s absolutely wild how much one’s experience can vary with the healthcare system in the US based on their insurer alone.


  • I’d actually argue the complete opposite of OP for developers.

    The picture I use for professional stuff is a shoulder up photo of me in front of a brick wall with some greenery in front of it. I’m wearing a black hat, plain shirt, glasses, and a backpack. I’ve gotten dozens of interviews and recently a new job with this photo that I’ve used since 2020. I’ve even received compliments on it being a, “not fake photo”.

    Being too much of a “suit” in the developer world can actually harm your chances IMO. Meta actively tells interview participants to come as they are and outright says to not wear a tie because in their own words, “we care about your abilities, not your clothes”. Obviously clean up and look nice, but that doesn’t mean you gotta stress about appearance. I’ve personally done all my interviews in various hoodies and it’s never been an issue or counted against me as far as I can tell.

    Obviously fintech and finance is gonna be a little more formal, but I don’t personally want to work somewhere where how people dress is anyone’s concern.






  • Surprisingly, Remarkable tablets, despite not being open source, you can do just about anything with. They allow root SSH access and the backend is a heavily stripped down version of Linux.

    I’ve been writing an application to allow customizing splash screens over SSH/SFTP and it’s actually been super easy to work with. The “jailbreak” scene is also super active, and the company has gone the opposite direction of most. They retroactively removed the need for a subscription to cloud sync on all devices, and seem to very much embrace the ridiculous things people have done with their tablets.

    The device is also no nonsense and does exactly what it’s designed to do extremely well and no more. No ads, no bloat, no constant internet connection. You could never connect the thing to the internet if you really wanted. Honestly one of the few devices I’ve bought in recent memory that I feel like I wholely own.

    Two big downsides are no Bluetooth, and you need a modified hardware device to unbrick the device if you fuck up (jumping type C pins to put the device into recovery). Overall really solid and would recommend.


  • I honestly don’t think we’re gonna go any further than we already have in the everything cloud connected direction. The software industry is already pushing against “everything in the cloud”, and on-prem with cloud/off-site backups is becoming more and more common.