igor47

I'm pretty mad about the blatant manipulation of index funds by spacex, and I looked into how to avoid the IPO. If you're like me, the amount of unrealized gains from index funds over the past few years mean the tax penalty of exiting these funds is just too big (the majority of my money is not in tax advantage accounts like a 401k). If not for the tax reasons, I would probably go with a direct index. Vanguard and Fidelity both have em. But honestly even at inflated float ratios this stock makes a very small part of the overall index. What I hate is forcibly being made party to such blatant manipulation, but like any other "concentrated benefits, diffuse costs" situation, it's probably too low a real cost for most people to deal with.

Multiple comments on this thread about how bad it would be if the main character was a woman. Why is that so bad?

Yeah agreed. I switched to kde from gnome a few months back, and it's amazing how much better it's been in a thousand little ways.

Are you me? I do feel like I'm starting to forget git as a result of my happy jj use. Thankfully some repos use git submodules, which keeps me at least a little connected

When I'm working in git, I always start work by creating a new branch with a name. Sometimes the branch becomes something different as I work and then I might rename it or more often just keep a stale name around. But in git commit descriptions come later.

In jj, it's the opposite. I start with a change, and I often describe it right away. Branches (bookmarks) come at the end.

You could, in jj, tag a new empty change with a bookmark as soon as you create it. You don't have to advance the bookmark -- that the first change in a sequence of changes is tagged with a bookmark is probably as much information as you need?

Code review is about getting a team on the same page more broadly. The question posed in the article -- whether the change should be part of the product -- is certainly part of that. It's important to have a shared understanding on a team re what the code actually does and should do. But you can also use code review to align on architecture, best practices, road map.

I also find that code review helps a team feel like a team. It's nice to know anyone actually cares enough to pay attention to what you do, and feedback is an amazing gift that's really underrated.

"just" is doing a lot of heavy lifting here. I always automate my repo setup, usually with mise tasks, but there's often assumptions baked in, eg around which ports the app will listen on. It's also still hard to get the agents to use "mise run test" instead of invoking pytest directly. This sounds harder to deal with when you have multiple copies of the test or dev stack running together on the same machine.

Nice. I ran into this issue a few months ago:

https://igor.moomers.org/posts/wled-christmas-lights#flashin...

Nice to hear it works now without the very hacky work around

Or maybe just an android emulator? I wonder if this would actually work...

Life without a smartphone increasingly challenging. You have to use either Google or Apple. I use a de googled Android lineage phone but this is always getting harder, as numerous threads on this site will attest. Plus literally every employer I've ever had has used Google services, plus lots of other sites I might have to use implement recaptcha or otherwise invisibly to me share my data or data about me with Google. Also, even if I do figure out a way to stay off Google's radar, they're a powerful force which shapes my world. They hire lobbyists to influence policy which affects me, build data centers which raise my cost of electricity, or sell killer robots to evil people.

I think where people go wrong is treating Google the way they treat their weird neighbor Bob. Bob's damage is limited. Google is an immense, powerful, alien entity, far beyond the control of any person, and with its own inscrutable goals which are the not goals of literally any person alive or dead.

I genuinely don't understand the desire to leave this entity unmoored to wreck what havoc it may.

But in a democracy, you at least have input! Google is also a coercive force with no real checks on its power, but it doesn't care about anything you have to say. That's the difference, that's it, right there. The answer to abuse of power is not to just unleash raw power, its to subordinate and restrict it. That's what government is for. When you find yourself arguing that power you participate in is bad and shouldn't restraint power you have 0 influence in, that's when it's time to wonder if they've gotten to you.

On this side of the wall, you and your friends are strong and happy and free in your garden. On the other side, a hellscape filled with giant monsters debating how best to filet you. You will keep ceeding them ground, your garden gets ever smaller. The monsters ate Brian, oops, well that's the consequence of freedom! But you're next, isn't it completely obvious you're next? Why would you unilaterally disarm against the monsters? Why for the love of God why would you say "no the monsters are good actually!"

Strong agree. That passage seems to me to be decrying the friction of the real world, whereas it's become increasingly clear to me just how valuable friction is in the world, and how inextricably tied the tech companies war on friction to the bad outcomes technology seems to engender.

No it's really bad

Yeah this is a killer app for mcp servers. My puppy has food allergies and the vet asked me to track his eating and pooping. I ended up building an mcp server to do this data entry and now I track his activity and planning to track also his training progress. It's very different when you can just tell your model "he ate 2 cups of his kibble" or "we practiced stay for 5 minutes" vs doing tedious data entry. As a bonus this helps my partner and me coordinate dog care so we can have fewer conversations about the dog

A long time ago, I introduced dogstatsd at Airbnb. We had already been using vanilla statsd (with no tag support -- cardinality lived in the metric name!) and this was a low cost migration. More than a decade later, I'm assuming it was difficult to track down and refactor all the places that statsd calls were emitted and using OTLP was an easier route. This is a great example of how technical decisions compound over time.

So is Victoria metrics?

Submodules work fine but yeah, it's frustrating that lfs is taking so long. But there seems to be some momentum recently https://github.com/jj-vcs/jj/pull/9068

I use vimium in Firefox and so my default key bindings are the plug-in ones. I push 't' to create a new tab, for instance. If I want to use the website key bindings I have to to into "insert mode" ('i'), or I opt into specific keys by site.

I do like when websites use ctrl-k -- it means nothing to my plug-in so websites always get it, plus it helps with key binding discovery.

Share your private thoughts and gratitude with your private group, plus the good people at grateful.so, their partners, their service providers, the processor of os notifications, anyone who buys the assets of grateful.so once it goes under, plus any script kiddie who cares to point an LLM at the vibe coded server base...

Damn I'm jaded by our industry...

Yes absolutely this. I've put so much effort into making sure I tick all the boxes and yet I constantly wonder if my email is getting delivered. This feels anti competitive to me. It's Google constantly telling me, give up, you know deep down you should just use Gmail.

Oh I made one of these once! In mine you play against other people. https://1dchess.igor47.com/

Oh cool I used to use https://web.hypothes.is/ for this but they never had a good Firefox extension so I stopped. How do you annotate with friends using Margin? Does everyone need to be on the same at proto server?

All the examples are fetching data from a server, and in such cases I think tanstack query already does all the hard part. I feel like people under-use react query and put too much state in their FE. This might be relevant if your app has some really complicated interactions, but for most apps they should really be a function of server, not client, state. Of course this exact reasoning is why I moved off react altogether and now use htmx in most of my projects

I've been waiting for people to have common sense in this domain for decades. The short term always wins

Well, reading the study, I'm not sure more patients could rescue it from methodological bias. They assumed the premise basically -- we should find a biomarker, which is kind of what this thread is discussing. Then they went trawling for biomarker in a sea of millions of biomarkers. They did this by training an model that produced the desired result, using a grid search for hyper parameters that even further expanded the available degrees of freedom here beyond what they had from the biology. No pre-registration; There are millions of places where the researchers could have made a different decision -- would they still have gotten a publishable result? Oh plus the authors mostly work for the company whose data they use, which is hoping to sell a diagnostic test.

I'm giving you a thorough response because I'm detecting a cavalier anti scientism which I think is sadly becoming more common. This stuff is hard; are you sure you understand it enough to have an informed opinion?

I cannot endorse mise more highly. I commit it to my repos to make sure every engineer has the same environment. I use it in CI for consistency there as well. I keep all commands that would normally be documented in a readme as mise tasks. I use mise to load the environment, independent of language specific tools like dotenv. I use a gitignored mise.local to put real creds into the environment for testing.

Dude get off the train. I run a self hosted immich instance for myself and some friends. I deleted all my Google photos from my Google account and I don't miss it. Alternatives exist. I love being able to automatically share my photos with my partner, or create shared albums that other people can upload into. All my photo syncing happens automatically via sync thing, but less technical friends use the immich phone apps and say it works fine for them.

Strongly agree! As someone who has been caring for a parent with dementia, it's definitely a use or lose it kind of situation. See also the studies on long term cognitive health in London cab drivers

https://www.statnews.com/2024/12/16/alzheimers-disease-resea...

Strong words. I wonder if the author has PTSD from poorly managed teams and has never had the fortune to work in a high performance well managed collaborative environment. I agree these are rare compared to the other kind, but they exist. Groups of people can produce more than lone wolves. One person didn't build the pyramids, the Linux kernel, or Amazon Web services. Even when responsibility for a top level domain rests with a single person, you still have to coordinate the work of people building the individual components.