jjice

I'm surprised you've had an apolitical experience up until two weeks ago. HN has always had politics creep onto the front page, less so during the off-season, but still pretty present IME. Especially prevalent over the last two years. Lots of side conversation bickering in random threads.

Maybe it is, but I know that lots of us used GitHub way before Actions launched. I think even if Actions wasn't free, GitHub would still have a massive advantage because it is the most well known place to host code, especially by less technical users (which is where I see a lot of finger pointing going with the increase in GitHub usage).

I would not be surprised if a serious amount of this usage is not user-intended and instead the LLM just inserts some sort of CI job off the bat and people roll with it.

Agreed. Happens on my personal Linux machines (debian based), but not my work Mac. Has for years I think.

Not everything is writing code and search. My uncle works in mortgages and the (very large) firm he works for is entirely Gemini. The way he talks about it is with wonderment in his eyes while he goes into detail on how it went through multiple documents and found unaccounted for cases. I think Google is doing fine in these kinds of industries.

I wouldn't even bet that Google is better than OpenAI and Anthropic for these categories, but they seem to be well used outside of our direct space.

To clarify, I'm not promoting the previous comment, I'm just clarifying what the date was

Searching the date yields this link: https://www.history.com/this-day-in-history/july-26/fbi-foun...

The FBI was founded.

Even outside of the human body, appreciating things that have aged, served a purpose, and have been repaired so they can still accomplish that goal is a beautiful thing.

Kintsugi and wabi sabi are probably the most convenient terms to capture this.

I find nothing more satisfying of clothing that I've personally patched up and repaired the sea on. I have memories tied up in that. The same goes for the scar on my wrist from falling off that longboard in college when I was trying to learn (that was the end of that).

I think I get where you're coming from. I've worked mostly on small teams (sub 30) so it's always been on us. The engineering leadership never prioritizes security unless it's something as big as the log4J stuff because it was on customers' minds.

That said, if they're software packages from something like Node or PyPI, I do think that it falls on the developer. If the point is that leadership needs to give devs the ability to act on that, that's very fair.

I fought hard about four years ago at a previous company to upgrade us from PHP 5 to 8 because I think 5 was going EOL or something like that (could be wrong about that now). Based on my experience in software, I have to imagine the vast majority of normal developers do not give a damn about security in being in security patch lifecycles on software.

The parent comment specifies they're existing solutions:

> There are three existing realistic solutions

I find that LLMs can write poorly structured code if not given a good foundation to do so. The existing code in a repo is what the LLM will see and generally output similar code to.

Given that, I'd recommend the mainstays on good general software design. One of my favorites is "A Philosophy of Software Design" by John Ousterhout. Good general design practices.

Other than that, I think a lot of books generally stay relevant. If it's topic specific, it'll give you a good background on understanding of that topic so you can make sure you or an agent are making sound decisions. Architecture of course is more valuable than ever.

It felt to me like it's more of a neat fact rather than a measure of quality.

I don't know about turning it off entirely, but I just did this after reading your post and I clicked the "not again" option and the whole thing went away.

Ah that makes this make way more sense. Thanks for the insight.

The idea of keeping my passwords in my work-sponsored password manager just sounds absurd. Who thinks that's a good idea? What happens when you leave the job?

Did you record over 30 seconds?

While I'm obviously not a doctor, dyshydrotic eczema is what my research came up with too. I do consume quite a bit of dairy, so it's very possible that's the case. The times I've seen it go away are during travel, which I usually associated with being an environmental change, but now that you mention it, I'm consuming less dairy when I travel.

Super interesting. I appreciate that. I'm going to take this into consideration!

Okeefes has definitely been good for the past few winters when my hands naturally get very dry, but this one is a big more consistent now and okeefes is helpful, but not a cure.

I really haven't considered telehealth because I figured it'd be easier to see in persona, but you're right that it's probably worth a shot and likely totally fine.

I appreciate your reply and I'll be taking quite a few of these suggestions into account this coming week. Thank you!

I generally agree with you, but one thing I _hope_ AI can lead to is me being able to actually consult something resembling a medical professional. It's so difficult to get time with my doctor, and when I do, I get swept out the door ASAP.

My hour long yearly physical takes 15 minutes and my miscellaneous questions get brushed off. I have some sort of eczema on my hand and he told me to use lotion. After telling him I do, he said to keep doing it. It had been six months then and now 18 months. Problem is still there.

Just today I finally found a dermatologist. Soonest appointment I could get from the ones I called is May of 2027.

At a certain point, I hope LLMs can provide at least basic medical help reliably because I can't get the opportunity to ask a doctor to begin with.

_virtually_ unlimited. Functionally for this individual, it is more than they could use. It feels unlimited to them. Not actually unlimited.

I generally can on Safari on iOS. Sometimes I have to force desktop mode. They're making it very difficult...

I assume I'm missing some context because you sound like you have a lot more experience here, but what kinds of capabilities are there that you can't do something like `interface LlmProvider { }` with a few basic methods for attachments and chat and then just implement from there? I'd even say you could not keep the ones you're not using (say, GPT is your go-to for a year) and update the other ones when you want to use them and swap out the `LlmProvider` implementation with a DI'd different implementation at your application's entrypoint.

I don't think the OP and your comments are mutually exclusive. If Anthropic usage is lower, and the OP considers it basically unlimited for their case, that just means that they would have virtually unlimited usage with other plans.

We're like 3.5 years into this new era - I'm not counting winners or losers yet.

I always get a kick out of YouTube and other social media people proclaiming what is "healthy" or not, when we all pretty much know. Eat fruits and vegetables and not too much junk food. Eating too much consistently isn't good for you. Drinking and smoking are bad for you. We've mostly known smoking is bad as a society for a good while now and outside of the old myth that a glass of wine with dinner was good, we know drinking isn't a positive for your health.

And you don't even have to be hard and fast with any of that. Drink on occasion and eat stuff that isn't good for you on occasion and you're probably still getting 95% of the benefits. Not sure about smoking though.

It's very simple at the end of the day. Simple != easy to do. Drinking is fun. People enjoy smoking. Junk food is appealing because it's relatively cheap and tastes good.

The best biohack is to eat like your doctor recommends while we all nod along and silently ignore. Just eat what is widely accepted as good for you and you'll have a good chance at living a happy, healthy life.

We're early on in the LLM era, so I have hope that it can crush Oracle's business model. That said, your database is probably the hardest part of your application to migrate off of...

Please do! Small, specialized models need more love and the time you spent would be a gift!

Ah this is good to know. Awful it's required, but glad there is a work around.

What does the future of something like F-Droid look like once Google does their lock down next year?

We're small enough of a team that everyone can review anything (with a bit of domain expertise in there). It's less of an issue of who is reviewing and more of an issue of getting anyone to spend a minute to look at it. More of a human problem. Would be curious what people to do remedy this.