gnyman

Also a recommendation for Strongbox if you're on a Mac. I've tried "all" since switching from 1Password a few years ago and it was the one I liked the most.

Scaleway and OVH are two other European companies who provide things like managed databases and blob storage, in addition to very competitively priced dedicated hardware and services.

One thing I personally like with OVH is that they have fixed prices for things like bandwidth, not pay-per-GB like most others.

sadly no, I just had a pair lying around at home which I got for biking I believe, they are unbranded plastic ones. But I don't think it matters as long as they are yellow :-)

In the other thread someone mentioned that normal "blue light" glasses do not work which is too bad, so maybe just grab some cheap plastic yellow glasses and see if they help. If they don't I'd like to hear so I can update my recommendation and try to figure out why mine work so well.

Have you tried yellow tinted glasses? I found it helped me a lot. I think night driving glasses are supposed to do the same, filter out the blue light but not sure if they do it as effectively as yellow lenses, maybe try both?

PSA, try yellow tinted glasses,they filters out a lot (all?) the blue light and it mostly solved the issue for me. There are special "night driving glasses" also and it might be that "blue light glasses" also work, I haven't tried. I just tried a pair of yellow tinted glasses I had lying around they improved things immensely.

I think it's just not great at Clojure, it's less popular and so there is less of it in its training data. Also Clojure seems to be kind of hard to get right.

I started trying to learn Clojure with this years advent of code, but got stuck and first tried to use ChatGPT to solve it. My impression matches yours in that it consistently produced non working code and even when told about the error it was unable to fix it.

Then I instead decided to let it use any tool or language it knows, and I'm now documenting how it's doing solving the things.

If you're interested, here is Day1 where I first tried to use it to help me solve it with clojure, but then I gave up and asked for any concise solution. So I got a working solution with `awk` https://blog.nyman.re/2022/12/02/chatgpt-does-advent.html

The second day I just let it pick anything, and it successfully solved the day2 puzzle using python which seems to be it's go-to language. https://blog.nyman.re/2022/12/03/chatgpt-does-advent.html

Definitely.

I'm using it to solve Advent of Code [1] and there a few others doing the same [2]. The training data biases definitely shines trough sometimes but thats mostly when the problem labels (R)ock, (P)aper, (S)cissor as A, B, C. It really wanted to use R,P,S as identifiers. But after you tell it to use the right identifiers it solves the problem.

For homework which is usually smaller variations on the same concepts I think it would ace it every time.

The [2] post is gathering some momentum on HN right now actually [3]. There are many more.

[1] https://blog.nyman.re/2022/12/03/chatgpt-does-advent.html [2] https://note89.github.io/the-advent-of-code-ai-edition/ [3] https://news.ycombinator.com/item?id=33842904

I was just rewatching Jonathan Blow's talk, where he discusses the Five 9s

> ... We don't use this anymore, in part because the number would be going down and we can't make it go up anymore ...

https://youtu.be/ZSRHeXYDLko?t=1704

I don't have a strong opinion of an hour of downtime is bad or not, and I think the concern about five nines for everything was silly. But it's an interesting talk I think it's worth watching. Is software getting worse and could we reverse the trend if we tried?

Yep, I paused at that also. Quite hypocritical of them to state they are "legally required" when the prompt they display isn't in compliance. The European Data Privacy Board (EDPB) has been very clear that the choice to opt out must be as easy as to opt in. Yes/No, not Yes/More Info.

But due to lax enforcement companies just keep ignoring this. Luckily it's (very) slowly changing. Even Google was recently forced to amend their practices but it took a fine of 150 million, and probably a threat of a lot more for them to finally do it. https://www.cnil.fr/en/cookies-google-fined-150-million-euro...

If anyone is interested, NOYB is doing good work to make sure companies are aware that they are breaching the GDPR, which is also very useful when it finally comes to enforcing it as they can't pretend they didn't know. If you like their work you can become a member to support them.

https://noyb.eu/en/noyb-aims-end-cookie-banner-terror-and-is...

I had heard the 1 ms number also and did some research some time ago. As far as I understood the 1 ms is for specialised low latency connections and devices not for the every day browsing.

Here are two sources which I found useful at that time https://broadbandlibrary.com/5g-low-latency-requirements/ https://www.linkedin.com/pulse/we-need-talk-low-latency-dean...

I do think Soulver would be a better choice for many. But it is much more powerful, allowing you to do conversions and labels.

So the simplicity of just having two calculators side by side and a fast way to transfer the results is probably exactly what some people need and want. Zero learning curve, just two calculators doing exactly what you expect.

I'm always fascinated by apps and programs that does exactly one thing, but does it really well (sometimes referred to as the Unix Philosophy). I feel it's harder to find tools and apps like these nowadays. Not sure if there are less or if they are just harder to find because of the huge number of apps overall.

[Edit] Writing this comment actually reminded me of this blog https://onethingwell.org/ which highlights software like this. Gonna try to contact the author and see if they want to add this one :-)

If you want a cordless vacuum designed for repairabilty, there is a new company called Lupe which is trying https://lupetechnology.com/

Vacuum wars were impressed at least https://youtu.be/rFbkGM9Lm18

No idea about battery balancing in them though, maybe I need to open it and try to find out. My gut feeling is that they wouldn't save out on a few resistors.

Last year when I looked into this there was no automated way to get this info; the normal audit log mentioned below does not contain any info about actions from oauth-ed applications afaik.

If you email GitHub support they can pull out detailed logs from oauth app interactions from their internal tools.

I would expect the GH security team to have relevant queries ready by now, maybe even do some proactive queries and start alerting anyone who had suspicious activity. (But this is just how I'd do it I have no special insight if they are doing this or something else).

Nice, I did the same, but in 280 characters :-) https://blog.nyman.re/2020/10/11/introducing-piss-a.html

I like your write-up of yours, re-reading my post I realise it’s severely lacking examples. Going to take som inspiration from you and update my post so it’s a bit clearer.

I had a similar experience with trying to switch a few times during the last few years and only within the last few months I've actually managed to keep using another search engine for more than a week, and it happened to be kagi.

If this is because Google has gotten worse or Kagi is better I can't tell, but either way I think we are at a point where Googles dominance is threatened for the first time, at last from my sample of one.

But there are other signals like how many of these "google isn't good anymore" posts we see on HN nowadays and the above example of a user not even noticing they aren't using Google.

Wow, thanks for the tip! This looks really useful, I've been using fzf for a long time and it works well but this seems like a clear improvement to that

Has HA broken without being touched for you? In my experience, as long as I don't touch it, it keeps on chugging along (with a notable exception for the recent winter-time bug).

Doing upgrades means things might break, but that is kind of understandable for such a extremely active project. ~10 new issues, ~25 merged PR's, ~4 new PR's per day for the last month.

I wish there was some way to better decouple the new things from the core though, hopefully they will figure something out so we can have stability without opting out of new features completely.

In a similar spirit there is http://spa.mnesty.com and https://twitter.com/rescambot , the first one which is more simple and the second which has been discontinued.

Re:scam seemed quite good at what it did https://imgur.com/a/rrHi2

Does anyone know of any other service? I've had varying success with spamnesty , if I recall it does (or did) just have a list of non-answers it picked randomly from. While the Re:scam seemed to use some kind of ML to generate more relevant answers.

This would be my first answer also. It’s traditionally been considered to be a highly (or the most) secure OS because of their focus on security. That means if you stay within their ecosystem you are unlikely to ever have to worry about zero days.

But it depends on what you want to do and what your threat model looks like.

As others has written, knowing your system is probably more important as I believe (mis)configuration is one of the more common ways to get hacked. And even the most secure OS won’t help you if you accidentally expose your elastic cluster to the internet.

To wrap, I’d say the best thing you can do to secure yourself is “defence in depth”. Use a locked down OS. Add a firewall (or two if one is a soft one which can be bypassed or disabled accidentally). Don’t bind services to internet interfaces unless you want it. Ensure you have authentication enables. Enable auto-updates for anything exposed.

If you do this you don’t have to worry about picking the most secure OS.

ok, I'm not following completely but I'm curious enough they I need to try to compile a ssh client with this, appreciate the explanation though

Super interesting, thanks for sharing.

You mentioning SSH makes me curious, could this be used for roaming SSH? With only? patching the client?

I find the character limit a interesting limitation, both negative and positive. On one hand, as you say, it's impossible to have a nuanced discussion in 280 chars. But on the other hand it forces (and allows) people to write interesting things in a very concise format.

Should be noted though that I only interact with twitter trough Tweetbot and with a extremely curated follow list. And in this specific way I find it's a great way to keep up to date on industry specific news and learn about new things (by following people in other industries).

It's far from perfect but I have not found any other way to keep up to date with the thoughts of people I look up to or find interesting. I had high hopes for micro-blogging but it has not caught on yet.

PSA: For those of us on iOS/Safari, Stop The Madness recently added the ability to add site specific css or scripts https://underpassapp.com/StopTheMadness/

I guess there are other apps(extensions) also that can do this with the new Safari extensions api I just happen to be a happy user of STM

There are a few companies who are focusing on sustainability/repairabilty, you just have to look (sometimes very hard) and then take the risk that the company won't be in business in 5 years.

One example I came across recently after hearing how hard Dyson vacuums are to repair is Lupe https://lupetechnology.com/

"Someone" should make a list of these companies, or maybe there is already some site or place which focuses on quality-repairability-longetivity?

I don't have a comment on this specific one but @jonathandata1 received his fair bit of critique before for being sensational when it comes to his finds.

Dan Borges (aka @1njection ) wrote a blog about him last year https://lockboxx.blogspot.com/2021/10/the-shenanigans-of-jon...

I personally have not looked in to either this or his previous disclosures, but my spider sense is tingling based on the sensational language he uses in his tweets like this one https://twitter.com/jonathandata1/status/1486466958992228366

So I would maybe be a bit careful in putting too much into this before there is some more validation, I think the citizenlab article linked elsewhere in this discussion had a more balanced take on it.

have you tried another client without the endless scroll? Also unfollow or mute everyone who isn't posting things you find real interesting (and also some that post interesting stuff but posts way too much, those I move out to a separate list which I review sometimes) :-)

I find having a manageable timeline a super useful tool to catch up on interesting or important news. But every now and then I notice that even with a old-fashioned timeline I spend too much time and I look over who I follow and cull people who post too things I forget the second I've scrolled past and instead keep the people who post well-thought-out things infrequently.

This is lovely! Is this your first N64 game? How much documentation is there for aspiring N64 developers?

I'm not an expert on DNS so I might be missing some important part, but why use this instead of a normal reverse dns lookup?

That is, `dig PTR 192.174.198.91.in-addr.arpa` or calling `gethostbyaddr`?

Hey, I usually upvote anything that tries to be a lightweight alternative.

But I don't think it's a good idea for anyone to enter any real slack credentials here unless you know the OP personally.

The hn account is 11 months old and only ever posted this link twice and there is a reddit account which posted the same which is 5 years and have a little tiny bit of history so that makes me think this might not be a scam or a phish.

But as it stands there are to many red flags.

If you (OP) want to get feedback on this you could publish the source and let people run it themselves. Or at minimum publish some info about who is running this and a explanation why people should put in their password, and why this can't be a slack app using the official api.

Nice concept, going to leave it on the phone with the other weather apps and see if I use it.

I guess you tried having just one clock and showing just the coming 12 hours? I'd think they might be even more intuitive for a first display because for a "glance at the upcoming weather" app I personally don't care much for the past weather :-)

Anyways I love apps which does only one thing well, bonus points for being under 1 MB. another favourite weather app is Sataa[1] (means rain in Finnish), which just shows the last hour of radar images for Finland.

[1] https://apps.apple.com/fi/app/sataa/id1350488142