flowerbreeze

The data sources listed for UFOs are from US and Germany:

National UFO Reporting Center (NUFORC) witness reports (US, via the CC0 TidyTuesday project, 2023-06-20)

and

GEP e.V. UFO/UAP Falldaten (Germany, Zenodo 18071015, CDLA-Permissive-1.0), 1925–2025.

They are unlikely to include data from many other locations.

I don't know if the above post had an edit, but it says "distributed on SD cards" now. My cheap ~5yo car is the same, I can only update the maps via USB. The navigation still works, but is a lot worse/more out of date than any smartphone app where I live. Although even google maps barely works around here.

Oh, that's a great name for it! I'll need to start using it.

When writing code for end-user applications, I think it's mostly true. When it's writing code for a database engine, a game engine, a 3d renderer, or anything else that involves heavy data processing, optimization is the core "thing" often and it might not even be a good enough solution without it. Although, a lot of time even then C++ is good enough even then when picking reasonable data structures to represent the data.

Thank you for the article! Do I see a Turbo-C screenshot there or am I imagining it? It was my first IDE (I didn't know that's what it was called) when I started programming. I sometimes miss it, it was really good, especially the help system.

I agree with the article of course. I think most confusion comes not having learned about the purpose of having them be defined based on the architecture in the first place. It took me a long time before I stumbled upon how they really worked and why, because while I started it was either x86 or nothing. When x64 showed up, suddenly it became relevant and everybody started learning about C types more in depth as they ran into issues with sizeof.

Also, misuse in data protocols is where I think the bad reputation of the flexible type sizes came from. stdint was desperately needed for that reason and it came a bit late.

I think being able to define additional rules in a language is a reasonable feature and sometimes would make life a whole lot easier, but I think when implemented through text macros rather than rewriting the underlying program graph along with accounting for all the constraints, that causes most issues.

Program graph rewriting is of course more complex, but can be better defined with regards to acceptable inputs and more defensive as opposed to letting the next step of compiler or interpreter handle it.

The other problem is that most IDEs don't offer an expand macro (or similar) option for the rewritten parts (or I haven't seen them doing that), so observing what the final code looks like requires additional steps. In some ways its similar to functions, but most languages have functions that have a well defined input/output structure whereas macros can do some unexpected things depending on the inputs.

Yes, especially people who subscribe to humanism as a philosophy. Studying humanities is of course great too! I didn't realize from the title that that was more what the article is about, not about the philosophical idea.

Since the article calls for learning more languages, I hope it's alright if I mention a still in-progress experiment of mine. I constructed a simple language (Not a programming language!) some months ago for my close ones, who do not share a common language.

It's intended to be very easy to learn, maybe a bit like a cross between Toki Pona and Ro. I can pinky promise it's much simpler than Mandarin. The core vocabulary consists of 70 single-syllable words with simple rules for compound word-construction to expand the language while speaking. It's somewhat limiting, but I've been usually able to bend it enough to convey what I want to express when trying.

I'd be very interested, if anybody wants to try to learn it (as recommended in the article above, to become more like the humanities!) and maybe try to chat in it online:

https://dupi.portia-fimbriata.net/overview

Please excuse the mistakes in the explanatory texts. After I made a major update to the language as a whole, left it alone for a few months and now looking at it again I'm seeing all the things that are wrong and remained without an update... I'll be correcting them soon.

I am not convinced mortgages is a net benefit to the society in any way.

Official docker registry, Kagi image search, and several blogs that I read were still blocked for me the last few nights at least. I'm using a local ISP, but I think they in turn operate over Orange.

Sadly I don't care at all about football. So I can't even boycott the football league that should not be named to make any difference...

Yes, it's possible. For me it went like this.

1. I learned enough vocabulary in a very short time to be able to more or less be able to read with the assistance of a dictionary on an e-book where you can click on a word. I spent ~2 weekends of Duolingo 8 hours a day, but I'd use flashcards in the future instead of that. It sounds a bit crazy in hindsight, but I feel like it was easier to stay strongly motivated for a short period of time than drizzling the words over a period of time. All other attempts where I did not do this, failed.

The limit of learning words for me is ~50-200 words per hour if using flashcards. It depends a lot on the language and how similar it is to the languages one already knows. 200, if it is similar to my native language, ~50, if I don't know anything similar at all. I'm also almost guaranteed to forget 80% of the words, if I don't use them in the next day or two, but its possible to cram a lot of them very fast and then move on to stage 2 to use them.

Important catch (for me at least), is that I can remember nouns far more easily than verbs or other more abstract/structural words. Those absolutely need to be in a sentence when learning. There's pretty much no point for me in trying to cram them on their own. I just cannot remember them in isolation at all and also which words they are used with differ per language anyway. Eg "to put" won't necessarily be used for all nouns in a different language even though a dictionary might say that's what it means. The good side is that knowing the nouns helps a lot in deciphering.

2. After the initial cramming I read a couple of short/simple books over some months in the evenings before sleep. I tried to listen to videos & to talk to some people a bit now and then, but it was extremely hard and boring as I understood too little. In hindsight I should have listened more to videos, but the good ones are surprisingly hard to find, if you don't live in the country that already speaks the language one is learning.

3. I tried several options for getting conversational after that. First a private teacher who knew the language well, but was not experienced enough (a university student) to teach somebody who tends to use weird methods for learning. She kept getting confused about why I understand complex words that were basically the same in English, Spanish and other European languages, but why I struggled with the basics when talking at the same time.

I also tried some learning groups. They were fine, but it's very difficult to find people at the same level after starting to learn a language and not following a preset path. I was felt either far ahead and far behind at the same time.

Eventually I found a native speaking private teacher who had years of experience. He could make the classes interesting and after ~1.5-2 years of talking with him 1-2 times per week for an hour about pretty much anything, I can hold a conversation and handle things in general. I still kept learning vocabulary in parallel and I'm not quite fluent as I don't use the language daily, but I'm not too far off either. Overall it took maybe 2-3 years, because I wasn't very consistent at the beginning.

If I were to change anything, it's getting the private teacher way sooner, skipping the learning groups entirely and using flash cards + asking for verbs/structural words as parts of sentences from either the private teacher or finding some other resource for it to start with. Using AI is possible too I guess for that purpose, but only with more common languages. Most AIs completely mess up my native language and invent words that don't exist, so I wouldn't trust them much when learning.

Being cynical, I'd think probably to squeeze out some personal data first. Best let the questions remain unanswered.

The guidance provided about CRA is difficult to follow and does not in any reasonable way cover proportional guidance for SMEs as it claims to do, but it seems websites/server side products are not subject to CRA as they are not considered digital products executing on consumer systems?

The only websites that should require it are the ones that provide downloadable software or software that is used on digital products. And it seems that European alternatives site lists primarily (pure) SaaS and only a few others?

It'd be the sensible non-surveillance option. All that would be required is that the OS or SIM sold is a tiny bit different with a flag that the browsers/apps respect and that each website that accepts underage users sends a header that browsers can check and allow/disallow access. Compliance with child friendly legal requirements would still be with the company owning the website.

Of course it's not as attractive to the governments as putting in place a full scale surveillance aparatus that puts KGB to shame.

Maybe it all works, but the paper is not trivial to decipher and the GitHub repository does not seem to exist. It doesn't seem to define what are the inputs to the system (what is a query? UTF-8 text? tokens?) and what are the outputs. It'd really help if the algorithm was written out step by step with all the expected type information included.

At first I thought it was similar to something I've built before as a long-term slowly degrading cache for augmenting an FFN by caching well-learned answers, answering by performing a beam search in the key space resulting in located key accuracy measure (how well it corresponds to the input query) and answer confidence (has it been a long time since verification?), but that's not quite it? It feels similar in some way, but is it?

Very interesting, thank you for making it and releasing it for free!

Do you get faster as you practice reading like that? I read ~50-60 pages per hour in English. That would be maybe 375-425 words per minute and also seems to be about the maximum I can more or less follow with ReadKinetic just trying the demo text.

Although, I don't feel like I understand quite as well what I have just read and it felt a bit aggravating? Almost anxiety inducing for some reason. Maybe it's just because it is a new thing and I'm not used to it.

Also, as others have mentioned, the controls took a bit of time to figure out.

All the verifications without being able to see what it even does fully feels like pretty a big barrier, even more so than being login gated. Gives it a big banking/gambling vibe too.

Then again, I was only curious how it compares to other GPU rent options in general. Somebody going to bid on something right away might not care.

It's fun, but a menu would be so much better instead of the weird scroll-jacking. It takes a minute to get to the bottom of it.

It's more like if you find one mushroom in the forest, it doesn't make sense to bring it home, get the knife, clean it, get the pan, oil the pan, fry the mushroom, eat it, clean the knife, clean the pan, put things away. It's not worth the effort for just one mushroom. If there are many, a lot of these actions only need to be taken once.

This is an interesting article, but I think most of it does not seem applicable to a reasonable civilization attempting colonization, that is aware of these problems.

  1. The population size in the colony will be very small. Maybe? It should be at least MVP (minimum viable population, not product) that is a relatively well researched concept. I'd argue that it won't be a colony below that. More of an outpost. And when talking about aliens and Fermi paradox, we don't even know what their MVP might be. It may not be even close to MVP of vertebrates on Earth.
  2. That selecting the colonists by some criteria is not useful. It is, it's just not selecting by the "best genes". It's selecting for more diversity in the first place. Lack of diversity in the colony is much less of a problem, if more of it is introduced in the first place. 
  3. That colonization is an isolated thing and people won't arrive/leave over time. Why wouldn't there be more than one colony ship, if sending them becomes viable overall? 
  4. The population does not grow and remains at the original diversity levels. Combined with 2 and 3, it may not even matter that much. Unless it experiences a series of catastrophes, it should keep growing and becoming more diverse over time again. This is of course the most questionable case. Will it? I think considering how humanity has spread like a plague on Earth, it won't be a problem.

What I think is that the table might look very different with the above taken into account. Maybe 99% of diversity in the first colony and none of it will necessarily remain static or diminishing.

It seems plausible. Less common sizes have a lower chance of being sold out, so if they can no longer be destroyed at the end and need to be further managed at lower quantities, it can become more cost effective to simply not make them. Whether it is true or not, I don't know.

From FTC website: Malware is harmful software that’s installed on your device without your knowledge.

So I think that is what we should continue to call it. LG monitors are installing malware, because they install the software silently and it harms the system by making it slower and disrupting the work of the user with advertisements.

In my experience in the UK aliases and names are used in a rather flexible way compared to other countries. Anybody can decide on a name and start using it. It's not illegal unless it is done for the purpose of deceiving people.

If people commonly use it to refer to them it makes it that person's name for most purposes. Changing the name officially also requires only an affirmation or an oath at a court, although I don't think this would be the case here.

I think nobody would care much and the majority of politicians would go along with Count Binface as the proper way of address, because it'd be rude to do otherwise. Not the least because picking a fight with Count Binface would be seen as getting themselves dirty in non-serious politics. Not a very attractive proposition to be perceived as such for most political parties no matter how ridiculous their actions might be otherwise.

It sounds like you're joking, but I've long dreamed of a different type of dishwasher. One that washes instantly. I don't need it to fit more than a single plate at one time. Just put something in from one end, and out comes clean and dry plate on the other end. Like a car wash.

I am quite certain these exist already large kitchens and I seem to remember one from a school diner from maybe 35 years ago, but I've always been wondering why they don't exist on smaller scale.

Different reasons. Mine is on the table and I use it more like a desktop. It will just idle when I'm not around, because I come and go often. My current uptime shows on Debian 30 days, 49 min.

Although... 30 days is maybe a bit misleading, because I ran some heavy shaders without thinking that triggered the GPU watchdog and forced me out of my session. I think killing all user processes is almost like a reboot, although not according to uptime.

Oh it comes with custom firmware? This is very interesting. I would love to be able to modify some UX and I am sorry, but I need to get the following out of my head. All the e-readers I have had have made it impossible to turn off features like:

  1. Selection highlighting... I never use highlighting when reading fiction, but whenever I am not careful enough when turning a page, it'll go crazy with highlighting. Flashing screen, need to close the popup that has added the highlight, removing the highlight again etc.
  2. Most of the time I don't want to click on a word to find out its meaning. It's sometimes useful, but I'd rather have it under menu to temporarily enable it. Same reason as before. My e-readers tend to prefer this often enough rather than taking the "next page" action.
  3. Make "previous page" be small and not-under-my-finger. Ideally let me choose its position in a fairly precise way.
  4. Easy access to accidental "scroll to page 900". I generally don't want it to happen and to be honest, I struggle to think of anybody who does. It can live in a single tiny faraway menu that is impossible to accidentally tap.
  5. Swipe-left for previous page. It almost never happens when I want it to happen, so I'd rather turn it off.

In fact, I would love my e-book reader to have no gestures at all. Pretty please let me turn them off! All I want is a tiny button top-right or top-left corner for "open menu", a "previous page" in the other corner and otherwise "tap anywhere" is "next page".

Personal request to any e-book reader software engineers. Please save the position in the book to persistent storage on each page change or every few. At least if the e-reader has any chance of crashing at all, which has been the case with all the ones I have ever had. Yes, not all of them save it...

That's not to say that all the above things are universally bad UX. I think many of these are very useful, if reading non-fiction or having a different goal when reading such as learning a new language. It's just that they are less than brilliant if the goal is to read a book for entertainment in the most comfortable way possible with the fewest things going wrong by accidental taps.

Opus said to me once without any poking at it something like, "Help Grok understand it better". Makes me wonder if they are all cross-pollinated to an extent.

The last option is the only correct one. A simple HTTP header would do the trick.

I've been through SOC 2 Type 2 in a company with ~100 people. I think it'd be in some ways simpler as a solopreneur, but still a lot of effort. You won't require as complex controls and you don't need to communicate between different parts of company, but it'll just be yourself doing it all.

On a positive side, you won't have to do 100% of SOC 2 Type 2. The only required part is security if I remember correctly. And a lot of it is best practices that need to be in place anyway. If you are using an established cloud provider a lot of it is in place through their certifications. Some of the controls can be "silly", but generally not hard to put in place. I'd try to figure out what are the minimum nr of controls required and see if that is doable. Pretty sure auditors will give a discount there if the scope is smaller.

It can be somewhat useful for the company if taken seriously, as it can point out weaknesses in processes. Although I agree with other comments that most of it is a checkbox exercise than something that provides any real guarantees to the client demanding it.

I also don't know if getting through it with <20k $ is something that is feasible. Before doing SOC 2 we relied on the clients' security questionnaires instead, so maybe something to always ask about. Usually they were able to make an exception and allow it, although the % started shrinking over time.

Edit: Also, the auditor makes a difference. Pick one that understands small companies. A corporation auditor will get confused with "segregation of duties" if you are the only person in the company.

I wish I had known that! Guessing and trying the answers worked too, given no internet and only having a faint idea that "age control" was not in fact part of the game itself. I learned that Bonnie and Ronnie was not in fact a thing. What is "Bonnie & Clyde"? Eh, probably some band name was my guess. It took some patience, but since it was one of 3 games I had somehow acquired (how exactly is a lost memory), I had to get past the starting quiz.

Since I also barely spoke English at the time, I got stuck in the game itself pretty soon anyway. Didn't manage to figure out how to say some things the right way. "Ken sent me" is the last thing I remember from it... and I never had any idea that the game was rather dirty until much later.

It is rather backwards. I've not seen things quite as bad as interviewers wanting to know how many agents you can run, but the attitude of "launch & fix later" is always present and kind of depressing.

Then I think of the companies (not necessarily software) that have had long term success and their products have been quite high quality at least at some point in time. The count of genAI instances someone can keep in flight is certainly a weird metric that I think will hurt the companies who choose to ignore quality.

Unfortunately it's a long process as it's possible to get very far with great marketing and sales with a poor quality product too. Then cash out before customers figure out that there's something else that is better. I have no idea if this pattern will ever self-correct.

Off topic: I followed your guides for network programming years ago getting my tiny C server/client setup working. Thank you so much for writing them!