xurukefi

As somebody who doesn't use Twitter actively, I don't understand why you would ever want to block someone. If you're sick of someone, why not just mute them? Blocking seems to be kind of a childish option because you want the other person to know that they are blocked. The idea that you want to actively disable someone from reading your tweets seems absolutely bananas to me.

Are there reported incidents where somebody was caught with such sophisticated techniques? It seems like every time I read about how the FBI caught some big darknet criminal it was pretty much always some trivial mistake on the criminal's side.

> So the main argument in the article is the claim that your total energy expenditure is more or less constant, and independent of "exercise". In the sense that if you spend 500kcal on running, your body will "not do" something else it would have done otherwise (e.g. stress about stuff) that would have cost 500kcals.

> In other words it challenges the idea that if you run today and not tomorrow, your TEE on day 1 will be 500kcal higher than day 2. The claim is that TEE on both days will be the same.

That's wild. Running a marathon will easily burn north of 2000kcals. That's a lot of "something else" the body will have to "not do" to compensate for.

"You can’t outrun a bad diet" seems to be the party line among fitness people these days. Anecdotally speaking that's not really true for a lot of people. I run > 100km/week pretty much all year round (most of it on extremely hillly trails) and despite my bad diet consisting mainly of pancakes, white bread, nutella, pizza, fries and burgers, I am not overweight with 78kg at 190cm. Before I picked up running I was moderately overweight. If I can't run for a prolonged period of time, then I start gaining weight again. I've met plenty of people who run similar mileage and many of them also have pretty bad dietary habits but none of them are oeverweight. Note that I'm not saying that this is healthy (it clearly isn't).

However, I still think that there is a lot of truth to the "you can't outrun a bad diet" statement. A lot of people who occasionally go for a run are likely to overestimate their energy expenditure and feel like they have to eat a cow to compensate. I think, however, that the statement becomes less and less true the more excessive the exercise gets. If you burn thousands of calories day in day out with exercise it seems to become harder to overcompensate this every time with excess food intake.

I think the bigger issue with exercise is a mental one. Weight loss seems to be a pretty bad motivation for exercising. Most people I know who picked up running for the sole purpose of losing weight eventually gave up or they are stuck with their 5km weekend park run, which, of course, is rather pointless for their objective. 100% of the avid runners I know (including myself) don't care about their weight. I could probably lose more weight (or improve my long term health) if I changed my diet, but it's simply something that I'm not interested in. I run because it's fun and a somewhat low body weight seems to be a by-product of that.

For me it also renders like this on Chrome/Windows: https://i.imgur.com/HCJTpfA.png

Really nice diagrams nevertheless

Apart from what others have said, signed binaries also pose somewhat of a challenge with reproducible builds.

> This website blog.polyhaven.com/how-we-handle-80tb-and-5m-page-views-a-month-for-under-400/ is currently offline. Cloudflare's Always Online™ shows a snapshot of this web page from the Internet Archive's Wayback Machine. To check for the live version, click Refresh.

oh the irony

See: https://www.arp242.net/curl-to-sh.html

Very interesting, thx!

Whilst we're at it: does anybody know what's the deal with the weird octal ASCII representation in the tar header? To store a file size of 10 bytes, all common tar implementations typically put literally the string "0000012" into the header, i.e., the size in octal in ASCII representation. I stumbled upon this when I had to write a parser because I wanted to rename files inside a tar without extracting the files first (turns out there is no tool for that). I found this format quite peculiar because I don't see any benefit of doing it this way.

In my experience, a regex can be a much more robust solution than an XML parser depending on the use case. Back in the days when I did some webscraping I often had parsers throwing exceptions becacuse of ivalid HTML. More often than not I switched to regular expressions, which always worked out flawlessly.

I'm a bit hesitant to believe that in the future low level systems programming will be done using languages like Rust, Nim or Zig. I would really love to see that but the programmers I know in that area are very reluctant to accept any additional layer between them and the hardware. It's not even just low level systems programming, but basically any large project with some sort of performance constraints still seems to stick to C++.

Probably the first tool I install on every Windows machine. It's beyond me how Microsoft has not incorporated something similar into Windows to replace its ridiculous native search.

Programmatically retrievable via

  whois -h whois.radb.net -- '-i origin AS32934' | grep ^route

Source: https://developers.facebook.com/docs/sharing/webmasters/craw...

I never understood why these package repositories don't include some (opt-in?) integrity checking option using digital signatures. If I download code that executes on my machine there should be at least the option to establish some level of trust. We have been doing that with linux distro package managers for decades. Seems like common sense to me.

I don't think this is a bad thing. Google, StackOverflow et al. are basically just extended documentation. What's bad about using docs?

> Unless Signal brings social features that makes these people experience FOMO of broadcasting their lives on one more network

That's ultimately what it boils down to. Making an alternative messenger that people will use is one problem, but making one that people will use *instead* of WhatsApp is an entirely different beast. The former can be solved with new (non-privacy related) features, but that's not gonna be enough to overcome the network effect.

The really important question here is whether people really don't value their privacy that much or if they just don't know any better. If people really don't care, then there is no problem to be solved. If it's an uninformed decision, then this is not a technological problem but a societal one that can be solved with awareness campaigns for example.

Signal is rightfully critiziced by people for their outdated server code. I'm just arguing that a) this does not make them as bad as WhatsApp or Telegram and b) updating the server code does not add as much trust as people tend to think. In fact, it's more of a symbolic act, because it does not give any strong technical guarantees.

1. The client is up-to-date on GitHub, which is a much better situation than with WhatsApp or Telegram. An outdated server code just gives you reason to distrust their metadata handling.

2. Trust is a bit more complicated than that. FB sells your data and Telegram is very shady in many more regards. The situation with Signal is completely different.

3. Even if the server code on GitHub was updated, what exactly would give you reason to believe that they run that version in production? Ultimately, you cannot trust the server. This can only be solved with decentralization.

> Users value their convenience, and security and privacy are inconvenient.

True. A messenger will be widely adopted if it offers features that WhatsApp et al. don't have. And by that I don't mean security or privacy, but something that people actually care about, like the stickers in telegram for example. And then that messenger just needs to be secure without interfering with functionality, convenience or UX ... problem solved.

I'm convinced that if you give the general public the "encrypt everything option", then too many people will opt in without being aware of the consequences. They will eventually forget their password, loose all their family photos and blame Apple for it. A disclaimer also wouldn't help here.

If anything, this should be some hidden developder mode kind of option to make sure that only those opt in who know what they are doing.

Looks really nice and educational. I'm a big fan of reeinventing the wheel for understanding certain primitives better. I did the same for ECC. Once it's done it's really liberating because you don't have to consider these things black magic anymore.

I always found it puzzling that DNS blacklisting for adblocking purposes works at all. I mean I'm not a webdev or anything like that, so what I'm saying might be a bit native, but couldn't you simply serve your ads using the same domain you use for your content? This immediately defeats things like PiHole completely.

As soon as as the cli becomes non-trivial I switch to python and use docopt for that (really cool module btw.). Pipelining processes is not as idiomatic in Python as it is in Bash though, but everything is better than getopts. I'd rather decapitate myself with a teaspoon than using getopts.

Can you elaborate on the ambiguity problem of Base58? I always thought that base58 was designed with this in mind as it does not include visually ambiguous letters such as "0", "O", "I", "l", etc..

It's pretty hard to make something that is not turing-complete. I've always wondered if it wouldn't be beneficial to have a turing-incomplete language given that Rice's theorem then doesn't necessarily hold and thus very powerful semantic analysis might be possible. But the language is then probably not practical due to limited expressiveness.

To get the size you would use the `_binary_FILENAME_size` symbol that the linker puts into the object file.

Didn't know about look. Seems good enough (tested on a HDD):

    $ dd of=pwned-passwords-sha1-ordered-by-hash-v5.txt oflag=nocache conv=notrunc,fdatasync count=0
    0+0 records in
    0+0 records out
    0 bytes (0 B) copied, 9.5864e-05 s, 0.0 kB/s
    $ time look `sha1sum <(echo -n password) | tr [a-z] [A-Z] | cut -d" " -f1` pwned-passwords-sha1-ordered-by-hash-v5.txt
    5BAA61E4C9B93F3F0682250B6CF8331B7EE68FD8:3730471

    real    0m0.137s
    user    0m0.002s
    sys     0m0.013s

dd is used to drop the file from the fs cache. Something the author probably didn't do given the unrealistic 49μs. It's simply not possible to fetch data from a HDD that fast.

You could also do the following to compute the overhead of the timing function:

    time_start = time.perf_counter()
    time_end = time.perf_counter()
    execution_time = time_end - time_start

Also recommended: Everything [1], a stupid fast file search engine for Windows. The initial index build takes less than a second for 100k files (no idea how it does that). Much better than what Windows provides natively.

[1] https://www.voidtools.com/