edwcross

Someone please help Framework sell to businesses, otherwise those of us who want but cannot buy them will have to keep buying user-hostile laptops.

I don't understand how to contribute to fix OCR errors. Even after registering and logging in, after searching a word and clicking on the image, then getting the annotation in the viewer, it does not show the "Edit | Confirm" buttons as they appear in the screenshot from the documentation. So I can see the annotation and its automatic transcriptions, but I can neither edit nor confirm it.

Because Framework does not work with resellers (so my company cannot buy it), Dell keeps doing stupid Linux-hostile things (like re-releasing accessory revisions that break Linux features), and Lenovo has no high-end laptops available (at least not on the business side). And because managers all love their Macs, they always find a way to get company policy to allow for them.

Or more likely, the thief just resold the stolen bike for $10, thus negatively impacting the economy (preventing legal resellers from operating fairly; forcing people to spend more money and effort on protection; minimizing the efficiency of the person having the bike stolen; raising insurance premiums) and the ecology (steering people towards cars, which are harder to steal, at least without insurance; forcing people to re-buy bikes and stolen goods). Proper wealth redistribution is not stealing, at least not this kind of "petty stealing".

Is there some tutorial about Fediverse for organizations, in a way that is really simple for non-experts?

I mainly use Twitter for town-level information (as a glorified RSS), and the local administration barely masters Twitter, let alone more complex tools. But if there is a simple way to copy their messages towards the Fediverse, that could help bootstrap it.

Maybe some day Software Heritage (https://www.softwareheritage.org/) will expose enough metadata between repositories so that people will be able to train their own recommenders for better discovery.

In theory, you are possibly right, but in practice, that is not always the case.

For instance, in France, currently some departments have a fixed number of interview slots per month, meaning people either have to hack their way into finding an available slot (e.g. using scripts to refresh the website in the middle of the night and hope to wake up when a slot becomes available; good luck if you don't know how to hack the system yourself), or they have to pay scalpers to get an appointment.

Then, you get assigned a pre-interview slot, which can take as long as a year of waiting. If you are lucky and everything works out, in the pre-interview, you will get assigned an actual interview slot. Which will be communicated by letter. And which can take yet another year of waiting.

Then, when the actual interview takes place, they give you no definitive answer for almost a year. You only know if it worked out at the end of the process. Overall, the "be in the country for N years" can thus become "be in the country for N+3 years, while not being sure that the next step will actually work or you will be sent back to the beginning"; also, hope that no politicians change the system in the meantime, since all of this is unspecified, so that they can change the rules at any moment and there's no one to complain to.

Of course, your mileage may vary; if you are rich enough to live in Paris, you will probably get better treatment.

So, the US is definitely bad, but the real situation in some Europeans countries is not as good as it looks like.

Unfortunately pCon.planner is Windows-only. I wonder if the file format is standard and compatible with other software?

There are probably others, but this is one such startup: https://trust-in-soft.com/

In practice, it's not that easy; fuzzers tend to be more effective for that.

Usually, the time required to properly setup and understand and refine the analysis will take most of the trial period.

These tools are better thought of in the long term, they may require some changes to coding style, and also some knowledge of the code itself.

While they are good for building defenses, they are not so effective for helping attackers. But in the end, yes, they _could_ be used for that.

Also Frama-C/Eva: https://www.frama-c.com/fc-plugins/eva.html

Note that Astrée was developed in France, by researchers from Inria. AbsInt sells it packaged as a product (with support, etc), but the bulk of the research was done in the 2000's by researchers such as Cousot.

This is also the case with CompCert, which is still developed mostly by French researchers (with an open-source, non-commercial license), and packaged in commercial form by AbsInt.

As an outsider, this description resembles that of Trump, except with less "tiny hands", and more "tech". Has Elon Musk considered doing politics?

The theory I heard worked in the other direction: if we assume scammers have a finite amount of time, it could be in their interest to minimize the amount of "likely good targets" in order to increase the amount of "very likely good targets". So all those untapped potential targets are just too similar to non-good targets for them to discriminate effectively, leaving them so far focused on the lower-hanging fruit.

I mean, with Google Translate, spellcheckers, etc, improving all the time, at least some of those messages should have been improving as well, no? If their grammar has not improved at all during the last decade, then there might be a hinge of truth to the theory.

Could it be for driver comfort? I sometimes have the impression that buses have their temperature tuned for the driver, who's staying in the bus for hours and has everyone's safety at least partially related to their comfort. But I find it impossible to confirm this by asking the bus companies why their buses are set this way (e.g. when windows were closed during Covid at about 20°C).

QuickDic (from F-Droid, with offline dictionaries) works for me.

On the one hand, I agree that paying attention to the spammer is bad; on the other, I do believe there might be some use in publicly stating that such PRs will never be merged and are frowned upon; hopefully other people reading (many of them likely beginner programmers) will get the message. But there's likely a better way to do the "teaching" without drawing any attention to the perpetrator.

The patch itself is indeed worse than useless, it's the kind of rephrasing just to say "I did something", but which actually makes it worse (adds useless words and English mistakes). If the "kid" is ready to send this kind of useless "contribution" (which takes some deliberate effort), they surely are ready for being reprimanded (or, more likely, they will be actually proud of it).

I tried using it several times in recent years, and whenever trying to upload files, I had lots of issues. Filed some bugs, never got them resolved, and ended up giving up after several of my photos and descriptions were lost (or never properly uploaded).

I really want the Commons Mobile app to work, and I don't know why it always failed with my Pixel phone, but it was a really frustrating experience.

A junior explorer pocket microscope (https://opentinkers.com/collections/toddlers/products/junior...) is about 10$ and comes without light, but the transparent plastic allows using a phone flashlight. In the end, I find it better than having its own (weaker) light source. Less prone to failure, no extra battery/weight.

We transitioned from a "high-trust society" (where software, Internet, etc, did not need to take into account so many security considerations, protect itself from enemies everywhere) into a "low-trust society" (encryption everywhere, hardware mitigations for Spectre-kind attacks, Javascript-based tracking, etc).

The overhead is similar to the one people face in third-world countries: you pay once for the government (taxes), then again for basic services (private education, private security, private health, private transportation), and you still end up in a worse situation than in a (high-trust) first-world country.

I tried using a Core 2 Duo laptop recently, but the simple fact that HTTPS is mandatory nowadays, and Core 2 doesn't have AES-NI, makes the laptop spin its fan much more than it should, even if I try to use an old Linux/Windows.

To be honest, USB 3 was already a loss in some aspects: USB 2 allowed for extremely thin USB drives (as thick as the inner part of the plug); several keyboards had plugs, cables were thin, all sorts of gadgets were made. With USB 3, the extra pins required extra thickness; the beefier requirements made many cables shorter and thicker, and connectors diminished in number (due to more expensive constraints). The extra pins introduced new failure modes (I had a pen drive with internal bent pins, very hard to see, impossible to repair), but at least most people respected the color conventions (blue).

If only color coding was used to help disambiguate some USB C features, that might help somewhat. When buying USB-C cables, I have between $5 and $125 for the same length; the vendor lists all kinds of features (power delivery, Thunderbolt 3, what else) but things simply do not work in the end.

Overall, the whimsical Cambrian explosion of USB 2 and all sorts of funny pen drives and gadgets started dying with USB 3, and I'm not expecting to see it return with USB-C. With increasing power demands and what not, maybe we'll soon see wall plugs with USB-C instead of ordinary plugs. At least international travelers won't need adapters anymore...

C is a pig of a language, so tools that show just how horrible it is, how many bugs are unwittingly written, and how hard it is to make its code safe, could also help pushing people towards better languages.

These tools are more like putting on X-ray glasses and seeing how your fast-food nuggets are made. You will still eat them if you have to, or if you're in a hurry, but you'll be more inclined to take the time to learn cooking, so you can eat healthier food later.

It would be interesting to know if at least part of the work could somehow be integrated into Software Heritage (https://www.softwareheritage.org/) or CodeMeta (https://codemeta.github.io).

Isn't that kind of what Brazil did in 1994 with URV?

From Wikipedia: "URVs were quoted in cruzeiros reais and its intrinsic value was pegged to three price indices and had a fixed parity of 1-to-1 to the daily U.S. dollar exchange rate."

Frama-C works at the C semantic level, without knowledge about processor instruction-level details (expect for some extended assembly syntax, and a few details here and there). If you are talking about unspecified sequences of C statements, then Frama-C can warn about some situations, as mentioned in the user manual (https://www.frama-c.com/download/frama-c-user-manual.pdf):

option -unspecified-access may be used to check when the evaluation of an expression depends on the order in which its sub-expressions are evaluated. For instance, this occurs with the following piece of code.

  int i, j, *p;
  i = 1;
  p = &i;
  j = i++ + (*p)++;

In this code, it is unclear in which order the elements of the right-hand side of the last assignment are evaluated. Indeed, the variable j can get any value as i and p are aliased.

In some cases, knowing the architecture is essential to get information such as the width of integer types. Ideally, one would like a completely system-independent, portable analysis, but this is extremely hard in C. For instance, the ISO C11 standard, in section 5.2.4.1 Translation limits, states that The implementation shall be able to translate and execute at least one program that contains at least one instance of every one of the following limits [...] 4095 characters in a string literal (after concatenation), 65535 bytes in an object, [...]. It also states that INT_MAX may be, for instance, as low as 32767. So, for a truly portable analysis, one would have to warn whenever any of these limits are reached, leading to an analysis useless except for some toy examples.

This seems mostly a guide from a 42 School student towards other 42 School students.

It shows one of the issues of having previous-year students teaching younger students: they don't always have enough experience to know the best way to teach things.

Overall, the title is very misleading, a better one would be "42 School Cheat Sheet", ideally with "by a former 42 School student" appended.

The Cartier Foundation for contemporary art had an exhibit on "geometric art" from South America some time ago, and there remains a 7-minute video (in French only, but the automatically-translated subtitles seem fine) on Youtube about Freddy Mamani, the architect: https://www.youtube.com/watch?v=0odbR7FOZpU

Very beautiful images from these buildings.

Counterpoint: when using a system with hybrid graphics which required me to use proprietary video drivers, switching to the latest Fedora version often brings issues with suspend, or random freezes, or lack of external HDMI output... After a few months, the release stabilizes and my issues disappear. But whenever I tried to be optimistic and jump to the latest release too soon, I got bitten by it and wasted hours trying to find workarounds.

In Paris and other French cities, you have "game cafés", where you pay per hour (or you buy food which is worth a "1-hour ticket"), and they bring you the games, so you don't have to own them. Good for trying out new games (which they of course sell you, if you want to take home), but neither free food nor free time.

Otherwise, some places have public "ludothèques" for playing board games, like public libraries; but these are non-profit associations, and they usually don't sell anything.