AlbertoGP

> “Their conclusion is that this strange piece of copper — 147 cm in diameter and dotted with rivets — is the upper part of a diving bell; …”

> “In any case, the copper bell in question, according to researchers, would not have traveled in that convoy, but on one of the salvage ships sent to recover the cargo. The promoter of that expedition was the Spanish soldier and politician Francisco Núñez Melián, resident in Havana and treasurer of the Windward Islands at the time of the shipwreck. According to a document dated 1630 and consulted by the researchers, Melián was aware of the lost riches and obtained royal permission to send a frigate to the area of the shipwreck in August 1625, piloted by a certain Francisco de la Luz. The fate of that vessel, whose name remains unknown, is still a mystery, but everything suggests that it was lost off the coast of Florida with the diving equipment it was carrying to attempt to extract the precious cargo.”

I can recommend this recorded live stream to get an impression of what happens when trying to use Fortran without previous experience: “Learning Fortran in 2023” https://www.youtube.com/watch?v=PvUQndB8R9s

And here is the reaction of the fortran-lang.org people to that and later streams where he went on to implement tic-tac-toe in Fortran using for the GUI the C library raylib (https://www.raylib.com/), for which he made his own bindings live on screen: https://fortran-lang.discourse.group/t/tsoding-on-fortran/61...

Could it be related to the recent news of diminshing sales of new iPhones?

“iPhone sales drop more than forecast, revenue falls for third straight quarter” https://news.ycombinator.com/item?id=36992251

“iPhone slump wipes £100B off Apple's value” https://news.ycombinator.com/item?id=37001689

If they expect people to keep using older phones, which would require repairs, they would prefer those to be iPhones.

A longer video by Stonebreaker (PostgreSQL) about DBOS (DataBase Operative System) on 2023-04-26:

“Michael Stonebraker - DBOS: A Database-oriented Operating System” https://www.youtube.com/watch?v=eB4bJqDzsU8

I’ve been thinking of building something like this for some years, and I read that IBM’s AS400 (now iSystems I think) worked essentially in this way.

As noted at the end of the video, their (DBOS) website is here, ane they are open sourcing some of the code: https://dbos-project.github.io/

> I am considering relicensing my tools under some sort of Attribution-ShareAlike license similar to the BY-SA the content on this site is licensed under. […] This would still promote community contributions, but would also require that any modifications be shared alike meaning that the modified source would have to be provided and I would have to be credited as the original author.

Wouldn’t the GPL be a better match, this being software, than BY-SA? https://choosealicense.com/

It is the Apollo Joypad (re-release of the SpeedLink Xeox Pro, http://www.apollo-computer.com/peripherals.php#Joypads) that I bought because it is compatible with the Vampire Amiga V4+ which means it has a very standard protocol, and it is set to DirectInput mode: it has a physical switch between Xinput and DirectInput mode.

I guess any proper DirectInput joypad would work just the same.

WOW! Just a `git clone`, download and unpack the .zip with the assets (end of the blog post), type `make` and it compiles on Linux without warnings into `wipegame`, run it and it works first time!

You need a joypad/joystick, I was surprised that I could plug it in after the game had started and it picked it up, insted of having to restart the game as I remember being the case with other games on Linux years ago.

Thanks, after doing that it still fails in the same way as when installing RTMIDI from source:

    CMake Error at cmake/FindRTMIDI.cmake:51 (string):
      string sub-command REGEX, mode REPLACE needs at least 6 arguments total to
      command.

I’ll see if I find some way of either compiling this one, or finding another one that does.

I went ahead and compiled and installed the current version of RTMIDI, and now it fails with something else that looks like some file is not compatible with current CMake:

    CMake Error at cmake/FindRTMIDI.cmake:51 (string):
      string sub-command REGEX, mode REPLACE needs at least 6 arguments total to
      command.

There might be issues even with normal linux boxes, just because of bit rot.

Coincidentally I was trying to compile this yesterday and it failed because it did not find the RTMIDI library, which INSTALL.md says is optional: “RtMidi/ALSA (optional, for Linux MIDI support)”.

The machine I’m using does not have that library nor any convenient way of installing it right this moment, so I just moved on. I don’t know whether there will be further snags.

One thing I like in Slashdot compared to HN is how stories there always contain some abstract.

Here is the one from that dissertation, which sounds intriguing:

> While modern software development heavily relies on versioned packages, the concept of versions is rarely supported in the semantics of programming languages, resulting in bulky and unsafe software updates.

> The dissertation proposes a programming language that intrinsically supports versions. To establish a basis of finer-grained version control in language semantics, the author proposes a language called VL, with core calculus for supporting multiple versions, a compilation method to the core, and an inference algorithm determining the version of each expression.

> The author proved the type safety of the core calculus to guarantee consistent versions in a program. The author also implements VL, a minimal but adequate functional language that supports data structures and a module system, and conducts a case study involving the simultaneous use of multiple versions.

AFAIK that originated with the video where he said this:

NSFW: https://www.youtube.com/watch?v=AbG6u86t4bA

Then the word “glowie” was coined in 4chan based on that, as others commented here already.

I had a vague idea of what the Golden Dinar was, but this article gives some concrete figures and references:

> [email dated April 2, 2011] "Qaddafi's government holds 143 tons of gold, and a similar amount in silver ... This gold was accumulated prior to the current rebellion and was intended to be used to establish a pan-African currency based on the Libyan golden Dinar. This plan was designed to provide the Francophone African Countries with an alternative to the French franc (CFA)." […]

> "According to knowledgeable individuals this quantity of gold and silver is valued at more than $7 billion. French intelligence officers discovered this plan shortly after the current rebellion began, and this was one of the factors that influenced President Nicolas Sarkozy's decision to commit France to the attack on Libya. […]

> Gaddafi's threatened attempt to establish an independent African currency was not taken lightly by Western interests. In 2011, Sarkozy reportedly called the Libyan leader a threat to the financial security of the world. How could this tiny country of six million people pose such a threat? […]

> … another enigma remarked on by early commentators. Why, within weeks of initiating fighting, did the rebels set up their own central bank?

> "This suggests we have a bit more than a rag tag bunch of rebels running around and that there are some pretty sophisticated influences. I have never before heard of a central bank being created in just a matter of weeks out of a popular uprising."

At the bottom of the page lemonade is made:

> Apart from the most obvious answer (that it’s useful for things exactly like <tgmath.h>), the best thing I’ve thought of to do with _Generic is to use it for deliberate error checking.

> The annoyance in all the previous sections was that it was very hard to avoid compile errors, when we wanted our code to actually compile, run and do something useful. But if what we wanted was to provoke compile errors on a hair-trigger basis, then perhaps we could use it for that more reliably?

[01:41] So I'm really excited to talk to him about that we're gonna bring in folks who are working on things that are within the Thunderbird universe that are exciting and interesting that folks might not even suspect. I will share a little bit of a teaser in that we do have some initiatives within the team

[02:06] to do things that require a server and you know that make email better by leveraging not just what you can run on your machine, but what we can do for you.

[02:16] This is breaking news but Firefox Send is being reborn within the Thunderbird family.

[02:24] The… We have a way to schedule meetings with folks.

[02:30] All… many products that are out there and make that more automated for our users.

I am using the C version of Blake3 after I decided some time ago to rewrite a prototype application from Rust to C and this is indeed very interesting for me. Some uses of this application would benefit from O_DIRECT in the way described in the article.

I'm going to try this out this weekend.

Reading the title I thought that the price was too low, but the linked page clarifies that it is just “for a small software without specific problems” and anything else requires a custom quote.

Lots of things might come up requiring investigation on compatibility with the library versions available in the taget Debian release. It’s not enough for the program to compile and run in an ad-hoc system where you install whichever versions are known to work.

I had forgotten that it was a fast linker:

> Mold 2.0.0 is a new major release of our high-speed linker. With this release, we've transitioned our license from AGPL to MIT, aiming to expand the user base of our linker. This was not an easy decision, as those who have been following our progress know that we've been attempting to monetize our product through an AGPL/commercial license dual-licensing scheme. Unfortunately, this approach didn't meet our expectations. The license change represents our acceptance of this reality. We don't want to persist with a strategy that didn't work well.

The article does mention that the general model it advocates is that of classic Internet forums, and I think the difference is just a better User Interface.

This does not seem difficult: combine the thread list with the thread view so that looking at the messages in one thread does not bring you away from the thread list. That way, taking a look at the contents of different threads is not as disruptive.

Take into account that old Model Ms tend to have their plastic rivets fail: those keep the sandwich plate-membrane-plate together, and after enough of them break some keys stop registering. There is a laborious fix, called “bolt-mod” which consists on drilling the rivets off and putting bolts with nuts instead.

So, you either get one whose keys have all been tested to work, or you should be ready to do some manual work.

As far as I know, internally all models (M122 terminal, standard Enhanced Keyboard, and Space Savers/TKL) are internally the same w.r.t. reliability, noise, and feel.

This is a traditional architectural feature also in my home city of Toledo in Spain: [in Spanish, could not find an English version] https://es.wikipedia.org/wiki/Patios_de_Toledo

That was however an ancient idea, already used for example by the Roman Empire (https://en.wikipedia.org/wiki/Domus) although the typical Toledo patio is a multi-storey construction like the Chinese examples in the linked article, most similar to the islamic “wast ad-dar” [“middle of the house”]: https://en.wikipedia.org/wiki/Islamic_architecture#Courtyard...

This is the abstract from a book dedicated to the subject titled “Courtyard Housing: Past, Present and Future”:

> Courtyard housing is one of the oldest forms of domestic development spanning at least 2000 years and occurring in distinctive form in many regions of the world. Traditionally associated with the Middle East where climate and culture have given shape to a particular type of courtyard housing, other examples exist in Latin America, China and in Europe, where the model has been reinterpreted. This book demonstrates, through discussions on sustainability and regional identity, and via a series of case studies, technical planning and design solutions, that the courtyard housing form has a future as well as a past. — https://books.google.de/books?vid=ISBN9780415262729

That summarization includes a video I hadn’t seen before where Calhoun explains the results in his own words and mentions the case of the Ik tribe which I didn’t know about:

> [00:01:55] at least one human group, the Ik tribe discovered in Africa by dr. Colin Turnbull (https://en.wikipedia.org/wiki/Ik_people#The_Mountain_People) has gone through a similar terrible process according to dr. Calhoun: the two conditions that they went through was that there was a decision on the then colonial government to make a wildlife preserve of this big valley which was their home as they are hunter-gatherers, they moved them up onto the mountainside and gathered together in much higher densities, they had been spread out small groups, all placed in one place and also required to engage entirely different way of life, which their social roles, the whole… everything they had learned how to do had no meaning. Under those conditions the social behavior broke down. Entirely they began living… they still live together but their activities was just directed towards themselves, everything you would think is human disappeared.

It wasn't obvious to me what this was about from the title but it's very relevant to my interests:

> In this post I will describe Use.GPU's text rendering, which uses a bespoke approach to Signed Distance Fields (SDFs). This was borne out of necessity: while SDF text is pretty common on GPUs, some of the established practice on generating SDFs from masks is incorrect, and some libraries get it right only by accident. So this will be a deep dive from first principles, about the nuances of subpixels.

Not, it’s clearly not him. I can tell by the pixels, and from seeing quite a few pictures and videos of Engelbart in my time. Just kidding, but even the voice is different.

Thanks for the hints, I did read the instructions at the URL I mentioned, but had assumed that GHC 9 would be backwards compatible with GHC 8. I have no experience with Haskell.

Ah well… tried to install the CLI client from source (https://github.com/simplex-chat/simplex-chat/blob/stable/doc...), installed the Haskell compiler GHC v9, the Haskell package manager Cabal, `cabal update` went fine, no problem until `cabal install`:

    cabal: Could not resolve dependencies:
    [__0] next goal: aeson (user goal)
    [__0] rejecting: aeson-2.2.0.0, aeson-2.1.2.1, aeson-2.1.2.0, aeson-2.1.1.0,
    aeson-2.1.0.0 (constraint from user target requires ==2.0.3.0)
    [__0] trying: aeson-2.0.3.0
    [__1] trying: base-4.15.1.0/installed-4.15.1.0 (dependency of aeson)
    [__2] next goal: android-support (user goal)
    [__2] rejecting: android-support-0.1.0.0 (conflict:
    base==4.15.1.0/installed-4.15.1.0, android-support => base^>=4.14.3.0)
    [__2] fail (backjumping, conflict set: android-support, base)
    After searching the rest of the dependency tree exhaustively, these were the
    goals I've had most trouble fulfilling: base, aeson, android-support

“android-support”???

Dependency hell again. I have no idea what those things are.

After a quick look the documentation is excellent: tells me clearly what it does, how, and why! A messaging system without user identifiers is something I had wondered about, why it hadn't been done as far as I knew.

This is interesting, but why don’t they link to the paper? Or at least give its title?

The paper is «“Low-Resource” Text Classification: A Parameter-Free Classification Method with Compressors»:

ACL Anthology, has PDF link: https://aclanthology.org/2023.findings-acl.426/

Arxiv, has PDF link: https://arxiv.org/abs/2212.09410

arXiv Vanity, HTML render of it: https://www.arxiv-vanity.com/papers/2212.09410/

ResearchGate, has PDF link: https://www.researchgate.net/publication/366423949_Less_is_M...

This is Listing 1: ”Python Code for Text Classification with gzip.”

    import gzip
    import numpy as np
    for (x1, _) in test_set:
        Cx1 = len(gzip.compress(x1.encode()))
        distance_from_x1 = []
        for (x2, _) in training_set:
            Cx2 = len(gzip.compress(x2.encode())
            x1x2 = " ".join([x1, x2])
            Cx1x2 = len(gzip.compress(x1x2.encode())
            ncd = (Cx1x2 - min(Cx1,Cx2)) / max(Cx1, Cx2)
            distance_from_x1.append(ncd)
        sorted_idx = np.argsort(np.array(distance_from_x1))
        top_k_class = training_set[sorted_idx[:k], 1]
        predict_class = max(set(top_k_class), key=top_k_class.count)

Abstract: In this paper, we propose a non-parametric alternative to DNNs that’s easy, lightweight, and universal in text classification: a combi- nation of a simple compressor like gzip with a k-nearest-neighbor classifier. Without any training parameters, our method achieves re- sults that are competitive with non-pretrained deep learning methods on six in-distribution datasets. It even outperforms BERT on all five OOD datasets, including four low-resource lan- guages. Our method also excels in the few-shot setting, where labeled data are too scarce to train DNNs effectively. Code is available at https://github.com/bazingagin/npc_gzip

Ah, this was already mentioned a few days ago in HN:

73 comments: “Gzip and KNN Outperforms Transformers on Text Classification” https://news.ycombinator.com/item?id=36707193

7 comments: “Plain old gzip+kNN outperforms BERT and other DNNs” https://news.ycombinator.com/item?id=36705472

7 comments: “A Parameter-Free Classification Method with Compressors” https://news.ycombinator.com/item?id=36707509

Do you mean this one of 2021-06-30?

> Many boxes of fireworks were stamped “Made in China.” The logo of a green space alien marked some boxes as coming from Area 51, a fireworks dealer in Nevada.

> [...] Officials had discovered what they called “improvised explosive devices” — or homemade fireworks believed to contain flash powder — mixed in with the commercial fireworks and piled in a trash bag.

> There were 44 devices about as wide and tall as a soda can, and 280 smaller devices about the size of a toilet paper roll. Each had a short fuse.

> All were moved to the shade — away from the commercial fireworks — and the containment vessel was called in. It was only about 80 degrees, but police used fans to keep the fireworks cool and prevent them from detonating.

> [...] The next thing Becerra saw, he said, was “the sun” — a burst of light as the containment vessel blew apart and the blast it was meant to control shot out in all directions, flipping cars and cracking foundations, smashing windows and injuring residents. Glass shards littered Silva’s face.

“From a fireworks tip to ‘catastrophic failure’ in 12 hours: How the LAPD blew up a neighborhood” https://www.latimes.com/california/story/2022-03-06/how-lapd...

> Mell Hogg, the lead technician — aka Bomb Tech A — had been on the bomb squad for less than three years. Following standard bomb-squad practice, Hogg did not weigh the powder from the fireworks the team intended to detonate in the LAPD’s containment vessel and instead grossly underestimated its explosive power, according to the inspector general’s report. He is now with the department’s training division, according to a recent roster.

> Mark Richardson — Bomb Tech B — helped Hogg X-ray samples of the explosives, the inspector general’s records show. Richardson still works in the emergency services division, which houses the bomb squad, but it is unclear in what role.

> Brendan McCarty — Bomb Tech C — had 18 years of experience on the bomb squad and warned his fellow technicians and Levesque that the plan was not safe, the report said. McCarty has since retired from the force.

> Thomas Deluccia — Bomb Tech D — arrived on scene that afternoon and helped load the commercial fireworks discovered at a home on East 27th Street onto pallets to transport away from the scene, the records state. Deluccia has also remained in the emergency services division, although it is unclear what his role is.

> And, finally, there was 12-year bomb squad veteran Stefanie Alcocer, or Bomb Tech E. Alcocer was primarily responsible for constructing a countercharge, the report shows, the explosive used to detonate the fireworks inside the department’s “total containment vessel,” which failed in the blast. Alcocer was suspended for 10 days for her role in the incident, according to police officials. At some point after the blast, she was promoted to sergeant.

“LAPD refused to ID officers behind a huge fireworks explosion. Now we have their names” https://www.latimes.com/california/story/2023-07-14/lapd-ref...

It’s astonishing to see how little comes in a bag of Doritos these days.