akubera

    const { performance } = require('perf_hooks');

    function fib(n) {
       if (n == 0 || n == 1) { return 1; }
       return fib(n - 1) + fib(n - 2);
    }

    function fibn(n) {
       if (n == 0n || n == 1n) { return 1n; }
       return fibn(n - 1n) + fibn(n - 2n);
    }

    var t0 = performance.now(); fib(35); console.log("fib:", performance.now() - t0);
    var t0 = performance.now(); fibn(35n); console.log("fibn:", performance.now() - t0);

    fib: 127.30008998513222
    fibn: 2134.1405459940434

(An equivalent version with Python on my machine: ~2657.3ms)

https://amzn.github.io/ion-docs/

Superset of JSON with many extra features that people in the comments here desire from their data-language, such as support for comments, timestamps and s-expressions.

This looks like a big improvement, with math functions auto-imported, and the --script option is a fantastic idea.

Good job.

[0.0] https://pypi.org/project/pythonpy

[0.1] https://news.ycombinator.com/item?id=8158976

(Relevant post https://news.ycombinator.com/item?id=14996715)

But on the topic: is there actually a dearth of APIs "these days" vs peak Web-2.0, or have the major players just restricted theirs due to abuse, and thus it seems like the whole world of possibilities have veen restricted? One can easily find lists of public apis (e.g https://github.com/n0shake/Public-APIs), but perhaps the video was more about the facilitators, like Yahoo Pipes.

Of course, isn't that convenient that they carry greater emotional weight...

[0] https://news.ycombinator.com/item?id=22290486

Seriously though, the question of "why isn't it done this way" is answered a couple times: "it's all just bytes in the end". Bytes & arrays-of-bytes are really the only thing that you can trust that all systems and all languages interpret the same (even 2-byte integers can be interpreted differently by different systems). Presumably the author just wants to be able to memcpy structures between programs or from disk, or maybe with some high efficiency pre-allocated heap stuff, but you really need to have code to validate the bytes, essentially requiring the parsing of your data-structure which the author is trying to avoid.

Running programs with some common argument notation does sound really nice, but ideally that's just calling a function from your language's binding to their library (as their executable should do, right?).

I'll try to answer it: I think you just want to move units around so h (Planck's constant) has units of Joules instead of Joule-seconds. While this works mathematically, it doesn't really make sense physically, where we care about the energy of a photon because that's the value that is conserved (along with momentum, also relating to h) during any interaction (e.g. the quantity gained by an electron if the photon is absorbed). It really doesn't make sense to want to quantify the "power" of a photon, which doesn't have any physical meaning, in favor of a simpler constant.

You are right about the constant 'h' having a "hard-coded" 1-second built in, but so do ALL derived physical constants . If we, say, changed it to two seconds, the number would be cut in half to reflect the scale change and preserve the energy of the photon (or any action) because that does not depend on scale.

As to "why is it Joule-seconds?": that's the discovered law of nature. If the energy was proportional to frequency squared, h would be in units of J*s².

> The Empire State Building. Construction was started and finished in 410 days.

> Unix. Ken Thompson wrote the first version in three weeks.

> Git. Linus Torvalds started working on Git on April 3 2005. It was self-hosting 4 days later.

Inspiring.

https://en.wikipedia.org/wiki/Baby_Gaga

I don't know how well that turned out.

I assume that if you only connect one computer (authenticator) and lose the device, you're either SOL or the service has some workaround where you pre-register an email address or a separate "Login With" service.

Also "foo['bar'] returns a KeyError", then saying sometimes this is resolved by "getattr(foo, 'bar', None)" -- the split between attributes and items are one of the best things about python.

I don't know why list comprehensions are "magic syntax", and you can't complain about having to type list(map(...)) and then complain that you don't like list comprehensions!

"You have to cast your data back to a list/tuple after using enumerate() and map()." -- I don't know what that means

"Different syntaxes for lists and tuples." -- Again, what? They're different types!

How is r'a\nd' less "goofy" than String.raw`a\nd` in JS or any other language with prefixed strings?

Python is clearly not for everyone nor for every problem, but in my opinion it is a great programming language.

I assume your "middle of nowhere" comment is just trolling.

There's the Rust repository, and the auto-generated documentation can be found at https://docs.rs/libp2p/0.11.0/libp2p/

I'm curious what about a C implementation is the requirement for you? Is it the ability for to link against your C/C++ code, or for reading and understanding the structures and algorithms with the simplicity of C parlance, or another reason?

If the first one, it is possible to create a C shared-library "cdylib" with Rust (included in cargo by default), and using cbindgen[1] you can generate the headers to use the c-api.

It doesn't look like anybody has done that yet, unfortunately. But know that it should be a lot easier to create C/C++ compatibility than doing a rewrite or port.

[1] https://github.com/eqrion/cbindgen

EDIT:

Or as the sibling comment points out, there ARE C implementations out there, so you can ignore this post.

To be clear, that's not a new feature in 3.8.

> It [The Internet/The Web] took off with hackers running independent servers in their basements.

Is that true? Maybe if we're talking "the internet", things like MUD, IRC, Usenet, or BBS servers were run by individuals, but looking at the "List of websites founded before 1995"[1] it seems like there were big players outside of labs/universities pretty early on in the history of The Web. I'm asking a legitimate question here: was www successful (at least in part) due to websites from individuals (let's maybe say "students on university networks" rather than "hackers in basements") over large institutions (like the magazines)?

> The informational landscape today shifted from people sharing personal websites to sharing social posts.

Again, I question the accuracy. Did more people have personal websites than BBS/Usenet/Web-Forum accounts?

Regarding the dpage.io site this article is promoting:

> As you can see, there are no ads, no clutter, only information that was created by the user.

This is a problem I think the "distributed web" community has to deal with: What if the user wants to monetize their content, perhaps with ads? Either people are going to do some "pay per view" of articles on your network (which seems to be antithetical to "information for the masses" that I have a feeling most decentralized developers identify with), or they use ads. Want your ads to do better? well don't fund the network that has reader anonymity built in.

> The domain name system enables any person on earth to have his human-readable address on the web. This system is decentralized by design and is owned by nobody.

So they want to fork DNS from ICANN so your identity is always linked to your website?

> Dpage.io is not a website builder, and it is not a blogging platform.

Design your own website, upload it to us and we'll host it for free.

Reading the contents of the strange little iPad on https://dpage.io/ it says:

> Your user profile details along with your data are by default stored on a free Blockstack's Gaia storage hub. DPAGE doesn't store your data on its own servers. You can run your own storage hub on a server of your choice.

So it is that kind of distributed server: the one you run yourself.

<aside> The phrase "a free Blockstack's Gaia storage hub" is strange and feels very corporate, being sure to indicate that it's not just any Gaia server, but the one OWNED by Blockstack. Sometimes it's the little things that show you how much control the marketing department has. </aside>

I'm not going to comment on their blockchain stuff.

I'm only criticizing because I think people are putting lots of effort into solutions that are misguided: "The people WANT to have their own websites! This is the platform they'll want for sure!". I very well may be missing the point here, but inaccuracies and romanticizing the past lead me to skepticism.

XAMPP[2] is still around after 17 years, (last updated 4 days ago), which I remember being super easy to setup your own site, and I have a feeling it's even easier now. CMS platforms like Wordpress.com, SquareSpace, etc, do allow you to own your own content, easily use themes/plugins, and backup/export everything if you want to move (for a price, because they're providing a service, which is reasonable).

I think the most interesting decentralized platform is dat/beakerbrowser, because

A) There's already a tangible working product (Beaker)

B) There's no login/user-admin: your browser upon startup is a node in the network

C) Backwards compatible with http(s), so it's not like there's a hard fork of the DNS namespace and links no longer work.

D) No concept of privacy/encryption (that I'm aware of). You really don't own your data, you publish it and the system is designed to allow anyone/everyone with the link to copy it, change it, host it, seed it. If you don't like that, don't use it.

Of course you have to seed your site, so again there's a chicken and egg problem of hosting your own server (but if it's just running the browser until your friend says they've downloaded the site, it's easy, right?)

I didn't mean for this to turn into a rant, but I guess it did.

I wish dpage/decentus/blockstack/gaia good luck because I think distributed webs do have a future and every player can bring something to the table, but I hope their message/intentions become a little more focused.

[1] https://en.wikipedia.org/wiki/List_of_websites_founded_befor...

[2] https://www.apachefriends.org/index.html

Actually it looks like I can get flake8 & pyright simultaneously; that's actually the ideal situation I was looking for.

Many projects need node, but not all; I don't know any which require docker.

I want to be clear: with pyright, it's not bad! I think it's the right way to do the LSP plugin: no dependency on the current python version (am I using the python in the venv? or ~/.local/bin? or /usr/bin/?) and great integration with the editor.

If it's a large project that will benefit from "5x faster" typechecking than mypy, then yeah, it's probably worth the dependencies.

If you have a 5-file python package I found on pypi that I want to contribute to, but there's nonstandard hoops to jump through to get `python setup.py test` working properly, I would have preferred you made a different call in this regard.

I'm glad we have choices and it's not mono-culture.

As I mentioned in a comment to that post, I used pyright for a few minutes and was impressed. You get red-squigglies under unexpected-type & no-such method errors, and I saw green squiggly under a "value may not be set" in a function like:

    def foo(x: int):
        if x > 4:
            y = 1
        return y 

Searching VSCode, I see no plugins for mypy & PyType, and the pyre-vscode plugin requires pyre-check to be installed, and a .pyre_configuration file in the directory? All I know is I couldn't get it to do anything, but pyright worked out of the box with no effort.

I haven't used this PyType project, but it looks like it competes with mypy. I like that it, too, uses your setup.cfg file for configuration (I'm all about single-file configuration projects). Someone else will have to speak to the differences between them. I'm sure they have the same functionality, so the differences will likely be in the specifics like what the error messages look like and how easy they are to configure. If you contribute to various python projects, I supposed it wouldn't hurt to have them both installed.

As for which to choose, mypy is Guido's project (or at least the one he's associated with) and until another is clearly superior, you can be pretty certain it will track changes in the python language closely for a while. I use mypy & flake8 for "static checking".

My question is: are there any plugins that show error-indicator squigglies for pep8 violations?

Only issue I see immediately is it does not use the default sys.path that the vs-code-configured python interpreter uses, so it cannot resolve some of the packages I'm using.

But overall, very nice.

The EU's document is "Charter of Fundamental Rights of the European_Union"[0], which does grant privacy as one. That pretty clearly does make it a "fundamental right".

Note the US Constitution's Bill of Rights does not offer right-to-privacy (but the 4th provides protection from searches-and-seizures without probable cause, which really doesn't deal with "privacy", especially from non-governmental actors).

Perhaps parent should have said "fundamental right granted by the EU", but in this context it's pretty clear.

[0] https://en.wikipedia.org/wiki/Charter_of_Fundamental_Rights_...

    methods = [method1, method2, method3, method4]
    for method in methods:
        try:
            method()
        except Exception:
            pass
        else:
            break
    else:
        raise NoMethodWorked()

    methods = [(method1, TypeError), (method2, KeyError)]
    for m, e in methods:
       try: m()
       except e: ...

I was able to upgrade the SSD and it was a pretty good experience, if you're into that.

It amounts to using CPU prefetch instructions with C++ coroutines to simulate hyperthreading in software by scheduling instructions around cache misses (but is potentially better than hyperthreads because it's not limited to 2/core)

Writing a REST API in front of their software would help with "service-discovery", usability, and of course reproducibility; but there's the issue of sending back many-GB of data per run, questionable amount of customization (but if you allow uploading an ini config file, that should be ok), no institutional history of doing tasks this way, and the fact that this is not what I'm paid to do.

For more on scientific computation differs from a more typical setup, I highly recommend this StrangeLoop talk by Jim Privars: https://www.youtube.com/watch?v=jvt4v2LTGK0