Nostr notes by THE MARTYR OF BUTLERIAN JIHAD

so, there are two ways to make this better.<li>Just ...

2026-10-04T12:14:03Z

In reply to nevent1q…09he
_________________________

so, there are two ways to make this better.<li>Just generate a random waveform (possibly, matching some hash of the message that contains the audio).</li><li>Download the audio and re-encode the waveform locally.</li>

Either of these is better than PCM-in-JSON.

i can almost guarantee they will display it wrong this way. So ...

2026-10-04T11:26:59Z

In reply to nevent1q…qhy5
_________________________

i can almost guarantee they will display it wrong this way. So you'd rather not, unless you expect shenanigans to happen.

again, what do you think .wav files contain, you big silly dummy?

2026-10-01T23:43:03Z

In reply to nevent1q…rw34
_________________________

again, what do you think .wav files contain, you big silly dummy?

read the matrix thing again. They literally put .wav file ...

2026-10-01T23:15:07Z

In reply to nevent1q…3a2m
_________________________

read the matrix thing again. They literally put .wav file contents into JSON. It doesn't really matter what reasons they pursue, because the very fact of what they did here is funny.

> thing encodes sound as amplitude samples > no codec ...

2026-10-01T22:27:08Z

In reply to nevent1q…5r3h
_________________________

> thing encodes sound as amplitude samples
> no codec related things here what so ever

you were saying? It is in the most literal sense PCM encoded as JSON.

Designing a good audio codec is hard. Designing a stupid one is ...

2026-10-01T20:48:02Z

Designing a good audio codec is hard. Designing a stupid one is even harder, and nothing can surpass Matrix people here.

> Under the m.audio_details content block, a new optional field named waveform is added. It is an array of (non-floating) numbers to represent the amplitude of the audio over time. Because floating point numbers are not allowed in Matrix events, integers should be between 0 and 256, inclusive. Though there is no limit to the number of entries in the array, senders should aim to have at least 30 and not more than 120.

We have reached peak webshit, ladies and gentlemen, because there are no applicable alternatives to encoding PCM into JSON arrays. None. Audiophiles can cope and seethe.

I will leave you with a link to this glorious proposal.

https://github.com/matrix-org/matrix-spec-proposals/blob/c4209de8c593a7880842bbe02a3da8b2938ab0d2/proposals/3246-audio-waveform.md

Nostr event nevent1qqsqpy4su0ykcssrvsyrncwl82k4c6fytqhlcc4sf4hfae9ex37tjjczyzhhcl3cfer4n0rtz9p4awve9laa6kckkg5pshjldz6yph9sgrdl2yafput

2026-06-09T22:32:31Z

I hate git submodules.

Daily reminder. We live in the world where most corporate ...

2026-05-22T15:04:39Z

Daily reminder.

We live in the world where most corporate produced pieces of software are indistinguishable from malware from 20 years ago.

Nostr event nevent1qqspepq5enju379rxcenr6cnqz6e8cyghgw4nxph8jw2w5ett8z96kszyzhhcl3cfer4n0rtz9p4awve9laa6kckkg5pshjldz6yph9sgrdl2h6n7v8

2026-05-14T14:01:56Z

From https://web.archive.org/web/20260514102012/https://thanosapollo.org/posts/bzr-saga/

This is the story in which :rms: forced the use of Bazaar for Emacs development over git for four years. Not because it was better–it was not. Not because it was freer–git was under the GPL from the start. But because it was a GNU project, as in the GNU foundation held the rights over code. Though, at the same time bzr was only developed by Canonical and not by GNU.

Four years later, Canonical nuked bzr development and suddenly nobody from GNU expressed the desire to pick it up and continue. And then Emacs switched to git.

This tale describes the core of GNU mindset. They will use worse tools developed by someone else, while being incapable of writing a similar thing from scratch or maintaining an existing code base, solely for weird political reasons proclaimed by their deranged cult leader. And then they will happily pat themselves on their backs, while their tooling continues being abysmal piss-flavoured rotting garbage. Until it all breaks apart and then they happily switch to something better, designed and handled by much more competent developers.

On a related note. Have you ever perceived medieval scholars ...

2026-01-14T12:00:49Z

In reply to nevent1q…qxxf
_________________________

On a related note. Have you ever perceived medieval scholars arguing about the exact number of angels on a needle as a completely nonsensical and ridiculous farce?

Now consider this. There are real people in this world arguing whether NULL - NULL should be equal to zero in C. Because right now, as the standard stands, this is apparently not the case. And yes, these people are paid to do this.

https://developers.redhat.com/articles/2024/12/11/making-memcpynull-null-0-well-defined#null_pointer_arithmetic

If there was ever any hope in you that the original C and UNIX ...

2026-01-14T11:31:32Z

If there was ever any hope in you that the original C and UNIX authors were the wise ones, the recovery of UNIX v4 code must have shattered it.

https://sigma-star.at/blog/2025/12/unix-v4-buffer-overflow/

They are an even worse bunch of fuckups that the todays C monkeys who can't write reliable C code for shit. The entire ecosystem was designed with "just throw it at the wall and use what sticks" mentality in mind. These people aren't heroes, they are a disgrace of computer programming industry, and the fact that so many worship them makes the situation even worse.

maybe. But hiding behind a MITM proxy isn't the best of ...

2025-09-29T15:31:36Z

In reply to nevent1q…87mf
_________________________

maybe. But hiding behind a MITM proxy isn't the best of solutions.

no, it doesn't. Almost 20% of all websites use Cloudflare, ...

2025-09-29T15:29:34Z

In reply to nevent1q…dzlk
_________________________

no, it doesn't. Almost 20% of all websites use Cloudflare, for one. And the number is growing.

https://w3techs.com/technologies/details/cn-cloudflare

doesn't really matter how many clicks. The only way to ...

2025-09-29T15:25:44Z

In reply to nevent1q…rq88
_________________________

doesn't really matter how many clicks. The only way to re-decentralise the internet with BSD is to put an OpenBSD CD into a nuclear missile and point it at Silicon Valley.

> RE-DECENTRALISING THE INTERNET WITH BSD > So... > ...

2025-09-29T15:03:48Z

> RE-DECENTRALISING THE INTERNET WITH BSD
> So...
> Ugh...
> Google and Facebook are evil
> But...
> You can set up your own mail, irc, and git servers on a $5/mo VPS with BSD

I just can't with these people...

https://events.eurobsdcon.org/media/2025/submissions/GBNXDL/resources/slides_3CWFtbd.pdf

Are there any internet forums focused on tech and programming ...

2025-08-29T11:57:50Z

Are there any internet forums focused on tech and programming left alive? I'm not talking about social networks with followers and like or karma-based garbage like Reddit and StackOverflow.

the compiler was entirely correct here. It's the code in ...

2025-02-23T10:36:13Z

In reply to nevent1q…j2r9
_________________________

the compiler was entirely correct here. It's the code in question that was broken.

Some day C-tards will learn to include headers directly instead ...

2025-02-12T11:43:55Z

Some day C-tards will learn to include headers directly instead of relying on transitive inclusions. Some day...

> It starts with JavaScript, or, I suppose, the lack thereof. ...

2024-12-01T23:20:27Z

> It starts with JavaScript, or, I suppose, the lack thereof. The community has become rather insistent that the language is to be avoided at all costs, and for good reason. By nature, it is vulnerability prone, and it provides many avenues for fingerprinting, despite Tor’s best efforts to prevent it. The website must be fully featured without a single line of JavaScript, which leads to some creative programming.

The only acceptable solution to the JavaScript problem is...

TOTAL JAVASCRIPT DEATH

Source: https://boehs.org/node/dark-web-security

and calling me a windows user is? I have no clue what parts of ...

2024-09-24T13:37:04Z

In reply to nevent1q…3fzq
_________________________

and calling me a windows user is? I have no clue what parts of ethereal astral world you even got that idea from.

I certainly do admire Windows from the technical point of view - NT is a much better designed kernel than whatever Eunuchs/L'Eunuchs offer today, mostly due to sheer hindsight and ability to learn from Eunuchs mistakes (did you know that Microsoft with their Xenix was a successful Eunuchs vendor even before they released DOS?) - but I don't use it.

so salty... ...

2024-09-24T13:33:33Z

In reply to nevent1q…df45
_________________________

so salty...

it was never even alive to begin with

2024-09-24T13:30:32Z

In reply to nevent1q…xe6m
_________________________

it was never even alive to begin with

Eunuchs is dead

2024-09-24T13:27:31Z

In reply to nevent1q…w3zu
_________________________

Eunuchs is dead

Just a gentle reminder. All this modern IDE tooling that ...

2024-09-24T12:37:01Z

Just a gentle reminder. All this modern IDE tooling that you've grown to rely upon makes you an objectively worse programmer.