Follow

@owl This is one of my utterly accursed, downright nasty opinions, but:

The modern web is unimplementable from scratch. And the chance we get locked into a single implementation permanently is very real. That should be grounds for treating the web as an illegitimate technology.

It's not possible to build simple software that pleases billions of people simultaneously. But if the web is an inevitable consequence of doing so, maybe we shouldn't please billions at all.

· · Web · 3 · 3 · 7

@cr1901 large scale expansion is bad actually and this dream tech ppl always seem to have of "winning" and becoming ubiquitous can get in the bin

the internet was a great idea and i love that we can all connect to each other

the web is bad

@alexisvl @cr1901 I think links were okay but forms are sus, maybe the web peaked with gopher

@scanlime @alexisvl I consider what I said "downright nasty" to the extent that it questions the narrative that "empowering the users is paramount at the cost of everything else".

I think "user empowerment at the cost of everything else, even if it makes software impossible to grok" is at least partially a backlash against shitty Unix RTFM culture. RTFM culture is/was wrong too.

I'm also bitter b/c "wire protocol by committee (HTTP/2)" is the reason I can't make old computers do new things.

@scanlime @alexisvl (No, I'm not bitter as fuck that Gemini requires TLS, why do you ask?!)

@cr1901 @scanlime ewww the infosec bros got to that too? i was just idly poking around at gemini to see if it was actually cool

@alexisvl @scanlime Hmmm, just realized this may not be accessible to you. My toot says:

"[Gemini's answer to] 2.14 [in their FAQ] is Not Good Enough. Gopher is already seen as dead, and unless they explicitly encourage Gopher in tandem w/ Gemini, Gopher will be killed off by Gemini."

@alexisvl @scanlime

>But it also makes no sense for the vast majority of internet users to sacrifice any and all privacy protection to facilitate this.

F*** off with this attitude, you could've added a non-TLS but you didn't.

@cr1901 haha the way this came up in my notifs i thought you were telling me to fuck off 😬

@alexisvl @scanlime Response from the person I was chatting to (I have permission to post w/ name redacted), who sums it up better than me:

>>"In fact, people serving content via Gopher right now are strongly encouraged to start also serving that same content via Gemini simultaneously."
>I like how they state this as if it's something that can be widely expected to continue rather than "hey look at all the people we got to move their shit over already"

@cr1901 I dunno, I'm not sure it *does* question that.

I'm feel pretty strongly that software _should_ empower users - it's a tool, screwdrivers don't tell me what I can do with them and neither should my computer

I think that's very very separate from whether it should aim to empower as many users as possible to do as much as possible at the cost of doing what's right, vs just being a reasonable tool for the people who need it

"Do what I say" is different from "do anything anybody says"

@cr1901 @scanlime @alexisvl Is the last part because cryptography requires a lot of calculations?

@x44203 @scanlime @alexisvl Well, a fast multiplier more than anything. And once we get past "no single cycle multiplier", you have to deal with "constant time multiply didn't exist back then".

There's probably hardware accelerators you can make. I want to make one that attaches to the parallel port that isn't "just" a Pi doing all the heavy lifting and making old iron look bad :D!

@cr1901 @x44203 @alexisvl this reminds me of a chip I used some years ago that had a Galois field multiplier peripheral to make AES practical on their 8051 core

Sign in to participate in the conversation
Mastodon

The original server operated by the Mastodon gGmbH non-profit