I loved Pegasus. Specifically because to move it to another machine you just had to copy the PMAIL folder and make a shortcut. No registry awareness, no dependencies.
I stopped reading once the author claimed it was a lie because the SecSrv knew technical terms, then claimed it was a lie because they didn't know the technical terms. It's too early in the morning to be purposely confused.
In Mass Effect, there is a distinction made between AI (which is smart enough to be considered a person) and VI (virtual intelligence, basically a dumb conversational UI over some information service).
What we have built in terms of LLMs barely qualifies as a VI, and not a particularly reliable one. I think we should begin treating and designing them as such, emphasizing responding to queries and carrying out commands accurately over friendliness. (The "friendly" in "user-friendly" has done too much anthropomorphization work. User-friendly non-AI software makes user choices, and the results of such choices, clear and responds unambiguously to commands.)
A bit of a retcon but the TNG computer also runs the holodeck and all the characters within it. There's some bootleg RP fine tune powering that I tell you hwat.
I mean it depends on what you consider the "computer", the pile of compute and storage the ship has in that core that got stolen on that one Voyager episode, or the ML model that runs on it to serve as the ship's assistant.
I think it's more believable that the holodeck is ran from separate models that just run inference on the same compute and the ship AI just spins up the containers, it's not literally the ship AI doing that acting itself. Otherwise I have... questions on why starfleet added that functionality beforehand lol.
I helped interview a guy recently who was (a) ESL, and (b) Typing in our questions and reading the answers at a breakneck pace to the point where it was almost pathological; we could not get a word in. In unspoken horror and synchronization, we got through the formal part of the interview as quickly as we possibly could.
I feel this in my soul. I work in higher education, and every major contribution I've made has been ripped from my hands and either dashed like the first copy of the ten commandments or handed over to someone shinier. I'm still proud of all I've done.
I have always thought the infinite proliferation of TLDs was a stupid idea. I'd be enlightened if I could think of one scenario that benefits from it outside of the registrars.
There are lots of people called John Smith. They all want a domain name. There's only so many variations of jsmith, j-smith, etc you can squeeze into .com, .net, and a few others.
Why shouldn't they be able to buy a domain name which contains their name?
Is it useful to be able to differentiate between McDonald's the restaurant and McDonald's the legal firm and McDonald's garage?
Why shouldn't each of those industries get their own TLD?
The original list of TLDs aren't some platonic good written by ineffable sages. It's OK for things to change.
>There are lots of people called John Smith. They all want a domain name. There's only so many variations of jsmith, j-smith, etc you can squeeze into .com, .net, and a few others.
>Why shouldn't they be able to buy a domain name which contains their name?
I fail to see how johnsmith[insert number here].com is any worse than johnsmith.[insert TLD here]. If anything a number is less likely to get mixed up than tlds, which have confusing pairs like ".tech" and ".technology", or ".engineer" and ".engineering".
It's not about typos or mishearing stuff, it's about words being jumbled in memory. Unlike a sequence of digits, people don't store words "engineering" in their head as a string (eg. "E-N-G-I-N-E-E-R-I-N-G"). It's stored as something like "[concept of engineer] + [present participle]". That's far more likely to get jumbled in people's head during recall.
The only actual answer would have been to drain the TLD swamp and open up the root zone. Give us john.smith, website.jsmith, and mc.donalds. It's just a label anyway, and one that normies don't pay any attention to—save that even if they did, it's hard not to fall for mc-donalds.com or mcdonalds-restaurant.com anyway.
If the whole EV certificates thing would have been set up in a way that it wasn't just a money extraction racket, that would be the way forward. Let user agents convey whether a site is trusthworthy, and what entity it is connected to.
Domains are the ultimate identity system for building a more trustworthy internet without handing over control to some kind of verified ID scheme or being forced into publishing your personal details to gain credibility.
You can build reputation and trust using a handle, even if it's not associated with your real world identity. For example, I know that if 'ryao' replies to a question about ZFS, the response can be considered trustworthy. I don't know who that is or even what country they live in, but I know they're a contributor that isn't speculating or guessing when they reply and that's all that matters to me.
Domains can be used as verifiable, globally unique handles which simplifies things for the average user because it makes it easier to help users avoid impersonation and confusion if you can point them to something simple and verifiable. For example, look at Bluesky [1].
I've been wanting domain based namespaces and handles for a solid 5 years because it just makes sense. Here's my oldest mention of it (asking why package managers don't use domain verified namespacing) I have on HN [2]:
> It seems like a waste to me when I'm required to register a new identity for every package manager when I already have a globally unique, extremely valuable (to me), highly brandable identity that costs $8 / year to maintain.
You can tell it's old because .com domains only costed $8 back then. IMHO, domain based handles are the #1 reason to use Bluesky over X/Twitter. People used to spend $10-15k buying "noteworthiness" via fake articles, etc. to get verified on Twitter. I can't find any links because search results are saturated with talk of X wanting $1000 per month for organization validation (aka a gold check mark). Domain validation is just as good as that kind of organization validation, at least for well known individuals and organizations.
Given that, I think there would be a bigger market for domains if domain validated identities catch on. It could even spawn specialty gTLDs that do extra identity or notoriety checks (if that's allowed) or maybe attestations would become a big thing if there were an easy way to do them against a domain verified handle.
I would actually think that consumers (of domains) benefit more than the registrars, because there is more competition. If I want a specific word as a domain, there are multiple options of TLD for me.
DNS should be a destination, not a utility. Every John Smith has a legitimate claim on smith.com.
DNS should offer disambiguation services. Instead, we have this awful system.
My dream is to fork a browser and replace the DNS component with an entirely new protocol that respects the notion that people in the real world share names.
have you gone through the process of naming and securing domains for startups over and over again because let me tell you, it's brutal. the more TLDs, the better.
I still remember looking at these with my mother when I was small. We loved to hunt for Gold Bug. She told me later in life that her favorite was, "Lowly worm washes his face and foot." (He always wore a sock and shoe on the end of his tail.)
