XMPP had rather bad name. Well-known design issues causing message losses, fractioned ecosystem due to varying implementation of extensions, unsuitability for mobile clients, absence of synchronization between clients, absence of end-to-end encryption. Most of these issues were (much) later fixed by extensions, but Matrix (or Signal for those who do not require federated one) was already there, offering E2EE by default.
Even today, E2EE in XMPP is rather inconvenient compared to Matrix due to absence of chain-of-trust in key management.
Sometimes I wonder if the endgame is each person having their own XMPP server for their set of devices. S2S is your E2EE then. Your chain of trust is your existing CA, unlike Matrix which starts from scratch. Cause XMPP wasn't designed from the start for clients not to trust servers, plus the fragmentation of C2S extensions was always a pain.
It's not a bad solution if someone can make it easy, even if it's a managed service that just lets tech-savvy users export it to self-hosting if they want.
One could argue that answers given by LLMs make sense. By assuming reasonability of the asking side, the answering side could assume that both options are possible and use abductive reasoning to conclude that the car to wash is already at the car wash station (and the question is about using another car to drive there).
600K request per day is ~ 400/minute. That is very low number. But seems to me that many webapps are so bad that even that small number causes significant load for them.
Most of NIMBY legislature and processes that block private construction also block public construction. So most YIMBY arguments to improve the situation apply to both public and private constructions. (Not to mention that public construction has a plenty of problems specific to it.)
Back around 1993-94 was a genuine gold rush in terms of domain names and network numbers.
My supervisor one day rushed into the bullpen and proclaimed that he had registered SEX.ORG, and presumably the only reason was to squat it awhile and then resell it for thousands. [Squatting and speculation were, in fact, quite legal and wise moves at that point in history, especially with a high-demand 6-character site!]
Personally, I discovered the registration process and forms for domain names and network numbers were fairly straightforward. I had seen a Usenet post where someone explained that you just had to write a description of your company, its structure and annual meetings, finances, etc. So I completely made up a fictional company and described those things in my application.
Hey presto, I was now the "owner" and "admin" of cthulhu.com and a corresponding 192.0.0.0/3 Class-C network. Now my coworkers at the ISP were savvy enough to arrange for the DNS servers to answer for their vanity domains. But having no appreciable homelab, or BGP peering of my own, my DNS domain and Class-C Network both languished, until ultimately they were reclaimed in a sweep of unused space by IANA and InterNIC.
I have been unable to recall the exact numbers or find them in a search, but I know that its moniker was related, such as "CTHULHU-NET" or something.
I went on to legitimately register under the .ca.us domain on behalf of my home network and my roommates. cthulhu.com has long been handed over to someone who uses it.
I had named it "HEARTLAND" rather than a Cthulhu-related name, which was hindering my searches. I had also asked Gemini and it hallucinated a historical record which it was unable/unwilling to link.
My original assigned user handle was: RE229 (a prime number, very on-brand)
My Netcom email address and a San Jose phone number are enshrined in the record. Don't bother contacting me through those! Interestingly, if you spell out the phone number, it ends in "NET", but does not spell anything compelling in its entirety.
This is great. You still "own" it, as it still exists in "whois" and ARIN records! The problem is it is assigned to an email address you no longer have access to. You might need to contact ARIN to get back control of it... seems possible since it's in your name and not a company.
That is baffling. I swear that I heard or received a direct communiqué, many many years ago, that ARIN and ICANN and IANA were sweeping out all of the unannounced networks and reclaiming them. Word went out during the initial pressure of address space exhaustion.
So if this is really still assigned to "me" as boss of "my company" I suppose nobody else has ever announced it. It has no BGP behind it. In fact, 192.160.0.0/16 has no BGP at all. That is a huge swath of space to be vacant.
So, in 34 years since its registration, no BGP announcement, no ASN has ever been associated with this Class-C, and it still "belongs to me", unpaid, un-rented? It boggles my mind. I had expected that it was easier to lose an unused IPv4 network than to lose a domain name from back in the day.
Now there is a lot of crazy contact information that is so, so old. It is credible but I barely recall even having some of those phone numbers. There's an email address at cts.com. Which appears to be utterly defunct now but it was a San Diego bboard run by "Bill Blue". I distinctly recall a lot of Usenet posters using "crash.cts.com" and it was a "shell account" provider. It would've been in-character for past-me to sign up there at some point. Some point, I don't know.
So it says they last attempted to contact me 16 years ago. Did they send a letter in the USPS? Did my family receive nothing? So weird. If I literally wrote to them with that return address, would they validate me?
I literally have no idea how I could even use a /24 network today. My ISP wouldn't accept it. I can't exactly run BGP from a Chromebook or Netgear router at home! I suppose the only way to use it would be on a VPS service? Would the VPS announce an old-school "personal" network?
When ARIN contacts you, it's through email. They send an email with a confirmation link, so that probably bounced.
You can definitely announce BGP from a VPS with some providers. I have been doing this for years. Vultr will do it. However, they will validate (through email) that you "own" the block.
My recommendation is you first contact ARIN and see if you can "recover" contact info associated with the class C.
There are some restrictions around legacy blocks that predate the existing of ARIN. For some reason, they cannot reclaim them easily. So they just sit there...
There is no requirement for compilers to be deterministic. The requirement is that a compiler produces something that is valid interpretation of the program according to the language specification, but unspecified details (like specific ordering of instructions in the resulting code) could in principle be chosen nondeterministically and be different in separate executions of the compiler.
We are not talking about deterministic instructions, we are talking about deterministic behavior.
UB is actually a big deal, and is avoided for a reason.
I couldn't in my life guess what CC would do in response to "implement login form".
For all I know CC's response could depend on time of day or Anthropic's electricity bill last month more, than on the existing code in my app, and the specific wording I use.
I do not talk about UBs, compiler can be non-deterministic even for completely valid and well-specified source code.
Seems to me that people just confuse determinism with soundness.
Determinism is just whether the output of the tool (machine code) is determined by the input (source code), i.e., for one input always returns the same output.
Soundness means that output is always consistent with the semantics of the input. i.e. returned machine code always does what is specified by the source code.
Compilers can be non-deterministic (although usually are deterministic), but they are always sound (ignoring compiler bugs).
LLMs can be deterministic (although usually aren't), but they are not sound in general.
It seems to me that it is much more symmetric. In both situations one side guess, due to fundamental uncertainties of human interactions -- each side knows their subjective costs/benefits, but not costs/benefits of the other side.
For 'guesser protocol', the initiatior guess whether the ask is appropriate (say initiator_known _benefit > responder_guessed_cost), while in 'asker protocol', the task of guessing is shifted to responder, as the responder has to guess whether the reject is appropriate (say responder_known_cost > initiator_guessed_benefit).
Even today, E2EE in XMPP is rather inconvenient compared to Matrix due to absence of chain-of-trust in key management.
reply