News for hackers. Stuff that matters.

keiyakins

Re:Tell HN: IPv6-only still pretty much unusable on Saturday December 10, 2022 @01:45PM (#33935440)
There's only a very few places where typing IP addresses actually makes sense. Configuring DNS is the main one. Trying to isolate problems is another - if you can ping 8.8.8.8 but not google.com you can reasonably infer that the problem is DNS. I really can't think of any others.

That latter one is admittedly kind of a pain, but a wallet-sized cheat sheet can solve it for you. Might be a good idea to try to convince vendors to include a reliable entry or two in the hosts file for that purpose? You can always add it yourself for now, when you're on your own workstation.

Re:Tell HN: IPv6-only still pretty much unusable on Saturday December 10, 2022 @01:36PM (#33935340)
How would changes not have been required in the ISPs? An IPv4 router wouldn't know what to do with a 4.4 packet. At best, it'd route it to the wrong place - 1.2.3.4.5.6.7.8 and 5.6.7.8 are totally different hosts that may well not even be on the same continent.

Additionally, the only reason so much code had to change for ipv6 is that Berkley sockets is a terrible, terrible API that has abstractions so leaky they might as well not exist. Sure, in other APIs (what few exist) low-level code had to be rewritten somewhat, but that's going to be true for any protocol change, because that's kinda what change means.

Re:Tell HN: IPv6-only still pretty much unusable on Saturday December 10, 2022 @01:25PM (#33935183)
Yeah, it's not always, but for most home networks it probably is. Ideally they'd give you a /56 or even a /48, but giving those on request and a /60 by default is fine.
It probably was, seeing as Plan 9 was named after a movie.
This article is kinda garbage, it gets simple facts wrong in the first sentence of the abstract (and again in the second sentence of the article): Plan 9 is not Unix. At all. It shares some ideas with it sure, but if that's enough to make something Unix so is Windows NT. It's also rather debatable that it "failed" - a lot of good research happened there and those ideas have gone on to influence production environments the world over.
Re:Tell HN: IPv6-only still pretty much unusable on Saturday December 10, 2022 @11:44AM (#33934063)
Not if you want SLAAC to work, no.
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @02:52PM (#33925728)
How are they blocking? You're just behind a NAT64, which is no worse than NAT44, with the bonus of having actual connectivity on the IPv6 side.
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @02:50PM (#33925717)
Even? /60 should be the minimum, even home users usually have a couple subnets (guest networks for example, sometimes one for the router's WAN link) and you want the boundary to be on a nybble boundary
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @02:30PM (#33925524)
IPv6 was designed in the late 80s/early 90s, before home users had internet access. At the time there was serious debate over switching to the OSI stack instead. You're showing your lack of history knowledge more than anything.
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @02:15PM (#33925330)
Why are you using IP addresses directly? That's what DNS is for. There's even mDNS for in simple home networks.

(The DHCP thing is valid but that's one single android dev being an ass who can't read deciding that it imposes limitations it doesn't and ruining it for the rest of us)

Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @01:57PM (#33925121)
NAT is neither a privacy nor security feature, what are you talking about? Have you actually tested what your CPE does when it gets packets addressed to internal IPs from the WAN port? Almost every time, the answer is just pass it on to the target host. Thinking NAT is a security feature makes your network MUCH less secure.

As for privacy - you can fingerprint individual devices pretty trivially, and with privacy extensions for SLAAC you can only tell what /64 network it's coming from, which is no more information than IPv4 (unless you're behind CGNAT, but frankly being behind a 4-to-4 CGNAT shouldn't count as internet access because you literally can't get incoming connections.)

Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @01:28PM (#33924750)
Because on the wire it's encoded as four bytes. If you can make eight binary digits count up to 999, you can do a lot more than just make IPv4 last longer.
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @01:25PM (#33924700)
Indeed, your system would require NAT4.44 as a transition mechanism, just like NAT64 is needed now. It gets no benefit over IPv6, and none of the other benefits like SLAAC.

So, what's the point? It's no easier to migrate to, and once we're migrated is worse.

Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @01:19PM (#33924613)
it's ::ffff:1.2.3.4, so that loopback can be ::1. But yeah same idea.
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @12:37PM (#33924097)
so somehow 0.0.0.0.8.8.8.8 is an extension of the legacy address 8.8.8.8, and ::ffff:8.8.8.8 isn't?
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @11:59AM (#33923605)
I suspect it's to make business plans artificially more appealing. After all, why offer a better service when instead you can just make your cheaper one worse?
Re:Tell HN: IPv6-only still pretty much unusable on Friday December 09, 2022 @11:51AM (#33923488)
Why do you care how long the addresses are? That's what DNS is for. Within a link - most home networks are only one, and those are the ones that need to be simplest - there's even mDNS.
Can we at least require you not to call yourself 'whatever news' if you're not at least trying to present factual information?
On what grounds? Caveat emptor was the rule for a long, long time. It was only later when regulation and standards started coming into play that it was a thing. The courts can't just declare a legal thing to be bad.
Or it's checking the monotonic time, not using the synchronized time (which is probably recorded as an offset anyway)
I'm comparing the lowest models that meet my needs. Why should I have to pretend there's not cheaper, less powerful handsets that can fulfill my needs just to make Apple look better?

(Granted, my strict needs could be met with one of TI's sub-$0.50 microcontrollers and a seven-segment display since they're "make phone calls" and "generate two-factor authentication codes")

So? The entire promise of computers is programability. If you can't make it do what you want it to do, you might as well just get an old nokia.
Apple is awful at providing a reasonable spread of devices though. There's no $50 iphone that does everything I need, I'd have to spend at least ten times that. Then I'd have to pay MORE to be able to develop and run my own programs?! At that point, why even have a microprocessor rather than using a simple fixed function ASIC?

Would I be happier if I could get a phone with the equivalent of BIOS and nothing else and vet and install my own software? You bet. If nothing else, I'd not have to install all those Play ____ apps I never use. But given the choices available, Apple is a bad one.

So, you want to put that much power into Apple's hands? How do you expect them to remain on your side in that case?
That's why we never had to do anything to address rampant fraud, cutting food with dangerous materials, straight up lying about what you're selling, con artists, snake oil salesmen, or irresponsible management of hazardous waste, right?
Re:Who controls glibc? on Tuesday May 08, 2018 @02:00PM (#17023068)
> This comment was document of its time, several decades ago. Software has history. Everything humans do has history.

So, should we never change documentation once it's written? When the technical details change, should the documentation become inaccurate because it's a record of the time? When the political and linguistic context changes and makes a joke cause confusion rather than laughs should we continue including it in new versions?

No, of course not. We should fix it. History is indeed important, that's why version control was invented. Access to the old versions isn't going anywhere.

Re:Who controls glibc? on Tuesday May 08, 2018 @01:46PM (#17022919)
Fuck 'em. I'm not going to bend over backwards to coddle people who hate me for daring to exist. As for laws, I think the late Dr. Martin Luther King Jr. said it best: "I would be the first to advocate obeying just laws. One has not only a legal but a moral responsibility to obey just laws. Conversely, one has a moral responsibility to disobey unjust laws."
Re:Who controls glibc? on Tuesday May 08, 2018 @01:38PM (#17022827)
If he was dead serious, he's an idiot. There was never any proposal to ban all usage of the word 'abort' in the sense of 'bring to a premature end because of a problem or fault' or as a noun meaning 'an act of aborting a flight, space mission, or other enterprise'.
Re:Who controls glibc? on Tuesday May 08, 2018 @01:17PM (#17022656)
It's never helped anyone understand the technical material and it's caused at least one instance of confusion (https://stackoverflow.com/questions/48445031/why-would-it-be...), that's reason enough. Jokes in documentation are good, but only if they serve the documentation.
Re:Who controls glibc? on Tuesday May 08, 2018 @01:09PM (#17022574)
I think you're looking at the wrong axis. This isn't an order-vs-chaos question, it's one of constructive vs destructive. This particular joke makes the manual worse, not better, at being a manual. Cut it and move on, and always keep an eye out for places it could be improved, whether that be through better descriptive prose or a good, understanding-building joke.

Older »