|
Lunar Suite posted:they were comparing the configuration files for the test environment and live environment. obviously this is a lot worse than using an actual diff tool, but it's honestly a pretty clever solution for a nontechnical user doing the best with the tools they know how to use
|
# ? Oct 19, 2023 19:02 |
|
|
# ? May 27, 2024 21:34 |
|
akadajet posted:I mean. it’s happened before for amd64. and windows has been built for different cpus for a long time now jumping to a new thing isn't the same as not requiring support for the old thing
|
# ? Oct 19, 2023 19:07 |
|
assuming one has zero build pipeline, how do devs safely sign code with the org cert
|
# ? Oct 19, 2023 19:21 |
|
Captain Foo posted:assuming one has zero build pipeline, how do devs safely sign code with the org cert I don't want you to research. I don't want you to try things. I don't want you to write to your architect, because I wouldn't know what to tell you to write. I don't know what to do about the deployment and the integration and the contractors and the bugs in the code. All I know is that first, you've got to get mad. You've gotta say, "I'm a programmer, goddammit! My life has value!" So, I want you to get up now. I want all of you to get up out of your aerons. I want you to get up right now and go to the window, open it, and stick your head out and yell, "Very Carefully!"
|
# ? Oct 19, 2023 19:26 |
|
Plorkyeran posted:obviously this is a lot worse than using an actual diff tool, but it's honestly a pretty clever solution for a nontechnical user doing the best with the tools they know how to use it is a diff tool, honestly. I like the moxie
|
# ? Oct 19, 2023 20:06 |
|
NihilCredo posted:I've seen that joke suggested in earnestness in the past (in the cosmetically different form of adding more byte blocks instead of widening the existing four) ty for explaining how how having wider bytes in the middle of the packet may have issues rather than be a genius protocol evolution
|
# ? Oct 19, 2023 20:17 |
|
what if you kept 32 bits but each bit could be 0, 1, or 2
|
# ? Oct 19, 2023 20:31 |
|
the addresses get too big that way. better to use 0, 1, and 1.5
|
# ? Oct 19, 2023 20:32 |
|
Athas posted:what if you kept 32 bits but each bit could be 0, 1, or 2 that'd be tits
|
# ? Oct 19, 2023 20:33 |
|
doesn't the packet header have a version field? i don't see how that would matter, you would just design it as a protocol that communicates over superficially similar looking packets routing devices would need an ipWhatever routing table. it doesn't have to intersect with the ip4 routing table
|
# ? Oct 19, 2023 20:33 |
|
if we used strings instead of bytes we could have emoji IPs which would be fun imo
|
# ? Oct 19, 2023 20:35 |
|
rotor posted:if we used strings instead of bytes we could have emoji IPs which would be fun imo you shouldn’t say such curses aloud, lest a trickster demon hear you and make it true
|
# ? Oct 19, 2023 20:47 |
|
used to be you could just use an integer as an IP address with some parsing stacks if I recall correctly. an emoji to represent a 32-bit quantity isn’t the craziest thing, but would conflict with punycode crap I suspect
|
# ? Oct 19, 2023 20:51 |
|
brb selling musk whatever ip 🤣 maps to
|
# ? Oct 19, 2023 20:52 |
|
even at their current clip it will take the unicode consortium a while to allocate 4 billion codepoints to idiotic emoji bullshit
|
# ? Oct 19, 2023 20:54 |
|
Subjunctive posted:used to be you could just use an integer as an IP address with some parsing stacks if I recall correctly. if by "used to be" you mean "today" and "some parsing stacks" you mean "all RFC-compliant parsing stacks" then yes the semantics of an ipv4 address are "1-4 dot-separated integers. all integers except the last are 8 bits. the last integer is however many bits are needed to finish off the 32 bit address". the integers do not have to be in base ten; hex and octal are also acceptable so, all of these are the same ipv4 address: (you can trivially verify this with ping or whatever on any OS)
they should absolutely amend the standard to allow the use of emojis
|
# ? Oct 19, 2023 22:49 |
|
yeah but not all parsing stacks are RFC compliant, because the behaviour is confusing and pointless in modern times. try a single integer as an address in the URL bar of a bunch of browsers
|
# ? Oct 19, 2023 22:55 |
|
Lunar Suite posted:when i asked one of the vendors programmers if the software is 2037 compliant they said "i hope to be retired by then" got an audible, hearty lol from me
|
# ? Oct 19, 2023 22:56 |
|
oh god is something happening in 2037 too?
|
# ? Oct 19, 2023 22:58 |
|
Subjunctive posted:yeah but not all parsing stacks are RFC compliant, because the behaviour is confusing and pointless in modern times. try a single integer as an address in the URL bar of a bunch of browsers browsers are pretty garbage but often still work if you just explicitly put http:// in front of it - most browsers will normalize to dotted-quad-of-base-ten-bytes before they do anything else, but they will nonetheless parse the goofy representations
|
# ? Oct 19, 2023 23:02 |
|
ntp craps out on feb 7, 2036 (ntp4 fixes this), don't know what's going on in 2037
|
# ? Oct 19, 2023 23:04 |
|
Subjunctive posted:oh god is something happening in 2037 too? in 2037 our promo packets finally go through and we're transferred into a critical path service team in mid-January of the following year
|
# ? Oct 19, 2023 23:07 |
|
Lunar Suite posted:when i asked one of the vendors programmers if the software is 2037 compliant they said "i hope to be retired by then" relateable as gently caress imo
|
# ? Oct 19, 2023 23:10 |
|
Subjunctive posted:oh god is something happening in 2037 too? anybody still using 32 bit signed ints for Unix timestamps breaks in 2038. maybe that one?
|
# ? Oct 19, 2023 23:21 |
|
ShoulderDaemon posted:if by "used to be" you mean "today" and "some parsing stacks" you mean "all RFC-compliant parsing stacks" then yes
|
# ? Oct 19, 2023 23:23 |
|
ShoulderDaemon posted:browsers are pretty garbage but often still work if you just explicitly put [url]http://[/url] in front of it - most browsers will normalize to dotted-quad-of-base-ten-bytes before they do anything else, but they will nonetheless parse the goofy representations I have not-infallible recollections of being yelled at by various losers when we stopped parsing all the non-dotted-quad IPv4 representations in Firefox (which would have been ~15 years ago) and I can’t imagine what would have convinced me to change my mind but it does work today on Windows at least. on iOS Firefox it behaves differently but I’m not quite sure what’s going on there. maybe I’ll remember better after a nap, or I’ll think of who to ask Chrome and Edge search for the string "http://12341234/" which is more what I expected to see these days how do those representations work with certificates, now that you can get them issued for IP addresses such as for DoH?
|
# ? Oct 19, 2023 23:24 |
|
Subjunctive posted:I have not-infallible recollections of being yelled at by various losers when we stopped parsing all the non-dotted-quad IPv4 representations in Firefox (which would have been ~15 years ago) and I can’t imagine what would have convinced me to change my mind but it does work today on Windows at least. on iOS Firefox it behaves differently but I’m not quite sure what’s going on there. maybe I’ll remember better after a nap, or I’ll think of who to ask 69696969 in edge does a search, but http://69696969 is converted to http://4.39.125.201 also the reason ios firefox works differently is because ios firefox is safari
|
# ? Oct 20, 2023 00:01 |
|
http://12341234/ does a search in edge, probably because its not routable?
|
# ? Oct 20, 2023 00:02 |
|
Shaggar posted:69696969 in edge does a search, but http://69696969 is converted to http://4.39.125.201 yeah but it doesn’t match Safari behaviour
|
# ? Oct 20, 2023 00:09 |
|
Subjunctive posted:how do those representations work with certificates, now that you can get them issued for IP addresses such as for DoH? browsers will normalize to dotted-quad-of-base-ten-integers before checking the certificate names. boring answer, but also the only possible answer with any even remote degree of sanity ipv6 has to do the same thing - pointing a browser at https://[some-ipv6-address] will convert the ipv6 address to normal form before it checks certnames. there's also some rules about how browsers decide if the port number should be part of the certname which i can't remember with enough certainty to recite but i assume are relatively straightforward and obvious
|
# ? Oct 20, 2023 00:20 |
|
Sapozhnik posted:more accuracy is better, right? x[0] = -6 x[1] = 64 x[n] = 82 − (1824 − 6048/x[n-2]) / x[n-1] so this sequence converges to 36, under half and single floats under doubles, it converges to 42, which is the wrong answer https://etna.math.kent.edu/vol.52.2020/pp358-369.dir/pp358-369.pdf
|
# ? Oct 20, 2023 11:25 |
|
ooops, i did my bfloat16 wrong and the paper actually covers it
Cybernetic Vermin fucked around with this message at 11:39 on Oct 20, 2023 |
# ? Oct 20, 2023 11:35 |
|
rotor posted:brb selling musk whatever ip 🤣 maps to Internet Janitor posted:even at their current clip it will take the unicode consortium a while to allocate 4 billion codepoints to idiotic emoji bullshit What if code points were NFTs.
|
# ? Oct 20, 2023 13:14 |
|
cross-posting from the buttcoin thread:drk posted:lol, the details on J.Zhong's crimes are right out of a bad sci fi novel
|
# ? Oct 20, 2023 17:02 |
|
I have been informed by someone with better recollection than mine that Brendan overruled me on simplifying to requiring dotted-quad. so it goes
|
# ? Oct 20, 2023 17:20 |
|
truly his greatest sin of all
|
# ? Oct 20, 2023 17:45 |
|
tef posted:x[0] = -6 lmao serves you right for doing math with a float
|
# ? Oct 21, 2023 05:39 |
|
floating point is a mistake and anyone who says otherwise is gaslighting you
|
# ? Oct 21, 2023 05:51 |
|
good for video games tho
|
# ? Oct 21, 2023 05:52 |
|
|
# ? May 27, 2024 21:34 |
|
just imagine all the pannenkoek2012 videos we'd have missed out on without floating-point numbers
|
# ? Oct 21, 2023 06:28 |