#archiveteam-bs 2013-08-09,Fri

↑back Search

Time Nickname Message
01:58 🔗 ivan` omf_: I hope you have time to look at the drive soon-ish
01:58 🔗 ivan` wouldn't want to keep the lazy gamer nonarchivists waiting for their ut-files
02:05 🔗 dashcloud Baljem: at least IE lets you temporarily trust a certificate- I can never seem to get Firefox to trust a certificate just this time- I always have to make mark it trusted
02:10 🔗 joepie91 DFJustin: the problem is that it doesn't tell you that it's a DNS error.
02:13 🔗 DFJustin yeah but anyone who knows enough to benefit from that information would know that anyway
02:14 🔗 joepie91 ... no, not really
02:14 🔗 joepie91 that's the whole problem
02:14 🔗 joepie91 if my shit stops working suddenly
02:14 🔗 joepie91 it's quite helpful if the browser actually tells me what's wrong
02:15 🔗 joepie91 so I don't have to fuck around with a bazillion commandline tools to figure out something *that Chrome already knows but doesn't tell me*
02:15 🔗 DFJustin is there a human being on the planet who can use command line tools to troubleshoot who does not know that "can't find foo.com" is a dns resolution error
02:16 🔗 joepie91 DFJustin: where in this error message does it make any mention of "DNS"?
02:17 🔗 DFJustin it doesn't but what else could it be
02:17 🔗 joepie91 DFJustin: if it doesn't, how do you expect _anyone_ to know with any _certainty_ that it _is_ indeed a DNS error and not some fancy generic error message that Chrome developers decided to implement as a catch-all for assorted other problems?
02:18 🔗 DFJustin because "can't find foo.com" is standard language used for dns errors since the dawn of time
02:18 🔗 joepie91 this is like saying "who needs error messages in a compiler, it should just say 'there was a syntax error' without furher info and you can figure out the line number yourself, after all you know the language you're writing in right?"
02:18 🔗 DFJustin and if you weren't sure, 5 seconds with ping would sort it out
02:18 🔗 joepie91 DFJustin: what part of "there is no certainty that no other meaning has been assigned to it by browser developers" is unclear to you
02:18 🔗 joepie91 I am genuinely having a hard time understanding how this is not obvious to you
02:19 🔗 DFJustin you can't be certain, but there is common language - I would expect to see "can't find" for dns and "isn't responding" for exists but does not respond etc
02:19 🔗 joepie91 EXCEPT YOU CANNOT BE SURE
02:19 🔗 DFJustin now if they've in fact used the same catchall for everything then fine
02:19 🔗 joepie91 this is the crucial issue
02:19 🔗 joepie91 you cannot be sure that that is what it means
02:19 🔗 DFJustin even if it did say what was wrong you can't be sure it's not lying to you
02:19 🔗 joepie91 it does not give any explicit indication that that is the meaning
02:19 🔗 joepie91 there is absolutely no way to know whether the meaning of that message may have changed
02:19 🔗 DFJustin so I'm not sure that's a useful criterion
02:19 🔗 joepie91 this means that they may as well not have told you anything
02:19 🔗 joepie91 because you have to figure out shit for yourself anyway
02:20 🔗 joepie91 so you can confirm that indeed the meaning of it has not changed
02:20 🔗 joepie91 what part of this is unclear
02:20 🔗 joepie91 I don't care that "that is how it has always been worded"
02:20 🔗 joepie91 I care that it actively refuses to give you any more information
02:20 🔗 DFJustin well the thing is with dns specifically I'm having a hard time even thinking of a program that uses the word "dns" in the error
02:20 🔗 joepie91 thereby adding more workload for ME
02:20 🔗 joepie91 to figure out wtf is wrong
02:20 🔗 DFJustin firefox doesn't, unix command line tools usually don't
02:21 🔗 joepie91 eh, how about every former chrome version?
02:21 🔗 joepie91 where there was either a directly visible error or a "more info" button
02:21 🔗 joepie91 that told you it could not be resolved?
02:21 🔗 joepie91 curl: (6) Could not resolve host: randomdomainthatdoesntexist.com; Name or service not known
02:21 🔗 joepie91 sven@linux-rfa7:~> curl randomdomainthatdoesntexist.com
02:21 🔗 DFJustin sure it's dumb to take the more info button out
02:22 🔗 DFJustin but it's literally five seconds of extra work to troubleshoot so it's hard for me to muster more than a "meh"
02:22 🔗 joepie91 DFJustin: it isn't "five seconds of extrat work"
02:22 🔗 joepie91 extra *
02:22 🔗 DFJustin open terminal, ping host
02:22 🔗 joepie91 it's "however much extra work that is necessary to pinpoint the problem"
02:22 🔗 joepie91 ...
02:22 🔗 joepie91 seriously
02:22 🔗 DFJustin I don't see "dns" in that curl message
02:22 🔗 joepie91 I've tried to explain this several times now
02:22 🔗 joepie91 ..
02:22 🔗 joepie91 dude
02:22 🔗 joepie91 please
02:22 🔗 joepie91 > could not resolve host
02:23 🔗 joepie91 there is no possible meaning for "could not resolve host" beyond "DNS issue"
02:23 🔗 joepie91 it is almost literally a synonym
02:23 🔗 DFJustin yes
02:23 🔗 DFJustin and so is "can't find"
02:23 🔗 joepie91 ...no, it isn't
02:23 🔗 joepie91 jesus christ
02:23 🔗 DFJustin to everyone except you
02:23 🔗 joepie91 I get the idea that you don't WANT to understand this
02:23 🔗 joepie91 "can't find" is an extremely generic term
02:23 🔗 joepie91 can't find what? the domain? the server behind it? the page?
02:23 🔗 joepie91 it doesn't tell you what the error relates to
02:24 🔗 joepie91 "could not resolve host" does
02:24 🔗 DFJustin if the page can't be found the server serves you a 404 page
02:24 🔗 joepie91 asdfasdf
02:24 🔗 joepie91 okay
02:24 🔗 joepie91 DFJustin
02:24 🔗 joepie91 please
02:24 🔗 joepie91 be quiet
02:24 🔗 joepie91 for a moment
02:24 🔗 joepie91 and hear me out
02:25 🔗 joepie91 I am a web developer. I use browsers not just to visit websites, but as development tools. When something goes wrong, I need to know why it goes wrong, as quickly as possible. Even one or two extra steps, over the long term, will greatly harm my productivity.
02:25 🔗 joepie91 I am not a browser developer. I am not involved in the decisionmaking process of the Chrome development. I cannot know what has changed in how they deal with error pages until I find out in practice.
02:25 🔗 joepie91 I have absolutely no way of predicting whether any kind of error will be handled in the same manner in the future as it is now.
02:26 🔗 DFJustin that's true whether or not they visibly change the message though
02:26 🔗 joepie91 I told you to hear me out
02:26 🔗 joepie91 you are not letting me finish
02:26 🔗 joepie91 I have no way of telling whether a 404 error is STILL displayed as a page served by the HTTPd, or whether it has been replaced with a generic browser-supplied error page, as SEVERAL browsers have been known to do in the past and even know.
02:26 🔗 joepie91 now *
02:27 🔗 joepie91 An error message is supposed to be informative exactly BECAUSE a browser is a black box to me that I rely on for my development./
02:27 🔗 joepie91 When an error page changes, I have no way of telling whether the same error page is now used for everything or just for DNS.
02:27 🔗 joepie91 The only way I can find out is by encountering every single other possible error, and seeing how the browser handles it -right now-.
02:27 🔗 ivan` does latest Chrome not have a More button?
02:27 🔗 joepie91 So that I can exclude any possibility of the same error page being used for all of those errors.
02:27 🔗 joepie91 ivan`: it does not.
02:28 🔗 joepie91 at least not for DNS errors.
02:28 🔗 joepie91 DFJustin: To continue, "can't find" is extremely generic and it is unclear what it refers to.
02:28 🔗 joepie91 As a developer, that is frankly a completely fucking pointless error message, akin to a compiler telling me "there's a syntax error in your code" without giving me a line number.
02:29 🔗 joepie91 Saying that "it takes a few seconds to find out" is, as I pointed out before, similar to saying "yeah well, you know the language, you can figure out where the syntax error error is for yourself."
02:29 🔗 ivan` Firefox experiences significantly less UI breakage than Chrome
02:29 🔗 DFJustin well in my experience it refers to dns errors 100% of the time. if they were using it for something else then I would be pissed same as you, but that's still a hypothetical
02:29 🔗 joepie91 While that is technically indeed possible, it's A. time-consuming B. very frustrating and C. completely fucking unnecessary because the browser/compiler already KNOWS the issue.
02:29 🔗 joepie91 DFJustin: I don't think you actually WANT to understand my problem
02:30 🔗 DFJustin I understand it I just don't see it as that big a deal unless they actually do the evil hypotheticals you've provided
02:30 🔗 joepie91 If, even after pointing out you're not letting me finish, you just interrupt my explanation AGAIN wih your personal judgment before you even UNDERSTAND my problem, I don't expect you to take me seriously.
02:30 🔗 joepie91 DFJustin: Which means you still don't understand it.
02:30 🔗 DFJustin how much more do you have to say
02:30 🔗 joepie91 But you're so convinced that you do, that you refuse to read half of what I say.
02:30 🔗 joepie91 If I am explicitly pointing out that your understanding of what I say is incorrect, what makes you think it's correct?
02:31 🔗 ivan` joepie91: Settings -> uncheck Use a web service to help resolve navigation errors
02:31 🔗 ivan` er let me actually check that first ;)
02:31 🔗 joepie91 ivan`: thanks, that seems to work.
02:31 🔗 joepie91 but wtf, why is that necessary.
02:32 🔗 yipdw I think joepie91's worry is that Chrome's trending towards genericizing HTTP 404 and DNS resolution failure in the same thing, which aren't the same thing at all
02:32 🔗 yipdw at some level
02:32 🔗 DFJustin sure
02:32 🔗 yipdw and that this'll spread to other error conditions
02:32 🔗 ivan` joepie91: you were leaking your navigation data for Google for 5 years ;)
02:33 🔗 DFJustin well the argument seems to be more that now he has to reverse engineer every error message to be 100% sure they haven't already done that without telling anyone
02:33 🔗 joepie91 ivan`: that's a different setting actually
02:33 🔗 ivan` well, that web service is run by Google and Chrome hits it when you get some < 512 byte error page
02:33 🔗 DFJustin which, fine, is a pain in the ass, although it strikes me as overly paranoid
02:33 🔗 joepie91 DFJustin; not just that; even if it isn't the case now, I have no guarantee that that won't happen in the future.
02:33 🔗 joepie91 it's not overly paranoid when your productivity relies on it.
02:34 🔗 joepie91 ivan`: I thought you were refering to address bar data
02:34 🔗 joepie91 in which case the setting is "Use a prediction service to help complete searches and URLs typed in the address bar"
02:34 🔗 yipdw I think the Chrome argument is probably "well why aren't you running with the Web Inspector"
02:34 🔗 yipdw in which case you get all information
02:34 🔗 yipdw it is kind of pain in the ass to have on-screen all the time, though
02:35 🔗 joepie91 yipdw: it also significantly slows down your page loads
02:35 🔗 DFJustin and as for the "why would they do this", obviously because most of the people who use browsers are not web developers
02:35 🔗 joepie91 and the situation I'm sketching is where for example I go to one of my own sites and it doesn't work
02:35 🔗 joepie91 unexpectedly
02:35 🔗 DFJustin and this way if they type something dumb there's a friendly google box to direct them where they want to go, although really it should be a both and thing with technical details
02:35 🔗 joepie91 DFJustin: yet that's still not a valid reason to do it
02:35 🔗 joepie91 that's the whole point of the 'More...' button
02:35 🔗 joepie91 I'm fine with not displaying it by defaultt
02:36 🔗 joepie91 as long as it isn't more than one click away
02:36 🔗 ivan` the only thing more futile than complaining on the Chrome bug tracker is complaining here
02:37 🔗 DFJustin I guess part what's colouring my thought process here is that firefox has said "can't find xxx" for years
02:37 🔗 ivan` I'm guessing Chrome needs to make Google even more money so they're probably happy with this
02:38 🔗 yipdw who needs DNS when you have Google
02:56 🔗 xmc who needs google when you have dns
07:59 🔗 SmileyG errrr
07:59 🔗 SmileyG has someone got the documentry link for the IA copy plz?
07:59 🔗 SmileyG my sewarching is fail.
08:36 🔗 ersi SmileyG: https://archive.org/details/DEFCON20Documentary
08:44 🔗 SmileyG ty
15:20 🔗 godane1 looks like my dumps of katproxy.com/community/ are in wayback now
15:33 🔗 godane1 so i'm looking at the katproxy stuff and there was some that needed login
15:33 🔗 godane1 turns out i don't have permission to view them ever logged in
15:34 🔗 godane1 those threads by my guess are trashed anyways
18:24 🔗 godane so i'm getting this: Y2K Family Survival Guide - With Host Leonard Nimoy

irclogger-viewer