[18:26:01.0000] variable: IDL attribute is the one on a DOM object, e.g. myDivElement.title = 'this is an IDL attribute'; [18:27:00.0000] variable: setting an IDL attribute that reflects a content attribute will set the content attribute, and vice versa [18:27:01.0000] Philip`: well... sorta :X [18:27:02.0000] Ah, I see. Can they *ever* differ? I often see in the spec that "the content attribute must remain consitent with the IDL attribute" [18:27:03.0000] I might have made the report sound like it was a spec error, but I don't know if that was validated or not [18:28:00.0000] is there a reason that this must be stated? Or is it just clarity [18:28:01.0000] *just for clarity ? [18:28:02.0000] I mean it would make-sense(tm) but I never actually looked to the spec to see if something like that is mentioned [18:28:03.0000] variable: yeah, e.g. is the same as HTMLInputElement.defaultChecked, and is different from HTMLInputElement.checked [18:28:04.0000] variable: also, some have no correspondance whatsoever, e.g. HTMLDocument.body is an IDL attribute but there's no element, let alone a content attribute, to go with it [18:29:00.0000] Hixie: you wouldn't happen to be Ian would you? [18:29:01.0000] YOU ARE! [18:29:02.0000] Wow what luck [18:29:03.0000] um hi [18:30:00.0000] is Acid3 still modifiable or any new tests need to go into the next Acid suite? [18:31:00.0000] acid3 is only modifiable in case of serious error [18:31:01.0000] Are there plans for an acid4 ? [18:31:02.0000] I have a test inclusion request for the next version then, what's is there a procedure? [18:32:00.0000] acid4 isn't something that is likely to happen any time soon [18:32:01.0000] fair enough then :) [18:32:02.0000] so if there's a bug that you think needs fixing, i recommend just filing a big rather than hoping for an acid test :-) [18:32:03.0000] yeah I've filled a bug with everyone but only Mozilla so far has taken action :/ [18:32:04.0000] Hixie, are there concrete plans for it? or is it just some vaporware idea atm? [18:32:05.0000] which will be a problem because they will be the "odd man out" [18:33:00.0000] even if their browser does a more correct rendering [18:33:01.0000] Alystair: feel free to use the whatwg blog as a soapbox :-) [18:33:02.0000] variable: define "concrete plans" [18:34:00.0000] Hixie, "this will get done on Towel Day, May 25, 2012. Or "This will get done with all the browsers release their next major version" [18:35:00.0000] oh the plan is to wait until all five major browsers have announced they pass acid3, more or less; same as with acid2->acid3 [18:46:00.0000] Hixie: it's hard to tell how webstandards.org and whatwg relate [18:47:00.0000] they are in no way related [18:48:00.0000] (why did you think they might be related?) [18:48:01.0000] Hixie: what about mobile browsers and Acid3 ? [18:49:00.0000] Rik`: what about them? [18:50:00.0000] Hixie: just my odd mind, because you're involved in both :) [18:50:01.0000] Is the "parse an address algorithm defined by the IRI specification" exposed anywhere to JavaScript? [18:50:02.0000] ah :-) [18:50:03.0000] or rather, saw your name on the acid test site [18:51:00.0000] Hixie: you don't want to wait for specific mobile browsers to pass Acid3 too ? [18:51:01.0000] Alystair: well, i guess there is a minor relation, which is that acidtests.org and whatwg.org are hosted on the same server :-) [18:52:00.0000] but that's just because i host them both myself [18:52:01.0000] (lots of stuff is on that server) [18:52:02.0000] (e.g. bully.org, hixie.ch, voidwars.com, etc) [18:52:03.0000] /me thinks he might end up getting involved in some way [18:53:00.0000] Rik`: any in particular you had in mind? [18:53:01.0000] AryehGregor: the url decomposition attributes expose it [18:53:02.0000] Okay. [18:53:03.0000] Hixie: no, just a thought that mobile browsers are turning into "major" browsers [18:53:04.0000] mobile browsers? [18:53:05.0000] Opera Mini/Mobile [18:54:00.0000] Apple's stuff.. [18:54:01.0000] what does blackberry even use? [18:55:00.0000] Rik`: indeed, but all the major mobile browsers correspond to major desktop browsers, so it's not a huge difference [18:55:01.0000] from my point of view there's no "mobile vs desktop", there's just the web [18:56:00.0000] to me too, but one mobile browser can have more marketshare than one of the "major" desktop browsers [18:56:01.0000] only limited by pixel width/height :P [18:56:02.0000] PPI [18:59:00.0000] Rik`: which one? [19:03:00.0000] Hixie: well, marketshare varies a lot depending on the area but Safari iOS or Opera Mini (if it counts) could be bigger than Opera desktop [19:04:00.0000] as I understand it Opera Mini intentionally breaks the specs, so it doesn't really count [19:04:01.0000] safari on iOS is something I look at, though [19:05:00.0000] yeah Opera Mini has a proxy system in place [19:05:01.0000] time to sleep, bye [19:06:00.0000] Has there been any talk about copy/paste buffers with html5? [19:06:01.0000] /me should go check mailinglist archive if there is one [19:13:00.0000] Alystair: not sure what you mean by copy/paste buffers [19:17:00.0000] well as applications move "to the cloud" you end up in the odd situation where an application cannot access or utilize the OS copy/paste buffer in a safe manner [19:17:01.0000] at the moment writing to the buffer is only possible via a cludge in flash or hacking configs in some browsers [19:20:00.0000] ah, yeah, there's some feedback on copy-and-paste... let's see [19:21:00.0000] only stuff about copy and paste events, looks like [19:54:00.0000] annevk: happy birthday btw :) [04:19:00.0000] I really have no clue why people are spending time on the polyglot thread. Isn't it a giant tar pit? [04:20:00.0000] fn + backspace [04:20:01.0000] well actually, just backspace suffices [04:21:00.0000] jgraham: http://xkcd.com/386/ is my best guess. [04:40:00.0000] It's a danger of the approach of allowing anyone in the WG to publish drafts [04:41:00.0000] Other members are likely to get dragged into discussing it even if they don't care about the subject [04:43:00.0000] and if they ignore it now then either they'll have to object to publication later when the editors are happy with the draft, or the WG will publish something technically poor [06:14:00.0000] http://lists.w3.org/Archives/Member/w3c-html-cg/2010JulSep/0010.html is interesting [06:14:01.0000] the 3rd paragraph in particular [06:59:00.0000] hsivonen: perhaps it would be best to discourage any further contributions of patches to make the "polyglot"-checking feature complete [07:00:00.0000] it seems like it might not be a very good feature with canonical validator.nu [07:01:00.0000] I think those who want to provide it should maybe instead be encourage to host their own instance and advertise that as a value-add or whatever [07:01:01.0000] it's never really been clear to me what the use case is for that feature anyway [07:02:00.0000] because I would assume that anybody who cares to make sure their documents follow XML well-formedness rules would most likely already be using an XML toolchain to author their documents [07:03:00.0000] unless there's something I'm not aware of that they want and that any XML parser would not already be able to report to them [07:49:00.0000] AIUI, what people what from polyglot checking is nothing like the strict level of checking that would be required for full DOM consistency. But rather, most seem to checking of a few XML-like talismans that give the appearance of cleaner and more consistent markup [07:50:00.0000] for instance, I'm pretty sure most don't particularly care about the optional tbody issue, unless they actually want scripts or stylesheets to interact with it [07:52:00.0000] and most would be happy to omit boilerplate markup like xmlns [07:53:00.0000] the important issues seem to be pretty much trailing slashes in void elements, explicit end tags, double-quoted attributes, always escaping & and < characters [07:54:00.0000] Lachy: wouldn't that be the "pedagogical" profile instead of "polyglot"? [07:55:00.0000] yes [07:55:01.0000] that's why I don't really think the polyglot spec is particularly useful [07:56:00.0000] and why I think the approach I have taken in my authoring guide in recent updates (not yet checked in) is better [07:56:01.0000] Lachy: I think you mean s/spec/draft/ :-) [07:56:02.0000] yes [07:56:03.0000] Lachy: are you updating the guide again? [07:56:04.0000] yes, I started doing that last week [07:56:05.0000] Lachy: cool [07:56:06.0000] I'd also done some other major changes a while ago that weren't checked in. [07:57:00.0000] significant changes are: splitting into HTML5 Reference (for syntax and element reference) and HTML5 Guide (for more tutorial like content, describing common use cases, etc.) [07:58:00.0000] on a completely different topic: does anyone happen to have pointers to innerHTML benchmarks? [07:58:01.0000] and also updating the syntax section to more clearly describe, and compare and contrast the syntax for both HTML and XHTML, [07:58:02.0000] I'm interested in benchmarks that represent benchmarks and benchmarks that represent marketing [07:58:03.0000] Lachy- Just commenting here: I like your "splitting" idea. [07:58:04.0000] -_- [07:59:00.0000] I also intend to publish this on html5reference.org soon too with a more liberal licence, so I can bypass most of the w3c politics on that issue [08:00:00.0000] that's another reason I stopped checking it in there, cause I want it licenced liberally there first, and then dual licence it like the spec. [08:02:00.0000] on another note, I also got my copy of Windows 7 today and tried out the IE9 preview for the first time. [08:02:01.0000] Lachy: I'd love to see the proponents of the polyglot draft to state their use cases for it [08:03:00.0000] seems impressive [08:03:01.0000] Lachy: from Amazon? [08:03:02.0000] yes [08:03:03.0000] .co.uk? [08:03:04.0000] yes [08:03:05.0000] seems slow compared to how fast they delivered Windows 7 to me [08:03:06.0000] after a month of waiting, they shipped me a replacement copy last friday by UPS and it arrived this morning. The first copy got lost in the mail [08:04:00.0000] if I recall correctly that it's been a while since you ordered [08:04:01.0000] ah [08:04:02.0000] they told me I had to wait till the 29th in case of delays before they would ship a replacement. I ordered it on the 2nd [08:06:00.0000] my one complaint about windows 7 is that running it in a virtual machine on my 2.5 year old MacBook Pro can be a little slow [08:07:00.0000] Works fine in a VM on my i7 [08:07:01.0000] (without the glass UI) [08:07:02.0000] it appears that VirtualBox OSE doesn't emulate a DirectX-compliant GPU [08:07:03.0000] yeah, i7 is a lot newer and faster than my Core 2 Duo 2.4 [08:07:04.0000] *2.4GHz [08:09:00.0000] also my graphics card is technically below the minimum requirements that VMWare says will support the Aero transparencies [08:09:01.0000] though it does work [11:12:00.0000] Interesting reading this Spolsky article from 2004: http://www.joelonsoftware.com/articles/APIWar.html [11:12:01.0000] Particularly the bit near the bottom about webapps, and the list of things you couldn't do well in a web application. [11:13:00.0000] I think (3) was doable back then with onunload, but we've since solved 1,2,4, and 6. 5, I dunno. [11:20:00.0000] 5 (having a keyboard-driven web app) is technically doable at least. Not sure how useful that'd be. [11:21:00.0000] Yeah, it's just annoying from what I know. [11:22:00.0000] There's a chrome extension that does just that, "vimperator". It hides every user interaction with the browser behind a key. For, instance, J scrolls down, K scrolls up, ... [11:22:01.0000] lots of keys to remember [11:49:00.0000] TabAtkins: O'Reilly wants to use your blurb on the back cover of my book, and they would like to know what your job title is [11:51:00.0000] Urgh, that's a hard one. Officially I'm "Software Engineer", but that's just because I haven't customized my title yet. [11:51:01.0000] no time like the present! [11:52:00.0000] Indeed. [11:52:01.0000] Standards Wrangler [11:53:00.0000] "Web Imagineer" [11:53:01.0000] /wrist [11:53:02.0000] "CEO" [11:53:03.0000] or are there some constraints on what you can choose? [11:53:04.0000] Chief Engineering Officer [11:53:05.0000] haha [11:53:06.0000] It needs to go through approval [11:53:07.0000] I suspect that "CEO" is on some list. ^_^ [11:54:00.0000] You could spell it out [11:54:01.0000] "Web Standards Hacker"? [11:54:02.0000] Chief Entertainment Object [11:57:00.0000] "Hack" ;-) [11:58:00.0000] Unless anyone has objections, go with "Web Standards Hacker" for now. [11:58:01.0000] well, whatever you decide, they need it by the end of the day [12:00:00.0000] in unrelated news, the most popular pages on diveintohtml5.org (in order): [12:00:01.0000] video - 39Kd [12:01:00.0000] detect - 33K [12:01:01.0000] forms - 27K [12:01:02.0000] semantics - 26K [12:01:03.0000] canvas - 23K [12:01:04.0000] past - 20K [12:01:05.0000] geolocation - 7K [12:01:06.0000] storage: 7K [12:01:07.0000] offline: 6K [12:02:00.0000] microdata: 4K [12:02:01.0000] in the past month [12:02:02.0000] Those seem fairly reasonably ordered. [12:04:00.0000] the ordering also tracks fairly closely to how much i give a shit about each particular topic [12:06:00.0000] man, IRC on an Apple //e is awesomeit [12:06:01.0000] it's like a text adventure [12:07:00.0000] I suspect it's roughly similar to IRC on irssi. [12:07:01.0000] go north [12:07:02.0000] There is a spec here. [12:07:03.0000] examine spec [12:07:04.0000] The spec is large; it contains multitudes. [12:07:05.0000] look behind the spec [12:07:06.0000] You spot a cabal. [12:08:00.0000] ^^; [12:10:00.0000] What does a cabal look like? [12:12:00.0000] examine cabal [12:13:00.0000] The cabal shifts and twists in the wind, appearing to each person exactly as they imagine it to be. [12:15:00.0000] go west [12:15:01.0000] There is a cabal here. [12:15:02.0000] examine cabal [12:15:03.0000] The cabal is everywhere. [12:26:00.0000] Your words have angered the cabal. It catches you and eats you. [12:29:00.0000] mpilgrim: You just reminded me of something beyond trivial and several years out of date. You adapted some script I wrote in "Greasemonkey Hacks" but my name never made the credits section :) [12:30:00.0000] At least I think that happened [12:30:01.0000] I could have imagined the whole thing [12:30:02.0000] Delusions of grandeur and all that [12:31:00.0000] You might think it wasn't much in the way of grandeur [12:31:01.0000] But it all depends where you started [12:31:02.0000] /me wonders if that was his Apple II catching fire [12:32:00.0000] s#I#//e# [12:32:01.0000] argh [12:32:02.0000] your regex-fu sucks. [12:32:03.0000] Yeah [12:32:04.0000] IRC magic regexp though [12:32:05.0000] Save me every time [12:39:00.0000] /me successfully upgraded to ADTPro 1.1.6 [12:42:00.0000] mpilgrim, do you have photos of #whatwg on your Apple //e? [12:45:00.0000] http://twitpic.com/2b16c2 [12:45:01.0000] oh my. beautiful. [12:45:02.0000] !!!! [12:45:03.0000] Yup, looks like a standard terminal client. ^_^ [12:46:00.0000] it's not through a terminal client program. [12:46:01.0000] Right. [12:46:02.0000] I have a real Ethernet card in the Apple //e, and a real IRC program (part of Contiki 1.3) [12:47:00.0000] I can also use it as a terminal, connected through a Super Serial card to the serial port on my Linux box [12:47:01.0000] sweet [12:47:02.0000] but this is not that [12:48:00.0000] jgraham: I apologize if I left you out of the acknowledgements for "Greasemonkey Hacks" (catching up with the backscroll) [12:49:00.0000] The acknowledgements and contributor list were several pages long. My O'Reilly editor had never seen anything like it. [12:49:01.0000] you should show him the html spec's [12:49:02.0000] lol [12:49:03.0000] 515 names currently [12:50:00.0000] At least two are the same person, though [12:50:01.0000] Hixie! By any chance, would you like to say something nice about my HTML5 book and get your name on the back cover? [12:50:02.0000] you didn't like the suggestion I gave you earlier? :-) [12:51:00.0000] (in general i'm trying to avoid getting on any html5 books, since there's so many of them) [12:51:01.0000] Wrong audience [12:51:02.0000] :) [12:51:03.0000] (and they all want me to write a foreword or whatnot) [12:51:04.0000] fair enough [12:51:05.0000] Philip`: yeah, i asked him about that, i'll remove one when i get to his mail in my pile [12:52:00.0000] Well, at least writing forewords wouldn't cause flamewars on the hybi list [12:52:01.0000] While we're bothering you, Hixie: What do you think about making the second parameter in pushState() modify document.title? Maybe that's better than the state we're in, where everyone (including, now, FF) ignores it. [12:52:02.0000] mpilgrim: It is really not something that has ever kept me up at night :) [12:55:00.0000] /me wonders how much of the book one is expected to have read to write a comment for the back or indeed a foreword [12:55:01.0000] jlebar: haven't thought about it, but i'm trying to go through the feedback pile and i know that is one of the options in the feedback [12:55:02.0000] /me guesses the answer is probably "none" [12:55:03.0000] jlebar: i recommend experimenting with it, see if it gives authors what they want [12:55:04.0000] jlebar: and then replying to the thread with your experiences from that experiment [12:56:00.0000] jlebar: if you do it and another browser picks it up and it turns out to be a good idea, it'll make my life easy when i get to the feedback :-) [12:56:01.0000] It is always mildly concerning when such comments are about the author rather than the book itself [12:56:02.0000] Or written by someone also mentioned in the acknowledgements [12:56:03.0000] Hixie, That sounds reasonable. I'll see if we want to try it out in the FF beta. [12:57:00.0000] jgraham: I guess you just need to read enough so that you don't say anything so blatantly untrue that somebody who does read the book will think you're an idiot and remember your name [12:57:01.0000] jlebar: if you can find a way to make the argument not be vestigial before the spec is even done, you'll certainly get my thanks ;-) [12:58:00.0000] I'll try. :) [12:58:01.0000] Philip`: That is the same as "none", as long as you are careful about what you say [13:37:00.0000] For the record, I've read "some" of the book. [13:37:01.0000] /me used the video chapter to remember what @type to give for video sources. [13:39:00.0000] TabAtkins: So is your quote a secret? Should we have a competition to see who guesses the most words correctly? [13:39:01.0000] No, I just posted it over in internal Buzz. [13:40:00.0000] "HTML5 is the future of the web, but Pilgrim shows you how to make it the present as well. Read this book right now if you want to know to use the most interesting features of HTML5 today, from videos to drawing to scripting, all the way down to the new elements that make your source code more beautiful." [13:44:00.0000] It's interesting how buzz sems to be popular inside Google but not really outside of it [13:45:00.0000] It is to Google what Lyskom is to Opera Sweden [13:45:01.0000] It works nicely as a Twitter+, once you have a decent critical mass. [13:46:00.0000] Hell, it's a decent Facebook News Feed+. [13:47:00.0000] jgraham: LysKOM is popular within Opera Sweden? [13:48:00.0000] gsnedders: Well it is popular in the sense of "widely used" [13:48:01.0000] Even if there are a healthy contingent of people who think it is a sin against nature [13:52:00.0000] jgraham: To be fair, the rest of Opera loathes it :P [14:01:00.0000] Workshiva: Most of Sweden does too… :P [14:02:00.0000] I don't think most of Sweden has had the displeasure of learning it exists [14:03:00.0000] How do you think we use it without knowing it exists? [14:03:01.0000] Oh, you meant Opera Sweden [14:03:02.0000] Not the country [14:05:00.0000] (to be fair to Kom, which I never am, I would prefer some of the stuff happens there rather than in email. Best to keep all the mostly-ignorable stuff together) [14:07:00.0000] You should just replace it with google wave [14:07:01.0000] They seem to occupy a similar niche :) [14:07:02.0000] Workshiva: Google employs UA sniffing to block Opera. [14:07:03.0000] Even better, reduces the amount of time people waste on it :P [14:15:00.0000] It is quite like Google wave in that there are no good clients [14:15:01.0000] But [14:15:02.0000] And this is important [14:15:03.0000] Wave doesn't store all the data as one giant sexp [14:15:04.0000] (I presume) [14:15:05.0000] Unlike LysKOM [14:15:06.0000] (allegedly) [14:18:00.0000] did i catch that the html wg voted on moving to an unversioned html specification? or am i wrong on that? [14:20:00.0000] paul_irish: there was a survey about versioning [14:20:01.0000] this one? http://www.w3.org/2002/09/wbs/40318/issues-4-84-objection-poll/results#xkeepnew [14:20:02.0000] the results: http://www.w3.org/2002/09/wbs/40318/issues-4-84-objection-poll/results [14:20:03.0000] yeah [14:20:04.0000] But no pronouncement from the chairs yet [14:20:05.0000] afaik [14:21:00.0000] The system allows the chairs to basically decide what they like although they have to back it up with reasons and stuff [14:21:01.0000] Roger that. Thanks [14:28:00.0000] btw, jgraham, before writing a foreword, I think you are required by law to read the entire book twice. Any technical errors or stylistic blemishes in the book then reflect on you, personally, forever. [14:30:00.0000] Harsh [14:30:01.0000] Are you sure that's not just in North Korea? [14:30:02.0000] I didn't make the rules [14:30:03.0000] hsivonen: any suggestions on a RelaxNG hacker who would be interested in helping the SVG WG put together a good (not converted-DTD) RNG schema for SVG? [14:47:00.0000] jgraham: I could tell you a lot about how things are stored, but I want to keep my job [14:51:00.0000] jgraham: I agree against version info in the doctype, but does that mean we should no longer talk about html5? Say, it would blend in the flow of continuous innovation around the draft for years and years to come, and never actually be final? Will be html5 or 6? [14:51:01.0000] will be html [14:52:00.0000] Hixie doesn't use version numbers anymore [14:52:01.0000] ... also, I can't spell device [14:55:00.0000] The survey only covered versioning in the doctype. [14:55:01.0000] Not versioning generally. [14:56:00.0000] The HTMLWG is only chartered to release one version of HTML, and W3C process doesn't allow continuous spec development. You have to freeze specs eventually. [14:56:01.0000] I guess you could charter a WG to let it only publish Editor's Drafts or some crazy hack like that, but the HTMLWG isn't, it's producing REC-track work. [14:57:00.0000] Even if you don't embrace global versionless development, the way w3c keeps splitting everything into smaller and smaller specs you get per-feature maturity tracking anyway [14:57:01.0000] espadrine, the WHATWG version of the spec mostly just calls it "HTML". It only keeps the "5" for now for buzzword purposes. [14:57:02.0000] We should just have one spec per paragraph, that would solve it. [14:58:00.0000] You don't need separate specs [14:58:01.0000] You just need per-paragraph tracking :) [14:58:02.0000] /me wonders why jgraham's objection is line-broken [14:58:03.0000] Core canvas is for all intents and purses CR already [14:58:04.0000] ... purposes [14:59:00.0000] I should go home and sleep [14:59:01.0000] A lot of HTML5 may as well be CR. [15:00:00.0000] I still like the almost-concluding sentence of my objection: "In summary, the proposal seeks to address a nonexistent problem, and fails to do even that." [15:01:00.0000] That is a pretty good sentence. [15:01:01.0000] Anyway, if the chairs pay even the slightest heed to the survey results, that proposal is completely dead. [15:03:00.0000] We had at least two people who objected so thoroughly to the proposal that they went over the 5 KBish limit in writing their responses, and the only ones objecting to the no-change proposal were 1) an academic who likes versioned doctypes because they help him date historical HTML pages, and 2) a two-sentence statement that doesn't state support of the other change proposal. [15:04:00.0000] To be fair, it was dead before the survey. It was a dumb idea in the first place. [15:04:01.0000] /me thinks that when it comes to serious issues, the change proposal procedure is a waste of time but not likely to produce particularly bad results. [15:04:02.0000] Well, yes. [15:05:00.0000] It exemplifies the problem with letting one person waste everyone's time with a giant pile-on of opposition when no one agreed with him in the first place. [15:05:01.0000] Yup. [15:23:00.0000] AryehGregor: emacs [15:24:00.0000] You just write stuff there generally in case web forms eat it? [15:24:01.0000] Or do you have some cool plugin that transforms textareas to emacs? [15:24:02.0000] I wish I had that for vim. :( [15:24:03.0000] Workshiva: But since you wouldn't say "s-expressions" it's not interesting [15:25:00.0000] AryehGregor: No I wrote it there because it is a superior text editor to my browser and it was a reasonable amount of text [15:25:01.0000] I would love to be able to use it seamlessly for textareas [15:25:02.0000] or at least I would like textareas to suck less [15:26:00.0000] The closest I've seen is a Firefox extension that lets you right-click on textareas to open them in an external editor. [15:26:01.0000] Awkward, but it's more convenient than cut-and-paste. [15:26:02.0000] It basically creates a temporary file, invokes the editor on the file, then updates the textarea whenever the file is written, and deletes the file when you submit the form. [15:26:03.0000] Something like that. [15:27:00.0000] I thought about writing that some years ago [15:27:01.0000] I didn't of course [15:27:02.0000] The only major thing that annoys me about vim is that there's no way I know of to easily move up and down between logical lines, which is really a necessity when you're dealing with something that's not full of line breaks. [15:27:03.0000] I mean. [15:27:04.0000] Visual lines, not logical lines. [15:27:05.0000] You know, the way up and down work in regular text editors. [15:28:00.0000] /me should go ask in #vim, although he thinks he's done that before [15:28:01.0000] Emacs' handling of visual lines is pretty abysmal [15:28:02.0000] As far as I can tell there is no way to get it to break on whitespace [15:28:03.0000] (visually) [15:28:04.0000] I don't mind that part so much. [15:29:00.0000] For code it is fine [15:29:01.0000] But when j and k jump me like a page at a time, it's really annoying. [15:29:02.0000] For prose, not so much [15:29:03.0000] With some HTML pages, etc. it's seriously obnoxious. [15:29:04.0000] Well, if you do version control on your prose, you want to line-break it anyway. [15:29:05.0000] (so you have to use auto-fill-mode and hope that people don't mind the line breaks) [15:32:00.0000] In vi, you can type "80 l", and you get down to the next visual line [15:32:01.0000] but there should be a builtin macro for that [15:32:02.0000] Oh, here we go: http://vimdoc.sourceforge.net/htmldoc/motion.html#up-down-motions [15:32:03.0000] gk, gj. [15:32:04.0000] Yay. [15:33:00.0000] /me didn't realize how many motions there are in vim, yikes [15:33:01.0000] nice! [15:34:00.0000] It's kind of crazy to have things like "+" when it's really just the same as "j^" or whatever, but if you get used to this kind of thing, it becomes a real time-saver . . . [15:34:01.0000] I think it's time for me to memorize a bunch more vim commands. I've gotten used to all the ones I know long ago. [15:36:00.0000] It seems like the _ vi move makes ^ redundant [15:37:00.0000] There's a lot of redundancy here if you look. [15:37:01.0000] Wow, now there are four different Chrome channels? [15:39:00.0000] AryehGregor: That textarea->texteditor plugin for Firefox was *awesome*. I used the crap out of it. It was great for long comments, and extra-great for code, since I can actually use tab in my texteditor. [15:45:00.0000] Maybe I'll try it again if I switch back to Firefox when they release 4.0. I'll have to see if I want to. [15:45:01.0000] when Chrome was released it was no contest for me, Firefox was massively less responsive than Chrome on Linux, but that's way better with 4.0. [15:45:02.0000] hmm. jailbreakme.com (new iPhone jailbreaking website) is nice, though somewhat scary in how it works. [15:46:00.0000] AFAICT, it's using some sort of PDF exploit to actually perform the jailbreak. [15:46:01.0000] I don't understand the content of the PDF file beyond that to know what exactly [15:47:00.0000] all jailbreaks are exploits [15:47:01.0000] but it does make me wonder what other hacks a malicious website could perform using a similar technique [15:47:02.0000] yes, I know that. [15:47:03.0000] But the fact that they can do it from just having the user load a web page [15:47:04.0000] that's the scary bit [15:49:00.0000] yes, remote code execution exploits are scary :-) [15:51:00.0000] to be pedantic that is a client side exploit, not remote ;) [15:52:00.0000] it's remote to the attacker [15:52:01.0000] wirepair: oh, we feel much better know ;) [15:52:02.0000] espadrine ehe [15:52:03.0000] but yeah, i guess it's not a remote code execution since you can't force it to occur remotely [15:52:04.0000] /me nods [15:53:00.0000] It's remote code execution because you can trigger it without access to the machine. You just have to get them to visit a URL, which is fairly trivial. [15:54:00.0000] As opposed to requiring that they install a program or something. [15:54:01.0000] well the reasoning for seperate terms is to differentiate between someone exploiting a service, versus client software [15:55:00.0000] Does a web page really count as client software? [15:55:01.0000] no but a pdf does [15:55:02.0000] It's supposed to be sandboxed. [15:55:03.0000] Hmm. [15:55:04.0000] PDF does. [15:55:05.0000] Er, that. [15:55:06.0000] I guess. [15:55:07.0000] It's sort of intermediate. [15:55:08.0000] in the end, it really doesn't matter, one requires client interaction, the other does not [15:56:00.0000] (I don't know why a webpage would not count but a PDF would) [15:56:01.0000] i'd consider a webpage as well actually [15:56:02.0000] the browser being the client software [15:57:00.0000] no but a pdf does (sorry meant adobe acrobat/whatever is reading the doc) [16:00:00.0000] http://validator.w3.org/check?verbose=1&uri=http://www.hp.com <- there's really some irony to this [16:02:00.0000] Why? [16:03:00.0000] who donated the hardware to w3c :P [16:03:01.0000] HP donates to w3c yet doesn't uphold the standard themselves? [16:04:00.0000] Oh, come on. [16:04:01.0000] There are web pages on mozilla.org that don't validate. [16:04:02.0000] there are pages on whatwg.org that don't validate [16:04:03.0000] e.g. the forums and maybe parts of the blog [16:04:04.0000] Probably on w3.org too. [16:04:05.0000]