| 04:22 | <Hixie> | jgraham: why is the core mobile web not the same as the core desktop web, core tv web, core print web, core AT web, core braille web, and core tablet web? |
| 04:30 | <zewt> | they're all the same, some are just more same than others? |
| 05:15 | <Von_Davidicus> | A serious question on some advice I've received: I wrote a tutorial on HTML 4.01, and someone suggested I should simply "update in places that need it". Would it really be that simple, or should I rewrite my book from beginning to end? |
| 05:30 | <sam___> | hello hello |
| 05:30 | <Von_Davidicus> | Hello. |
| 05:32 | <sam___> | um, im encountering a weird html problem |
| 05:32 | <sam___> | mind taking a look for me? |
| 05:32 | Von_Davidicus | MIGHT be able to help.... |
| 05:32 | Von_Davidicus | is still an old-school HTML 4.01 coder, though |
| 05:36 | <sam___> | no problem |
| 05:36 | <sam___> | the problem is that the contents in a p tag is breaking out of the container somehow |
| 05:36 | <sam___> | i'm generating the html using python, it doesn't happen if i use pure html... |
| 05:37 | <Von_Davidicus> | That's a new one... |
| 05:37 | <sam___> | how come i cant send msg in #html???? |
| 05:39 | Von_Davidicus | has never used Python in his life, so can't answer that question. |
| 05:39 | <sam___> | k here's a link to the project i'm working on, it's a school project, so Im using a free domain as a redirector |
| 05:39 | <sam___> | http://www.luxuryparadise.co.nr/ |
| 05:39 | <sam___> | search for "a" and click on the result, check the right hand side where it says "recent messages" click on "show more" around that area, and you will see the problem |
| 05:41 | <Von_Davidicus> | I don't see a "showmore", nor recent messages. |
| 05:44 | <sam___> | are you sure??? do u see the account details then? |
| 05:45 | <Von_Davidicus> | Asdfasdfdf Asdf |
| 05:45 | <Von_Davidicus> | Email: asdf⊙ac |
| 05:45 | <Von_Davidicus> | Sex: Not specified |
| 05:45 | <Von_Davidicus> | That is what I see. |
| 05:45 | <sam___> | oh u gotta click on it |
| 05:45 | <sam___> | click on the supposed user name |
| 05:46 | <sam___> | i think ur still on the search result page |
| 05:46 | <Von_Davidicus> | Ah. Where you have a very, very long "word |
| 05:46 | <Von_Davidicus> | "? |
| 05:47 | <sam___> | do u see a bunch of messages? |
| 05:47 | <sam___> | where it says "fasdfasd......." |
| 05:47 | <sam___> | lol |
| 05:48 | <Von_Davidicus> | Yes. |
| 05:48 | <sam___> | the div's height is not set, it's supposed to grow as more contents are put in there |
| 05:48 | <sam___> | why is this happening? |
| 05:48 | <sam___> | i tried copying all the code including html css and js into jsfiddle |
| 05:48 | <sam___> | and suddenly everything works fine |
| 05:49 | <Von_Davidicus> | Thing is, it's all ONE word, and in comments (at least from what I've seen), such overflow is a headache on many sites. |
| 05:49 | <sam___> | btw i copied the code from chrome inspector |
| 05:49 | <Von_Davidicus> | Seriously, if someone makes a long word like that, they are being a pest and deliberately breaking the layout, so I don't think you really need to worry about it. |
| 05:50 | <sam___> | oh! i think ur right. I just added some messages that are more like words instead of a long string of text |
| 05:51 | <Von_Davidicus> | Yeah, really long words often do that. |
| 05:53 | <sam___> | another thing is Im using a jquery plugin that collapses words or elements when it gets too long. I used it on the divs, that's why the "show more" link is there |
| 05:54 | <sam___> | but when I use the plugin on the messages, which is p.sendermessage, it's not working any more |
| 05:54 | <sam___> | again, it works if I used pure html, instead of this python generated one |
| 05:54 | <sam___> | i don't know if it's python that messes thing up, but that's the only reason i can think of |
| 05:55 | <Von_Davidicus> | *Nods.* I do stuff on a lot of online art galleries, and one thing that annoys a lot of artists if a comment is something like "loooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooool" |
| 05:55 | <Von_Davidicus> | It messes with the layout. |
| 05:58 | <sam___> | haha |
| 05:58 | <sam___> | ok i get how it's messing up the layout, but it shouldn't affect jquery |
| 06:13 | <sam___> | oh well, that's not a big deal now. |
| 06:13 | <sam___> | Thanks Von! |
| 06:30 | <Von_Davidicus> | I've read some HTML5 posts that say that JavaScript is the only scripting language on the web--but the most popular browser (IE) also supports VBScript. That being said, is VBScript indeed so uncommon that JavaScript could be considered the only scripting language on the web? |
| 06:31 | <annevk> | yes |
| 06:32 | <Von_Davidicus> | Since IE is so common, one could almost put aside the argument "well, only one browser supports it", so why else is VBScript that rare? |
| 06:36 | <annevk> | why bother if there's JavaScript that works everywhere |
| 06:37 | <Von_Davidicus> | True. I just wondered if VBScript was more limited in its capabilities as well. |
| 06:43 | <Von_Davidicus> | Certainly, I never could get it to do much. |
| 06:51 | <j_wright> | you could actually add other languages to IE, i remember a TclScript and a PerlScript |
| 06:52 | <Von_Davidicus> | I've heard of PerlScript. Never tried it, though. |
| 06:52 | <j_wright> | i think it was more used, still not much though, in iis |
| 06:53 | <Von_Davidicus> | A serious question on some advice I've received: I wrote a tutorial on HTML 4.01, and someonesuggested I should simply "update in places that need it". Would it really be that simple, or should I rewrite my book from beginning to end? |
| 06:54 | <j_wright> | update in places seems fair |
| 06:55 | <j_wright> | its not like its xhtml 2 |
| 06:55 | <annevk> | seems kind of hard to say without having seen the book |
| 06:55 | <annevk> | you can have a read through http://dev.w3.org/html5/html4-differences/ |
| 06:56 | <Von_Davidicus> | It was written to be the kind of tutorial I wish *I'd* had when learning HTML. My book starts at the level of "<" starts a tag, ">" ends a tag. |
| 07:07 | <Hixie> | Von_Davidicus: then you weren't teaching HTML4, you were teaching what browsers implement, and that's actually closer to the HTML spec today than it is to HTML4. :-) |
| 07:08 | <Hixie> | (in HTML4, tags start with < and end either with >, or with /, or with <.) |
| 07:08 | <Von_Davidicus> | Well, the rest of the book went into the elements of HTML 4.01/XHTML 1.0, explaining the differences between the two where they cropped up. |
| 07:11 | <Von_Davidicus> | I didn't know tags could end with < or /. |
| 07:12 | <Hixie> | <p/This is a paragraph/ is equivalent to <p>This is a paragraph</p> in HTML4 |
| 07:12 | <Hixie> | not that any browser implemented it |
| 07:12 | <Hixie> | and you can say things like <p<em>This</em> is valid</p> |
| 07:12 | <Von_Davidicus> | I knew that /wasn't/ the case with XHTML. |
| 07:12 | <Hixie> | also <p<em/This/ is valid.</> |
| 07:12 | <Hixie> | (note </>) |
| 07:13 | <Hixie> | also you could use "<>" where only one element would have been valid |
| 07:13 | <Hixie> | none of this stuff was really ever implemented by browsers |
| 07:13 | <Hixie> | i think i got mozilla to do the <a<b thing for a while, but that was a mistake that we fixed when we defined parsing later |
| 07:13 | <Von_Davidicus> | So my statement of "</> is nonsense to browsers" was correct? |
| 07:14 | <Hixie> | to most user agents, yes |
| 07:14 | <Hixie> | most validators supported it, confusingly |
| 07:14 | Von_Davidicus | based a lot of his book on either experience or experiments. |
| 07:20 | <Von_Davidicus> | Another question: How's this for a demonstration that valid code != good webpage? http://www.mrinitialman.com/Experiments/Horror/horrors.html (Completely valid HTML 4.01 and CSS) |
| 07:31 | <Von_Davidicus> | ??? |
| 07:48 | <Von_Davidicus> | ... you don't need the SVG namespace for an SVG graphic to be embedded? |
| 07:49 | Von_Davidicus | is finding HTML5 more and more baffling.* |
| 07:51 | <Ms2ger> | The namespace is still there, but you don't need to write it out |
| 07:52 | <Von_Davidicus> | Does the xmlns attribute still have some use, though? |
| 07:54 | <Ms2ger> | Nope |
| 07:55 | <Von_Davidicus> | That's depressing to know. |
| 07:56 | <Von_Davidicus> | Well, at least my SVG file goes sideways when I remove it. |
| 08:29 | <smaug____> | hmm, still today time to think about whether to attend webapps f2f |
| 08:30 | Von_Davidicus | is almost afraid to read the HTML5 spec and see what ELSE it screws up. |
| 08:36 | <smaug____> | HTML5? I hope you read HTML |
| 08:38 | <Von_Davidicus> | I'm a longtime user of HTML 4.01 and XHTML 1.x . HTML5 is a scary, scary thing; I mean... embedding SVG should *require* XHTML, XML namespaces and, if at all possible, a Doctype that takes up half the document! |
| 09:12 | <jgraham> | Hixie: It is, more or less (specs like device orientation are more useful on mobile than on TV though) |
| 09:13 | <jgraham> | I see the "mobile" thing as an oppertunity to ride on the coattails of fashion to get useful work done |
| 09:13 | <jgraham> | It also provides some semi-objective criteria |
| 09:13 | <jgraham> | e.g. "what features do we need to repliacte the top 100 mobile apps as web apps" |
| 09:25 | <Von_Davidicus> | And, apparently, HTML5 won't let me throw in my own XML namespaces. Drat. |
| 09:27 | <Von_Davidicus> | ... 400+ XML files are taking FOREVER to upload. :-@ |
| 09:29 | <smaug____> | if you need namespaces, use XHTML form |
| 09:30 | <smaug____> | http://www.whatwg.org/specs/web-apps/current-work/multipage/the-xhtml-syntax.html#the-xhtml-syntax |
| 09:31 | <Von_Davidicus> | Well, I'm poking at HTML5, and seeing if anything I remember that was needed in XHTML 1.x still applies. |
| 09:44 | <jgraham> | Hixie: (also the market on mobile is noticably different to the market on other platforms, and not really in a better way) |
| 09:47 | <Von_Davidicus> | Question: Are the XHTML, SVG, and MathML namespaces the *only* ones allowed in HTML5? |
| 09:57 | <Von_Davidicus> | Or is there at least a limited set of namespaces allowed? |
| 10:26 | <Von_Davidicus> | I've been thinking about the way SVG and MathML are embedded in HTML5. It seems to me that they're not tacked on, as they are with XHTML 1.1 + MathML 2.0 + SVG 1.1 where you have "islands" of different XML languages in an encompassing one, but rather that HTML5 includes the lot. Am I correct, or am I off-base? |
| 10:28 | <Philip`> | Von_Davidicus: The HTML parser is hardcoded to recognise <svg> and <math> and switch to a different parsing mode that recognises <circle> and <mfrac> etc |
| 10:29 | <Philip`> | The DOM generated by the parser is the same as what you'd get from old-fashioned XHTML1+SVG+MathML |
| 10:30 | <Von_Davidicus> | So it's still islands of different languages--but ones the parser knows how they work, and so doesn't need the namespace to understand? |
| 10:31 | <Philip`> | Yes, it doesn't need the explicit xmlns="..." declaration, since it knows that <svg> should implicitly trigger the SVG namespace etc |
| 10:32 | <Von_Davidicus> | So it just "assumes" that namespace is there? |
| 10:33 | <Philip`> | Yes, effectively |
| 10:34 | <Philip`> | (Explicit xmlns and xmlns:* attributes are always completely ignored by the HTML parser, and usually are conformance errors) |
| 10:35 | <Von_Davidicus> | I must admit I find HTML5 baffling, since it seems to let you get away with so much more. I find myself wondering if ANY of my experience with the older HTML and XHTML is of any worth. |
| 10:36 | <Philip`> | It's designed to not force authors to jump through needless hoops for concepts like modularity and extensibility that don't actually exist in reality |
| 10:36 | <Philip`> | (hence the simpler doctype, lack of xmlns, etc) |
| 10:38 | <Von_Davidicus> | I thought extensibility did exist--unless I have the wrong idea of what it means. I thought it meant you could add in other XML languages, and I do that all the time. |
| 10:39 | <Philip`> | You can't add other XML languages in a way that makes web browsers do anything interesting with them - you can just add opaque lumps of data that scripts can read later |
| 10:40 | <Von_Davidicus> | So, basically, XHTML + SVG + MathML was the closest to that extensibility that the browsers ever got? |
| 10:41 | <Philip`> | They're the only languages that browsers interoperably do interesting things with |
| 10:43 | <Von_Davidicus> | That's basically what I meant, so I'll take that as a yes. |
| 10:43 | <Philip`> | If someone wanted to add a new language to the web platform, extending the hard-coded HTML parser rules would be a trivial task compared to the other effort required in getting consensus and decent implementations, so xmlns-style syntactic extensibility is optimising the trivial problem and not the hard ones |
| 10:44 | <Philip`> | I didn't say "yes" because they're not really examples of 'extensibility' - they're just examples of web browsers implementing features in a centralised fashion :-) |
| 10:45 | <Von_Davidicus> | What I said was the -closest- that browsers got, not an actual example. :) |
| 10:46 | <Philip`> | (unless you count the SVG/MathML plugins for IE, which I suppose are examples of real implementation extensibility) |
| 10:46 | <Philip`> | ((though seemingly not hugely successful ones)) |
| 10:46 | <Von_Davidicus> | I hadn't known those existed. |
| 10:47 | <Philip`> | (Adobe SVG Viewer and MathPlayer, I think) |
| 10:49 | <Von_Davidicus> | Oh. Those. Never tried them. |
| 10:51 | <Von_Davidicus> | What the heck was modularity? |
| 10:52 | <Philip`> | As in http://www.w3.org/TR/xhtml-modularization/ ? |
| 11:05 | <Von_Davidicus> | Yeah, what was the whole idea behind that? |
| 11:06 | <Philip`> | As far as I'm aware, some people thought it would be fun to mess around with writing unnecessarily complex DTDs |
| 11:06 | <Philip`> | and that's about it |
| 11:06 | <Ms2ger> | Lots of booze |
| 11:08 | <Von_Davidicus> | Well, if you like *simple* DTDs.... http://www.mrinitialman.com/Experiments/showcode.php?project=ElementML&filename=elementml&filetype=dtd :D |
| 11:08 | <Von_Davidicus> | It was a joke markup language I wrote. |
| 11:10 | <Philip`> | I think the policy here is to dislike all DTDs of any kind whatsoever :-) |
| 11:12 | <Von_Davidicus> | ...BLASPHEMY! |
| 11:13 | <Philip`> | Blasphemy is also a policy here |
| 11:16 | <Von_Davidicus> | How about goofball? Is goofball an allowed policy--specifically, the kind of goofball that would do a website with its own markup language, including DTD and XML Schema? |
| 11:18 | <Philip`> | Such a person might not be a lost cause, if they can be convinced of the folly of DTDs and XSDs |
| 11:18 | <Von_Davidicus> | Proof for the DTD part: http://validator.w3.org/check?uri=http%3A%2F%2Fcoachrandom.zzl.org%2FChronological%2Fcomic-005.xml I had so much fun tinkering with that website. |
| 11:20 | <Von_Davidicus> | DTDs have their place; I still use XML files as databases (though I *do* know SQL). XSDs: 20 miles south of useless, really. There's not much they actually *do* that DTDs don't, and theres a few things they don't that DTDs *do*. |
| 11:26 | <Von_Davidicus> | I probably had *too* much fun, considering that site used both DTD and XSD. |
| 13:04 | <annevk> | http://commandcenter.blogspot.com/2012/04/byte-order-fallacy.html guess he never heard of WebGL :/ |
| 13:30 | <annevk> | http://plan9.bell-labs.com/sys/doc/utf.pdf is pretty interesting |
| 13:31 | <annevk> | apparently the unification in Unicode was done because they tried fitting it all in 16-bit code units |
| 13:37 | <zewt> | heh |
| 13:38 | <zewt> | seems like han unification is the main excuse for the resistance unicode still sees in cjk, or at least japan |
| 13:38 | <zewt> | (though i don't believe it's actually legitimate, more like habit) |
| 13:38 | <annevk> | I think that's mostly stopped now, although the algorithms to determine which font to pick are still not defined |
| 13:39 | <zewt> | i've had that argument relatively recently (in here, if I remember correctly), but I don't know how common it is |
| 13:41 | <annevk> | prolly here with me :) |
| 13:42 | <zewt> | no, with rniwa i think |
| 13:45 | <annevk> | http://krijnhoetmer.nl/irc-logs/whatwg/20111217#l-390 |
| 13:46 | <zewt> | we weren't having the han unification argument there :) |
| 13:46 | <zewt> | http://krijnhoetmer.nl/irc-logs/whatwg/20111201#l-198 |
| 13:47 | <annevk> | oh, I thought we were discussing fonts |
| 13:51 | <zewt> | i mean the argument that "cjk people can't use unicode because something something han unification" |
| 13:52 | <zewt> | which is just false so long as you have language tagging, since you don't lose any information (it'll 1:1 to the legacy encoding) |
| 13:55 | <Philip`> | Is language tagging a less maligned concept than pre-Unicode code pages? |
| 13:55 | <annevk> | language tagging is also not feasible in a number of constructs |
| 13:55 | <zewt> | should be, since unicode depends on it for CJK |
| 13:55 | <annevk> | e.g. title="" |
| 13:56 | <zewt> | should probably do something like use the lang tag of <HTML> for that? |
| 13:56 | <zewt> | (i'd say <body>, except title is in head, so that seems weird) |
| 13:57 | <annevk> | I meant the title attribute, or the alt attribute for that matter |
| 13:58 | <zewt> | use the (effective) lang of the element it's on |
| 13:58 | <annevk> | it has the language of the element it is specified on, but if you mix things you are out of luck |
| 13:59 | <Ms2ger> | If you ever need to make something really obvious, but only for Netscape/Gecko users... |
| 13:59 | <zewt> | iirc there are also language selection control codes in unicode (but nobody will ever know how to use those) |
| 13:59 | <Ms2ger> | <font font-weight=900 point-size=50> |
| 14:00 | <annevk> | zewt: really? |
| 14:00 | <zewt> | i vaguely recall it, but I forget what they're called |
| 14:01 | <zewt> | (and I doubt anyone in the world implements them...) |
| 14:01 | <annevk> | I think Unicode only has bidi characters |
| 14:01 | <annevk> | and emoji flags |
| 14:01 | <zewt> | http://en.wikipedia.org/wiki/Unicode_control_characters#Language_tags |
| 14:02 | <annevk> | deprecated... whatever that means |
| 14:03 | <zewt> | could be made much easier to use with &names; that translate to them (not that it'd be worth it) |
| 14:22 | <Ms2ger> | <a href=foo><font color="red">text</font></a> |
| 14:22 | <Ms2ger> | What color should the underline have? |
| 14:23 | <Philip`> | Ultraviolet |
| 14:23 | <Ms2ger> | Correct |
| 14:24 | <Philip`> | Overexposure to <font> can cause skin cancer |
| 14:29 | <zewt> | Ms2ger: none. modern web pages don't underline links, figuring out what's clickable is supposed to be like playing a sierra game, click everywhere, see what happens |
| 14:29 | <zewt> | (and die if you click the wrong spot) |
| 14:29 | <Ms2ger> | Also correct |
| 14:30 | <Ms2ger> | !summon zcorpan |
| 15:18 | <jgraham> | Ms2ger: could be hard on a Sunday |
| 15:18 | <Ms2ger> | Pff, slacker |
| 15:25 | <jgraham> | Oh my |
| 15:26 | <jgraham> | The ES 5.1 official testsuite uses sync XHR to load the tests |
| 15:36 | <Ms2ger> | I'd fix it, but I'm not an ECMA member |
| 15:39 | <gsnedders> | Anyone know an ECMA member? |
| 15:39 | <Ms2ger> | I don't |
| 15:40 | <gsnedders> | (To be fair, Dave-whose-surname-I-forget tends to fix stuff quickly given bug reports) |
| 15:40 | <karlcow> | https://duckduckgo.com/?q=%22Anyone+know+an+ECMA+member?%22 → No results |
| 15:41 | <Ms2ger> | Herman? |
| 15:41 | <Ms2ger> | Hyatt? |
| 15:41 | <jgraham> | Herman seems more likely in context |
| 15:42 | jgraham | wonders if there are any Daves whoese surname doesn't begin with H |
| 15:42 | <Ms2ger> | Daron? |
| 15:42 | <Ms2ger> | With a B, even |
| 15:43 | <gsnedders> | Fugate. |
| 15:43 | <jgraham> | Does anyone call dbaron Dave? |
| 15:45 | <jgraham> | That just seems wrong |
| 15:45 | <Ms2ger> | It does |
| 15:48 | <odinho> | jgraham: My CORS testsuite uses sync XHR all over the place :P I have a mustfix on that one though. |
| 15:49 | <karlcow> | jgraham: I didn't think about it… but maybe I will start. If he were wrong it would be the surest way of proving it. |
| 15:49 | annevk | has used sync XHR too when sorting out stuff related to encodings |
| 16:06 | <MikeSmith> | so what engine code is Firefox mobile for android built from? |
| 16:07 | <MikeSmith> | I see the http://hg.mozilla.org/mozilla-central/file/tip/mobile/android tree |
| 16:20 | <Ms2ger> | MikeSmith, plain Gecko below that, I think |
| 16:21 | <MikeSmith> | Ms2ger: so that's C++ code while the android app is Java app.. I'm wondering how the Java app gets built from C++ engine code |
| 16:22 | <Ms2ger> | Around http://mxr.mozilla.org/mozilla-central/source/widget/android/AndroidBridge.cpp I guess |
| 16:22 | <MikeSmith> | aha |
| 16:23 | Ms2ger | tries to avoid that stuff |
| 16:25 | <gsnedders> | Probably the same way as a lot of Android apps: a minimal Java app to spawn the C++ code. |
| 16:26 | <Ms2ger> | Not terribly minimal, I'm afraid |
| 16:29 | <smaug____> | The whole UI is done in Java, I think |
| 17:20 | <Hixie> | i just noticed... we never did add a version number to the W3C HTML Canvas 2D Context specification |
| 17:21 | <Hixie> | i wonder why the people so emphatic about not removing version numbers from HTML5 aren't clamouring to add it to other specs... |
| 17:22 | <Ms2ger> | Isn't the "2" a version number?! |
| 17:22 | <Hixie> | hah |
| 17:28 | <karlcow> | Ms2ger: and it is even followed by Canvas 3 |
| 17:29 | <karlcow> | I'm impatient about the Canvas 4, enabling timetravel in Web pages. |
| 17:29 | Ms2ger | wonders how he got 'SUPERSCRIPT MINUS' (U+207B) into this file name |
| 17:39 | <FedorEmelianenko> | hi all |
| 17:43 | <annevk> | Hixie: logic appears to be missing in more places than just the platform |
| 17:43 | <annevk> | (also explains the state of the platform) |
| 17:44 | <zewt> | Hixie: if all change is bad, then adding a version number to them would be just as evil as removing it from html! |