tag:blogger.com,1999:blog-17809600565735611.post5706841311762001112..comments2023-03-25T12:39:46.494+01:00Comments on Itpastorn's Webdev + Thinkpad update & maintenance blog: Validation and doctype myths and (inconvenient) truthsLars Gunther (itpastorn)http://www.blogger.com/profile/11544012919049072827noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-17809600565735611.post-63798230740856091082009-08-19T00:40:00.128+02:002009-08-19T00:40:00.128+02:00@Anonymous:
My point exactly. The doctype does no...@Anonymous:<br /><br />My point exactly. The doctype does not affect anything of importance except styling in a browser.Lars Gunther (itpastorn)https://www.blogger.com/profile/11544012919049072827noreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-83900839243727151202009-08-18T23:46:31.931+02:002009-08-18T23:46:31.931+02:00There are certain elements that display differentl...There are certain elements that display differently by default depending on the doctype, such as the line height of an image element in XHTML Transitional or XHTML strict.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-2449167016121624052009-07-13T10:59:01.055+02:002009-07-13T10:59:01.055+02:00OK, I have re-phrased my post, to make it more acc...OK, I have re-phrased my post, to make it more accurate. Many thanks for your input.Lars Gunther (itpastorn)https://www.blogger.com/profile/11544012919049072827noreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-28199544594896177882009-07-13T10:26:29.368+02:002009-07-13T10:26:29.368+02:00My point was that your step 1 and step 2 are reall...My point was that your step 1 and step 2 are really the same step: the parser builds the DOM by reading the network stream. The second step would be rendering the DOM+CSS.zcorpanhttps://www.blogger.com/profile/02707063734332204359noreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-55998649603557770862009-06-28T19:51:00.300+02:002009-06-28T19:51:00.300+02:00This comment is from Dean Edridge and was sent to ...This comment is from Dean Edridge and was sent to my by e-mail:<br /><br />Good article, just one minor correction. There is no such thing as the "HTML5 doctype", a lot of people refer to it as the HTML5 doctype, but it's really just a generic doctype for (X)HTML. This new doctype for (X)HTML that was introduced in the (X)HTML5 spec "<!DOCTYPE html>", is simply called the "HTML doctype", (or I guess you could call it the "(X)HTML doctype") as it is not exclusively for just one language like doctypes of the past were (such as HTML4 Strict, XHTML1 Transistional). The "HTML doctype" will be used for all HTML and XHTML flavors from now on, such as: HTML6, XHTML6, HTML7, XHTML7 etc.Lars Gunther (itpastorn)https://www.blogger.com/profile/11544012919049072827noreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-77880265003892897472009-06-28T12:08:27.608+02:002009-06-28T12:08:27.608+02:00Zcorpan: I was trying to explain this from a web d...Zcorpan: I was trying to explain this from a web developer perspective. Simplification was necessary. Perhaps I am using the word "parsing" to narrow. You are describing it as involving everything, including painting "boxes on the screen"!<br /><br />I am basing my thoughts primarily on remarks by David Baron (in a talk at Google) and Henri Sivonen (why HTML is just as speedy as XHTML: "Parsing is cheap"). Right now he is building a HTML 5 parser for Mozilla, and as I comprehend his work (not having looked at the source code), it corresponds to my first step above.<br /><br />If I should use another word for the actual process of "reading the code", what would you recommend?<br /><br />Swenglish? Hoppsan! Fixed.Lars Gunther (itpastorn)https://www.blogger.com/profile/11544012919049072827noreply@blogger.comtag:blogger.com,1999:blog-17809600565735611.post-66180406074918124982009-06-27T21:13:35.683+02:002009-06-27T21:13:35.683+02:00> Building the DOM is the second step.
Buildin...> Building the DOM is the second step.<br /><br />Building the DOM is what the parser does, so should probably be part of the first step. :-)<br /><br />> It involves the simple mapping of the HTML markup, but also the applying of CSS, handling events, etc.<br /><br />Applying of CSS is part of the rendering step (which itself is a series of steps: parse the style sheet, match elements against selectors, build up the rendering tree based on the value of 'display' for each element, lay out the boxes of the rendering tree, paint those boxes on the screen).<br /><br />On effects of validation: it might be worth drawing the analogy to spell-checking here. You run a spell-checker to find typos and grammatical mistakes. A validator will similarly find typos and other mistakes.<br /><br />> No you must not.<br /><br />Swenglish? :-) Certainly it is allowed to validate.zcorpanhttps://www.blogger.com/profile/02707063734332204359noreply@blogger.com