"should we make this only useful for AV streaming" - NO! type='fs' is a godsend, can there be type='eth' or ='net' with multi-protocol support (tcp, udp, dccp, sctp, etc)? that would rip the door open on the issue of websocket falling short in areas.
[Bug 10028] Adding informations about the bitrate of the video to the source element could allow user agent to select the right video in case the same format is available in different levels of quality.
[Bug 10044] Should allow the SCOPE attribute on TD elements so do not have to create a class to override the TH element styles set automatically by browsers.
[Bug 10057] I do not like the "or do nothing" part, as the label is an really good aid for aiming at small form elemnts like a checkbox or radiobutton.
[Bug 10062] "If the attribute is specified, the attribute's value must be the ID of a . . ." No algorithm is given for what it means to "be the ID" of an element.
[Bug 10071] there should be a 'stop()' method or a 'clear()' method to clear the currently playing media because pause() doesnt clear it and setting the src to an empty string or null doesnt work
[Bug 10077] The <embed> should support an alt attribute. While the embedded content should provide for accessilibility of the embedded object, the developer doesn't always have control over those content, some embeddable formats are inherently less accessible, and th
[Bug 10081] Okay, so could you add a note explaining what kind of active Document is *not* "fully active"? After reading very carefully for a few minutes and pondering a bit, I *think* it would be a Document that is the active Document of its browsing context, but w
[Bug 10091] As far as I can tell, nothing specifies what the intrinsic width/height/ratio for iframes should be. You'd assume it's the height/width/ratio of the document, but of course that's wrong unless the iframe is seamless.
[Bug 10092] "The transitive closure of all the browsing contexts that are directly reachable browsing contexts forms a unit of related browsing contexts." Since a browsing context is not a directly reachable browsing context of itself, this means that top-level brow
[Bug 10114] Rather than having inline documentation for external scripts, what would be infinitely more useful would be the contents of a script tag with a src attribute should be executed upon the complete loading and execution of the remote script. Therefore <scrip
[Bug 10133] Use of this element requires passing a contextID string to getContext, but the standard does prescribe either (a) an enumeration of valid contextIDs or (b) a method to enumerat the contextIDs supported by a given UA. (b) should be a requirement of the sta
[Bug 10137] What criteria should browsers use to determine whether a URL's hash fragment differs from the previous one? Should the comparison be case-sensitive? Currently, all implementations of hashchange use case-sensitive comparisons except IE8, which uses a case-
[Bug 10204] The algorithm is unclear here whether we mean the "current node" at this point in time or the "current node" at the time the insertion occurs (which might be different). As an example, consider "<table><p><p>Foo". I think you mean the current element at
[Bug 10223] Presumably button shouldn't appear in this list given that we have a notion of "in button scope" which explicitly adds button to this list...
[Bug 10280] New: "the user agent must associate ... ***before*** inserting it wherever it is to be inserted." - does this imply the insertion might change the form owner by invoking "reset the form owner" algorithm? If so, please make it clearer.
[Bug 10281] New: Confusion bug in space ship game example: "Notice how the hyphenated attribute name becomes capitalized in the API." The word "capitalized" may confuse authors because it usually means something else than camel-cased. I suggest to use "camel-cased" instea
[Bug 10282] New: Typo: "overrride". Your nanny program told me "Please ensure that your comment is descriptive enough that the editor can understand it", so I'm padding out my two-word report with meaningless garbage to trick it.
[Bug 10282] Typo: "overrride". Your nanny program told me "Please ensure that your comment is descriptive enough that the editor can understand it", so I'm padding out my two-word report with meaningless garbage to trick it.
[Bug 10287] New: The definition for an ambiguous ambersand excludes known named character references, but fails to exclude valid decimal or hex numeric character references.
[Bug 10287] The definition for an ambiguous ampersand excludes known named character references, but fails to exclude valid decimal or hex numeric character references.
[Bug 10292] New: For XML documents, attributes from other namespaces (##other) should be allowed, but not validated (lax), on all html elements. This is to allow validating using XML schema for custom attributes (dojo:type, dojo:attach-point)
[Bug 10310] If I have a 100 by 100 pixel image and draw a filled rect of the same size at point {10, 10} the result is the bottom and right 90 pixels followed by the top and left 10 pixels. A way to offset the image so that it draws cleanly would be appreciated. A mo
[Bug 10310] New: If I have a 100 by 100 pixel image and draw a filled rect of the same size at point {10, 10} the result is the bottom and right 90 pixels followed by the top and left 10 pixels. A way to offset the image so that it draws cleanly would be appreciated. A mo
[Bug 10314] New: "in cell" (secondary) insertion mode shows strange behavior for e.g. <table><tr><td><svg><td></td><td>, <table><tr><td><svg><tbody></tbody>
[Bug 10315] New: please allow "command" and "base" in "in head noscript" insertion mode so all void elements that might be handled in head-related insertion modes are treated in the same manner, which would simplify parser implementation a little
[Bug 10331] New: What do "defined to contain a URL" and "defined to contain one or more URLs" mean? There are some cases that are unclear. For instance, useMap can only contain (a subset of) valid URLs AFAICT, does that count? Also, obsolete stuff like applet.src has n
[Bug 10331] What do "defined to contain a URL" and "defined to contain one or more URLs" mean? There are some cases that are unclear. For instance, useMap can only contain (a subset of) valid URLs AFAICT, does that count? Also, obsolete stuff like applet.src has n
[Bug 10332] New: Shipping browsers don't have form.action reflect the DOM attribute exactly. Specifically, if the content attribute is the empty string, they generally return "" on an IDL get, not the current address. This contradicts the spec, but it would be confusing
[Bug 10332] Shipping browsers don't have form.action reflect the DOM attribute exactly. Specifically, if the content attribute is the empty string, they generally return "" on an IDL get, not the current address. This contradicts the spec, but it would be confusing
[Bug 10343] New: width and height are said to be DOMStrings in this interface, but in fact Chrome, Firefox, and Opera all output "number" on this test. data:text/html,<!doctype html><script>alert(typeof document.createElement("video").width);</script>
[Bug 10343] width and height are said to be DOMStrings in this interface, but in fact Chrome, Firefox, and Opera all output "number" on this test. data:text/html,<!doctype html><script>alert(typeof document.createElement("video").width);</script>
[Bug 10344] col.span and colgroup.span should not be limited to only non-negative numbers greater than zero. Chrome, Firefox, Opera, and Safari all agree that this alerts "false" (didn't test in IE because it doesn't support data URLs properly, too lazy): data:text/
[Bug 10344] New: col.span and colgroup.span should not be limited to only non-negative numbers greater than zero. Chrome, Firefox, Opera, and Safari all agree that this alerts "false" (didn't test in IE because it doesn't support data URLs properly, too lazy): data:text/
[Bug 10345] New: hspace and vspace on HTMLImageElement are signed long in DOM 2 HTML. Was there a good reason for the incompatible change to unsigned long?
[Bug 10351] New: """The stack of open elements cannot have both a td and a th element in table scope at the same time, nor can it have neither when the insertion mode is "in cell".""" - the second part of this is wrong in the fragment case
[Bug 10352] New: The range of unsigned long should be defined to be [0, 2147483647], since no browser allows setting to values outside that range. See <https://bugzilla.mozilla.org/show_bug.cgi?id=586118>. In the following test case, Mozilla, Chrome, and Safari output 0
[Bug 10352] The range of unsigned long should be defined to be [0, 2147483647], since no browser allows setting to values outside that range. See <https://bugzilla.mozilla.org/show_bug.cgi?id=586118>. In the following test case, Mozilla, Chrome, and Safari output 0
[Bug 10354] "Zero is not allowed" is unclear as normative text. What happens if you do an IDL get on size for <select size=0>? Does it return 0 or 1? Browsers disagree: Gecko and WebKit return 1, IE and Opera return 0. At least in the case of <select size=0>, ret
[Bug 10354] New: "Zero is not allowed" is unclear as normative text. What happens if you do an IDL get on size for <select size=0>? Does it return 0 or 1? Browsers disagree: Gecko and WebKit return 1, IE and Opera return 0. At least in the case of <select size=0>, ret
[Bug 10355] As far as I can tell, no browser but IE actually implements half of these weird marquee properties. Firefox 4.0b3 and Chrome dev don't implement behavior, bgColor, direction, height, scrollAmount, scrollDelay, trueSpeed, or width. Opera 10.60 doesn't im
[Bug 10355] New: As far as I can tell, no browser but IE actually implements half of these weird marquee properties. Firefox 4.0b3 and Chrome dev don't implement behavior, bgColor, direction, height, scrollAmount, scrollDelay, trueSpeed, or width. Opera 10.60 doesn't im
[Bug 10365] New: popstate event fires after onload (first page visit) -- intended behavior? would be annoying if there was a script that ajax loaded a page at onpopstate, and hand typing in a url, the page is returned by the server, and then onpopstate fires after onload,
[Bug 10370] New: RFC 3339 date-time format, which looks very similar to HTML's date and time strings, gives special semantics to "-00:00" time zone. HTML should either follow the convention or disallow use of "-00:00" time zone.
[Bug 10372] New: Authors should ensure that in each set of radio buttons that one is initially "on". Because some have made bad UI choices you should not recant this rule! At all times, exactly one of the radio buttons in a set is checked. Radio button Only one item
[Bug 10377] New: The form elements (input, select, textarea) seem like they should be included; dynamic data is still "data", and extending machine-readable capabilities to these elements would be a benefit.
[Bug 10380] "the content attribute must be set to the conforming value associated with the state that the attribute would be in if set to the given new value". Does this imply in canonical case? Browsers seem to preserve the case when you do an IDL set. This alert
[Bug 10380] New: "the content attribute must be set to the conforming value associated with the state that the attribute would be in if set to the given new value". Does this imply in canonical case? Browsers seem to preserve the case when you do an IDL set. This alert
[Bug 10383] DOCTYPE token: The text here seems to require lower-case "html" as the node type; i.e. "<!DOCTYPE HTML>" would be invalid. This contradicts Section 10.1.1, which says both are valid (and in fact gives the latter for all its examples).
[Bug 10383] New: DOCTYPE token: The text here seems to require lower-case "html" as the node type; i.e. "<!DOCTYPE HTML>" would be invalid. This contradicts Section 10.1.1, which says both are valid (and in fact gives the latter for all its examples).
[Bug 10388] New: Should the canonical case for GET/POST/PUT/DELETE be lowercase instead of uppercase? IE (8 and 9PP4) seem to treat it as lowercase, and they're the only ones that canonicalize case at all here, as far as I can tell. This alerts "get" in IE8/9PP4 and "Ge
[Bug 10388] Should the canonical case for GET/POST/PUT/DELETE be lowercase instead of uppercase? IE (8 and 9PP4) seem to treat it as lowercase, and they're the only ones that canonicalize case at all here, as far as I can tell. This alerts "get" in IE8/9PP4 and "Ge
[Bug 10410] New: On getting, should the UA percent-decode? Firefox seems to say yes, Chrome/Opera no, at least for the hash. Test case (save and visit with some percent-encoded fragment): data:text/html,<!doctype html><script>alert(document.location.hash);</script>
[Bug 10410] On getting, should the UA percent-decode? Firefox seems to say yes, Chrome/Opera no, at least for the hash. Test case (save and visit with some percent-encoded fragment): data:text/html,<!doctype html><script>alert(document.location.hash);</script>
[Bug 10411] New: Why are all the RFCs referenced in .txt versions? The HTML versions are more convenient (and it's silly for HTML to avoid the HTML versions!).
[Bug 10411] Why are all the RFCs referenced in .txt versions? The HTML versions are more convenient (and it's silly for HTML to avoid the HTML versions!).
[Bug 10415] New: This is the only place where online and offline events are referenced in this section and they are marked as UA text. If authors hide UA text they won't see anything about online/offline events in this section
[Bug 10415] This is the only place where online and offline events are referenced in this section and they are marked as UA text. If authors hide UA text they won't see anything about online/offline events in this section
[Bug 10417] New: the example images for DETAILS (osx finder file info?) ... does this indicate form fields could be marked up with details, using summary for the label? would summary > label be required for accessibility? if so, might be good to include sample code (promo
[Bug 10460] New: Proposal about doctype. I think correct doctype must be <!DOCTYPE html5.0> or <!DOCTYPE html5> or <!DOCTYPE html5.> or <!DOCTYPE html 5>. Not <!DOCTYPE html> because unknown version of HTML. (Eugene.Mirny@gmail.com)
[Bug 10490] New: "If node is not an element with the same tag name as the token, then this is a parse error." <- Tag name comparison must use ASCII case-insensitive match
[Bug 10505] New: A dataTransfer object should be able to be passed to an <input type="file" id="recipient"> object to simplify form processing where ajax is not required or desired.
[Bug 10512] New: im trying to make two buttons, on top of each other, when i click or hover on either one, i want the a blank button with info to appear under the one i clicked on. pushing the other one down, please help. Julio.Avila@live.com
[Bug 10517] New: The size attribute should default to 20. This is what IE8 and WebKit do; Gecko and Presto default to 0 per spec. 20 makes more sense, because all browsers render <input size=20> identically to plain <input>. Test case that alerts default input size and
[Bug 10517] The size attribute should default to 20. This is what IE8 and WebKit do; Gecko and Presto default to 0 per spec. 20 makes more sense, because all browsers render <input size=20> identically to plain <input>. Test case that alerts default input size and
[Bug 7543] Consensus on ietf-http-wg was that repetition of relations is meaningless and we should rather use numbering (e.g. "up2"), names (e.g. "great-grandparent") or an attribute (e.g. level="3").
[Bug 7545] Atom (RFC 4287) defines a "logo" in addition to the "icon": "The image SHOULD have an aspect ratio of 2 (horizontal) to 1 (vertical)". Consider doing the same here.
[Bug 8966] Consider "<svg><desc><div><svg><desc><b>X" â Should we change "Pop elements from the stack of open elements until the current node is in the HTML namespace." in the in-foreign-content mode to actually be "Pop elements until there's no element in scope t
[Bug 9204] The onload event is missing a description on when it fires. Particularly on elements downloading auxiliary documents (img, script, link) it is important to specify when exactly that event is fired. Same for onerror and onabort events (and similar).
[Bug 9576] just say "an ASCII case-insensitive match for "URL"" instead of "either a U+0055 LATIN CAPITAL LETTER U character (U) or a U+0075 LATIN SMALL LETTER U character (u), a U+0052 LATIN CAPITAL LETTER R character (R) or a U+0072 LATIN SMALL LETTER R character
[Bug 9587] It is not clear whether document styles then scoped styles should be applied to the scope or only the scoped styles, ignoring the document styles.
[Bug 9601] ose corners are the four points (sx, sy), (sx+sw, sy), (sx+sw, sy+sh), (sx, sy+sh). If the source rectangle is not entirely within the source image, or if one of the sw or sh arguments is zero, the implementation must raise an INDEX_SIZE_ERR exception.
[Bug 9602] That autofocus attribute will wreak security havok. What an ignorant idea to bring more logic to HTML. I think I know a couple of ways to abuse it, since it actually is some sort of flow control, which only scripting languages should be capable of. I hope
[Bug 9603] If the origin parameter does not match the Document origin the Referer header should also be excluded. Alternatively a way to omit the Referer header would be good for CORS.
[Bug 9629] It says that UAs should provide controls to enable/disable closed captions, but it doesn't define how to load/set the available closed captions or how to render them
[Bug 9643] <rp> should be able to used outside of rubies, for cases where CSS is used to present content. For example, in a chatting application: <span>John Doe<span><rp> [</rp>12:30am<rp>]</rp></span><span> where the inner <span> is floated right and the <rp> is hi
[Bug 9664] Detailing each of the parameters or the type of transformation matrix used or whether or not the matrix transforms the entire canvas state is required for context.transform and context.setTransform entries
[Bug 9690] Redefining elements can cause compatibility issues. What if some browsers read it the old way, and some the new way. This can lead to problems if someone uses the element the new way, only for it to be read the old way, or for old documents to require rew
[Bug 9695] Maybe there could be a function to specify a cache manifest in the API so that instead of the application asking for offline storage when the page loads, some JavaScript can be used to request storage when the user clicks on a "Use Offline Mode" link.
[Bug 9705] The use of the colon in the first sentence (â...within the page: a section with navigation links.â) may be confusing. Suggestion: âwithin the page. It is a section with navigation links.â
[Bug 9711] "The wbr element is expected to override the 'white-space' property and always provide a line-breaking opportunity." -- "In testing, I found that Gecko and WebKit allow <wbr> to override white-space: nowrap, but not white-space: pre." (from http://www.w3.
[Bug 9727] We told people for years that I and B are a bad idea as they are only visual but hint at making things more prominent. Can we just let them die?
[Bug 9733] one reason for having this not restricted to audiovisual streams would be the option of, say, reading a barcode scanning device to fill a shopping cart. This is something you currently need Java or a custom plugin for. rs232 and fs are the perfect thing f
[Bug 9736] Shouldn't be: "create a new subpath whose first point is the same as the previous subpath's last point"??? This whole sentence seems unclear to me reading explanation of "A closed path" in the beginning of this section.
[Bug 9743] I misread this initially as pubdate or pubdate="pubdate" were allowed, then as pubdate="" or pubdate="pubdate" were allowed. providing an example of all three styles would be great
[Bug 9748] There should also be a absolute time, a time from the media playing. For sync issues etc. See <https://bugzilla.mozilla.org/show_bug.cgi?id=566375>
[Bug 9760] "It might not be zero if the clip's timeline is not zero-based" - please make startTime start at 0 even for clips that are not zero-based, to make the API simpler to use for authors. If this change is made, startTime seems to be equivalent to buffered.sta
[Bug 9770] There should be a way to exclude the page that calls the manifest from the cache as well, for those who want to use the cache manifest to speed up a site, but not use it as an offline application.
[Bug 9773] There is not a clear difference between "subtitles" and "captions". These are mostly used to describe "closed captions" (binary transmissions in TV broadcasts) vs "subtitles" (text files), as far as I know.
[Bug 9774] "consecutive lines displayed below each other" - since subtitles tend to be rendered at the bottom, it's actually better for new subtitles to be rendered higher up. Of course, the individual lines of each subtitle still extend downwards, but it's still po
[Bug 9775] "positioned to a multiple of the line dimensions of the first line of the cue" - enforcing the same line height for every line ihurts text rendering appearance for no apparent reason. This is especially true if text styling is supported (bold text is larg
[Bug 9776] Percentages for text positioning don't seem better than using the actual video dimensions. If you require integer percentages, then it doesn't work well for videos larger than 100px high. If not, then it's just an arbitrary resizing of the dimensions (sli
[Bug 9777] "the cues are being displayed over the video as appropriate" - this says "cues", so multiple cues are supported at once. But many SRT players don't actually support more than one SRT subtitle at once - they either play the first one in the file or draw th
[Bug 9784] I think input elements associated with datalists should have a "must-be-in-list" attribute which should cause a validation error when the value is not found in the datalist
[Bug 9785] Add a "filter" attribute. If set, the user agent will filter the datalist. If not, filter won't be filtered automatically (unless you do some sort of filtering by javascript or/and server-side)
[Bug 9800] The wording for the suspend event ("but does not have the entire media resource downloaded") puts it in the same boat as the removed load event. The description should reflect that the event is fired when the networkState changes to the state of NETWORK_I
[Bug 9813] bgsound: Use "audio" instead isn't helpful for back-compat. Proposal: Use "audio" instead, with "embed" fallback for backwards compatibility
[Bug 9814] There are many systems that could use this: barcode readers, rfid readers. So, it would be very useful to let this generic. HTML5 Apps will use it a lot to permit industry automation with web environment
[Bug 9822] We should be allowed multiple CanvasRenderingContext2D Objects for multitouch devices. By allowing only one context object, we can only draw one element at the time
[Bug 9825] Right now the keyword "media" allows audio and/or video data, but why not allow image data as well? It would be useful on many devices (especially phones) to be able to take a still photograph which can then be manipulated or transmitted (assuming the sa
[Bug 9839] "then set the foreign flag to true." - this would be easier to understand if it just said "set /foreign/ to true" since it is not called a flag when it is defined
[Bug 9860] Browsers will need to handle nested microdata. Ie. an event-item containing a location-item and a list of vcard-items of peoples attending the event. give an example so that browser vendors will know about.
[Bug 9880] noticed the comment about per-source mutexes for storage. If that's not desired, an approach that has worked well in other API designs is an application-selectable mutex per context. Thus, each localStorage object could have a "_mutex" property, which by
[Bug 9885] Need to clarify that "User agents must treat plaintext elements in a manner equivalent to pre elements." and other similar lines added in r3502 only refer to the semantics and not the IDL interface or the tag name in the DOM.
[Bug 9917] Often times the first non-blank frames are the least interesting. Poster images should do just that: show the most interesting part to entice the user to watch the video.
[Bug 9924] Apparently a form can be submitted before it is fully loaded (close-tag not yet seen, close-html-tag not yet seen). This is insane behavior, but should be documented.
[Bug 9926] when lots of keywords (tags) you waste lots of bandwidth in html. A meterGROUP looks more usefull where max is the total amount of tags, min is 0 and value is the current instances of the tag. example: <metergroup max="100" min="0" high="60" optimal="80"
[Bug 9938] No API for fullscreen? Really? Can we please have better options other than a totalitarian "you can't do it?" I thought the idea was to be better than Flash. Yet, you still don't allow simple functionality such as that. Tie it to a button. Ask the user's
[Bug 9939] If we have a concept of "WebDevice", manufacturer could start to create amazing web devices (Example: www.poken.com). These devices once connected could interact with the website using a REST PROTOCOL. Event based (The device send information when we do a
[Bug 9942] The sentence "The shadowBlur attribute specifies the size of the blurring effect." is misleading, It should say "The shadowBlur attribute specifies the level of the blurring effect."
[Bug 9947] In the start tag whose tag name is "select" clause, the special behaviour if the insertion mode is "In Column Group" doesn't seem to be possible; that is you can never reach this clause in that insertion mode. Also there is a missing quote before in table
[Bug 9948] In "Any other end tag" the generate implied end tags should probably be excluding the name of the token otherwise we tend to pop everything off the stack
[Bug 9977] It shall be valuable if you add a "forcedownload" link type which forces the client to download the target document in the download window rather than opening the document in the same window of the browser as what is hapening right now
[Bug 9986] add ability to add noscroll 'property' for objects like video. Right now when you scroll and your mouse hovers a video object it stops scrolling because it tries to scroll inside the video tag.
[Bug 9987] Attribute values should be allowed to contain ambiguous ampersands since large amounts of content on the web uses them in URLs and most or all web browsers already interpret them consistently.