02:44
<Rastus_Vernon>
Could by any change validator.nu happen to be down?
02:44
<Rastus_Vernon>
http://downforeveryoneorjustme.com/validator.nu
02:52
<MikeSmith>
Rastus_Vernon: it's been down a few times recently. there's a mirror at http://sideshowbarker.net:8888/
02:53
<MikeSmith>
Rastus_Vernon: http://validator.w3.org/nu/ runs the same backend
02:55
<Rastus_Vernon>
MikeSmith: I believe the Unicorn validator uses validator.nu as well for the validation of HTML5 documents.
02:56
<MikeSmith>
Rastus_Vernon: I guess Unicorn does but I don't use it so I wouldn't know.
02:58
<MikeSmith>
Rastus_Vernon: btw the UI at http://validator.w3.org/nu/ has features that aren't available elsewhere. It lets you filter out messages persistently and it also can show you an outline that conforms to the HTML(5) spec
02:59
<Rastus_Vernon>
MikeSmith: I believe validator.nu can give such outlines as well.
02:59
<MikeSmith>
Rastus_Vernon: also http://validator.w3.org/nu/ is pretty much always the most up to date these days. I push changes to it weekly
02:59
<MikeSmith>
Rastus_Vernon: validator.nu does not show outlines yet
02:59
<Rastus_Vernon>
MikeSmith: I’m sure it does.
03:00
<Rastus_Vernon>
In fact, let’s be honest, here: I couldn’t be more confident that validator.nu shows outlines because I’ve actually used them regularly.
03:00
<Rastus_Vernon>
However, it’s great to know that validator.w3.org/nu is more up to date. I’ll use it, in that case.
03:01
<MikeSmith>
well the only way that it would be showing outlines at this point is if I actually updated the UI source for it myself and checked it in
03:01
<MikeSmith>
which I guess I must have, and just forgotten
03:01
<MikeSmith>
http://sideshowbarker.net:8888/
03:01
<MikeSmith>
should be the same
03:01
<Rastus_Vernon>
Well, I know there’s a “Show Outline” checkbox.
03:01
<Rastus_Vernon>
And that shows a neat outline.
03:01
<MikeSmith>
yeah I see that it does have it
03:02
<Rastus_Vernon>
I find it very useful to verify that my table of contents is actually complete.
03:02
<MikeSmith>
so yeah added it seems
03:02
MikeSmith
pats himself on the back
03:02
<MikeSmith>
Rastus_Vernon: I'm glad you're using the outline thing
03:02
<Rastus_Vernon>
Well, it’s useful.
03:02
<MikeSmith>
Rastus_Vernon: I think you may be the first person who's ever told me they're using it
03:03
<MikeSmith>
Rastus_Vernon: yeah, helping you catch those kind of issues is what I had in mind when I added it
03:03
<Rastus_Vernon>
I use section elements, so I more or less need it.
03:03
<MikeSmith>
ok
03:03
<MikeSmith>
if you find any problems with it, please file bugs
03:04
<Rastus_Vernon>
MikeSmith: I use this CSS code to generate my table of contents: https://hastebin.com/ixegubafaf.css
03:04
<MikeSmith>
it's not 100% in conformance with the current spec, because Hixie made a change to the ouline algorithm a while back that I still need to implement. but that change was for a corner-case I think
03:05
MikeSmith
looks at Rastus_Vernon CSS
03:05
<Rastus_Vernon>
The actual HTML code is this: https://hastebin.com/tisacekeja.xml
03:05
<MikeSmith>
hmm hastebin has a bad cert
03:05
<Rastus_Vernon>
I use Prince XML because it’s the only user agent that supports the kind of stuff I’m doing.
03:05
<Rastus_Vernon>
Oh, sorry, remove the s in https
03:06
<MikeSmith>
ah you're using nav I see
03:06
<Rastus_Vernon>
All that CSS is valid according to the cgpm specification, but pretty much no browser supports it.
03:06
<Rastus_Vernon>
The advantage is that I just need to add ids like id="1.5" to header elements and I let the CSS get the name and page number by itself
03:06
<MikeSmith>
ok
03:06
<Rastus_Vernon>
It generales a very beautiful result too.
03:07
<Rastus_Vernon>
But I need to add some HTML manually, and thus I use the outline to make sure I left nothing out.
03:07
<MikeSmith>
cool
03:08
<Rastus_Vernon>
And since I frequently make tag mistakes without noticing, the validator tells me about them and I fix them. It is, for me, the equivalent of a spell checker, but for tag soup instead of spelling mistakes.
03:08
<MikeSmith>
yeah that's what it's intended for
03:09
<MikeSmith>
it's really more of a linter than a validator -- except that the linting rules or normatively defined in a specification
03:09
<MikeSmith>
it's meant to help you catch mistakes you wouldn't have caught otherwise
03:10
<MikeSmith>
rather than telling you you're right or wrong
03:12
<Rastus_Vernon>
It’d be really useful to have CSS selectors for selecting headers of a certain level, because it’s currently impossible to get it right with CSS.
03:12
<MikeSmith>
I guess you'd want to e-mail www-style about that
03:12
<Rastus_Vernon>
Of course. Although I’m sure selectors level 4 already has something for it (it has something for everything).
03:26
<MikeSmith>
Rastus_Vernon: yeah I don't keep up with selectors 4 much. I just take it on faith that it's in good hands
03:28
MikeSmith
finds http://drafts.csswg.org/selectors/
03:30
<MikeSmith>
TabAtkins: it'd be nice if CSS specs had link somewhere to version-control commit history, as whatwg specs have these days
03:30
<TabAtkins>
That should probably be possible to automate.
03:30
<MikeSmith>
yeah
03:30
<TabAtkins>
Mind filling an issue on Bikeshed?
03:31
<MikeSmith>
will do
03:31
<TabAtkins>
Thanks.
03:35
<zewt>
an amusing way to cause the whatwg server to combust: automatic git blame when viewing the spec
09:22
<IZh>
Hixie: Hi. Consider following paragraph from "Image maps" -> "Processing model": "If the shape attribute represents the rectangle state, and the second number in the list is numerically less than the fourth number in the list, then swap those two numbers around."
09:25
<IZh>
Hixie: Isn't the condition inverted? According to "The area element": " In the rectangle state, area elements must have a coords attribute with exactly four integers, the first of which must be less than the third, and the second of which must be less than the fourth."
09:27
<IZh>
Hixie: I suspect that "less than" should be changed to "greater than" in the comparison condition in "Processing model".
09:46
<IZh>
Hixie: And what you think about adding support for floating-point coordinates for shapes. These coords would be in the interval of 0.0 to 1.0, and this will work for any resolution of image and any zoom level.
10:40
<IZh>
Hixie: What's the purpose of allowing circles with zero radius that will be considered empty and eleminated at the
10:40
<IZh>
Hixie: later steps of parsing.
15:27
<mathiasbynens>
hsivonen: http://html5.validator.nu/ is down again
16:48
<MikeSmith>
dglazkov: I don't really understand your response at https://twitter.com/dglazkov/status/472764490748919808
16:50
<caitp>
it's just a value to check, totally non-normative, it's cool if different vendors check different things
16:53
<MikeSmith>
caitp: what different things? different attribute names other than "is"?
16:55
<caitp>
I'm not being serious, I really hope it doesn't come to that
16:55
<caitp>
that would be terrible
17:11
<MikeSmith>
caitp: terrible is just a synonym for "work in progress"
17:12
<caitp>
sure, but if people start shipping terrible, it becomes harder to make it less terrible =(
20:33
<erlehmann>
is there an implementation of the sortable table thing?
20:47
<IZh>
Mmm... Sortable grid view. That would be great.