| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12
Comments:
<0> meh...
<0> everyone's broken sites will display as broken
<1> I'm Looking for style opinions.
<0> proper web authoring would prevent a new user agent "breaking the web"
<0> broken web authoring will be revealed if IE7 "breaks the web"
<2> bewest, except that there is no "proper web authoring"
<3> bewest: yes. Now to discuss whether that is a GOOD thing or not.
<1> http://dotsandloops.net/headlines/tags/culture,news
<0> Windrose: hmmm well every system has its constraints
<1> The "Links" section in the sidebar ... it's supposed to allow you to look up those tags on other websites, but it's ugly as ****, isn't it ?
<0> Windrose: it could be that by makign the web's requirement of "working everywhere" encomp*** such a large scope that it's not a good thing
<0> Windrose: but I think it probably /is/ a good thing
<4> bewest: what is 'proper web authoring'?
<0> here's another thing that "breaks the web"
<2> bewest, if we all followed standards exactly IE wouldn't work, even with conditional comments we still end up with broken pages when Trident changes and the "fix" no longer works, or breaks what we were trying to fix in the first place
<0> www.myspacetoolbox.com
<0> that site's business model breaks the web
<3> bewest: I don't know. It isn't very good for the users. I'm torn.
<0> Zeros: umm... the web isn't about how things look...
<0> Zeros: proper web authorship includes stewardship to make sure that a representation of a resource is always viable
<0> not that it always appears the same
<1> I'll just trust myself on this one I guess. It's ugly. :)
<0> so coding your site for looks is inherently a weak authorship practice
<4> bewest: so... "am i doing it right?"
<2> you didn't qualify what "proper web authoring" meant bewest, I had to infer it
<0> poignantly demonstrated by IE7's release "breaking the web"
<2> Either way, no matter what we say the web is about looks
<0> it's not breaking the web... it's revealing weak authorship
<5> Zeros: I'm aware I can't modify the contents with CSS, even if I can use display:none if I really need too
<4> bewest: isnt it revealing weak browser authorship?
<0> no
<0> when you code hacks into your site
<0> that is weak authorhsip
<4> bewest: sometimes there's no other way
<2> so we should just ignore IE?
<0> anyway have a look at myspacetoolbox.com
<4> bewest: we can't just ignore IE
<4> i already looked
<0> you see what they do?
<4> yes
<0> they provide bits of HTML for users to copy/paste into myspace
<4> yes
<0> hidden in the HTML
<0> are paid advertisements
<0> so they'll generate html to display an image
<0> this is very helpful to the myspace crowd, I imagine
<0> but in the HTML is a link to reebok sneakers
<2> sure
<4> interesting
<0> where reebok is paying for inbound links
<0> very dangerous to web architecture
<5> Zeros: I'm looking at csszengarden.com and I see I can in fact do a lot with just CSS
<2> bewest, I don't see how thats dangerous
<2> advertisements don't _have_ to relate to the contents of the page
<4> retro_neo: note that their html is poor
<0> Zeros: but the ad is embedded in the content
<2> so what?
<0> as in "I'd like to put a funny rabbit on my site"
<4> how about, i think Zeros is http://c.myspace.com/Groups/00007/34/70/7070743_l.gif
<0> ok, so this place provides the HTML for the funny rabbit
<4> :P
<0> you copy/paste the funny rabbit onto your site
<0> but the image is actually a link to reebok
<4> bewest: thats not really much different from the adware model
<0> and the alt text and all meta data says reebok
<0> it's very different
<0> so an agent processing this page
<0> will look at this picture
<0> and say "this is reebok"
<0> because the meta data is describing reebok
<4> right...
<0> in reality it's a funny bunny with a pancake on its head
<0> but that's just where it starts
<0> the next step is when someone else wants to do the same thing
<4> bewest: but myspace users "Have no idea what you're talking about" so...
<0> and they put different alt text
<0> hax: has nothing to do with myspace useres
<0> hax: I'm saying this company's business practice is harmful to web architecture
<4> bewest: yes
<4> bewest: i agree with you
<4> bewest: this kind of cancer can't be removed from the net though
<0> well
<0> you say that now
<0> we'll see :-)
<2> bewest, I don't think that has anything to do with the web architecture
<4> Zeros: it does
<2> I don't think so, the web architecture is a hardware thing
<0> umm
<4> what?
<2> that's like saying that the elevator that says "going down" when its really moving up in a building is harmful to the foundation of the building
<4> Zeros: you win the award for worst analogy ever
<2> heh
<2> its bad for one page, myspace, and myspace's HTML is horrible anyway
<4> its a bad practice
<2> sure
<2> but it doesn't relate to the architecture of the web at all
<4> if myspace is a succesful business and they do it
<4> others will do it
<2> speculation
<0> Zeros: it's exactly web architecture
<0> http://www.w3.org/2001/tag/ doesn't mention anything about hardware
<0> Zeros: the architecture is more about nodes and graphs than hardware
<0> Zeros: what I'm saying is if you think of the web in terms of graphs and nodes, this specific practice breaks how nodes connect to one another under the web model
<2> does the funny bunny link to rebok too?
<0> funny bunny is just an image
<0> some bloke wants it on his website
<0> but doesn't know html
<2> so he copy and pastes it, sure
<0> so this bloke copies and pastes from another site that generates the html
<0> this other site embeds the image into a link that goes to reebok, because reebok is paying for these links
<0> all the meta data says that this image is somehow connected to reebok
<0> when in fact it is totally irrelevant
<4> http://www.greenbtn.com/bunny.jpg
<4> thats the bunny, for reference
<2> thanks hax :P
<4> just trying to help
<0> yes :-)
<2> ok so the link is misleading visually
<0> it's worse than misleading
<4> it breaks the semantics of the page
<4> and therefore breaks the web
<0> it breaks the way web nodes are supposed to link together
<2> no its not, the link goes to reebok, the alt text references reebok
<4> its subversive
<4> Zeros: but the picture does not
<2> only the visual image is a bunny
<2> sure
<0> and the page does not
<0> and the image has nothing to do with reebok
<4> Zeros: do you agree that its misleading?
<2> but the node is reebok node, it links to a reebok page
<2> the web is fine
<0> hax: he said it's misleading
<4> Zeros: what about if you're google image search?
<0> Zeros: no, there are 3 nodes in question
<0> hax: correct
<0> hax: if you google imgae search for bunnies
<0> hax: reebok will come up
<4> no
<0> hax: and like... wtf...
<4> other way
<0> both ways
<0> :-)
<0> but yeah
<4> how would google know what a bunny looks like?
<4> thats silly
<0> hax: that's how this came up
<4> no, what i'm saying is
<0> hax: we started getting some funny results in an image search we are developing
<4> bewest: heh
<2> hax, alt text is rarely descriptive enough to let google really know whats in an image
Return to
#web
or
Go to some related
logs:
#math
#math
#ubuntu
Firetech filesharing
PHP Domain name required for sender address
easyubuntu no-network
#python
postfix catchall-address and virtual alias
#lisp
#ubuntu
|
|
