- From: Aurélien Lheureux <contact@aurelienlheureux.com>
- Date: Mon, 16 Mar 2015 19:15:34 +0100
- To: public-html-comments@w3.org
Hi, With your explaination I would not against the rel attribute. There is a chance that this request is taken into account one day ? Aurelien Le 10/03/2015 21:14, Michael A. Peters a écrit : > sitemaps should be dynamically updated or they go stale quickly. > > But back to rel - I know it is currently used for relationship with > external document, but reality is, it's instructions to indexers. > > things like rel="nofollow" are so specific to search engines they > cause confusion to some users who don't want to use the attribute > because they psychologically want their users to follow it (e.g. > affiliate marketing) and the phrase "nofollow" is counter to that. > > I don't think a second attribute to instruct search engines is needed, > just expand the definition of rel to also encompass the relationship > of the node to the document when it isn't a node that generally links > to an external source (such as a[nchor] and link) > > With the source element, the meaning of the attribute (and what > attributes are legal) changes depending upon the parent node, so there > is precedence for context influence attribute meaning already. > > On 03/10/2015 01:03 PM, contact@aurelienlheureux.com wrote: >> @Andrea : >> Could you give me an exemple of code ? but this way need to specify >> height and width ? The advantage of image tag is the autodetection of >> these properties. >> >> e.g. I have a character in my html page animated with jquery and i want >> that the page to be responsive and i can't use breakpoint. >> I have just to specified percentage width and let the height adjust >> itself. >> When I resize the window, the character resizes itself, and jquery >> calculates news coordinate for the animation. >> >> Anyway, aside from my example that was just here to bring my thoughts, I >> supposed that others persons had maybe encountered this problems of some >> unwanted indexed images. >> >> By the way, concerning the intellectual property, I agree with Michael. >> You can reproduce an illegal situation of a copying a book on a html >> page, use the noindex meta tag and diffuses it even if you are not the >> author, it is probably illegal, but you can do it. >> >> @Michael >> Rel attribute is used for a relationship with another external document >> from a source to a target, but considering that an image is already >> included in the content, this would make it to do a relationship from >> the source to the source, do not you think ? >> >> A sitemap image node could be indeed another approach. >> The subtilitie is that with an attribute you can change the src of image >> without impact the fact that you don't want that this image particulary >> is indexed. Moreover with a dynamic content displayed with a loop, you >> can specify this attribute without take into account the image source. >> Finally you can apply the non indexation on a fully range. >> >> Whereas the sitemap need to be continually updated with the sources >> unwanted. >> >> >> >> Le 10/03/2015 13:32, Michael A. Peters a écrit : >>> What about something like rel="nofollow" ?? >>> >>> That's what we do when we don't a link to be construed as related to >>> our page content itself (e.g. an advertisement) >>> >>> But perhaps the right place for this isn't in the html itself but in >>> the sitemap file - maybe an image:priority child of the image:image >>> node. >>> >>> As far as intellectual property implications : >>> >>> A) It is not the job of HTML itself to protect intellectual property >>> >>> B) If it isn't being indexed in the image search, then it may still be >>> an intellectual property violation, but at least isn't one that is >>> bringing new business to the site away from the real intellectual >>> property owner. >>> >>> On 03/10/2015 03:42 AM, Andrea Rendine wrote: >>>> Hi Aurélien >>>> I haven't tested any case yet, but I usually prefer to maintain >>>> graphic >>>> elements and page content separated in my pages too. >>>> Now, it depends on what you mean by "no choice but to use an image >>>> instead of ...css". >>>> I incorporate graphical elements in form of transparent 1px gif images >>>> and use the "real" graphic as a CSS background element. >>>> Then I wrap all the content which is to be replaced by the image in an >>>> element with display:none, so that only the graphic appears. >>>> Would this solution help you? >>>> BTW, I don't think that such an attribute could be implemented. >>>> As a very basic caveat, consider the malicious use that could be >>>> done of >>>> such an attribute, hiding potentially illegal content present in the >>>> page. >>>> Regards, >>>> Andrea >> >> >> Le 10.03.2015 13:32, Michael A. Peters a écrit : >>> What about something like rel="nofollow" ?? >>> >>> That's what we do when we don't a link to be construed as related to >>> our page content itself (e.g. an advertisement) >>> >>> But perhaps the right place for this isn't in the html itself but in >>> the sitemap file - maybe an image:priority child of the image:image >>> node. >>> >>> As far as intellectual property implications : >>> >>> A) It is not the job of HTML itself to protect intellectual property >>> >>> B) If it isn't being indexed in the image search, then it may still be >>> an intellectual property violation, but at least isn't one that is >>> bringing new business to the site away from the real intellectual >>> property owner. >>> >>> On 03/10/2015 03:42 AM, Andrea Rendine wrote: >>>> Hi Aurélien >>>> I haven't tested any case yet, but I usually prefer to maintain >>>> graphic >>>> elements and page content separated in my pages too. >>>> Now, it depends on what you mean by "no choice but to use an image >>>> instead of ...css". >>>> I incorporate graphical elements in form of transparent 1px gif images >>>> and use the "real" graphic as a CSS background element. >>>> Then I wrap all the content which is to be replaced by the image in an >>>> element with display:none, so that only the graphic appears. >>>> Would this solution help you? >>>> BTW, I don't think that such an attribute could be implemented. >>>> As a very basic caveat, consider the malicious use that could be >>>> done of >>>> such an attribute, hiding potentially illegal content present in the >>>> page. >>>> Regards, >>>> Andrea >
Received on Monday, 16 March 2015 18:16:04 UTC