Re: rel="nofollow" attribute (PR#7676)

Mark Birbeck wrote:

> Anyway, there are plenty of solutions better than "nofollow", and ultimately
> it's just daft to put a qualification on a link (which is what @rel is) that
> says the type of this connection is that there is no connection.

[ I removed w3c-html-wg@ crossposting; HTML WG members are supposed to follow ]
[ the public list too anyway                                                  ]

Plenty of solutions eh?

I read everything and its contrary about this "nofollow" value for rel. I'd
like to remind the readers of this thread a few facts:

1. nobody ever came up with a solution for the problem raised by Google, even a
    more complex one.
2. a solution is needed. And even if you disagree, that's a nice and harmless
    try.
3. the W3C's (X)HTML WG has placed HTML 4 in "errata mode" so no evolution of
    the spec, good or bad design, good or bad wording, good or bad idea, is
    likely to happen. Conclusion: you need to add something to HTML 4 and it's
    harmless to web browsers ? Do it!
4. legacy browsers just do NOT care about "nofollow" or any new rel/rev value.
5. when the HTML WG designed the rel/rev attibute, we had in mind only
    object-to-object references. But we're now more than SIX YEARS later.
    It's not totally crazy to say that the spec is outdated, that 2005 brought
    a new interesting server-side usage for this attribute. A standard is not
    something engraved in marble until the end of the universe. It's a living
    object, that needs corrections, extensions, with extinct portions and new
    proposals. With that in mind, and with my former HTML(4) WG member hat on, I
    see _nothing_ wrong in Google's extension.
6. blink or marquee are not only "poorly designed (...) elements" as Alexander
    Savenkov said , they are elements used by MILLIONS of pages in asian
    countries. We would all love to have in our products' basket a poorly
    designed object used by millions.
7. the word "nofollow"... aaaahhhh.... Keepers of The Temple, raise your hands.
    Think a few seconds about it : you're screaming for a word that you say does
    not fully represent what it means, added to a spec where A is at the same
    time the source and the target of a link, where LINK represents the same but
    inactively, where ADDRESS is not an address, where a list can be
    instantiated in two different element types, where there's a totally useless
    and semanticless element called BODY, etc. Please, give us a break about
    "nofollow"... Fantasai said that assigning a poorly-chosen name can lead to
    confusion. Because you think the vast majority of the world speaks english
    and will see a confusion where YOU, as an english speaker, see one? I am not
    a native english speaker, and trust me, Google could have used rel="foobar"
    that I would still find it clever - and funny :-)

Please, all, don't be more royalist than the queen.

</Daniel>

Received on Monday, 24 January 2005 09:04:03 UTC