W3C home > Mailing lists > Public > whatwg@whatwg.org > April 2007

[whatwg] Target Attribute Values

From: Spartanicus <mk98762@gmail.com>
Date: Sat, 28 Apr 2007 09:18:03 +0100
Message-ID: <n2m-g.mqt533hropcc49cgbps1d7p6t9tebbhg87@4ax.com>
Ian Hickson <ian at hixie.ch> wrote:

>> 2) Afaik currently any attribute value for the target attribute which 
>> hasn't been defined opens a new window. If _blank were made non 
>> conforming authors would imo resort to using non defined names which has 
>> the same result in practice, but which makes filtering such methods out 
>> on the user end much harder.
>
>If people widely blocked _blank, then authors would start using the names 
>anyway. So that doesn't really change anything in practice.

People blocking _blank would indeed be rare, even more so author's
awareness of this, but I'd put the number of authors who test for
conformance as significantly higher. Those authors are imo likely to
look for a conformant alternative that works.

>> I've argued my socks off trying to convince authors that they should 
>> leave opening new windows to users, but there are an awful lot of them 
>> who for various reasons insists on doing just that.
>
>It would be interesting to hear the needs of these authors. Can anyone 
>elaborate? We might well need to re-allow it in the end, I'm curious to 
>hear why people use it.

In the many discussions with authors I've had on this issue, I've not
yet encountered a reason for this practice that stood up to scrutiny,
but regularly the lack of rationale and pointing out the negatives
doesn't stop people from opening new windows despite it having been
pointed out that doing so is bad practice. The most common reasons given
are:

1) I like [often off-site] links opening in a new window, others will
too.
2) When moving to another site, not opening a new window would cause my
site to disappear (sometimes accompanied with the argument that this
would confuse people).

-- 
Spartanicus
Received on Saturday, 28 April 2007 01:18:03 UTC

This archive was generated by hypermail 2.3.1 : Monday, 13 April 2015 23:08:34 UTC