[whatwg] Proposal: target="_tab"

João Eiras joao.eiras at gmail.com
Fri Jun 13 01:03:58 PDT 2008


2008/6/13 Borek Bernard <borekbe at yahoo.co.uk>:
> Hi João,
>
> you're right that everything important has been already said. I
> have withdrawn this proposal because it has been pointed out that it
> is not backwards compatible and the correct solution will be part of
> CSS3 anyway (which is much more flexible - we will have not only
> target-new, but also target-position; I guess you strongly dislike
> both of them).

What's lovely about css, is that features like this can be easily
disabled with local stylesheet. Overriding _tab would requiring
running local script, which greater performance impact, and migh thave
unforseen consequences. So css gives greater control, with less
effort.

>
> But the discussion has been interesting anyway. There is probably no
> point in carrying on because we see the problem from two different
> standpoints - you want to have the specs as "pure" as possible while I
> want them to be as flexible as possible so that it can accommodate any
> use case you can think of.

It's not about being pure, it's about not giving more control to the
webpage than it should have. For a webpage running in a tab or
separate window is exactly the same thing.


> I kind of understand why simpler standards are better than the longer
> ones but on the other hand, the lack of "_tab" or something similar
> makes my user experience on some websites suboptimal (heck, even
> frustrating sometimes). If you can't appreciate that different users
> can have different user preferences ("I can't honestly come up with a
> single reason why a webpage should open a new window instead of tab"),

This is not a user preference. It's the complete opposite. The spec
would allow a user preference to be broken by spaning windows or tabs
accordingly to the webpage author's liking.
Historically, we've seen that giving webpage control over the user's
browser in someway can be abused: alert(), open(), oncontextmenu ...

> you will probably have hard times understanding my points. But your
> view is quite common as I've learned :)



> Regards,
> Borek
>
>
>
> ----- Original Message ----
> From: João Eiras <joao.eiras at gmail.com>
> To: Borek Bernard <borekbe at yahoo.co.uk>; whatwg at lists.whatwg.org
> Sent: Thursday, 12 June, 2008 7:17:19 PM
> Subject: Re: [whatwg] Proposal: target="_tab"
>
> Hi ! I didn't saw that reply.
>
>> I'm not sure why you keep insisting that it's up to the browser -- IMO, it's
>> up to the USER.
>
> You're not understanding me:
> when I say browser, obviously I mean client, client-side, browser,
> user or whatever you want to call it, as opposed to web application or
> server-side
>
>> Also, having means to open new tabs as opposed to new windows in the specs
>> is nothing against the user preference, in fact, it helps to express the
>> user preference if the browser fails to provide it.
>
> Then we're going to bloat a specification due to browser idiosyncrasies ?
> Allowing a page to control such behavior would be bad. Currently
> browsers with tabbed interface manage to unclutter the taskbar and
> desktop, while aggregate pages inside a single window, which is
> overall good for the user's experience, good for performance, good
> usability.
>
> We'd be providing a mechanism that is not backwards compatible with
> the current state of the art user agents, although we've seen that new
> features heavily demanded get implemented quickly, and we'd be
> providing, again, authors with mechanism to degrade the user's
> experience.
>
> I can't honestly come up with a single reason why a webpage should
> open a new window instead of tab. All use cases you can come up fit
> better if new tabs are open. If you don't like the fact that a tab
> fits the entire window, you can either detach it, your use a user
> agent that support MDI interface.
>
> With all this say now your going to tell me I'm contradicting myself
> by supporting windows now. No, you'd be wrong. I'd expect a browser to
> always open tabs if there's a _blank target. Having _target and_tab
> would require UA's to support two different way of opening new pages:
> tabbed and detached ones.
>
> For me it's all a matter of letting the user control the web page.
>
> Considering this discussion is still going to last a bit, and
> everything significant that could be said by me and others was said, I
> rest my case.
>
> Cheers.
>
>
> 2008/6/12 Kristof Zelechovski <giecrilj at stegny.2a.pl>:
>> Programmatically controlling the containment of a new window is a two-edged
>> sword: you can provide for the lame user agent but you can also override
>> user settings.  The latter possibility is more painful; upgrading the
>> browser is easier than dealing with an impertinent Web site.
>> IMHO,
>> Chris
>> P.S.: If you want your answer to go to João only, just send it exclusively
>> to him.
>>
>> -----Original Message-----
>> From: Borek Bernard [mailto:borekbe at yahoo.co.uk]
>> Sent: Thursday, June 12, 2008 11:25 AM
>> To: Joao Eiras; Kristof Zelechovski; Ian Hickson; whatwg at lists.whatwg.org
>> Subject: Re: [whatwg] Proposal: target="_tab"
>>
>> Hi João,
>>
>> I'm not sure why you keep insisting that it's up to the browser -- IMO, it's
>> up to the USER. Please read all my arguments before, it's not true that a
>> user using a tabbed browser always prefers opening new tabs instead of new
>> windows. That's just your user preference.
>>
>> Also, having means to open new tabs as opposed to new windows in the specs
>> is nothing against the user preference, in fact, it helps to express the
>> user preference if the browser fails to provide it.
>>
>>
>>
>>
>
>
>
>      __________________________________________________________
> Sent from Yahoo! Mail.
> A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html
>



More information about the whatwg mailing list