Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

On a touch screen it’s typically something like long press -> Open in New Tab. Or you’re saying that’s beyond most people?


I'm quite confident that you wouldn't have to interview many people to discover that even though it is not beyond people that it is unknown to many.

I know several non-technical people who find the distinction between window and tab difficult to manage for instance. As well as people who have had Android mobiles for years and still don't understand those three little buttons at the bottom of the screen: back, home, 'task' list.


This is the conversation I think we’ve just had:

  d: Represent resources only as standard hypertext documents.
  h: Progressively enhance with nonstandard features. The traditional
     affordances of hypertext will remain available—for example,
     middle clicking to branch browser history.
  k: But screen readers can’t middle click.
  a: Branching is afforded by *some* interface though, right?
  w: Not on touch screens.
  a: Not even by using the touch interface?
  w: Nope. You can’t expect everyone to know how to use it.
So we’ve established that many people are unable to effectively navigate the web using the native features of their web browser. It’s not clear to me how that informs the surrounding debate. Are browsers’ navigation features inadequate? In context that sounds like an argument in favor of websites implementing their own navigation.

Personally I’d prefer us to prioritize building a web of addressable resources rather than applications, and to the extent that our user agents aren’t meeting our needs, improve them rather than work around them.


> to the extent that our user agents aren’t meeting our needs, improve them rather than work around them.

Most of us are in no position to improve our user agents only large organizations like Google or Mozilla can do that. In fact calling them user agents is extremely misleading in most cases.

If they were genuinely user agents they would obey the user not the server. They would assist the user to a much greater degree than they do now; for instance there is no native way of saving the rendered page in either Firefox or Chrome so sharing a page is exceptionally difficult if the URL points at a resource that require logging in and also risks that the resource changes before the recipient can read it. One has to install an add on to do that and hope that the author doesn't get bored of maintaining it or piss off the Chrome Store or whatever it is called.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: