14 comments

[ 3.4 ms ] story [ 46.4 ms ] thread
I don't know what will come out of this, but I do think that experimenting with different ways of viewing the web, and seeing what does and doesn't work is vital for making things easier to use.

If we don't push the boundaries far enough to fail then we never find the unlikely things that work.

I don't think this article accurately represents the scope of Pancake's ambitions. Optionally hiding the url bar is a fairly trivial UI consideration. Clicking through to the Google group shows a more interesting (if a bit superficial) exchange about visualisation of the web's underlying graph.

That sounds a lot more interesting to me, but still pretty superficial. I hope they're exploring more fundamental and more drastic changes. Like how about a distributed commenting system, in which blog posts can be tagged as responses to each other and the browser does the work of collecting responses. The browser could then display a tree of responses (both up and down) in a manipulable way, i.e. a user could control threadedness, collapsedness, sort order (based on social network?), etc.[0]

Or how about a better navigation paradigm? (If I were to have my way, a strictly keyboard based one. [1])

It seems like the Pancake team is in a unique position to dream big. I hope they do.

[0]: http://blog.byjoemoon.com/post/3357516295/end-of-comments-ii...

[1]: http://blog.byjoemoon.com/post/9325300749/a-different-kind-o...

Just read the small blurb, and it doesn't really give much away. I do think Mozilla are right to challenge current paradigms, and I long for a bit of innovation.

The browser UI needs a chronic refresh, and thinking about what else the browser can do, is certainly worthy of consideration.

I personally would like to see the browser take on the challenge of navigation better.

I was confined to a 133MHZ text based machine for six months - and I happily learnt to navigate the web with Lynx - by using its keyboard shortcut-ted numbered links! It was amazing how easy it all felt, and it's something I've longed to see as an option in a browser ever since. This one thing alone would make using my Kindle's browser - bearable.

It's an interesting point, and I do think making browsers more keyboard-centric would make them more accessible, too.

As an interim solution, I use the Vimium extension for Chrome. You can hit 'f' to open in current tab and all the links will have short key sequences appear next to them.

I have used the keyboard navigation extension with Chrome, not sure if it's the same one.

The Chrome extension works well by default for a touch typist by setting the shortcuts to be available by typing two letter key codes, but you can change that from what I remember to numbers too.

Give Lynx and W3m a whirl they are both worth a look in. Lynx's numbered links mode is fantastic, to set it, you need to go to options>keypad mode.

The other browser, that makes link navigation a lot easier is Opera, use shift and arrows, to move around the links.

I love the above interface so much, it's so simple, and could easily be mimicked with a trackball and a touch pad such as that on the Blackberry. The Blackberry utilises a similar method for it's nav and menus and it's very easy, quick and intuitive to use.

Of course you could place the directions on the home row, for touch typists too. But I guess you can do something similar with Vimperator etc.

I read your blog post, and firsts thoughts were that shift and arrows could be used for OS UI navigation as well. Tiling window managers work well for me, but I have a lot of trouble with keyboard navigation in applications, mainly due to inconsistencies. Something as simple as moving from text pane, to toolbar, to sidebar - can require - some dexterous moves using function keys.

Hitting escape, and then moving around the UI with arrow keys would be much easier. Hit escape twice to move up the application stack. Hit the menu key - for context sensitive menu, hit escape->menu for app menu and so on.

Tabbed browsing for me is a real curse, as there is so much inconsistency between applications. This should be the preserve of window managers. So back to your blog post. Why haven't the best parts of some of the esoteric window managers propagated through to the mainstream OSs?

Interestingly tabbed browsing is what won me over to Opera and Firefox in the first place - the desktop OSs should have picked up on this earlier. It's a hack, to get around a poor desktop UI.

Back to Mozilla, why have they barely touched or been creative with the UI for the last decade? The native implementation of tabbed browsing is nothing but horrendous. I'm hoping some of this will be picked up with 'Pancake'!

Wouldn't hiding the url bar/link make it a lot harder to identify phishing?
I see hiding the URL as a big step backwards.

We're on the edge of a "Web 3.0" revolution where people use URIs as global unique identifiers for concepts. It makes sense that Google hides URIs because they're threatened by the semantic web (by increasing the amount of meaning/byte, the kind of 'big data' analysis that Google does will be possible for smaller startups and eventually desktop users.) Firefox doesn't have a reason to hold back the future though...

I think we've pretty much ruined URLs in terms of usability. Sometimes they are logically organized URI's which point to a specific resource but a lot of the time they are a jumble of characters for session state, tracking info, script names, and all kinds of other garbage the user can't make heads or tails of.
We're on the edge of a "Web 3.0" revolution where people use URIs as global unique identifiers for concepts.

Machines and browsers do that. People don't treat URIs as anything. They could not care less, actually.

What makes you think people understand URI bars? Often times they become fairly unintelligible as the number of resources go up. Even if we slugify titles into the URI, there are almost always other path hierarchies and parameters to handle paging, sorting states, or otherwise. The user has already clicked through, he can easily see the title in the tab; the rest of the URI is probably meaningless or not worth understanding for the 99%. 99% of the time, it is probably even a waste for the other 1%; it's not that URIs are going away.

I'm not even sure what you mean by a "Web 3.0" revolution. The following is the future?

* Youtube link off my Youtube homepage: http://www.youtube.com/watch?v=0lvMgMrNDlg&feature=g-all...

* Twitter status link: https://twitter.com/#!/snookca/status/164424086309715968 -- though I'd wager a majority of users simply hit retweet and never see the resource URI

* Facebook photo off my wall: https://www.facebook.com/photo.php?fbid=343550758997176&...

I wouldn't agree that you can easily see a title in a tab! I'd rather a meta key that would bring up the info about the page easily - which listed the URI etc. But the title should remain at the forefront.
(comment deleted)