The web is changing the way we think.

We are learning to process information in a hierarchical way – we don’t need to translate it into small bits, each with a linear structure. Not anymore.

That’s why something like Twitter is possible – and enjoying enormous growth in the face of its many shortcomings. The Twitter experience implies participating in different simultaneous conversations – asynchronously; having to linearize each conversation every time we add to it, or consume a piece of it, would be a gargantuan effort – somebody describes this as ‘being overwhelmed’ by Twitter. Yet everybody catches up quickly – the Web trained our collective brain in using the brand new Hyper-Information Processing Framework. How would’ve people reacted to Twitter ten years ago?

We are presented with increasing amount and variety of information per time-unit and per medium-space-unit (think about what’s on your computer screen at any given time – it is a goal pursued consciously even when detrimental to good design) – yet the increase is beginning to lag behind our brain’s “processing power” growth.

We have become literally faster than our own browsers – we scan a web page in a glimpse but then have to wait a second or two for each link we click to open. More and more the medium-space is acting as a bottleneck of the information-consumerist era.

Why? There is still one linear element in our Web 2.0 life: the ‘Web Page’ concept (I’m talking about the semantic structure of course). Granted, the vast majority of the web pages we stumble upon, nowadays, aren’t linear at all. Such pages have multiple information consumption paths (columns, boxes, and more importantly meaningful and pre-processable links), yet all these structural tricks only manage to do marginally better than a newspaper over a book.

A very good example (and an awesome app anyway – precisely because it appeases our inclination for the good ol’ times) is feedly – one might think it’s incommensurately better than a printed magazine, but analyzing the whole process from start (browsing and scooping up all the RSS feeds, one at a time) to end (consuming the aggregated information, one piece at a time) you can see how it is barely different from casual reading at the dentist’s waiting room.

Granted, the whole point of the Open Web (2.0) is to allow anybody to filter, process and regather information from the whole Web to her heart’s content – but all this eventually must end up onto an intrinsically linear medium, the page. And this isn’t satisfactory anymore.

Of course we had the infrastructure all along – the link. Googling, following links, recursively opening cascades of tabs is fundamentally the same thing as mashing all that information together a-la Web 2.0; but since we lack a hierarchical presentation medium we must do the mix in our minds instead of on the medium itself.
The problem is that the context switching overhead (the context might be simply web pages) has become too much. Our brains can envision structured information much better than the limited dimensionality of the page allows, and we find ourselves in that recursive tab-opening scenario while looking for a single piece of information – but what about the incredible amount of bookkeeping the brain has to do? It’s exactly the kind of (inefficient) bookkeeping that GTD has taught us to outsource. We’ve invented software to take care of this kind of problems, yet it’s the software that creates the problem for us, far from solving it.

I find it rather interesting how everybody is focusing on trying to augment the speed and interactivity of the mashup creation process (they seem to forget that the fundamental steps of scooping up information sources and presenting the results is doomed to remain linear, hence inefficient) instead of researching hierarchical alternatives to the page as the presentation medium.

That’s why an apparently trivial feature, like having links open in an adjacent tab as opposed to the far right, makes a huge difference. It lets the underlying link hierarchy show through a little more, instead of brutally destroying it. Those of us who are heavy consumers of the web noticed it seconds after deciding to give Google Chrome a try.

That’s why Mozilla must follow the trend.
That’s why the biggest favour you can do your surfing self is to get one of the extensions that make Firefox behave like Chrome and beyond, taking the concept the logical step further by introducing groups of tabs.

But what’s really going on here? Aren’t we simply putting some overworked makeup on the inherently ugly, dimensionally limited page? Any external toolset that helps us in coping with pages is welcome, of course. But these tools don’t solve the problem – the problem is the page itself.

We need the multi-dimensional hyperpage.

We need some enabling concept to get there, the same as AJAX took us from the “old web page” – which was little more than a linkable “plain old (paper) page” – to the “container page for web apps”. AJAX is a great tool to interact with information, and made web apps possible. But we need something new to represent information and navigate through it – we must reinvent the browser.