World Wide Web

The death of browsers

About a year back in one of the internal 'prediction' wikis of IBM, I had predicted that 'folders' and 'browsers' are going to be a thing of the past in about 2 years' time. I in fact blogged about the folders prediction here.

Now after a year, I am still confident about the latter. However, it might take a couple of years for the concept of browsers to change.

I read a blog post today, which convinced me about it. Alex said in his blog,

“Detecting information in Web pages and handing that information off to other applications changes the role of the Web browser from being solely a HTML renderer to being an information broker.”

There is more about it here and looks like an interesting post.

This is radically different from our current usage of browsers. Browsers would be more like information harvesters rather than a medium to 'browse' information.

This is already true to some extent. Compare the browsers of today with the browsers of the past. If you look at the features of Firefox 2, you will notice features like Live titles, Live bookmarks, not to forget the widgets of Opera.

However, it seems obvious that end users will notice a drastic change. It is going to be a gradual change from the current to the future. And when the time comes, you will have browsers as standard Operating System applications (why were people criticizing Microsoft for making IE part of Windows?!) and a platform to connect and use the web. (Remember, usage of the web is going to be quite different from the present).

And now regarding the folders stuff. Well, that is not the point of this blog entry, however, I would like to conclude by saying that 'folders' in their present form have limitations, but there seem to be better ways of doing away with these limitations than doing away with this approach altogether. By the way, I am still looking for the day when my system is going to boot up and take me to secondlife and I work from there.


Bye, bye folders

It occurred to me when Gmail used it. Many other sites/products/services were using this same concept, albeit with different names.

I am talking about the concept of tagging (or labeling). There is a very interesting study on tagging here.

It made me wonder, why at all have folders? Why not have just tags? How about a world without folders?

Well, if you think I am fantasizing then you should look at evidence that people are moving in this direction.

* The much hyped Reiser4 filesystem supports Semantic filesystems.
* WinFS supposedly has this.
* MAC OS's Spotlight search is a work-around to this.

2 years down the line, you might see that “Folders are history”! But the question is what will replace them? Is it tags? Well, in its present form, tags are not quite a replacement to folders. While tags have several advantages like automatic rule application, there are some disadvantages of tagging compared to folders. One that I can describe right-off is “Context awareness”.

Suppose I tag a file by name “Project” does it mean it is MY project, or does it mean SOMEONE ELSE'S project or does it mean PROJECT RESOURCES? In folders, it is possible to give a context to the data, while tags in their present form, will not help us here.

A good solution is to have the best of both worlds. The concept of tag clusters would probably be what would actually replace folders. It is not yet unclear how tag clusters would look, so it would be early to comment on it, but you can expect something soon!