Ask HN: What's the future of the file browser?

4 points by MatthewRayfield ↗ HN
If desktop applications go the way of web and mobile apps, will we be using a file browser in the future?

With web apps it becomes difficult to pull a user's cross-app data together with such varied APIs and data formats. But do we even need our data in one place or will we just open the program that has our data "in it"?

What do you think?

12 comments

[ 4.2 ms ] story [ 39.0 ms ] thread
Hopefully the file browser and the hierarchical file systems die. I've never liked them. It is a single limited view on how to organize your information. Severely limiting. Within a year of using a hierarchical file system, I've been waiting for something better to come along and get traction. Can't wait for it to happen, I just hope I like everything else that comes along with it ( and the privacy and security implications of the cloud give me pause ).

i still hope that something like lifestreams ( http://cs-www.cs.yale.edu/homes/freeman/lifestreams.html ) takes off or a file system w/ the database like aspects that the BeOS had ( WebOS has a decent amount of that ).

I certainly agree with that.

But do you think that in the future people will use one central application to organize their data, or will we organize it inside of the apps we use?

I find the inside the app to be easy for an intro but fundamentally limiting. I've always been excited by technology that features common, expandable data stores...

OpenDoc from Apple years ago featured some of this. BeOs's file system had some as well but it still gave you a hierarchical file system. WebOS has it some. The Newton had it. And I'm sure there are tons more examples that I never came across or don't come to mind now.

The problem w/ standard shared data storage for applications is that everyone has to agree on what a contact is or you need to be able to add additional data to it that other applications could tap into. Getting everyone to agree on formats, having them hold up over time etc is... problematic. Still, I hold out hope that pooled data storage will come into being.

That's where I get caught up. At the moment it's very hard for me to imagine all the makers of web apps agreeing upon such things. But at the same time, I can't imagine a future where apps can't swap data with the ease they could on the desktop.
I think an ideal persistent-data system would let you mix and match your methods. Sometimes you want a stream, and other times you want a hash, tree or graph type of structure, or a full relational database. A file system that included all of these would give you the essential data structures of programming in a from-the-ground-up persistently stored form.

But the biggest issue is really a standardization one. Hierarchical is everywhere and it's hard to pull away from.

Don't you think that having so many ways of organizing data might just lead to confusion? For you or me it might seem great to have so many options, but for the average computer user I feel like it could be pretty overwhelming.
If you provide a standard pool then the application could decide how to display by adding specific data to it. Want to just use metadata, that is there. Want to use hierarchy? Provide a way to do. Each app can do what it wants.
I really like that idea. That way it would be completely up to the application creator to make the experience of organizing data make sense in context.
That was the general idea behind the newton's 'soup' and the db parts of the BeFS.
These were all things before my time. I appreciate you bringing them up, Lifestreams too, gives me interesting things to read and think about.
It is sad is how little we have progressed since then.
It's true. It seems strange to me that there hasn't been any innovation in the way we organize files in a long time, especially with the trend towards touch-based interfaces and focus on user experience.