« Developmental lassitude | Main | FInding the value of a FileUpload form element using Javascript - easy? no. Well... yes! »
June 15, 2004
New email paradigms
Just thinking about the Google Mail offering and I suppose it also applies to the upcoming release of Longhorn (smirk).
The new paradigm I am talking about is of course meta data over folders.
I was thinking that it would be a very difficult task to swap from one paradigm to the other until I realised the following - folders impart meta data.
Its a very rigid meta data and usually void of the more meaningful name=value pairing of true meta data, but its still meta data.
The big problem is not adapting to a new mental model, but breeding new habits.
To make meta data effective it must be created along with its object - watever form that might take - a word document, a phtoshop file, a text file. This is really an issue for application developers.
Currently the only metadata that is saved with a file - at least on a PC - is a date stamp, some info about the file permissions (which is the same for all documents anyway...) and a little bit of 'hard-coded' meta data - the file extension.
Well OK. Most applications now make a stab at at least parsing out a title for the object. This will not be good enough for when the paradigm switch comes.
Pushing the idea of Longhorn file system to extremes would mean that all your files are esentially saved into one big folder. There will be no naming conflicts because you wont be naming your files (!). Meta data and a feck off search engine will be what you use to retireve your files.
Its a big step and once taken and everyone understands it we wont see how we could live any other way. Everything we do on the computer will suddenly have context. Excellent. This is what happens in real life anyway - its natural for us to associate objects - thoughts, feelings, people, situations, smells.... - with nebulous connections - again thoughts, feelings, people, sights, sounds... - that only become obvious when context is applied, its the very essence of sentience and hopefully bringing it to the world of computing will shine lights in deep corners of AI research and finally stimulate development of something that will be the great-grandaddy of a mchine that can help us answer the question - 'what is it to be alive?'
Or maybe I'm reading too much into this... :)
Posted by dottie at June 15, 2004 8:28 PM