To tag or not to tag ... choices in digital clasification of items

A couple of days ago I had a discusion regarding the options available for organising items in a system.
Initially the system had followed the traditional way of using a tree and attaching item to the tree branches. It worked ok but new times give new ideeas and the tagging concept was put on the table.
Nothing new ... a lot of flame wars betwen the tagging fans and the rest of the world. Everyone thinks he has the answer.
Is there really an answer?
Is there really a choice?

If you think a litle .... you will see ... there is no real choice.
I think that it is more like a matter of understanding the concept of organising things, how we do it and how we visualize things.
We are naturally organising phisical things ... books, underwear, CD's, and so on.
We are used to put those things in physical repositoryes ... a box, a drawer, on a shelf etc
Those physical items and the physical repositories we use for them are right there at hand .. one book existing on that specific shelf ... one CD siting in that specific box.
There is one instace of the book and one instance of the shelf in the same place and the same time.
We make at any one time a choice to store the item in one specific place clearly identified.
When you go digital things change.
Now you work with files or with records in a database.
You can easily create a new instance of the digital item or even smarter you can use links to have the same item in different places at the same time. Wow .. that's really diferent.
The fact is that we humans are very stuborn and prefer to use concepts that are already in place. Against the usual thinking we are not very adaptable.
In order to manage things we first created the ideea of the tree ... each file was attached to a branch and we were happy with that.
We know that with a physical item this is the way to do it. Everybody considered the ideea to attach a file to multiple folders and subfolders on your disk drive as stupid and here we were ... with limited options of using a digital repository.
And here is how we had the hyerarchical way of organising digital item ... and it was usefull for organising documents based on a taxonomy.

Then somebody said that this is not flexible enough becouse it did not let us attach the item to multiple branches at the same time and a great new ideea was imported ... most probably from the library world ... let's attach to each item a number of tags that describe it in all details.
And we created the tagging system of organising electronic things and it was usefull when you needed to know what subjects a web article was covering.

Then somebody came up and said that if we use tags we cant organise the items in a hyerarchical way so we have to come back to the tree ... but he wants also the flexibility of the tags .... and here we have a mixed system.

Does it make sense for you to consider that an evolution?
From a technical point of view nothing stop a programer display the hyerarchy of the tree in a tagged interface and viceversa. Tags or trees they are the same ... you either atach the item to a property (the tree branches) or atach the property (tags) to the item. Saying that this is different is like saying that 2 + 3 is bigger than 3 + 2.

I wil not say that this is the definitive article on that subject. It will be to easy. I just hope that somebody out there will get the ideea and realise that you just change the interface and have the advantages of the other system.

The only matter in discusion is to take a closser look at your taxonomy ... this is the real issue. Is the taxonomy that you intended to use for tagging different in any way than the classes defined in the tree?


Have a good day ... and only deal with great taxonomies :)

Digital library metadata

I wrote some time ago a study on digital libraries and there is something that keeps jumping thru my mind. Lot's of people are saying that there is a lack of standardization in the field. In some aspects I have to agree. On the other hand ... it looks like there are enough XML schema standards in the metadata area.
MODS is accepted in libraries and archives, and can easily conquer many other areas if needed...
PREMIS is really doing what it should in preservation ...
METS is already considered to be the universal glue in the library world ...
If you need anything specific you can add it as a finding aid and reference it.
Why should anybody need to design a different metadata architecture.
It looks like we have the standards ... I am missing anything?

Digital libraryes news ... no news

I'm doing a research for a project that involves reviewing everything that's new and bright in the digital libraries and it looks like there are not too many news in this area.
The area is developing but without the librarians. It looks like very few people in this branch realize that the concept itself has changed in the past few years. They still discus about classification and metadata but that's old news. You have to think about YouTube and P2P about how to integrate those in the library. There are lots of blogs and there are no legal deposits for them. If you exclude Internet Archive from the equation there is nobody to do this. Where are the true librarians?