Recently, I had an idea for a new site of mine. It was a nice idea, and while I was thinking about different aspects of implementation, I tried to outline for myself how would I like to create it.
It was fairly simple. I wanted to post an informational articles/essays on a chosen subject, with categories/labels/tags. I wanted readers to able to comment. I wanted readers to contribute (as actively as possible of course). I wanted it to drive traffic.
I would be ready to invest my time around this, as it is on subject of an interest for me. I would pet this site and care for it as long as it doesn't require too much effort, because otherwise I have rather more interesting/important things to do with my life.
So, while I was thinking about it, I finally decided that the easiest way to start this would be a blog. I'd start a simple blog using one of multiple blog hosting platforms around the web, and then,
if when it drives enough traffic or I'd like features that platform would not offer, I'd move it to more private infrastructure, with more features, etc.
Easy, isn't it?
But then it hit me: tags/labels would not do. At least not always.
Let me start with a little bit of obvious stuff: what are tags? Citing wikipedia,
"A tag is a (relevant) keyword or term associated with or assigned to a piece of information (like picture, article, or video clip), thus describing the item and enabling keyword-based classification of information it is applied to".
In general, 'tagging' could be defined as appling a set of semantic labels on different types of fairly complex objects of data, such as text (say, on more text than one word - essays, articles, posts, etc.), video files, audio files, pictures and combinations of all these. We do it for 3 major reasons:
- To find things faster. This one is self explanatory. We "label" things in our mind according to our way of thinking. So when I create/absorb a peace of information, I associate few concepts with it. When I want to retrieve that peace of information, it is very probable that I'd use the same associations, so here it is - I semantically create a way for myself to retrieve information I might be looking for.
- Help other people to find things faster. The same thing as before, but other way around. I create categories, and when someone looks at all sorts of things I created/found, she'd like to find something particular. So, by looking at list of different categories, she could do it relatively easy. Just like search in library archives.
- It's easy. This one is the real deal. I'm starting to believe, that people can develop an addiction to tagging. For seasoned blogger/writer in so-called "Web 2.0" era (definition from Wikipedia), tagging is like second nature. We don't think when we're doing it. It is so easy, that not doing it probably considered weird and strange. Its like we can't help it.
- Blogging/Blogs or News reading
- Thunderbird 2.0 This is a new one. Released practically few days ago, this is an email client software. This new 2.0 version supports email's tagging. What is that? Well, instead of putting my mail message to one specific folder, tagging allows me to keep all my mail messages in one folder, hidden somewhere in the system, but at the same time it allows me to have "virtual folders" defined by tags. This, in turn, allows me to see/find the same message within multiple "virtual folders". This is the feature I've been always missing; I was craving to find a program which can provide it. Such emailing software exists for awhile now, and it is, of course,
- Gmail. The problem with tags/labels in Gmail was greatly resolved. You only have one "Inbox" folder and multiple "label" folders, and every message can appear in any such folder you like. The problem with Gmail though, is a limitation on a number of labels (which may not be a bad thing after all - how many folders do we use anyway?), and, of course, it is not connected to my corporate Exchange server (yeah, I know....please don't comment on this).
- RSS readers. Recent development of "Web 2.0" bubble, and as a result of it - all sorts of content management systems, we came to a development of aggregations.
I can get a short list of recent changes on a particular site in a text file. So, when I follow multiple sites for essays, news, articles - or just changes, I just "sign-up" to that file, and then I read everything in one place using a special program which aggregates all these files in one centralized place. It is convenient and it is very easy.
Almost every time we use this technique, each of these files will be created automatically, based on chosen 'tags' (or labels, or categories).
I will explain myself here. I like wandering around the city (it doesn't actually matter which one, all cities around here are bordering each other), looking into stuff, such as graffiti, streets, buildings, and people. I like to see things that other people don't, or just don't bother.
So recently, I started noticing some sticker in different places. On bus stops. On power stands. On power closets on streets. On trees. It was a sort of armadillo picture, and that's it. Nothing more.
A couple of weeks later, I was walking down the street in central Tel Aviv, and saw the same animal on the window. You know what? It was a bar - coffee shop (cafe as we call it here). I really liked the "commercial" - they stuck all those cute stickers around, without saying what they are, and then when I finally saw it somewhere else -I was really and genuinely intrigued.
Whether I ate at that cafe or not, is besides the point. What I really digging to, is this:
If I were to blog about this cool "sticker" campaign, and would like to "attribute" tags/labels to it, I'd be really tempted to use this exact image. Not some general armadillo picture - this specific one.
So, here it is. This is what had hit me. I would like to use nearly the most basic "feature" of most recent and up to date content management software available, and it is not enough. Because all of them allow only text tagging.
So this got me thinking. If we're using tagging to simplify our own minds way for searching things - even that we had barely started doing it en masse, we might be already overloaded the feature many times over. We may have overgrown it much faster, than it got really adopted.
How can it be "resolved" (if there's anything to resolve at all)? I was thinking about that as well.
If modern content management software regards tags/labels as metadata, maybe this metadata needs its own metadata as well. So, instead of using text tags/labels for 'tagging' our content, we could use metalabels/metatags to 'tag' our tags. So, in the end, I could use a picture (or any complex object) to tag my information.
Wouldn't that be handful and interesting?