It’s May 1st and that means CSS Reboot day wherein folks around the world redesign their websites. I’m not one of them, however I have been playing around with a new version of this blog, which I am calling Clagnut 2.0 beta. Beta because it is utterly unfinished, by which I mean the functionality is nowhere near all there and the design, such as it is, will most likely change beyond recognition.
The interesting bit, for me at least, refers to the 2.0 part of the label as I have started playing around with APIs and tagging. This has led me to think more about my approach to tagging. In a recent blog post, Thomas Vander Wal, who coined the term folksonomy explained that:
Tagging [lets] users describe the world in their own terms as taxonomies become “folksonomies.” Putting information in each person’s vocabulary is important. We lose so much information and having the means to pull it all back in and refind it is an incredible tool to have with in our reach.
I’m finding that the information I’m losing is spread across different websites, but consistent and detailed tagging can be the magnet to collect this data back together. Now when I tag a photo in Flickr I try to think about other contexts for that information. For instance I would ask myself: to which blog posts would this photo be relevant? and tag accordingly.
A good example of what I’m talking about is a post I made last year about a trip to the West Country. In Clagnut 2.0 beta, I am plugging the post’s tags into the Flickr API and displaying the matching photos. By ensuring I have included pertinent tags such as Cornwall and Devon, the returned photos are all from this trip.
I also geotagged the post. The geotagged tags tells the blog engine to display a Google map and the geo:lon and geo:lat tags provide the location coordinates. The structure of the tags is exactly the same as the geotagging you might see on Flickr.
Finally I created a third format of tag. In the post I wrote about the book Penguin by Design so I have isbntagged the post. Using a compound format similar geotagging, the isbntagged tag tells the blogging engine to hook into Amazon’s API and the isbn=0713998393 tag says which book is being referenced.
Obviously lots of different and specially formatted tags like this would make a tag cloud look fairly ridiculous, but I’ve never found tag clouds particularly useful anyway. This is all about personal tagging; about recombining dispersed but related information into one view. And I want to do more of it.
What other sources of information are there that I could pull in by this manner? Where else might my stuff be (I have links on delicious for example)? Is it actually useful to supplement a blog post with third party information?
Blogging · Information design · Clagnut news · Flickr
Mike Stenhouse wrote:
I’m finding that tagging gets out of hand very quickly without some kind of controlled vocab. The problem is that you need to do a lot of tagging before your vocab emerges and then you need to remember it across sites… I really like the delicious popup’s recommendations and as-you-type suggestions. Those seems to help keep me in line, but the other sources of my tagging are all over the shop.
I added tags to the posts on my site because I couldn’t think of categories to cover everything but it’s also a good way of adding ‘aboutness’ to an entry – words that describe the subject that may not occur in the body copy. At the moment I use my post titles to generate a list of ‘possibly related’ posts but the plan is to feed the tags in instead. I just need to finish tagging the final 200 or so posts! I like the idea of using them to link to Flick etc too… It’d be really good for pulling in potentially interesting links from delicious, as well.
JH wrote:
Please don’t be one of those people who starts showing only one blog post at a time. It’s so hostile.
Rich wrote:
JH – Clagnut 2.0 beta as it stands is just the blog post page. There will be a proper home page showing more than one blog post, but I haven’t built it yet.
Incidentally, what do you think about the current Clagnut home page? I now show only one post in entirety, with summaries for another 9 – good or bad and why?
Rich wrote:
Mike – I agree with you about tagging getting out of hand when it is used as a categorisation or findability mechanism. I’m very much of the opinion that a set of generic categories to organise content can work with tagging when tags are used to add aboutness (nice word). Of course knowing what categories to chose in advance is tricky, especially with something as potentially varied as a blog.
Small Paul wrote:
JH: why is it hostile?
To provide a vote the other way, I’m a big fan of one HTML page per blog post. I appreciate the focus.
Loving the geotagging and ISBN tagging. Tagging’s the one thing I’ve found since I started using computers that seems to use their specific capabilities to really increase efficient information management. That wasn’t meant to sound so jargon-y. But tagging itself is pretty quick and easy, and everything you’ve just noted would be so labour-intensive to do without it.
People must have been doing this for years in some fields, right? Libraries?
With the Flickr Devon example, however, what about if you have another trip to Devon? Will the wrong photos be with the wrong blog post? Or are you date matching too?
Hauke Rehfeld wrote:
Sorry, I’m a bit… I don’t know, this stuff matters to me ;)
I always thought beta was used for almost feature-freezed versions – see http://en.wikipedia.org/wiki/Development_stage#Beta
ps. no UTF-8 symbols? … & – dont seem to work in the preview
Rich wrote:
Hauke: The ‘beta’ badge was really just a playful dig at all the Web 2.0 apps out there that claim to still be in beta. And you’re right – what I have created is little more than a prrof of concept, probably pre-alpha.
Small Paul: Truth is I’m highly likely to head to Devon again as I have relatives there. Quite how I’ll sort out the tagging to ensure the correct photos ar matched, I’m not sure. But then tagging is always a bit vague and I’ll think I’d be happy with photos of Devon being matched, regardless of when they were taken. If I need to match anything more specific than that I can add some more tags.
jimmy steel wrote:
I also don’t like one blog post on the homepage. I understand why its good to archive one per page. But I like to be able to go back to older posts and quickly look at them from the home page.
Tom wrote:
Tagging for me, is about putting in some words that I can use to recall the post/photo/link later. I like that a related post will show a photo by using a similar vocab, that’s really pretty.
I like the isbn tag, but perhaps you should adopt a smimilar style as geotagging. e.g book:isbn=0713998393
You could also pull in Upcoming data if your talking about a specific event, or perhaps data from iCal/Google Calendar.
Rich wrote:
Tom – I thought about the three-part format for the isbn tag but really an isbn is an isbn. What you’ve suggested is ‘booktagging’ of which isbn would be a part. I guess you could add
book:author=Baines,Philor something like that. For me a simpleisbn=nnnwill suffice as it’s still suitably open and uniquely identifies the book in question (or at least a unique edition of the book).