Tag: tags

Posted on
by

Bookmarked Thoughtware by Paul Bricman

Paul Bricman’s ‘thoughtware’ tool Lexiscore, a nutritional label for food for thought, was mentioned in the Obsidian discord channel on knowledge management. His works seems of general interest to me, so I added his writings to my feed reader. I mailed him with two elements that are important in my information strategies that I don’t immediately see covered by his description. One is: In between individual subscriptions and engagement by the masses (likes, shares) to surface what others curate, there is the level of communities of practice and interest. Subscribing to multiple people within a community and doing so in many communities allows a focus on patterns in what people are talking about (what are Berlin coders ehthusiastic about these days, what’s going on in the [your fav topic here] scene in Argentina?), beyond just focusing on individual pieces of shared content. Two is that being able to see how other people differently describe (in tags e.g.) the same pieces of content that caught my attention, gives me a measure of distance to other groups unknown to me, yet with interesting overlap. They are interested in the same thing but use very different words and language to describe it, representing different view points which is valuable information (Surprise). This is how I used Delicious bookmarking when it still showed you how other people tagged the things I bookmarked as well, and who those other people were. Because tags are not just descriptors but also navigational way points.

My research is focused on extending human thinking with artificial ways of thinking. An important part of this venture is bringing to life actual tools which incorporate the artificial affordances I’m designing, and then taking them for a spin. I call this family of tools thoughtware

Paul Bricman

I have changed the way I add date tags to my PKM notes. It used to be in the form of #2021- #2021-02-26. This as my main viewer on these notes, Obsidian, only supports search on full tag names, so searching on #2021- does not surface #2021-02 as tag. In December Obsidian introduced nested tags, which you can do by adding a / in between, like in #maintag/subtag/subsubtag etc. Normally I am adverse to sorting tags into hierarchies, tags are not categories after all. But for dates a nested hierarchy is useful: now I can add #2021/02/26 as tag, but in search that will return results for #2021/ and for #2021/02 too. It took a bit of time, but I’ve now replaced all my old date related tags with the new nested tags. An added benefit is that it cleans up my taglist enormously, as all tags related to a year are collapsed into one.

Posted on
by

Earlier this week I came across a Lifehacker posting “Get a Better Creative Workflow in Evernote by Ditching Tags” by Melanie Pinola, quoting Tiago Forte who’s into productivity, which proposes one might as well get rid of tags in Evernote because :

  • “When you rely heavily on tags, you have to perfectly recall every single tag you’ve ever used, and exactly how it is spelled and punctuated.”
  • “The real problem with tags, and why they not only fail to help, but actually even hurt people’s creative self-esteem, is that they give the impression that keeping a useful collection of personal notes requires nothing less than a heroic feat of comprehensive planning, followed by years of meticulous, unwavering cataloging and annotating”

This does not make much sense to me at all. For me tags are a key ingredient in provoking serendipity, as well as a navigational aid. Both play a strong role in my creativity process. If you think tags limit your creativity, I think it is likely because of how you use tags.

Tags vs Categories
It seems both Forte and Pinola see tags as categories. Tags aren’t categories. Yes, categories do require you have a good understanding of how they are organized, and need you to stick with it thoroughly, as otherwise everything ends up in the ‘miscellaneous basket.’ Categories are things you make up before you start categorizing. Tags work the other way around: you add tags to things as you go along. Over time a structure may emerge from the tags which you could adopt as categories, but that isn’t the purpose of tagging. With tags everything starts out as miscellaneous. This key difference is the difference between approaching information from a hierarchical perspective (categories) and from a connected perspective (tags). In the networked age, Everything Is Miscellaneous, as David Weinberger put forth in 2007.

Categories in Evernote
Evernote has no explicit categories functionality, but allows you to work with categories in 2 ways.

  • One is dividing your notes in different notebooks. This is something you can use for fixed and mutually exclusive categories. I have different notebooks for different areas of responsibility.
  • The other is using the tagging functionality. These can be used for non-exclusive categories (as a note cannot be in more than 1 notebook at the same time, but can be in several categories). I use tags like this as categories as well, for instance to indicate project status, or that a note is related to a specific project. However those tags as categories are just a small part of the tags I use.

How I add tags (e.g. in Evernote)
My tags do not form a structure of categories / a taxonomy. They are reflective of my associations with a piece of information. I add tags to capture what a piece of information means to me, what I associate it with, or how I might use it. All of this in a non-prescribed way, and not as a ‘must’ either. There’s plenty of stuff I don’t tag at all, and there is no planned consistency in my tagging. It simply evolves with my own internal dialogue and idiom (something I would have tagged socialsoftware in 2002, would maybe have been tagged socialmedia in 2009 and socmed in 2015).

Key here is that with my tags I do not try to capture what something is “objectively” about, like the echo of systematic categories, but why I saved it. A piece about an animal may be tagged with collaboration or with business_models based on the associations I had while reading it.

My tags may very well not be used or present in the information I tag with it. (In general if you ask people to tag stuff or title it based on what it means to them, there is a good chance they use words not present in the tagged information itself).

I also save material in about half a dozen languages, and then tagging is a way of connecting material together and make it findable in ways that full-text search cannot do, as search is monolingual.

There is likely a power-law distribution in my use of tags: most will get used maybe once or twice, some will get used heavily. The more heavily used ones, if I notice it as a pattern, can become a sort-of de facto category. So I don’t need to remember all my tags and how I used them, as suggested in the linked article above, I usually only remember the less than 10 I use frequently. I am not bothered if I don’t use them.

How tags help my creativity
There are two ways in which tagging aids my creativity.

The first is that it aids my serendipity. If I search my notes it surfaces things not only based on the content of those notes, but also on the associations I used as tags, and other words I used as tags that are not in the content itself. That way unexpected search results, but nevertheless relevant to or overlapping with my search question, can pop up. So that when I search e.g. for business models the example article about the animal I mentioned above will pop up. That way I find things I did not realize I was looking for.

The second is that tags allow me to navigate and pivot through my collected material. I see social software / networked tools as working in triangles (see my 2006 posting Social Software Works in Triangles).
Such a triangle is formed out of an information item (a Flickr photo, a Delicious bookmark, or indeed a note in Evernote), the person that created/shared it (in Evernote usually myself), and one or more descriptors (tags, locations etc.).
The point is that tags are not just descriptors, they are also turning points on the path through my data. These pivots or forks in the road, allow me to hop-step-jump from an article to other things within the same context through a tag, like another article, and then through to the author of it and maybe onwards to one of their other writings, to somebody’s bookmark collection of which it is a part, to that person’s blog etc.
It allows for navigation and triangulation that way, bringing me places I didn’t know about. That is a richness in association, multiple viewpoints etc, that a category system cannot produce. ( I even dreamt about tags and pivots once, in 2007)

So, don’t ditch tags because they cramp your style. Uncramp your style so you can use tags fruitfully.

Posted on
by

Dreaming of Tags
Shortly before waking up one morning, while camping in the Austrian Alps in the past weeks (as seen above) I had a dream about tags. Or rather I dreamt that my brother in law had created a database in which each data item was treated and useable as a tag as well. In my dream I was very enthusiastic about this idea. When I woke up Elmine asked me ‘what does that mean, that everything is a tag?’. We kept coming back to the topic, and at the end of the day had a conversation around it over a couple of Weizen beers.

Tags do Double Duty
Tags serve two functions. First they are descriptors, and in that sense subservient to the piece of data they describe. But they are also pivots, i.e. turning points in your path through data. A pivot allows you to see the same set of data, or a different set of data which overlaps the current one, in a different view.

If you go to the picture above and follow the tag ‘huben’ on the right, you are presented with a number of pictures that are also tagged ‘huben’, so you can navigate to a different photo within the context of ‘huben’ and so on. Pivots are the forks in the road of your surfing.
When everything is a tag as in my dream, then everything is a pivot as well. This reminds me of the view on data-items the people of Mediamatic have: everything is a thing. So a thing can be a tag, but also a list of tags, or the entire Flickr-database, or any part thereof. In my dream everything was a tag, a pivot.

Pivots in Social Software Triangles
My description of social software as triangles, which got quite a good response at the time, put tags as pivots in the center view: social applications allow you to navigate from one app to another through their tags as pivot points.

From a Flickr photo to a point on a Yahoo or Google map, to a location in Plazes, or to photo’s taken geographically nearby. The thing is, I cannot directly jump from a Flickr photo to the corresponding location in Plazes. I could if the Plaze itself was the tag. In my 2006 posting I already indicated that where now usually a tag is in the triangle, there basically could be anything. As long as the other two points are a person and a object of sociality. So you could theoretically jump directly from a picture to an event to a place to a review to the author back to the picture again. If tags could be more than just a descriptive word they would be better pivots.


(image from the triangle posting last year)

Need More Pivots!
Hence I concluded that my dream was basically a call for more pivot points in social media. So that we can navigate our web apps better, and build better personal information strategies.
Question remains who has the rights to the concept of a database where everything is tags: me because it was my dream, or my brother in law as he came up with it in my dream and showed me a working prototype 🙂