The Tao of Topic Maps (2000)
30 points by Tomte 3 days ago | 5 comments
  • brettkromkamp 2 days ago |
    To this day, topic maps still provide me with my go-to (meta) model.

    Topic maps truly provide a straightforward organizational metaphor. The foundational topic maps concept —comprising topics, occurrences, and associations— achieves a balance between simplicity and clarity while offering sufficient structure. This allows you to effectively translate your mental model of a domain into a corresponding topic map representation.

  • sam_lowry_ 2 days ago |
    I did research using Topic Maps. One of those meaningless R&D projects where EU burns millions to no avail. It paid well though, and I could attend conferences around the world.

    As for the usefulness of the Topic Maps, Mind Maps and ontologies in general... they are useful instantly for the person who composes them as a way to organize thoughts. Pretty much what Notion, Evernote etc are doing, but with much less elegance.

    They are utterly useless as a way to share knowledge.

    Also, even 20 years ago, the field was full of old farts wasting everyone's time and consuming salaries. I doubt it changed for better.

    • prox 2 days ago |
      When I started working and got hold of code and research, I made an app to make a knowledge system, but at some point I got the same realization, they don’t really have any advantage to a group of people. They are hard to read as well. I actually sold my first iterations to another company, but don’t think anything came out of it, other than a nice visual in their main app.
      • burnt-resistor 2 days ago |
        Personal notes have value, but probably only to you, and there's no real need to implement a super-specialized subset when many wikis, brainstorming/mindmap, and note-taking apps already exist with many of the same features. Sorry that you spent so much effort on a category that was already/became commodified, but I guess you learned from it. That's the main risk of being an entrepreneur because most ideas tend to be fatally-flawed somehow for the current market.
    • burnt-resistor 2 days ago |
      > They are utterly useless as a way to share knowledge.

      How so? Facebook's graph is implemented as a a triplestore where the edges have values rather than entity-relationship of traditional DBMS. GraphQL also exists.

      Maybe RDF and getting humans to structure their statistics, transactions, memories, and annotated knowledge rigidly and methodically in a universally-composable manner was a "failure", but now LLMs have the potential to transform freeform content such as from web pages or APIs for us automatically in both materialized and presentation forms that would be mechanically-consumable.