Re: Using TM to help define the future of XML

Patrick,

> More seriously, do you mean we need one or more tutorials with odd
> sounding titles?

Well, I like it when tutorials include nice pictures and some jokes. This makes them more attractive as good night litterature than academic papers. It's hard to find easy-accessible litterature for TM beginners.

> Or just more tutorials in general?

> I suspect part of the problem is that unlike the Haskell tutorial, there
> isn't a command line interface (or other) that gives you immediate
> feedback. 

Well, that's where my secret secret plan is. But I need CTM import for tmjs first. Details will be posted on this list. ;-)

> That is if a tutorial started off talking about topics and you
> enter/copy the code for a topic, yawn, now what?

Isn't that the way many programming tutorials start? Enter some data, start querying the data. This is how you store information, and that is how you retrieve information.
The Ruby on Rails community did an excellent job. Everybody started writting blog engines with Rails. They were pretty useless, but it was an easy accessible way to the technology. Everybody knows what a blog is, and you only needed some lines of code to create a blog by yourself. Wow!

> See the problem? 
> 
> The *bang* in topic maps takes a bit longer to become evident. 

That's for sure. But you have to get started somehow. I prefer practical introductions.

> My thinking being that the create ontology/types/topics -> write file ->
> load file -> navigate -> merge with other files -> still text interface
> approach interests a small minority of users but just isn't appealing to
> the game boy/netflix users. If it doesn't jump, fizzle or pop, they lose
> interest pretty quickly. 

Agree.

> PS: Can you say what would be of interest to the nodejs community?
> Thinking that tutorials that touch on items of interest to particular
> communities are likely to be more successful than others. The Haskell
> tutorial for example touches on concerns meaningful to programmers but
> not rose gardeners. Topic maps tutorials for either of those groups
> would be somewhat different. Yes?

Yes. As a developer I was thinking about tutorials for developers. Management and rose gardeners can be convinced later (maybe by other developers).

As for the nodejs community, it turned out that they are not interested in databases as much as other communities (e.g. Ruby on Rails). Usage of NoSQL storage is pretty widespread. But still, nobody knows where the development of all those NoSQL databases will end. Would you invest money into such storage solutions? If you use TMs as a layer between your application and a NoSQL database, migrating to other NoSQL databases will be quite easy. Write a new backend. Export XTM from the old backend. Import XTM. Done. You get the benefits (flexibility, scalability, etc.) of NoSQL but your information is always accessible using an ISO standard.

Also, it turns out that node programmers like functional programming. A case for path languages?
And: node is still new and hot. Much more easy to get attention for anything, at least I think so.

- Jan