Scribd
Upload
79 views1 page

Implementing TMQL For Teaching: The Straightforward Approach With Emphasis On Understandability and Modularity

here are a lot of voices claiming, the current TMQL draft is too hard to understand and to implement. We argue that it is not the content but the presentation thereof. A clear user-centric documentation with lots of examples and a straightforward, modulary, publicly available implementation in a mainstream programming language will greatly help the understanding and acceptance of the current efforts. Additionally, this modulary implementation can show possible ways to modularize the standard itself.

Uploaded by

tmra-conference
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
79 views1 page

Implementing TMQL For Teaching: The Straightforward Approach With Emphasis On Understandability and Modularity

here are a lot of voices claiming, the current TMQL draft is too hard to understand and to implement. We argue that it is not the content but the presentation thereof. A clear user-centric documentation with lots of examples and a straightforward, modulary, publicly available implementation in a mainstream programming language will greatly help the understanding and acceptance of the current efforts. Additionally, this modulary implementation can show possible ways to modularize the standard itself.

Uploaded by

tmra-conference
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Implementing TMQL for Teaching

Benjamin Bock, Sven Krosse {bock,krosse}@informatik.uni-leipzig.de, Topic Maps Lab, University of Leipzig

Goals of TMQL We‘re all strangers – the target audience


TMQL is thought to become the preferred way to Learning is work and should be fun. It's essential to
access and handle Topic Maps data, independent of experience results as soon as possible, to make or
the programming language used, the underlying keep the learner motivated. It's essential to minimize
engine, or the type of storage. From a technical distraction during a often-needed setup phase. Existing
perspective, the main addressee is TMAPI and its knowledge needs to be reused where-ever possible.
spin-offs. From a conceptual perspective it’s meant to Besides the abstract understanding of TMQL, examples
replace tolog, Toma, and other Topic Maps query should be provided to the users. Examples may serve
languages.
as templates for first experiments
What SQL has become for relational data,
and enable users to check
TMQL should achieve Topic Maps
for Topic Maps data.
Newbies experience Pros their understanding.

exploit experience
programming-only and leverage knowledge
Learn TMDM background tolog
databases or
XML background TMAPI Toma

SQL XPATH
use transformations

FLWR
RTM-TMQL Toma2TMQL tolog2TMQL

adapt knowledge to TM walk the object-graph experience similarities and differences


(Top-Down) (Bottom-Up) (Top-Down)

Something new?! People are different


The Topic Maps Query Language has formally been It’s common sense and still it needs to be repeated
around for a while. The standardization process over and over again: One can’t please everybody.
started in 2003 and went through several iterations. The pure syntactic transformation from the existing
The current draft was created under the lead of (or a more specialized) Non-canonical level to a
Robert Barta and controversially discussed. People standardized canonical level is implemented
where expecting something easy to understand and modularly (or can easily created for new types of
easy to learn, something simple. NCL).

simple != simple
prefer words? prefer symbols?
They got something simple, but from another
perspective: The inner structure of the query
language is organized and generalized to a heavy NCL or (another
extend, to make the grammar simple and uniform CL or (new) NCL
new) NCL
and remove exceptions. They got something simple,
but not what they expected.
Theses two perspectives for simple need to be
matched to satisfy all stakeholders. The theoretical transform all to a
simplicity of the inner workings of the language must common Canonical Level
be brought to the top - both in terms of readable
syntax on the technical side and educational effort The parse tree is your friend
on the human side. Special non-canonical levels Some queries are hard to
which can be added to our existing implementation understand – no matter how fitting
in a modular way can be used to better fit the the style of the language used may
needs. be.
The parse explains a query
by displaying its parts
hierarchically.
It also shows optimizations done.
Montag, 9. November 2009

You might also like