|
Tanveer
Patel strongly
believes in the Philosophy of XML - the "eXtensible Mark-up
Language", a semi-structured data environment. He has
made education and knowledge transfer a part of his daily
life in metalayer. He leads and educates the "metalayer
1.0" development team.
Tanveer joined metalayer in September, 2000. He is a multi-skilled
employee, 'inspired' by .NET, client side JavaScript /
DOM(Document Object Model), XSLT(eXtensible Stylesheet
Language Transformation) and the core of our latest Java
initiative. Tanveer is an excellent knowledge transmitter,
working towards constant technology evolution.
Interview |
XML(EXTENSIBLE MARKUP LANGUAGE) - COMPUTABLE AND UNDERSTANDABLE
THE
DIMENSIONS OF INFORMATION
Professor
Von Krogh divides information and knowledge into the following
three dimensions:
- DATA
- The basic building blocks of information, i.e. sales, industry,
consumer and financial data.
- INFORMATION
- Information is data that has been arranged into meaningful
patterns. It may be in the form of sentences, policies models
or tools.
- KNOWLEDGE
- Knowledge gives the potential to act and also gives the capability
to transform data and information into ideas and decisions.
It can be either explicit or tacit in nature.
Some authors add as a fourth dimension "wisdom",
but for our purpose these three are sufficient.
STRUCTURED
AND NON STRUCTURED INFORMATION
With
a focus on computers, we could extend these three dimensions by
defining where the three dimensions reside:
-
Data resides on hard disks - within files and relational databases.
-
Information is the representation of data, mainly screens, print-outs,
charts etc.
-
Knowledge resides in human brains.
What
are the dominant standards of representation for different dimensions?
-
Data is stored in a STRUCTURED FORMAT - the dominant form are
tables and relational records.
-
Information is stored best in a SEMI-STRUCTURED FORMAT - XML,
HTML, SGML.
-
Knowledge is represented and transferred in NON STRUCTURED FORMAT
(or maybe better: less structured format): Natural language
is the dominant standard here, but also other human communication
forms, like symbols etc.
The
relational data model has been widely accepted as the dominant
data standard for structured data. Its tight rules and rigid structures
ideally support transactions and predefined processes.
Co-creation,
human communication and knowledge transfer are unstructured and
chaotic and it becomes difficult to model and support co-creation
with relational data structures. The "eXtensible Mark-up
Language" fits more naturally to the creative minds.
PROCESSING
STRUCTURED INFORMATION
Machines
have been extremely successful in "data processing" - in the ability
to work with and on STRUCTURED INFORMATION. The history of computing
can be read as a success story of automating all processes and tasks,
which could be represented within the structured model. Relational
data stores, Enterprise Resource Planning (ERP) software, order
generating and processing, production planning and finance and trading
systems are common examples.
Top
AI
- AN ACADEMIC EXPERIMENT
The
history of Artificial Intelligence is as old as the history of data
processing. Since its earliest days, computer scientists are experimenting
with the processing of non-structured information and knowledge.
Until today, we have seen very little commercial success of Artificial
Intelligence. Computers today have been able to replace millions
of people calculating in banks, insurance companies and the finance
departments, but even the best computers today are not able to understand
the simple language, which even a 5-year-old child speaks. No correct
translation system is available today and computers are simply not
able to get any meaning out of our form of knowledge representation.
No computer system is able to "understand" speech, a picture, symbols
or any other format we use to transfer knowledge.
THE
WEB - MOVING FROM DATA TO INFORMATION
In the 90s, the World Wide Web was the big success story
in the computer world. Even though there were many attempts to implement
a data focused philosophy on the web (the Internet as a global network
of databases), the INFORMATION paradigm became the dominant model
for the web. Not data records, but information in HTML and XML are
transferred over the web. Computerization is moving from the Data
and the structured information model to a semi-structured information
model - from a Data processing to an Information processing paradigm.
It was the first time in history of computing that the Information
paradigm became dominant in a wide area.
Top
HTML
AND THE INTERNET
In the early days of the Internet, the network was used
to connect and synchronize databases. The dominant data processing
model was soon taken over by the information model, as it was much
more natural to what people used the web for - to communicate, work
together, exchange information and knowledge. The real success story
of the web though was not through the exchange of relational structures,
but through HTML, a semi structured format. Today, if you think
of the Internet, you think of the web and HTML - even database exchanges
have adopted a semi-structured model, as relational structures were
just too complex and restrictive for distributed environments on
the web.
ALL DATA STANDARDS ARE EQUAL
Goedel and theoretical computer science tells us, that
"all computers are equal". This means that there are clearly faster
and slower computers, but there are not 'more powerful' machines.
Any algorithm which can be deciphered by one computer, can also
be understood by all other computers (Von Neuman architecture).
The same is true for information representations: All representations
are equal - whatever you represent in relational data, is also representable
within any binary standard, XML, HTML, the natural language or any
other representation. All common information representations are
universal. So, choosing a representation is not a question of what
can be represented, but much more of choosing a standard, which
fits naturally to the domain you would like to apply it in. Financial
information, order lists or customer databases fit very well in
tables and relational structures, and for processing and automating,
relational databases are great tools. But a study though can be
represented within a relational structure, fits much more naturally
in natural language. HTML and XML are becoming the dominant standards
on the web because they fit much better into the domain of information
that the web transfers and exchanges; information and knowledge
rather than data.
Top
XML - A COMPROMISE BETWEEN COMPUTERS AND HUMANS
- The
tight structure of the relational model makes it very simple
and straightforward for a machine to process. So, the relational
model has come to be the best model for computing.
- Unstructured
(or less structured) information like the natural language is
the form in which humans store and transfer knowledge, but this
language is not 'processible' by computers.
- The
semi-structured model of HTML / XML is a "compromise" between
the two standards. It is "processible", as there are tags to
process it, and it gives enough flexibility to cope with the
unstructured human language.
XML
PHILOSOPHY
Other than most of today's state-of-the-art applications,
which use XML as an Information exchange standard only, metalayer
is using this new technology to store knowledge and as an application
philosophy throughout the product. This provides the metalayer platform
with a flexibility and eXtensibility which no current technology
can provide.
metalayer
is using XML throughout its product - not only to exchange data
between different layers, but also to store information and as an
application-processing model. Or from a technical perspective: As
today's applications load and manipulate records of a database,
metalayer loads and transfers XML within a DOM model (Document Object
Model).
Co-creation
and collaboration are not fixed processes, but rather chaotic -
and they have to be unstructured to "cope up" with unstructured
and chaotic idea generation and innovation process. The early decision
to fully build metalayer not only on XML technology, but also on
an XML philosophy provides metalayer today a key advantage over
the competitors. XML is the best available technology for processing,
and is still very close to the human thinking process.
With XML the next paradigm shift to open architecture solutions
that support human conversations becomes possible, allowing people
to acquire knowledge in the manner they know best - the marvellous
and chaotic manner of human communication.
Top
|
