Analogy DBS - connectivist/LaaN-based learning system

The implementation of a connectivist/LaaN-based learning system can be similar to the implementation of database systems (DBS). I don't mean that connectivist/LaaN adopts an objectivistic view of learning where knowledge is viewed as an object and as a consequence can be stored in a database. It's just that we can borrow some of the concepts used in the implementation of DBS and apply them when trying to implement a connectivist/LaaN-based approach to learning.

Let me first give a brief overview of the architecture of a DBS. More information about this architecture can be found in the following references:

- H. Garcia-Molina, J. Ullman, and J. Widom: Database Systems: The Complete Book. Prentice Hall, 2001.

- R. Ramakrishnan and J. Gehrke: Database Management Systems. 3rd Ed. McGraw-Hill, 2003.

- T. Härder, E. Rahm: Datenbankssysteme - Konzepte und Techniken der Implementierung. Springer, 1999.

A DBS consists of three main components: (1) Data System, (2) Access System, and (3) Storage System.

Data System:

The main component of the Data System is the query processor. The query processor’s task is to turn user queries into a sequence of database operations and execute those operations. The query processor performs two operations: (1) query compilation and (2) query execution.

At query compilation, the query is parsed and a parse tree for the query is generated. Then the query optimizer, which is the most important component of query compilation, first, converts the parse tree to an initial query plan, which is usually and algebraic representation of the query, and then transforms this initial plan into an equivalent plan that is expected to require less time for execution. Second, the query optimizer turns the generated equivalent plan (logical query plan) into a physical query plan by selecting algorithms to implement each of the operators of the logical plan, and by selecting an order of execution for these operators.

At query execution, the algorithms selected at the query compilation phase are executed. These algorithms fall mainly into three classes: (1) sorting-based methods, hash-based methods, and index-based methods.

Access System:

The Access System uses auxiliary structures to speed up the retrieval of records in response to certain search conditions in order to avoid scan of all pages or segments or even the whole DB. Indexes are often used in speeding up queries. The indexes commonly used in commercial DB systems are B-trees and hash tables.

Storage System:

The DB data is stored on external storage devices such as disks and fetched into main memory as needed for processing. The unit of information read from or written to disk is a page (block). The basic abstraction of data in a DBMS is a collection of records, or a file, and each file consists of one or more pages. In other words, a DB relation (table) consists of a set of records and is stored on one or more pages of a disk file.

The table below shows the analogy between a DBS and a connectivist/LaaN-based learning system.

DBS	Connectivist/LaaN-based Learning System	Example
Relation	- Personal Knowledge Network (PKN); - Knowledge Ecology (KE)	- Network around an Edublogger - Network of Edubloggers
Record	Knowledge Node - Tacit Knowledge Node - Explicit Knowledge Node	- Edubloggers - Blog posts, videos, audios, slides produced by edubloggers
Index (e.g. B-tree, hash table)	- Social Network Analysis - Long Tail Aggregators and Filters	- Blogwatcher, NetLearn - Google Blog Search, Delicious, Digg
Page	Personal Knowledge Network	Network around an Edublogger
File	Knowledge Ecology	Network of Edubloggers
Disk	Collection of Knowledge Ecologies	Blogosphere, Web

In LaaN, in order to learn, we build, maintain, and extend our personal knowledge networks with new explicit/tacit knowledge nodes. The diagram below depicts the processing of a simple query in the different layers of a DBS and a connectivist/LaaN-based learning system.