| Project Name | Database Evaluation Project, Phase 2 |
|---|---|
| Project Manager | John J Karcher |
One of the things that the Amiga platform is really missing is a decent DataBase Management System. At the moment there are quitea few custom and proprietary Data Storage File systems, some with their own front ends etc but most of them are just for hobbyist Data Collections or, at a push a small inventory control system.
If the Amiga is to push out into the real world then what it needs is a cheap, industrial strength, fully scalable database solution. We know just from our plans for this web site that we will need such a tool within a year if the site is to become more than just a few directories of HTML & GIFs.
What will such a system give us? The ability to produce commercial applications that can rely on the speed, efficiency, data integrity and security of a DBMS without having to worry about designing all those things ourselves. The ability to code full strength Java applications and distribute them across the web for other platforms to use. The ability to build a powerful, responsive, distributed web site and provide the basis for the beginning of our virtual community. The ability for all Amiga users, from the single user right up to the biggest developer to be able to use the same Database.
We want to see our site migrate towards a Data Repository Model. This provides for a far more powerful architecture, allowing for all data to be stored within data files and then Java Apps or Rebol scripts extracting the data and dynamically generating HTML, VRML etc for delivery to the client. Data replication, site distribution and security become much more simple and correspondingly better while the powerful native searching, indexing and enquiry engines of the DBMS become directly available to the users of the system.
As an example of the power of this model, consider making our site multi-lingual. Without such a model, multiple directories with multiple pages for the same content will have to exist. With it, we store the different language texts as data records, build a Style Sheet for the article/page and then at request time, simply take the client's language setting, extract the text records and merge them with the Style sheet.
This project however is not about developing a database. While we see only good reasons to do such a project, there may be bad reasons or it may even be unnecessary if Java DBs become available. Also there is even now much argument and debate in the commercial world about the future direction and application of databases. We have the standard Relational DBMSs, the Object Databases, the Hybrid approach of RDBMSs' adding so called "object extensions" via modules, cartridges & datablades, OnLine Transaction Processing versus OnLine Analytical Processing (OLTP v OLAP), the problems of BLOBS (Binary Large OBjects) and other complex datatypes. And ultimately, if people do produce Java DBs then we might not even have to worry about writing our own at all.
This project then is a research paper for the society. It should attempt to provide an introduction to the current state of Database technology on other platforms, terminologies and trends, links to useful resources, investigate the current state of Data Storage and Retrieval on the Amiga and then make some suggestions and possibly future actions that we should be making as a society. Hopefully this document will lead to the creation of a full DB design and development project.
John J Karcher - Project Manager
Not yet done.
Project spec completed.
All pages on this web site are copyright 1998 Jay Miner Society.