DATAdesign Manual : Introduction

PROGS, Professional & Graphical Software
Dr. Frans Hemerijckxlaan 13 /1
2650 Edegem
BELGIUM
tel : +32 (0)3/ 457 84 88 fax : +32 (0)3/ 458 62 07 e-mail : joachim@club.innet.be
www : http://www.club.innet.be/~year2827

Throughout this manual you will find the following symbol *. If you are reading this in the html Reader clicking on this item will take you immediately to the index.



* Section 1 : Disclaimer and Copyright Messages

DATAdesign software and documentation are copyright with all rights reserved. No part of this documentation or the DATAdesign software may be copied, reproduced or stored on any electronic or other media except for personal use.

The ptr_gen, wman, hot_rext and config files are copyrighted by Qjump.

The menu_rext file is copyrighted by Jochen Merz.

The ConvOLD, Import, and associated text files are copyrighted by Wolfgang Lenerz.

In no circumstances will PROGS, be held liable for any direct, indirect or consequential damage or loss arising out of the use or inability to use this software and its documentation.

* Section 2 : Introduction

This is the manual for DATAdesign. We strongly advise you read it as it contains important information relating to the use of this program and because it may help avoid many of the problems that may otherwise occur during its use. Most of the enquiries we get about our programs could be solved by reading the manuals.

This manual does assume that you are acquainted with the Pointer Environment. Don't worry if you aren't, we have included an introduction with this document

* Motivation

DATAdesign is actually a special kind of database because it is 'free-form'. This means that the lengths of the fields are not limited, and that fields can be added and deleted when necessary. The main advantage of this is that you don't have to know in every detail how the data in your database will be arranged when you create it.

For instance if you created a database with four lines for the address and you come across someone who needs five, what would you do? These kind of problems will never occur in DATAdesign.

On the other hand if you suddenly realise that you also have to keep the VAT number of some customers, you can just create a new field which contains just that.

Of course the fact that DATAdesign is a free-form database also has some negative implications. It is the case that databases where all fields have a fixed length are faster to access. You always know where your record is, and if you change the record, it can never grow larger than it was, because all records are equal in length. A fixed-length database can have shorter files, which also load faster, even if all fields are almost filled, because they don't have to know how long a field is (it is always the same). But free-form databases can also be much smaller and more practical in some cases. Can you imagine what a database of letters would do in a fixed-length database, or even worse, a file which contains chapters of a book.

* Section 4 : History

This is actually the fourth incarnation of DATAdesign.

It is a program which has gone through a (r)evolution.

It all started with DATAdesign v1.00. This version of the program was not much more then a simple cardfile database. Of course it already used the Pointer Environment, but that says it all.

After almost a year, and mostly because Wolfgang Lenerz asked for it, we split the original program in two parts: the main program and the engine. This had the main advantage that the DATAdesign database management system (dbms) was accessible by everyone, so that there was finally an alternative to Archive.

This however had two problems, we had to rewrite big parts of the main program, as we had to call all the routines in a different way, and because we also added some extra features like filter, mark, duplicate and shortcuts to access some routines. We didn't write it properly though, so DATAdesign v2 relied heavily on the internal structures of the engine.

The other problem was the engine (v1). We supplied it in two parts, the engine itself, and the Basic interface. The engine lacked some commands to access some of the embedded data, and the Basic interface was quite stupid. You had to explicitly type all the parameters, and parameters were quite type sensitive (it would not pass an integer as a float).

About seven months later we released the second version of the Basic interface. It was a major improvement because the commands were more readable and because you could omit the parameters you didn't need. The interface was also more friendly about accepting parameters of different types, but it had problems with array elements. We also made sure that the files were accessed more easily, with the disadvantage of having more problems when accessing several files simultaneously in a program.

About 16 months after that version we released DATAdesign v3. It was still sold in two parts, as not all DATAdesign users needed the programming interface. It also included the new Basic interface directly into the engine, and the engine became v3, to give the engine and main program the same figure in front of the dot in the version number.

The DATAdesign main program was now finally a program which accessed the engine properly, without accessing the internal structures, and we redesigned the window. Actually this was (another) complete rewrite of the program and we tried to make it better than it ever was.

The major difference was actually contained in the engine. The Basic interface was now as it was intended to be. Very flexible with respect to the the parameters, and with an improved method to handle errors.

It was now also easier to access several files in the same program, thus, together with unique numbers for fields and records, creating a powerful relational database for the programmer.

More noticeable to the average user was the introduction of disk-based files and indexes. The first allowed for large files to be accessed; this should prove very handy for all those people with large databases like doctors, lawyers and people who compile 'Qlaw' databases. The latter provides a fast search within a program and better handling of sorted and/or filtered files.

The engine also allows types to be given to a field, thus allowing numeric values and also raw data like screens or fonts or anything you want.

1996 saw the release of DATAdesign v4. This was yet another re-write of some of the code undertaken by Bernd Reinhart and featured an improved user interface in the shape of the Icon Bar and this manual arranged into _html format and partly re-written by Roy Wood.

* Section 5 : Credit's due

There are some people who have been a great help during the development of DATAdesign. To start we would like to thank our parents, as we wouldn't be programmers without them, they have always supported us and continue to do so. Thanks.

We would also like to thank Tony Tebby and his team at Qjump, They have made the task of programming a user interface much easier, and provided a powerful interface for the engine with the thing system.

Jochen Merz by writing the Menu extensions provided some practical windows for things like loading a file, reporting errors, confirmation requests, etc.

We would also like to thank some people who gave some help and good advice to improve the program and/or manual (no special order):

A special thanks has to go to Wolfgang Lenerz. Partly for beta testing this program, and mainly because he gave a lot of guidance in what the next step in the development of DATAdesign could/should be. He also wrote a utility program, ConvOLD, which is provided on the accompanying disk to convert v2 DATAdesign files to v3 or v4 and giving you the option to convert the data types. He also wrote a similar program, ConvExp for converting Archive export files to DATAdesign.

Another special thanks has to go to Sohail Bhatti from QLAW. He used the DATAdesign engine to compile its database and thus came across some problems that he reported and we tried to solve. He is a big fan of DATAdesign and tells this to others.

Yet another special thanks has to go to Dilwyn Jones from Dilwyn Jones Computing. When he was our U.K. dealer he saved us from a lot of work marketing our program and allowed us more time to concentrate on programming.

As mentioned above DATAdesign V4 only exists because of the work done by Bernd Reinhardt who has improved the user interface and, in the process, fixed some of the bugs.

Also for DATAdesign V4 Roy Wood has done a great job in updating the manual to reflect all the changes, making it more readable, and converting it to hypertext format

Finally, thanks to you all, we need your help and support for this program (and any other program we (and others) write). So if you have any suggestions, don't hesitate to write or call us, and we can try to help you, and you can help all the other users of DATAdesign. Worse: if you find a 'b-u-g', you have the moral duty to tell us when-where-and-how you found it, so we can help you all by correcting it.


This document is mostly based on the original text for the DATAdesign Manual revised and converted into html format by Roy Wood
PROGS, Professional & Graphical Software
last edited Saturday 24 August 1996