[Gambas-user] Future Of Gambas (Re: Baze Project)

[Nigel Verity]

Hi Markus et al

One of the big attractions of MS Access is that all the data, forms and queries can be delivered to the user in a single file. This is very attractive when it comes to distributing standalone databases to be run on the desktop. LibreOffice Base, at its simplest level, seeks to do the same thing, but it is seriously hampered by having such a rudimentary form designer, and its implemention of the Basic programming language is more difficult to use than VBA. When you add to that the default HSQL database engine, which is both slow and inefficient, you have a tool which I cannot believe is widely used for anything other than local "quick and dirty" databases, where performance and appearance are of little importance.

Of course, both Access and Base can be used for far more sophisticated purposes. Through the use of ODBC Access can connect to a number of different types of database and, combined with the use of local tables, can extend the core dataset. Again, Base can do much the same, but it remains hampered by having such an unsophisticated form designer and the tedious programming language.

With Gambas it is not possible to achieve the "single file" approach to database application delivery that you can with Access or Base (well, I've not found a way), but it is still very simple to deliver a Gambas executable (assuming the runtime environment has already been installed on the target) and an associated SQLite database in just 2 files. This makes distribution by email, subject to size constraints, a practical proposition in way that is not possible if the data is hosted in a server-based architecture.

Developing a database front-end in Gambas is more like developing such an application in VB than Access, as you still need to create the links between the controls on the form and the database which, of course, is done for you automatically in Access. For all that, though, it is very easy whether you use the Gambas databound controls or prefer to write a few generic read/write procedures long-hand.

>From the description of your Baze project in Google Code, rather than duplicating an existing application I believe that you will end up with a high performance tool which achieves the flexibility aspirations of Base combined with the ability to create a sophisticated, professional-looking user interface. By including support for NoSQL you will be ahead of the game.

It's an ambitious goal, but I do think that as long as ease of use and implementation are always at the forefront of the design, then it will eventually find an enthusiastic user base.

I wish you every success.

Nige