BibWorm Core Class Reference Growing

As some of you may have noticed, the class reference for the BibWorm core is steadily growing. I have also improved the PHP code behind the class reference page a little, but these improvements are minor, concerning mainly the structure of the script and displaying additional info such as generics (especially allowing crossreferences to generic classes by separating type parameters from the class name) and calls to constructors of super classes (base in C#).

I'm not sure whether I've said this anywhere before, but the class reference generally follows C# syntax e.g. for class headers, method signatures, properties (accessors), and especially the notation of events. I do realize that these things are somehwat differently put in Java, but firstly I'm more familiar with C# and secondly I believe that C# notation is a little less complicated for the purpose of this reference, which is essentially a design document. I am also vastly in favor of C#'s convention concerning the naming of interfaces since in my humble opinion IComparable is so much clearer than Comparable, or IDatabaseWrapper to DatabaseWrapper. At least you know straight away that you're dealing with an interface, whereas Java had me a little confused about these things at first (I'll just say List, which in C# refers not to an interface but a class).

Started Publishing Class Reference for BibWorm Core

Finally I've come up with a solution to the publication of the class reference for the BibWorm Core. Last night I figured out a managable way which has a PHP script assemble the whole thing from data contained in the project database, and this also allows me to have crossreferences dynamically generated. It's technically neither impressive nor difficult, but it's the easy solutions that seem to evade us most of the time...

I only just started entering all the class details, so please don't be disappointed at the lack of information. More will be added continuously as I find the time. I am also going to produce a little diagram to illustrate inheritance. Stay tuned for more! :)

Web Survey Online!

Finally I managed to put a user survey online. Its intention is to gather further information from users to provide the kinds of features they need. I am not expecting, however, that its results will have much of an impact on the fundamentals such as the data abstraction layer. The survey's main impact will rather be on the design of user interfaces, plugins and other modules.

At this point very many thanks to Dan from Cambridge University's CARET for his input while drafting the questions. For all those interested, the survey has been realized with phpESP. Due to the peculiarities of sourceforge.net's web hosting environment, I had to configure the survey on my local XAMPP server and feed the SQL dump into BibWorm's MySQL database. Voilá!

Over the next days and weeks the main focus will be to update the development documents to include the core's interfaces and the data abstraction layer classes. Drafts for all of these exist, but I am as yet uncertain how to present them. If all else fails, I'll try to finish coding them (probably in C# contrary to my former intentions of completing the Java core first - I just can't afford losing too much time right now learning a new language as term is about to start), so that there will at least be something in our subversion repository.

The Website is Growing

Slowly but surely this website is growing. The development section now has at least a figure of what I like to call the BibWorm 'universe', illustrating how manifestations of BibWorm slot in with one another and with whatever other services and software are out there. Check it out in the development docs section.

Mostly, however, I have been working on a bit of PHP to allow turning my verbal eruptions in this blog (and those of fellow developers who may hopefully one day find this project worthy of their time and effort and join up) into an RSS 2.0 feed for wider circulation. It's all about spreading the word. Quite a good thing to learn from Jesuit missionaries, actually. ;) Even techies can learn from history (and historians from techies, of course).

Once again my attempts to come to grips with an as yet unfamiliar language has illustrated the immense power of Google. It is an extremely helpful service (as you all are already aware of) but let's face it: it's a huge filter on information. Millions of people use Google exclusively every day, and what their index doesn't list - and more importantly, what the engine doesn't throw up on the first two pages - is hardly ever seen by anyone. Makes you think. Being somewhat naive, however, I prefer to believe that what Google's search turns up is actually what is most relevant by objective criteria so let's not question this any further!

Number 1

Everything has to have a beginning, right? So this is the first entry in my very own BibWorm development blog. In it I hope to share as often as possible (but irregularly) the occasional thought related to this project, its design and particular aspects of coding.

Currently, however, I am occupied with building this website (and studying the history of the Jesuit mission to China, in case anyone cares). Here's some trivia: Originally, the plan was to use Joomla. But since sourceforge.net mounts webspace as readonly I had to learn PHP instead and build this from scratch. I have to admit that I am enjoying this. :) Some of you who are familiar with Joomla will have no difficulty seeing how much the design for this site has been influenced by it. But why reinvent the wheel when imitating it makes so much more sense?

In the next couple of days I will hopefully manage to get started on publishing all the design documents for BibWorm so that you people can get an idea where this is headed. So please do check back soon!