Coding and design

I have managed to arrange for a chance to present the XMLoverHTTP project in a short presentation at work, hopefully it will generate some interest in the properties of the technology.

While I have been slowly making progress on the code of the project itself (invested relatively little time in it these past few weeks due to inevitable circumstances) I have also taken some time aside to think about designing some of the accompanying applications.

Specifically, when I was coding up a quick website project, I realized that while I have spent a lot of time programming complex back-end programs that are ultimately required to produce data that machines can process, I should also have spent some time to implement trivial front-end applications that display the data harvested in this way. The technology (XML) and why it is so interesting is much easier to explain once you have a simple example to show how everything just works.

Creating the aforementioned website has also shown me, again, the value of creating frameworks and how they can save a lot of time in subsequent programming. So I was looking to design a framework that would serve to create the kind of simple user interfaces that I'll need to make in order to create the XML-based accounting software I am ultimately aiming for. My current idea is creating a framework that allows the creation of XML+XSLT based "Internet Spreasheets", which's cells can be linked to XML data available on the Internet if necesary and the results can be either displayed as the spreadsheet itself, or read directly in XML, for processing in other similar spreadsheets or other XML-based webpages. The cell math equasions would be XPath based for processing in XSLT and I have also been looking to the xsltforms project to simplify communication between the client and the server to directly usable XML.

The aim of these procedings is to create software that people can be comfortable enough working with, to get used to the idea of making calculations based on direct sources of accurate data rather than of one-time copied averages or approximates in instantly outdated books and such. Once a network of such data exchanges exist I persume having this data on the Internet will seem much more interesting.

As a side effect I think the back-end projects I was working on before will be able to use some of the functionality created here (for example feedback from XForms can be used as input controls for manual hardware regulation). And although I am somewhat confused as to what my priorities should be (there appears to be insufficient time, no matter how I look at this), I will always keep working on something.

As a side note, I have also moved my Senzor project to a Mercurial CVS, since due to certain network restrictions in effect I have been syncing the code using an USB key, this should proove quite valuable.



More meetings are scheduled to occur on Friday (3. 7. 2009 in Izola; meeting with Raffael of NET), during the weekend (4. or 5. 7. 2009 in Maribor; ZM meeting) and Tuesday (7. 7. 2009 in Ljubljana; meeting with Gorazd of BrainBox Community). I am not sure if I will manage to take part in the weekend one, but everything has been arranged for the other two.

At the same time, I am continuing development (learning along the way) required to complete the XMLoverHTTP project. I have spent time programming a batch of hobbyist applications, which purposefully included an asynchronous pipe mechanism, which I intend to use in XMLoverHTTP, to overcome the problem of pushing commands from a web-based interface (PHP; with very limited permissions) to a hardware control program (Assembly; running with root permissions for hardware I/O).

I am noticing a difference in pace between the social and technical aspects of the progress we need to do and I fear that they may represent a problem on the long run. In simpler terms, we have been having a lot of meetings with people, a lot of discussion taking place, a lot of knowledge being exchanged in a short amount of time; on the other hand the software development, hardware repairs and construction projects have been naturally progressing at a substantially slower pace, and they cannot go any faster due to purely practical reasons. I fear that in expecting results from the practical projects, we will soon have nothing more to talk about and interest in the movements will fall. These differences in pace may even lead people to believe that we are "all talk and no work".

Regardless I hope for the best and intend to keep bussy.