Sound advice - altblog

Tales from the homeworld

My current feeds

Sun, 2004-May-09

My old weblog, as it stands

I mentioned in my first blog post this era that I've been a blogger in the past. I've collected as many fragments of the old page as I could lay my hands on through wayback machines and the like and I've collated them for the viewing audience. Most actual blog entries (the writings pages) are missing. The occasional one still does exist. If anyone reads this who has copies of old pages that I didn't manage to revive I'd sure welcome their return to be put back up.

Anyway, without further aduour: my old blog

Sun May 9 22:23:56 EST 2004

Sun, 2004-May-09

A first post

I'm actually a blogger from way back. It's been a long time. I used to do it all the time at uni, about five years ago. I've been inspired by the creation of planet humbug to start putting up a few entries again. Perhaps an old listener or two will tune in again.

I'm not very good at completing things, particularly at the moment. I've been working fairly intensively on work-related things for the past few years, and haven't used the weekends for much but recouperation. This should be understood as a backdrop to anything I say in the next few paragraphs:

I would like to write an accounting package. I've had a passing interest in personal accounting for quite some time and have records stretching back to about 2001. The records I've kept have used Quicken, and old version I picked up in I think 1999. As an open-source zealot who only runs linux on my PC's these days it would be remiss of me not to look at the free alternatives out there.

The current alternatives with the most promise are probably gnucash and kmymoney2. Both have many good points, but don't match some requirements I have for the software and neither appears particularly well-suited from an architectural perspective to adapt to something new.

My requirements are these:

  1. For practical adoption purposes, the software must run under modern Microsoft Windows variants
  2. For idealogical purposes, the software must run on a free software stack. This stack must match a definition of free that permits inclusion into the debian linux distribution
  3. The software must cater to multiple financial jurisdictions, and be built in a modular fashion that allows me to say "I live in Australia" and have it understand my tax system and gaap
  4. The software must provide excellent reporting functionality. It must surpass Quicken and other commercial products
  5. The software must be as easy to use as ms money but as easy to tune as gnucash.
  6. The software must strongly geared towards a future where an application such as itself will be able to contact local banking or financial organisations online to perform online reconciles of bank accounts and to lodge, modify and manage, scheduled (future-dated) transactions to be undertaken on the user's behalf by the institution.

I've started to draw up diagrams of how I want various interfaces to work, and I've begun to settle on a broad architecture and software framework. I think I'll be using c#, targeting the mono/gtk# platform. I figure that will give me the best combination of free platform capability and running sufficiently well under windows. I'd actually like to see this kind of thing as a way for free software to penetrate the consciousness of windows users.