Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
Software

Journal fucksl4shd0t's Journal: More on accounting software 10

Journal by fucksl4shd0t

I've worked out some basics for my accounting package, and I really haven't had time to do this as quickly as I wanted.

I've chosen Python and wxPython as my platform, with Metakit for the file-based database. Since Metakit appears to support multiple processes accessing the same database, I'm not going to worry a lot about having a separate accounting server right now. I will move first to construct the UI while creating a DB-abstraction layer with Metakit implemented for the first versions. It should be a simple matter to add to the DB-abstraction layer later to support other databases.

I've done all the basic research I need, and I've formulated the db schema, so I'm ready to start coding.

Is there anybody else that wants to jump in now, before I get started and make some potentially irrevocable decisions?

Long-term goals are to extract the db-abstraction layer to a C++ library, and from there it will be possible to split the library into an accounting server and a client library if that is still what I want to do. Honestly, I'm so impressed by the metakit api that I don't think I'll need a separate accounting server to provide multi-user access for home users, although such a server might still enable the app to scale to very large enterprises in the future. Since my immediate focus is on home accounting, and then small business accounting, I think that my plan provides suitable room for making it scale upwards while simplifying design for initial implementation.

This discussion has been archived. No new comments can be posted.

More on accounting software More

More on accounting software

Comments Filter:
  • Wish I could offer coding help because it sounds like an interesting project. Not a developer though, and I never will be. I can offer some small help though. I work for one of the big five accounting firms doing IT auditing. Basically, a big part of my job is to audit accounting packages in use at our clients and sign off on whether our accountants can rely on the data they spit out.

    Mostly I look at program change management and user access management - but one of the things I've never seen is a decen

    • YES! That is exactly the kind of help I need right now! I know basic double-entry accounting, and all the literature puts it as fairly straightforward stuff, and since it's a well-known, well-practiced, and well-tested method, converting it to code is the easy part. Making it useful for both accountants and home users is the hard part.

      Go to my website and send me an email through the web form, and I'll be in touch. Basically, as soon as I have some basic code worked up (I was hoping to do it tonight, b

      • Cool. Glad I can help, and I can probably beat the bushes at work for a little more. I checked the website for a form or email and didn't see anything - but I'll dig deeper tomorrow.

        • I'm sorry, I just recently rebuilt my server, and I didn't realize that I had placed FormMail in /var/www/cgi-bin (one of the partitions that got formatted). I have fixed it now, but I haven't yet put the link back onto the website (still some extra work).

          The URL is this [servebeer.com]. It's not pretty, but what do you expect for 5 minutes of work to get the form back up? I'll make it pretty later, bigger fish to fry right now. :)

  • Have you looked at all the existing accounting-project and concluded that what you want is so different from all the others that it's not more productive to help one of them out ? Gnucash is pretty decent these days for personal accounting. For more business-oriented features (allthough those are seeping into Gnucash too) there's Gnu Enterprise, and several others.
    • You want 3 layers in your application:
      • The GUI, possibly there can be several guis. (i.e. one X11-based, one web-based, one text-based)
      • The accounting logic. This is the main part of the application, it needs to know all about things like accounts, transactions, splits, users ...
      • The backend storage. This way there can be also more than one storage-modules which is good because some people want Postgres, some people want nothing but Oracle, others hate Oracle and some single-user installations migth prefe

    • I have looked long and hard, and actually have a history, however small, with GNUCash. Basically, and this is in another journal entry, I went to GNUCash and offered to help them. I offered budgeting, offered proposals, discussed it on the list, and then the guy who was really interested told me he would provide a UML diagram of what he was saying and then disappeared. About two weeks later my window for adding budgeting to GNUCash expired and other responsibilities kicked in. Soon after that, GNUCash p

      • Okay. So you made an effort to try to contribute to an existing project. Good ! Applause to you, and shame on the developers on the Gnucash-list that didn't give you a nicer reception.

        From what you say though, it sounds as if your main problem with Gnucash was not the codebase, but the people. Did you seriously consider the fact that you don't have to deal with the people at all to make improvements and/or add-ons to their codebase ?

        Nothing stops you from importing their latest CVS and then opening up e

        • From what you say though, it sounds as if your main problem with Gnucash was not the codebase, but the people. Did you seriously consider the fact that you don't have to deal with the people at all to make improvements and/or add-ons to their codebase ?

          Have you looked at the GNUcash codebase? It's a C monstrosity! Man, oh man did I think long and hard about forking. I generally oppose forks (they can divide the community in a bad way), but occasionally forking makes a lot more sense then starting from

          • Have you looked at the GNUcash codebase? It's a C monstrosity!

            Sure I have. I agree the choise of C is unfortunate. It *does* however help somewhat that larger and larger parts of the actual accounting-logic are being moved to guile, which has garbage-collection and all that other new-fangled stuff.

            The thing is, there are specific benefits to starting from scratch.

            Sure. I know. The benefits and the drawbacks are essentially the same: You get to choose how to do everything yourself, and don't need to

Slashdot Top Deals

Intel CPUs are not defective, they just act that way. -- Henry Spencer

Close