There are several directions that the project can move in. These directions serve different constituencies, different communities of users. Many of these seem incompatible, and so the tension must be resolved, or a choice must be made. The list below is ordered along the lines of 'most obvious' to 'far out there'.

Small Business Version

This is probably the #1 most requested 'enhancement' to GnuCash. The use of GNU/Linux is sky-rocketing in small businesses; yet there are no easy-to-install, easy-to-administer small-business accounting programs, at least not of the 'open' kind. The list of desired features dates back to the dawn of computing half a century ago: payroll, accounts-payable & receivable, invoicing, billing.

There are some real problems with moving in this direction. Not the least of these is alienating GnuCash's core constituency: the casual/home user. If we build something with whiz-bang features, the home user may conclude that this is an application that is too complex, too hard to use, too hard to understand. They'll go looking elsewhere, and meanwhile, we will have done GNU/Linux (and *BSD) a dis-service. The adoption of GNU/Linux by average (home) computer users is gated by the availability of the basic desktop cornerstones. Personal Finance Management is one of those cornerstones. We've got to have a simple, easy-to-install, easy-to-use personal financial manager.

Sophisticated Investor

GnuCash downloads stock prices and currency exchange rates from off the net. The current beta version has some sophisticated graphing capabilities. How fancy should one get? The new GnuCash reporting structure should make it easy, or at least straightforward, to create a variety of sophisticated graphs, charts and reports. These might range from simple technical analysis such as 90-day moving averages trading volume and beta, to arcane indicators like 'Bolton-Tremblay', 'Zweig Thrust' or 'McClellan Oscillator'.

Downside? The 'investor' community is smaller than the home user community. And stock technical analysis doesn't necessarily require accounting features. Thus, such needs may be best served by other systems: and there's a variety of these to choose from. Should GnuCash reject stock-oriented patches? No. Should core developers focus on stock trading? Probably not.

The Financial Browser/The Ultimate Home User Experience

As the user interface gets slicker, and the druids become more and more helpful in guiding the novice user, a natural direction for GnuCash would be to become the ultimate home user financial experience. That is, to build the 'Financial Browser for the Web', the 'Nautilus of Financial Transactions'. For starters, this means using OFX and supporting on-line bill-pay and banking. In some distant future, this means integrating with on-line shopping, so that your shopping spree shows up in GnuCash without your having to hand-enter all those transactions.

Some of this integration is mundane: e.g. using an appointment book (for example, the calendar in Ximian's Evolution) to show upcoming bills and budget items. Or even using the calendar to enter bills into GnuCash. More broadly, providing a set of accounting interfaces such that any other package (for example, a stock technical analysis package) can push data into, and get data out of GnuCash: the holy grail of desktop integration.

The Financial Browser, Part II: the PDA

GnuCash, the Gnome Desktop, and the KDE desktop have been flying in the face of 'common wisdom': the coming extinction of the general-purpose PC. According to this theory, lithe, portable, special-purpose hand-held devices will make PC's irrelevant. Of course, the theory is flawed: a large keyboard and a large screen provide convenience that no small palm-top can provide. None-the-less, one must look over ones shoulder. Can GnuCash be slimed down and fit onto one of those Linux hand-helds, e.g. the YOPI? Right now, it depends intimately on a large number of packages for many features, and so a sliming exercise is hard to imagine. But then, a financial manger is a cornerstone application for the desktop, it is equally so for the palm-top. Someday, not too long from now, wireless, hand-held 'financial browser' will be as ubiquitous as the wallet.

The Financial Browser, Part III: the ASP

Gnucash has a built-in web browser. This browser can not only show standard web pages, but it also understands special markup for interactive graphs, and even embedded checkbook registers (for example, take a look this page with a recent build of GnuCash). GnuCash uses these features to create and display reports. There is no reason to have these reports be local to the desktop machine: they could be on 'any' website, any server.

GnuCash uses XML as a file-format, but this XML could just as easily be served up by a cgi-bin running on a web server. There's some prototype demo code in src/experimental/cgi-bin, or you can try loading this link with GnuCash. In other words, GnuCash can offer a nice, slick, web-integrated front-end to a web-based financial ASP server. A front-end that spanks any other financial ASP out there.

What's wrong with traditional ASP's? Hmm, let me count the ways. (1) World Wide Wait. No desktop app on some old 386 ever took 15 seconds to respond when you hit the 'enter' key. (2) Those 'beautiful' web forms. They might be pretty enough for asking you about airline flight info or filling in your name and address, (yes, CSS makes web pages prettier), but they still don't hold a candle to desktop-specific apps. Imagine being able to use a real ledger interface for working financial info in a web site. (3) Disconnect from the net. Even if your ASP never crashes, and you never lose your network connection, its still conceivable that you might want to fiddle with financial info on your laptop somewhere where there's no handy Internet connection. We think we know how to make GnuCash work in 'disconnected' mode, using cached data, which can then sync up with the big 'ol server when it gets back on the net again. That's what the GnuCash engine is designed to do. (Demo soon?)

ERP

These days, it's inappropriate to talk about business accounting without also mentioning ERP in the conversation. GnuCash is not, and probably will never be an ERP system, but might become the component of one. The basic idea behind ERP is that different businesses have different business practices: no one does things in quite the same way. And so, there is no 'one-size-fits-all' application; instead, one hopes for a framework with extensibility, easy integration, and rapid-development features. The hope is that one can quickly hook up that brand-new ecommerce package to that decades-old billing application. GnuCash doesn't do this. Nor is GnuCash easy to modify, extend, or enhance. It takes some considerable sophistication to add new features to GnuCash. In particular, this means that it's hard to integrate GnuCash into your business processes: you can't just tell it to suck out financial info from here, and dump it to there. If you want to do this, go check out GnuE at http://www.gnue.org/: it was specifically designed for this.

On the other hand, the GnuCash infrastructure is becoming general enough that it wouldn't be hard to mold it into a variety of related financial tools. And if GnuCash really is to succeed in the business accounting world, it will have to become savvy in talking to other systems.

Napster for Money

If you thought the music industry hated Napster, just wait till you see how the feds and the banks react to its financial equivalent. Suppose there really were an infrastructure for putting secure, anonymous eCash into your eWallet. Or rather, of giving it to your friend or business partner, and they in turn could spend it on real goods and services. Now, don't confuse the idea of an eWallet with the idea of eMoney. There are any number of efforts underway to create eMoney: some with pretend-money, such as flooz, or frequent flyer miles, some with real money, such as e-Gold, pay-pal, or the latest ecommerce scheme from your credit-card company. All of these schemes require a centralized server (traditionally called a 'Bank'). This centralized server validates your identity, and confirms that you have the money you claim to have. And actually moves that money around. And works in concert with the IRS to make sure you don't have unreported, taxable income. And with the DEA, to make sure you're not money-laundering. And with... well, that's a matter of national security.

By contrast, my wallet is a 'bank' in the sense of holding money; it just doesn't charge me a service fee for doing so. And when I use it for some anonymous peer-to-peer money sharing, such as paying the kid down the block to mow the lawn, it doesn't tell the IRS that the kid owes 'em money any more than gnutella tells Metallica about song downloads. So, where were we? Oh yes, now try to imagine an 'eWallet'.

Now, GnuCash will probably never be the Gnutella for money. For several reasons. Whatever it is that you exchange on-line, its just funny-money, the electronic equivalent of chain-store coupons, until you have a broker, an under-writer, a market-maker who will guarantee the exchange of your eCash for that other kind of money that banks accept. And as you may guess, the resulting shadow economy would prompt the Federal Reserve Bank to shut down said underwriter faster than Alan Greenspan can say 'the expansion in the M2 Money Supply Index'. None-the-less, the cypherpunks made a promise in 1995 that someday, everyone really would have an e-Wallet, and this concept is still fun to think about even as it remains futuristic.

The last word?