Okay, so this is way off topic for the stated mission of this blog (except that it’ mine and I can write about whatever the hell I damned well please! 
) but. . .
If you’re a linux user (and I am), finding a decent accounting program is enough to make you throw up your hands in despair! Oh, it’s not that there aren’t a lot of choices for the small business person (which I also am), it’s just that none of them quite “fit” for various and extremely frustrating reasons.
Let me begin by explaining that for nearly seven years I had used a proprietary accounting program written by my grad-school brother. It was a DOS program, however, and a program that, after completing his Ph.d he discovered he didn’t have the time to update and convert to universal code the way he wanted to. Besides, other programs were emerging, and they were produced by majors in the computer sciences. So that left me kind of hanging out there on my own. At the time (this was a few years ago), save for a few check book programs, there were no programs that I could find that would run under linux. GnuCash was the oldest, and it was double entry (required for businesses, which run under the accrual system), but it still seemed pointed entirely toward individuals and their retirement accounts.
Frustrated, I started using OOcalc to keep records that would satisfy the IRS, but little else.
Unfortunately, the burden of trying to keep up when everything is done manually was (is) exhausting when you also work all day. And, there are many jobs that require “standard bookkeeping practices” capable of external audit. And while OOcalc is a beautiful piece of work, but it just can’t replace a real accounting program when it comes to audit requirements.
So, the hunt was on. It’s not done yet, but I have to say that, after 3 weeks I’m frustrated. Here’s what I’ve learned so far:
GnuCash: Of all the program’s I’ve looked at, this is my favorite. Not in the functionality department (which in many ways sucks weenies if you’re trying to manage a business), but ease of use. Give me 15 minutes and I can do not just one day’s bookkeeping, but 2 or 3. It has a beautiful user interface (if you understand basic accounting) that “remembers” your entries in various journals. So in one or two key strokes you can re-enter similar transactions. Also nice is the semi- automated on-line banking feature. By various means you can download your bank’s transactions and GnuCash will do its best to figure out how they match to your existing entries. Once you’ve helped it out and made sure everything is the way you want it, clicking “okay” automatically “clears” the matching transactions. A great time saver in keeping track of how much money is going into and out of that busy business checking account.
And of course, GnuCash has a lot of features for investors. The investment community was where it originated, so that’s where a lot of its strengths lay. In what other GPL program would you be able to find the ability to automatically calculate a stock split?
Less impressive is the newly added “business” module. The module is very simplistic. Though it has a “job” function, so that you can track expenditures and charges based on a specific “job” (or project, as other programs call it), the “job” function is not tied into the GL. So anything actually paid for in cash in regards to a “job” is immediately lost. Almost equally annoying is the fact that every transaction requires you to start over. Select vendor or customer, find the vendor or customer (which requires entering a search string), and then you can finally get down to doing whatever you wanted — for that transaction only. For the next transaction you get to start all over again at the pull down menu.
GnuCash also lacks basic “virtual accounting” functionality. Meaning a program that automatically adjusts for (at a minimum), deferred and accrued income and expenses within a given time period. GnuCash does not do this. In fact, GnuCash it has no way to “close” its books at all, either literally or dynamically! And that in and of itself is its death knell for business users.
But perhaps the most frustrating aspect of GnuCash is the attitude of the development team. They treat their user base as an annoyance, rather than a necessity. GnuCash seems to be more of a play thing to them than a project on which other people’s livelihoods might depend.
I’m afraid that until or unless GnuCash is taken over by a serious team or company (like Open Office or the Gimp), it’s never going to be more than a toy with which wealthy investors can play with their money.
TurboCash: When I found out about this piece of software I got exited. Here, I thought, was GnuCash for the entrepreneur. And so it is. It adheres to the GPL school of thought (though it’s not GNU software), it’s overseen by an international team with help from hundreds of others, and seems to be responsive to the businesses around the globe that are dependent upon it for their business bookkeeping needs.
And then the bug landed in the ointment: The developers chose a Borland (read: Micro$oft lacky company) programming language. TurboCash is a Windoze only program. I mean, to the point that it won’t even run under CrossOver Office! And of course, since the original coding was done in a Borland language, switching to a GNU/GPL form of that language is going to be a lot of work. (Though there’s apparently enough demand that the team is trying.)
So if you’re a Windoze user, take a good look. If you’re a linux user, hit the NEXT button.
Compiere: When I found Compiere I thought I was in love. On the one hand, the program is entirely too powerful for my needs. On the other, it provides functionality that no other accounting program I had (or have) found provides, all in one place, all integrated, and all available “on line”.
But Compiere is not just an accounting program. It’s billed as an “ERP/CRM” program. ERP stands for “Enterprise Resource Planning.” CRM stands for “Customer Relationship Management.” So Compiere’s involvement with a customer and their project begins at the first phone call and lasts. . . Well, forever technically. But at least through the full life cycle of the project, tying the project cycle to the invoicing, the inventory, the payroll, the returns, the payments by the customer, and so on and so forth.
Compiere assumes a multinational business, which is why I said it’s in some ways way too big for lil’ol’ me. But man what a piece of work!
Then the bad news set in: After figuring out the interface, I thought using the program would be easy. But it’s apparently not! I can’t even enter a store receipt to adjust inventory without getting an “invalid” error. Manual changes to inventory don’t “take”. Payments by customers for services rendered on work orders or invoices don’t seem to be being processed. Obviously I’m missing something!
As it turns out, Compiere isn’t really GPL software, as it’s billed (though you can get the source code for at least one product). It’s really cripple ware: Here’s a massive and complex piece of software that you’re able to download and install on your computer (and it will run on almost everything, Windoze, Linux, MacOS), but. . . If you want a user manual or any other kind of operational information beyond installation instructions (which they gladly help you with) well, that’ll cost you $50.
And don’t even bother to ask for help in evaluating the software. I wanted to be able to put the generic “company” they provide you with in their “GPL” version through its paces to see how well it would fit my day to day routine. What did I get back?
- We’ll gladly sell you the manual for $50.
- We offer lots of online flash seminars — for $50-$300 each!
- We offer lots of instructional seminars. We’d be happy to enroll you — for $800-$1,200 each, or more.
Well, I might just spend money on some of that if I could determine whether or not the program was going to fit my needs. I mean, even Quicken provides enough information for users to determine how well, and which product, will fit the user’s needs without trying to pick their pocket!
But don’t ask that of Compiere. They’ll rip the fabric tearing that wallet out of your pocket!
ADempiere: At one time there was only Compiere. But for many of the reasons I stated above, some of the Compiere developers got disgusted and set out on their own. They named their Compiere fork Adempiere.
Adempiere is now developed in much the same way GnuCash is: By a team of volunteers around the world. Given some of the high-powered names I saw associated with it, it is apparently quite stable and functional. Unfortunately, like a lot of GNU software, everything, including the documentation, is incomplete, fragmented, and disorganized. They have some great ideas, but how they’re going to play out remains a huge question mark — one I don’t advise betting one’s business on.
Also, and somewhat ironically, ADempiere is dependent upon on postgreSQL and has no support for the GPL version of the SQL databases provided by either mySQL or OpenDB.
Tiny ERP: This was another program that looked deliciously good on the surface. For one thing, it’s a linux native program built on the GTK+ tool kit. For another, rather than requiring one of the massive (and clunky) SQL databases of one sort or another, they chose the much smaller and more versatile XML format for their database. Finally, like Compiere and ADempiere, Tiny ERP allows remote access via the web.
If you live in the EU, let me advise you to give Tiny ERP a serious look. If you’re anywhere else and have no time for development work, forget it! They don’t even support OFX imports from your bank.
OpenBravo: This application has a top notch website and sounds really, really good, until you dig deeper. Like Compiere they offer the software free. Also like Compiere, ADempiere and Tiny ERP, it’s web based. But. . .
If you want to know how to use it, pay! Also, although they bill themselves as an ERP/CRM application, they’re really, really light on the CRM part of the equation. You could probably do as well or better with a paper dayrunner type application. Also, they don’t seem to offer any kind of online banking support.
SQLLedger: This application looked like a promising replacement for GnuCash. Not as full featured and sweeping as Compiere would have been — it’s not an ERP/CRM — but it does allow remote access through a secure http connection. Unfortunately, like ADempiere, it’s postgreSQL based and offers no support for the GPL version offered by OpenDB. Also, in a simple query about OpenDB support the developer wasted no time in informing me s/he “doesn’t code for free.” So despite the adverts on the website, don’t expect support unless you plan on tapping the bank account.
LedgerSMB: Like ADempiere, LedgerSMB is an open source, community based version of SQLLedger. And like ADempiere, while it offers features that its parent doesn’t, it’s so disorganized that the documentation is nothing more than a useless skeletal outline. Don’t look for a clue as to where the project is headed.
So, the search goes on. Right now I’m so frustrated I’m ready to give in to corporate greed and plunk down the money for a copy of QuickBooks. (QuickBooks is a CrossOver Office supported application. It would certainly make my accountant happy if I did so.)
Of course, I could also plunk down that $50 for a user manual to Compiere so that I could fully evaluate the software — but therein lies the rub. If I want to know about QuickBooks, all I have to do is talk to my accountant, or take a look at their long and very public track record. I could talk to dozens of local business owners about how QuickBooks has worked for them. I can’t do the same for Compiere or any of the other above mentioned products. And while $50 isn’t much, I’ve got a half a dozen sales people every week that want to take “only” $20 or $30 or $50 out of my company for some “good reason” or another that’s going to “improve” my business one way or another. So as much as I like the idea of Compiere, dropping $50 on a bet is something I’m very reluctant to do.
So what’s next in the Agony of Accounting?
Fuck if I know!
I’ll let you know as soon as I figure it out.
Technorati Tags: accounting, ERP, CRM, Compiere, ADempiere, Tiny ERP, TurboCash, OpenBravo, SQLledger, LedgerSMB, linux, open source, GPL, GNU
Powered by ScribeFire.
[FYI]
Curious what you might think of
http://www.postbooks.org/
And also here’s a link to more: http://en.wikipedia.org/wiki/Comparison_of_accounting_software
Yes I’ve been to the wiki site.
And I looked at PostBooks shortly after getting thoroughly fed up with OpenBravo and it’s “user’s manual” — bright and shiny version 2.3 even. Which is VERY helpful (not!):
Given this blindly clear and detailed explanation of the difference prices, it only took me two days to figure out that the “standard price” is what the rest of the world calls the “purchase price” and the “list price” is what the rest of the world calls the “selling” or “resale” price, and that the “price limit”, or “maximum discounted price” to the rest of us, is really not overly useful.
And the rest of the user manual is equally perspicuous.
But anyway, to answer your question about PostBooks: It looks nice. But I was/am too tired of ERP programs to worry about it. And, PostBooks like all the other ERP programs (save tinyERP) has no shared language for importing or exporting financial transaction.
Yes, it does (if memory serves) do CSV (as does Compiere, OpenBravo, Adempiere, etc., but CSV is lame compared to OFX.
Also, all of these ERP programs seem really, really slow when it comes to data entry. For all its faults, I can do 3 days worth of book keeping in about fifteen minutes in GnuCash. After figuring out the order of operations in OpenBravo I timed a few simple transactions — like a tank of gas for the service truck: About 5 minutes each if I used my check card! And then I’m not done, because all transactions are “suspensed” until they clear the bank, which since there’s no OFX processing, also has to be done by hand. (Oh, and did you know that a tank of gas is INVENTORY in these programs? Actually, every transaction that goes through a bank account HAS to be run through the inventory system.)
PostBooks didn’t look much sleeker when it came to data entry of basic transactions.
In summary: ERP looks fine if you’re running a store or some other business that’s entirely inventory based. But anything else, stay clear.
I came across your blog while googling for ADempiere.
Just what you’ve mentioned about ADempiere “everything, including the documentation, is incomplete, fragmented, and disorganized” was true in my experience in installing it.
It took me several days figuring out how to make it run compared to other accounting/ERP system that I’ve explored on GNU/Linux.
The project is quite young and still trying to break mold from its Compiere root.
Here are two other ERP/accounting program that I’ve found while trying to explore and learn some basic accounting/bookeeping.
WebERP and FrontAccounting both are LAMP based.
Like LedgerSMB a fork of SQL-Ledger, FrontAccounting trace its root from the discontinued OpenAccounting which was a fork of WebERP. So, FrontAccounting is an indirect fork of a WebERP.
With some basic PHP you can modify their interface to only show basic accounting functionalities.
I am also a Linux user. I even use Linux professionally, I am also certified in Linux.
But, I have to ask, is it really worth all that trouble to avoid QB?
IMO: here is a brief listing of what foss financial software seems to be missing. Note: all of the following is based my limited understanding, and my opinions. Please correct me if I am wrong about any of this:
* Cost advantage: QuickBooks simple start is free:
http://quickbooks.intuit.com/product/accounting-software/free-accounting-software.jhtml
Or I can buy the full version of QuickBooks in only $128:
http://www.qbpro2008.com/quickbooks-2008-coupons-for-amazon/
Seems to me that any cost advantage of using a foss alternative is negligible.
* Ease of use: Somewhat debatable. But some people site this as a primary reason for Intuit’s amazing success with QuickBooks – supposedly 87% of small businesses use QuickBooks. Although, I have to wonder how the number of foss users can be accurately counted?
* Integration with online banking: my understanding is that only intuit or msft products can easily integrate with online banking. Not absolutely sure about that.
* Payroll: very regional, and changes often == not well suited for foss.
* Taxes: somewhat regional, and changes often == not well suited for foss.
* Wide acceptance: I think most businesses are much more comfortable using products that are accepted standards.
* Wealth of available add-ons: Intuit has a very active community of 3rd party developers. You can buy practically any kind of an add-on you can imagine. These add-ons cost money, but at least they are available.
* Major company: I think a lot of businesses are not comfortable with a product unless there is a major company behind that product. I have to admit, even I am not comfortable with software products that are essentially one man operations.
* Support: I can always hire somebody who knows quickbooks, or find a “ProAdvisor” consultant, or I can get support from the company, and there are hundreds – if not thousands – of developers who specialize in developing for quickbooks. I can not see where that is true for any project.
* Training availability and costs. I can hire people who already know quickbooks. If I hire somebody to work on some foss alternative, then there will be a significant training expense. Of course, there is also the issue of training availability.
* Documentation: If I had to pick one thing that kills the usefulness of more foss projects than anything else, this would win in a slam-dunk. Of course, this varies among projects, some foss projects have great documentation. But, I can always find plenty of books, or other documentation for popular proprietary financial apps.
* Many accountants, maybe as many as 200,000, use QB and recommend it to their clients. Some accountants will charge much more for files that are not in QB format.
* QB has much better 3rd party integration. For example, ecommerce packages like oscommerce, and magento, work with quickbooks, not foss alternatives. Msft accounting works with ebay. I can not find that sort of integration with foss software.
I’m not sure what your beef is against PostgreSQL. I’m a pretty big GPL booster myself, but if a project wants to use a license that is more free than the GPL, why are you looking to get them to use OpenDB or MySQL?
Don’t even get me started on MySQL having a GPL version. They also have a commercial licensed version which they want you to use if your application touches non-GPL code. Check out their FAQ sometime.
PostgreSQL is free to use, distribute, modify, etc. Your code doesn’t even have to be GPL. It has NO RESTRICTIONS that I or anybody else can find. Why is the lack of GPL a hindrance, when it is replaced by a license that has less restrictions?
Perhaps you don’t understand it’s license?
By the way: PostgreSQL is also a much better database than MySQL. (Coming from a database administrators point of view). It is the only free database that can even come close to enterprise databases like Oracle.
R
I noticed you mentioned that OpenBravo doesn’t provide any sort of online banking. Do you know of any that do? I can’t seem to find anything but Quickbooks that offer this for business accounting.
Hello! I hope that my reply is not too late. Have you tried WebERP? I think that it is a good accounting software, and its free and open source, runs on different platforms, too. I have tried GnuCash, too, but my experience with WebERP is so much better than with GnuCash.
I hope this little information about WebERP could help you in your quest of finding the software that suits you best!