Looks like a nice, reliable, extensible, easily-customized business accounting package, doesn't it?
Other people thought so, too, including William Grimm of Dublin, Ohio, who emailed NewsForge to say:
I have been an Appgen reseller since the Spring of this year. I paid my reseller fees and diligently learned the product and its programming language, and went about attempting to sell Appgen. I had 2 sales which would have finalized this month, had Appgen not closed its doors.
Here's the rest of Grimm's email to NewsForge:
Appgen closed its doors on 3 October 2003. The VARs were not notified of the closing. The only mention of its closing we received was a brief e-mail from Steve Elliott, who used to direct Appgen Tech Support. The e-mail was sent to us via a mailing list, appgen-dev@mirror.org. Last week, many of us were clamoring for what we had been promised in the eventuality that Appgen went out of business - the product's source code. Most if not all of us were told that the source was placed in escrow with a law firm in Dallas, TX, and would be handed over to the VARs if Appgen went out of business. Evidently some VARs have this promise in writing. The source code was never placed in escrow.Grimm's out-of pocket loss -- essentially the cost of his developer kit -- pales besides other Appgen resellers' investments. One reportedly paid $10,000 to have Appgen ported to "SCO and UnixWare" in September, only weeks before the company went away.Last week, most VARs were clamoring for the source, and wished to open source this product. A small handful of the older VARs kept telling us to "wait for Steve and Errol's proposal." (I have mentioned Steve before, Errol was Appgen's CTO). When Steve and Errol called the VARs, they had little planned except for a vague desire to sell keys to us, and "fix problems in the code," something Appgen has not done well since I became a VAR.
To date, we have received no official communications from the officers of Appgen regarding the status of the company- whether it is in bankruptcy or what. None regarding the legal status of the product or of the legality if we attempt to resell the product. Without license keys, end users are in big trouble if their Appgen installation requires re-installation.
I have received several nasty e-mails from a small group of VARs who are telling me and the other VARs who want some answers to back off and "trust" Appgen. I do not trust Appgen, and believe that as a group, the VARs were defrauded. Appgen closed its doors on 3 October, but kept its sales efforts in high gear until the end of September 2003. Since I became a VAR, I have never received a single sales lead from Appgen; other VARs have told me that Appgen neglected its VAR channel and was selling direct to interested companies.
Other resellers have staked their entire businesses on Appgen. Some have dozens, possibly even hundreds, of corporate Appgen users who depend on them for support.
Messages posted to the appgen-dev email list over the last few weeks show plenty of fear, uncertainty, and doubt, not to mention a certain amount of factionalism.
In general, it looks like "senior" Appgen resellers are hoping for an amicable resolution, possibly including former Appgen employees Steve Elliott and Errol Allahverdi, while some of the newer, less-established resellers want to take legal action.
There are questions about who now owns the Appgen code and the right to distribute activation keys. Appgen apparently didn't declare bankruptcy, but simply ceased to exist as a corporation. Some resellers claim the code is held in escrow by a law firm, but others say it is not; that this promise was made but never kept. Some want to open source the code if they can lay hands on copies of it, while others want to keep it proprietary and perhaps form some sort of consortium or new company to handle its distribution and support.
This isn't the first time Appgen has abandoned a product
Appgen Personal Software, a "sister company" to Appgen Business Software, once published Moneydance, a well-regarded cross-platform personal money management program. Appgen Personal Software then laid off all employees except Sean Reilly, the original Moneydance developer, and Appgen Business Software took over its marketing and support -- until early 2002, when Appgen totally abandoned Moneydance. Bye, bye, Moneydance.
(See this April 17, 2002 post to the gnome-office-list for a little more of the story.)
It wasn't until February 2003 that Sean Reilly managed to reassume full ownership of Moneydance. Soon afterwards, this NewsForge story appeared. Favorable mentions on NewsForge and other tech news outlets helped Moneydance sales bounce back and grow into what Reilly calls "a nice part-time income."
Perhaps some of Appgen's newer VARs should have checked out this bit of history before committing their money and time to the company.
His last $2,000 down the drain
When he decided to sign up as an Appgen VAR in 2001, Darren Remington didn't mind spending $2,250 for a developer kit and VAR license he hoped would give him the base he needed to write a financial application specifically for law offices that could challenge the leading legal industry financial management package, Juris.
But almost immediately after Remington sent Appgen his money he was laid off from his day job, and lack of that $2,250 suddenly hurt terribly. "It was like I spent my last $2,000 on this," he says. "My wife is still angry about it."
Remington says, personal situation aside, Appgen gave him heartburn from the start. "I got the development kit, but no development manuals," he says. "I had to call and wait three days for my license keys."
He says he never did get development manuals, just excuses about why they couldn't be sent to him, and then -- just to make things worse -- he says the incomplete Appgen VAR page that was his only source of information not only never got completed, but totally disappeared from the company's site in the spring of 2002.
But Remington's biggest complaint is that he claims he was promised access to the source code for Appgen's core products (under NDA) but only got source code for a couple of modules, which wasn't much help for his anticipated Appgen-based custom software project.
Remington has never made an actual Appgen sale. The big customer objection he ran into, he says, was that for most businesses, "QuickBooks was cheaper and would do the same thing."
What if Appgen had been open source?
Moneydance developer Sean Reilly says he'd be happy to help Appgen VARs with any unresolved code issues "a few hours a week, anyway," if he was sure of the legal status of Appgen's code -- a barrier that wouldn't exist if it was open source.
Many of the VARs have enough programming expertise to carry the support flag, and if Appgen's source had been open all along, or if there had at least been a solid escrow arrangement of some sort that would have given the VARs code ownership when Appgen went out of business, Appgen VARs and users wouldn't be biting their nails today.
Reilly says he wonders how the company could simply cease doing business; that while he hasn't had any recent contact with them ("They stopped returning my calls some time ago"), he believes they have creditors they have not paid.
Those creditors could argue in court that if the company's source code is one of its remaining assets, they should own it or at least force its sale at auction to help satisfy any debts Appgen left behind.
If the code had been open source from the start, this would not be a possibility. Perhaps Appgen's failure should serve as a cautionary tale to any person or company that considers buying proprietary software. How many software buyers think about this sort of problem? How many make sure they have contracts that give them source code access if their software vendor goes out of business? How many Windows users realize that Steve Ballmer, under oath, threatened to stop selling Windows if the Microsoft antitrust trial had produced a final verdict the company found unacceptable?
I often suspect that Ballmer's chilling threat to withdraw Windows from the market and, presumably, stop supporting it, was the tipping point that made so many large financial institutions decide to start moving their critical infrastructure to Linux.
The open source business model?
Let's start with a closed source business model:
1. Invest time and money to become a software VARNow contrast this with an open source business model:
2. Software publisher goes broke.
3. Big loss, no profit!
1. Modify an open source software package to fit a niche marketA growing number of independent software developers and resellers seem to be choosing the open source direction.
2. Sell installations, manuals, customization service, and support to that market
3. Profit!
After looking at what is happening to Appgen's resellers (and users), do you blame them?
Note: Comments are owned by the poster. We are not responsible for their content.
Did I mention the new site sucks?
This is more relevant today than most people realize and than many others are willing to admit. If Microsoft goes to bankruptcy, the format specifications, XML schema and other secrets necessary to migrate from MS-Office 2003 will be locked up as "assets" so fast you'll hear a sonic boom as the lawyers arrive. What then? "It's in XML", squawk the marketeers. Sure but if you look more closely, the schema is not available and the file itself is still binary and encrypted.
No big deal for businesses, they can file an insurance claim for the damage caused as their entire vertically integrated network -- from data to DRM-crippleware to Palladium to LaGrande CPUs -- gets thrown on the dung heap like so many Wangs from yester year.
It'll hurt governments foolish enough to lock away court records, tax records, medical records, school records or charters in proprietary formats. While the data is locked away, citizens and businesses alike dependent on the information are S.O.L. Want a transcript to go to college? Tough. Your tax returns to get a small business loan? Tough. You Dad's military records for health care? Tough.
So, reverse engineering the file format would be the immediate work around. But not with the DMCA/EUCD. For U.S. users, add in the EEA, too. See OASIS's work on the OpenOffice.org XML schema for a way out.
Redhat has stated publicly that they will never hire any significant part of the Linux developers.
Folks, he's trying to argue quantity, not quality nor effect. He's trying to determine whether an effort is "major" based on counting heads. And he also doesn't have much info, so he guesses the number is between "a handful" and "50" -- a rather wide range, and I've worked in several software development companies which earned millions of dollars a year with less than 50 developers.
The reality is, there is no reality. IBM isn't selling software, so there is no open source software business model involved.That's funny. Here I thought that my paycheck, which is from IBM, was for supporting software which was developed and sold by IBM. And I thought that one of the components of the software package I supported was the IBM HTTP Server, a branded version of Apache. I guess I'll need to check with my boss to find out what I'm really being paid for, since IBM isn't selling software.
Have the appgen box, opening for other projects
Posted by: Fonze on October 24, 2003 05:22 PMThis gives an opening for other foss projects. The article talks about others wanting to do development for appgen products. Time to look at foss projects, especially now that this serves as a prime example to others.
<A HREF="http://www.sql-ledger.org/" TITLE="sql-ledger.org">SQL Ledger</a sql-ledger.org> was a foss accounting project I was aware of last year. Are they still around? Is it still Free Software? Want to do development?
Maybe it deserves a look...
The sweet part about Appgen failing, besides the fact that they will be the poster boys for why not to rely on closed source, is that their software ran on (puke)java. That's another reason to rejoice in their failure.
Did I mention the new site sucks?
btw, if you limit stories to only one word per line, you can squeeze in even larger ad$...
#