Developers Lose With Proprietary Software
An anonymous reader writes "Appgen looked like a nice cross-platform accounting program independent software developers could use as a base for custom applications, and lots of them paid $2000 or more for the company's development kits. Then Appgen went out of business and left all those developers stranded. They can't even generate license keys, and their support has disappeared. Nobody knows who now owns Appgen's code, so it looks like all those developers and their clients are screwed. This couldn't happen if Appgen was Open Source. There's a strong lesson in this story for those who choose to listen." Newsforge and Slashdot are both part of OSDN.
sPh
We gave our source to our paying customers when we decided to drop the product and switch directions.
Everyone seemed pleased with the arrangement, even though I doubt they were pleased when they got the gazillion lines of C++ code without support.
And to think my idea of going open source was ridiculed by management 6 months before we flamed out.
Sheesh...
#1 - serious question - how many serious accounting packages are being worked on in the open-source world? It's exactly the kind of software hackers usually denigrate...
:)
#2 - I think this is as much about poor planning (contract negotiations on the part of the developers, defining and/or selling and/or making a "will" for the software on the part of the owners) as it is about IP. And I'll bet somebody knows who owns it, they just haven't settled it yet.
#3 - How many abandoned Open Source apps are there? I mean, sure, you won't have the key problem, but still. The grass may be greener, but it isn't self-mowing, self-watering, and immortal!
Obligatory Criticism from Merovign.
1. Modify an open source software package to fit a niche market
2. Sell installations, manuals, customization service, and support to that market
3. Profit!
Unfortunately I haven't seen many Open Source businesses manage to achieve point 3.
I'm sure everyone can mention a few that have done so (Redhat and Cygwin spring to mind) but there are vastly more that have either fallen by the wayside or are resorting to begging for money (Mandrake?).
Open Source is great, but it too isn't perfect ...
Avantslash - View Slashdot cleanly on your mobile phone.
Er.... the Java programming language is a specification, and an open one at that (IIRC). If Sun went under, IBM still has a kick-ass VM and SDK. And a great IDE in Eclipse, too. If Sun went under, Java would continue unabated; it's a programming language, not a library or modifiable application.
From the article it seems to me that Appgen (which I'm not familiar with) is either an IDE/Library/ProprietaryLanguage, or a full-blown application that developers can modify for their own use. It's a far cry from Java.
"Times have not become more violent. They have just become more televised."
-Marilyn Manson
#1 - serious question - how many serious accounting packages are being worked on in the open-source world? It's exactly the kind of software hackers usually denigrate...
In my analysis, it isn't that open-source developers don't want to work on this sort of thing, it's that there is a certain amount of infrastructure that needs to be in place before projects like this can proceed. Several enterprise-class accounting projects have been started, but few finish; it's because the tools aren't in place yet.
The FOSS community doesn't avoid doing corporate-type projects, as a lot of people claim. FOSS software is written because it is positioned properly to fullfill a need. Until very recently, FOSS was not accepted in the enterprise. Now, as more and more corporations are depending on various FOSS software, you will see many projects targetting medium-to-large corporations.
For instance, look at the relatively-new GNU Enterprise project. This is a major undertaking which has begun by creating the tools required to build an enterprise management infrastructure.
As FOSS software penetrates various markets, you will see many FOSS projects building finance/hr/materials-management/analysis tools. I predict that 2004 will be the year of the enterprise for FOSS (Linux,*BSD,GNU). You'll see prepackaged medical management software, ERP software, etc. By the end of 2005 I believe you will see a complete enterprise management system, from supply chain to finance to HR to payroll.
But maybe I'm just a pollyanna.
Microsoft is to software what Budweiser is to beer.
Meet gcj, smart guy.
There are no trails. There are no trees out here.
ugh.. as nice as it would be to say proprietary software is bad, these are the results. That just is not the case.
/. analogy, when I buy a ford car, I have no guarantee that I'll be supported after ford goes out of business.
Proprietary software failed in this case because the people using it (stupidly) paid a lot of money for software that had no contingency plan or guarantee.
To use a popular
When you pay a lot of money for something with no service guarantee of any kind, stuff like this happens. Sure using OSS may have helped with this problem, but OSS has a whole slew of other problems.
If its merely a license key issue, I'm sure these "developers" could get around that. Judging by the number of keygen programs for other software packages that come out the same day a program is released, this is a non-issue.
Yes, the lesson is: don't buy a propriatary app without a 3rd party source-code escrow agreement. That was figured out around 1965.
So, you have a contract that specifies software escrow. And when the company goes bankrupt and you find the source is not in escrow (or not all of the source is in escrow, or there is third-party IP in the escrowed source, or ...):
who are you going to sue?
An escrow agreement is likely to be enforceable right up until the moment you need it.
The difference with open source is that you have the source in hand now and so if the company disappears you don't have to sue a non-existant entity to get the code.
But this has already been covered once WRT Java, which is precisely the same thing. Java and .NET are both open specifications. There are a couple replacements for each (Java: IBM JDK, gcj; .NET: Mono, DotGnu) so if one of them should drop off the face of the earth there is a free and open source replacement.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
I am sure there are those in the mega-corp world would see this as an example why you need to use Microsoft,Oracle, etc. because going with a smaller startup company could leave you stranded. Just a thought.