Lemon Laws That Apply To Bug Filled Software?
from the on-the-way... dept
Here’s an issue that shows up every few years, but really hasn’t received that much attention lately. However, now that people are predicting that virus writers will increasingly focus on exploiting applications rather than operating systems, more questions will be raised about whether or not “lemon laws” should apply to software. Of course, the article is a bit ridiculous in a few ways. It’s not really true that viruses attack operating systems. In fact, at this point, it seems like most viruses are targeted at an application: Microsoft Outlook, rather than an operating system. Also, up until this point, software developers have been able to defend themselves against lemon law type claims via the end-user license agreement, which basically says “you get what you get, live with it — and we’ll try to patch stuff if it gets too bad.” While there may be a reasonable claim to be made about intentionally placing security holes in software, or making claims about security that the company knows is false, expecting software developers to produce perfect software all the time goes too far, and would pretty much decimate the software industry by creating a huge liability for anyone to ever release any software publicly.
Comments on “Lemon Laws That Apply To Bug Filled Software?”
Software Warranties
I’m a professional programmer, and would certainly oppose any law requiring perfect software. However, I do think the existing “warranties” that only warrant the media against defects are ridiculous from a consumer perspective. Warranties of merchantability and fitness for a particular purpose should always be applicable.
Also, remember that lemon laws for cars don’t require “perfect” cars. They allow people who have a car with several problems to get relief. If a program repeatedly crashed my computer, I don’t think expecting a refund is out of line. If a program trashed my data, maybe more relief would be useful (paying for a data recovery team, perhaps).
I would also encourage a listing of testing results to help people understand a program’s limitations. For example, a database company might say they tested the database on files with 1,000,000 records of 10,000 bytes each. Users using files smaller than that would get stronger protection than users using files that exceeded the developer’s published testing.
In fact, those test results could actually be used as a marketing tool. We might even see QA wars between developers trying to publicize the “best” tools. After all, would you buy the photo tool that claimed to work with 5 megapixels or the one that claimed to handle 16 megapixel photos if they cost about the same and had the features you wanted?
Hmm... yes and no
I both agree and disagree with your point. While there is a limit to what one can expect of software, and license agreements have/do provide some defense; software is increasingly controlling what we like to think of as mechanical systems.
My 2005 Chrysler 300C supposedly includes many seperate computer systems that control almost every aspect of the car’s operation. Whether a problem relates to going, stopping, crashing [safety systems], navigation, steering, lights, wipers, or convenience systems like locks and windows, there are computer systems and software involved. So while applying lemon laws to Microsoft Internet Explorer seems problematic, lemon laws are likely to increasingly be applied to vehicles and appliances that are now heavily controlled by computer chips and software thanks to 21st century technology.
Since other news items have suggested that current programmers of “imbedded systems” are much less likely to respect the power of their software, it may be that lemon laws will increasingly relate to software flaws in these imbedded systems. If that becomes commonplace it won’t be such a stretch to apply such laws to other software.
If a recent thread on /. concerning lone developers has any merit, patent and other IP issues are quickly driving non-corporate developers out of the market anyway. If only huge corporations are releasing software, then it is likely that they can absorb any losses due to lemon laws for bad code. On the other hand, if a company’s code is so bad that they can’t stay in business due to such expenses, then aren’t we all better off if they fail anyway?
lemon laws
I’m not sure that the underlying principle of lemon laws can apply to software.
A lemon law applies to a single unit with a “serious defect or abnormal condition”, like a new BMW that turns out to have a malfunctioning transmission. On the other hand, if you bought a brand new Ford Pinto that worked just like every other Ford Pinto, you can’t make a claim under the lemon laws just because you now own the shittiest car ever manufactured.
My point is that if Outlook sucks that doesn’t mean that you got a ‘lemon’ copy of Outlook. Just like if you bought a Kevin J. Anderson novel you couldn’t return it as a lemon, because all Kevin J. Anderson novels are like that. Software is more like a novel than it is like a car. (This is my perspective as a fluent programmer, your software consumer admittedly probably has a different relationship with his software than I do.)
However, other considerations may apply. ‘Fitness for a particular use’ has been mentioned, and it’s always struck me as odd that commercial software companies can just wave their hands and make that one go away. There may also be some class action kind of stuff going on with regard to particularly flagrant examples of bad software. But I think that mostly happens in the ‘false advertising’ area rather than in the ‘malfunctioning device’ area.
I DO think there should be accountability. I just don’t think that the lemon laws are the right model to adopt.