New Microsoft Patent may hurt CityDesk

On Tuesday, the USPTO awarded a patent to Microsoft for its Method and system for reporting a program failure.

http://slashdot.org/article.pl?sid=03/10/02/1218208

Happy to be working
Thursday, October 2, 2003

There must be lots of prior art for this. I'm sure that lots of us have written something similiar (I know I have).

John Topley (www.johntopley.com)
Thursday, October 2, 2003

FogBugz too: http://www.fogcreek.com/fogbugz/help/UsingFogBUGZtoGetCrashRep.html

Although i'm pretty sure Netscape had developed something pretty similar long before Microsoft had.

John Rosenberg
Thursday, October 2, 2003

The patent doesn't cover automatic generation of error reports, it builds on top of that.  The program, upon encountering an error will connect to a server at Microsoft and scan the bug database to see if that error had a bug associated with it (that was already filed).  If so it checks to see if a fix has been posted.  If the fix has been posted it helps the user download the fix.

Now that's a nifty feature.  Is it worthy of a patent, I don't think so, but its not anything I've seen before.

Lou
Thursday, October 2, 2003

If it's what you are describing, the patent seems ridiculous.
Or maybe I don't know what a patent is.

GP
Thursday, October 2, 2003

Agreed with Lou; the patent is for a system that sends a bug report to a remote server, collates the data on the remote server, and recommends downloads based on the collated data.

I don't see how that would hurt CityDesk.

The Pedant, Brent P. Newhall
Thursday, October 2, 2003

I think we hit a /. moment and didn't RTFA. 

Software patents are really tricky things.  The USPTO probably shouldn't be handing them out, as they can be overly restrictive.  Software patents are a way of preventing your neat idea from becoming the next software meme. 

1-click checkout is a prominent example of a software patent.  It was developed by Amazon, and Apple bought a license to use it.  But 1-click checkout isn't a real mind-blowing idea, its just sort of neat but not a real programming leap.

And that's really where the debate starts.  Do software patents encourage development or harm it?  I would expect those in favor of OSS to be against software patents, both for the freedom aspect and because OSS is relatively notorious for borrowing features from other applications.  If those features were all patented we'd run into trouble.

That said, the more I think about this particular piece of code the more I think its pretty worthy of a patent, or at least more worthy than any other piece of software that has been awarded a patent.

I don't think there have been any real challenges in the courts over the validity of software patents, but I'm sure greplaw has some good documents on it.

Lou
Thursday, October 2, 2003

The courts keep wrongly-awarded patents in check -- if the patent holder takes somebody to court over it and the judge finds it invalid, anybody they've collected from in the past can now sue to recover fees they already paid. This is why if one of these patent extortion companies tries to collect $1000 "fee" for your use of their "system and method to simulate pressing a button on a computer display", you just ignore them. They're not going to risk having their patent thrown out and lose everything they've made so far for $1000.

(Or so my patent lawyer friend tells me...)

Rick
Thursday, October 2, 2003

Patents are ideal for areas like drug research, where a companies can spends billions of dollars in research to find  a solution which can be administered for pennies  (ie combining pepsi and apple juice cures cancer).  No drug company will spend those research dollars if their findings  can be used and profited by other companies.  Result, no research gets done, and the consumer loses.

This has no equivalent in software.  If Amazon comes up with a 1 click payment solution, most of their work is in the implementation.  Another company copying the process still has to spend the time to implement the idea.  In other words, the cost of research + implementation is not much more than the cost of copying + implementation.  Therefore, govt does not need to provide additional motivation by granting patents.  The consumer loses when only one company has a monopoly on the one click payment process.

Therefore, I totally disagree with the statement "the more I think about this particular piece of code the more I think its pretty worthy of a patent". 

notAlGore
Thursday, October 2, 2003

Ever hear of defensive patents?  Given the recent IE plug-in fiasco, I imagine that Microsoft might want to try to get a patent on anything they can think of. 

SomeBody
Thursday, October 2, 2003

I believe that there is a big, blue company that has some prior art on this (it also has a legal department that does a much better job of checking for prior art than the drones at the USPTO).

sell out
Thursday, October 2, 2003

Yeah, I know IBM had this on the AS/400. The system would download software fixes, or if it detected a hardware problem, would order the replacement parts and have them sent to you, then dispatch a CE to do the repair. It was pretty neat.

Bill
Thursday, October 2, 2003

I'm sure FogCreek are shaking in their boots.

pb
Thursday, October 2, 2003

I don't know what an AS400 did but this a patent on an automated debug session to allow gathering of additional information and possibly then automatic patching for the end user.
Personally I feel it's marginal on patentability, however if MS kept as a trade secret then someone else had the same idea and decided to patent it MS are screwed. They have to document the idea, so why not go for another patent for the portfolio?

Peter Ibbotson
Friday, October 3, 2003

  Recent Topics

  Fog Creek Home