Debugging in Plain English? 274
sameerdesai writes "CNN is carrying a story about Researchers from Carnegie Melon: Myers and a graduate student, Andrew Ko, have developed a debugging program that lets users ask questions about computer errors in plain English: Why didn't a program behave as expected? I guess with recent exploits and bugs that were found this will soon be a hot research topic or tool in the market." We recently did a story about revolutionary debugging techniques; the researchers' website has some papers and other information.
Hal do you read me? (Score:5, Insightful)
Not to appear smug but... (Score:5, Insightful)
Andrew Ko's invention is just another tool. It won't do the debugging for you. Just like modern cars have diagnostic computers, but somehow it appears you still have to fork off $30/hr for the workmanship to get it fixed...
pointless (Score:4, Insightful)
Plain english error reporting? (Score:3, Insightful)
me: why did the program leak 1GB of memory then segfault?
computer: because you don't know how to program, you idiot!
Fundamental logic problem (Score:3, Insightful)
In essence, for there to be a "English debugger" (one that speaks more english than current debuggers, that is), it would essentially need to know how to program itself on top of being able to follow the flow of code and find where it breaks, so as to be able to tell you precisely what the problem is.
Sounds a bit fictional to me.
How do they answer these questions (Score:5, Insightful)
I don't expect this early research tool to catch all of these, but I'd like to hear the researchers' response on how their system might (after years of development) answer questions about some of these bugs:
- Why did the Mars Pathfinder software deadlock (priority inversion)
- Why did the Mars Polar Lander crash (improper state management)
- Why did the Ariane 5 blow up (arithmetic overflow in a register)
- Why did the Patriot missiles miss in the 1991 Gulf War (accumulated time error)
- Why did a radiation therapy machine zap patients with the wrong doses (inconsistent state between GUI display and internal software state)
I'm sure there are some others on comp.risks and elsewhere.
Another point: this approach is still "just" a testing tool. In other words, it can only find errors on paths that have actually been taken in tests, which means the testing program must cover enough cases to generate the runtime errors in the first place. In all of the above cases, it was the testing program that permitted the bugs to be fielded.
Error messages: understandable vs. informative (Score:2, Insightful)
In the area of error messages, we need a better balance between "understandable" and "informative". Today, it's usually one extreme or the other. Either error messages are too "friendly" to the point that they're rather meaningless (as in the notorious "Web site not found"), or on the other extreme, they're so "complicated" that the novice user has no idea what happened (as in "an xxx exception occurred at xxx, here is the stack trace, register states, and many many more confusing numbers").
It would be nice if we (as a society that happens to use computers) could adopt a guide that gives the technically-savvy enough info to go on, but still helps the novices understand what went wrong in terms they can understand too. A balanced approach.
Such an error reporting scheme, if successful, could almost certainly NOT be designed by marketing weenies nor by geeks alone, but through mutual collaboration and willingness to compromise on both sides. Hmm, no wonder it's so challenging.
Re:"Why didn't this program work as expected?" (Score:4, Insightful)
Computer: "How should I know, I just do what I'm told."
Re:plain English development.... (Score:3, Insightful)
Unfortunately it's because COBOL is excrutiatingly bad at expressing the sorts of things programmers need to express. It's not so hot for expressing business information either.
I am reminded of the attempts to cross horses with zebras -- it worked, but the hybrid inherited the worst traits of each animal.
Deja vue (Score:5, Insightful)
This reminds me of back in approx 1985 or so, someone "invented" a human language programming environment called "The Last One" or something like that. This would supposedly make it simple to write programs without having to learn C etc. Some programmers quaked in their boots. However the real issue with programming is learning the contructs, not the language (ie. if you understand what a linked list is, then writing one in C vs Pascal is pretty simple). Anyone that thinks that programming in English is easier is seriously misunderstanding programming. The ultimate test is to look at the languages that have survived: The more "human readable" languages like COBOL have not survived, but the more cryptic ones like C have. The big "killer app" for making programming simple for the non-programmer was the spreadsheet and that's hardly a natural language.
Now debugging is pretty much the same deal. Verbose English debugging interfaces might make it simple to learn to do very basic debugging, but once you get into things a bit deeper (and get more experienced), English becomes a huge liability and you'll be wishing for more concise and expressive languages.
What about plain French? Or Russion? (Score:2, Insightful)
Then why all those "natural language" thins are in English?
Seems to me that is easier to learn a programming language then to learn English for a non-native English speaker: much less ambivalence.
Re:Deja vue (Score:4, Insightful)
No. It's actually rather trivial to machine translate individual C++ statements into valid assembly code. The resules of doing so are inconvenient, because anyone with a little practice will find that 90% of the English text is boilerplate that can be more concisely presented as *&+=.;{[->, etc.
Verbose English debugging interfaces might make it simple to learn to do very basic debugging, but once you get into things a bit deeper (and get more experienced)
But what is a waste of time for experienced coder might be just what an end-user needs to help him better decide how to go about solving an unanticipated problem. It'd be nice if an untrained person could proceed through the following dialog (BEFORE having to contact a programmer).
X11 Window connection closed on SEGABRT
"Why did it seg?"
Deferencing invalid pointer 0x0
"Why was it invalid?"
Pointer was assigned as return value of OpenForWrite function call
"Why did the function return 0x0?"
Drive D: does not exist
Capabilities like that could help fullfil the Open Source promise of "Every user is a (competent) QA"
Actually, I've seen more than a few professional "software engineers" who could've benefited from something like that. A C++ guy transitioning to ADA, for example...
Cardboard Programmer (Score:2, Insightful)
I was going to write up a cardboardprogrammer.com site with a flowchart with 20 questions to ask about a bug to clarify your thoughts. [I suppose both linux and windows can have a REBOOT as the first directive and what was the difference between now and when it was last working as the second, are you using the latest version of the code as the third has anyone else been working on these files the fourth.. well you get the idea, [it is like shooting fish in a barrel, but I have never seen it the full flowchart] that was in the dotcom era, when even a tiny good small idea like this was in someones mind a possible next big thing].
Re:Deja vue (Score:3, Insightful)
The problem isn't that the concepts in programming are hard to express in English. In fact, it's quite easy to express programming concepts in English. Why do you think pseudocode is so much easier to write and understand? The problem is that English does not have a context-free grammar. Without a context-free grammar, it infinite orders of magnitude more difficult to write a compiler for a given language.
If someone were able to write a compiler for a natural language like English, it would be the most amazing advance in the field of computer science in years, possibly ever. Natural language recognition is some heavy stuff. But just because natural languages are more difficult for a computer to process, don't let this make you think that using computer languages makes it easier to "describe" your programs.
Getting good field reports. (Score:4, Insightful)
In the products I'm involved with I often get stupid reports from the field of the form "framing error causes unit to reset". When I get one of these, first thing I do is get back to the user and figure out exactly what they saw and what heppened withouth them trying to figure out the symptoms. In the "framing error" problem, what was really happening was that a power glitch was being caused when the RS232 cable was attched (because of bad grounding). This caused a reset. However, the user was a "super user" who knew bad things happen to serial data when you plug/unplug cables. One of the buzzwords he knew about was a framing error. So he "half solved" the problem by saying that a framing error caused the problem.
There is a big difference between observing and fixing problems. QA is about observing, not fixing, problems. It is better to provide a good way for users to make accurate error reports (eg. backtrace/log/whatever) than try have them try explain what went wrong.
Re:Not so... it is the scientific method. (Score:3, Insightful)
To go from 1 to 2 is the hard part. But, you can use tools to instrument the problem and determine where the problem may be.