Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
07/07/05 18:52
Read: times


 		 #96792 - another article of interest
Responding to: ???'s previous message


http://ganssle.com/articles/crashburn.htm

I have to do my code walk throughs with my boss, who knows nothing about software, but we have a software design specification and a software verification test plan. We go through the code line by line and I have to explain to him what the software is doing. It is at this time that software bugs are caught. Theoretically, testing should catch all of the bugs, but in practice, the testing either would take too long or the resources just are not available to do a thorough test of the code. By thorough, I mean that you are testing every possible input and recording the result and doing this for every function that exists in the software. Even then, you couldn't be sure that some external hardware condition wouldn't trigger some unknown sequence of events causing your software to crash.

Jacob


List of 105 messages in thread
TopicAuthorDate
disciplined clear process for debugging            01/01/70 00:00      
   Use a scope!            01/01/70 00:00      
   Design for debug            01/01/70 00:00      
      comments            01/01/70 00:00      
         Comments on comments            01/01/70 00:00      
            In the eyes of the beholder            01/01/70 00:00      
               budget            01/01/70 00:00      
                  budget            01/01/70 00:00      
                     that is when it will be needed            01/01/70 00:00      
                        unless...            01/01/70 00:00      
                           you do not get the point            01/01/70 00:00      
                              shoud I?            01/01/70 00:00      
                              Yes.            01/01/70 00:00      
                                 inconsequent            01/01/70 00:00      
                                    no, no, no...            01/01/70 00:00      
                                       then why            01/01/70 00:00      
                                          ...which means...            01/01/70 00:00      
                                             which you must verify which is more work            01/01/70 00:00      
                                                Coding style...            01/01/70 00:00      
                                                   that's impossible            01/01/70 00:00      
               9 out of 10 cats prefer it            01/01/70 00:00      
                  corrections            01/01/70 00:00      
                     Experience, not correction.            01/01/70 00:00      
                        I'm curious            01/01/70 00:00      
                           no ICE            01/01/70 00:00      
         I like Serial Debugging            01/01/70 00:00      
         re:ice            01/01/70 00:00      
            Don't put the bugs in!            01/01/70 00:00      
               your brain is your best debugger            01/01/70 00:00      
                  not always so            01/01/70 00:00      
                     blindness            01/01/70 00:00      
                        Tell it to the gatepost            01/01/70 00:00      
                           are you calling            01/01/70 00:00      
                           Simpler help.            01/01/70 00:00      
                              clerical errors            01/01/70 00:00      
                                 Re:            01/01/70 00:00      
                                    when quoting, be complete            01/01/70 00:00      
                        re:blindness            01/01/70 00:00      
                           testing            01/01/70 00:00      
                              debugging is NOT testing            01/01/70 00:00      
                                 not according to NASA            01/01/70 00:00      
                                    What?            01/01/70 00:00      
                                       unnecessary code            01/01/70 00:00      
                                       read all about it            01/01/70 00:00      
                                          that's the difference ...            01/01/70 00:00      
                                             I agree with NASA, you do not            01/01/70 00:00      
                                 terminology            01/01/70 00:00      
                                    definintions?            01/01/70 00:00      
                           when designing devices that can harm lif            01/01/70 00:00      
                              re:don't do harm            01/01/70 00:00      
                                 another article of interest            01/01/70 00:00      
                  WYWIWYS            01/01/70 00:00      
                     An old programmer's law            01/01/70 00:00      
                        An Old Programmer's response            01/01/70 00:00      
                           perfect communication            01/01/70 00:00      
   hmmmm            01/01/70 00:00      
      Learning.            01/01/70 00:00      
      Sorry Mr. Professional            01/01/70 00:00      
         don't read if they are boring            01/01/70 00:00      
            Allright            01/01/70 00:00      
               as I said            01/01/70 00:00      
   Thanks Payam!            01/01/70 00:00      
      Don't let Erik see that!            01/01/70 00:00      
         Bad engineering            01/01/70 00:00      
            I don't think it is bad engineering!            01/01/70 00:00      
               Calm down - it's only a joke!            01/01/70 00:00      
                  Yes and no            01/01/70 00:00      
                     Testing            01/01/70 00:00      
                        succesful testing does not prove the abs            01/01/70 00:00      
                           The proof is in the pudding            01/01/70 00:00      
                              you can't            01/01/70 00:00      
                              Proving            01/01/70 00:00      
                           successful testing!            01/01/70 00:00      
                           Bad testing            01/01/70 00:00      
                              bad, maybe, but possibly complete            01/01/70 00:00      
                                 Perfect vs Good Enough.            01/01/70 00:00      
                                    the slippery slope            01/01/70 00:00      
                                       90% agreement.            01/01/70 00:00      
                                          then why do you discuss releasing progra            01/01/70 00:00      
                                             I don't "release".            01/01/70 00:00      
                                                Is repeating part of the question an ans            01/01/70 00:00      
                                                   Yes.            01/01/70 00:00      
                                                      not authority but knowledge            01/01/70 00:00      
                                                         Where...?            01/01/70 00:00      
                                                            do you really think that anyone consider            01/01/70 00:00      
                                                               The $3.000.000 machine            01/01/70 00:00      
            Yes, but...            01/01/70 00:00      
   I have one more ...            01/01/70 00:00      
      CVS or SVN            01/01/70 00:00      
         ....            01/01/70 00:00      
   re:ISR            01/01/70 00:00      
      defeating a purpose            01/01/70 00:00      
         re:ISR            01/01/70 00:00      
   Wikipedia:: debugging (etc.)            01/01/70 00:00      
   Another one for your list            01/01/70 00:00      
   PC-LINT            01/01/70 00:00      
      anyone not using it?            01/01/70 00:00      
         re:anyone not using it            01/01/70 00:00      
            PC lint cost about $200            01/01/70 00:00      
               PC Lint            01/01/70 00:00      
            PC lint cost about $200            01/01/70 00:00      
               Not Always            01/01/70 00:00      
                  amen to that            01/01/70 00:00      
   All nice and pretty...            01/01/70 00:00      
      Noise problems?            01/01/70 00:00      

Back to Subject List