Debugging

If you have ever wrote a program larger than “Hello, World!”, which worked without debugging – it doesn’t mean that you’re a genius, it just means that out of a thousand of programs you’ve wrote, you’ve got lucky once.

Most of developers spend most of their time debugging, so we probably should pay some attention to this subject…

Client-Side. On Debugging Distributed Systems, Deterministic Logic, and Finite State Machines

December 7, 2015 by “No Bugs” Bunny

Finite State Machine, Hare Style

Quote:

After your logic has failed in production, you can “replay” this inputs-log on your functionally identical in-house system, and the bug will be reproduced at the very same point where it has originally happened.

Another Quote:

You can implement your Finite State Machine as a deterministic variation of a usual event-driven program

Filed under: Book: D&D of MOGs1st beta of Vol. I-IIIOn.System ArchitectureDistributed systems(Re)ActorsOn.ProgrammingDebugging

Read more

Multi-threading at Business-logic Level is Considered Harmful

September 7, 2015 by “No Bugs” Bunny

Last Straw: Business Logic+Multithreading

Quote:

However, the window for triviality is very narrow: for example, even going into two interrelated mutexes instead of one, can easily make multi-threading non-trivial

Another Quote:

Technically you are able to jump to any point of your program, but the variables you see may (and if you have a multi-threaded bug – will) differ every time you jump there.

Filed under: On.System ArchitectureDesign decisionsOn.ProgrammingDebugging

Read more

Debug Complexity: How Assertions Affect Debugging Time

November 1, 2014 by “No Bugs” Bunny

Debug complexity

Abstract:

An approximate model on how ASSERTs reduce amount of work required to debug a program.

Quote:

ASSERTs can reduce debugging time dramatically

Filed under: On.ProgrammingDebugging

Read more

A Model for Debug Complexity

May 1, 2013 by “No Bugs” Bunny

Yes, it is indeed obvious...

Abstract:

A humble attempt to make some approximation of debug complexity.

Quote:

Our model shows that coupling is a bad thing for debugging complexity, so we can consider this sanity check to be passed

Filed under: On.ProgrammingDebugging

Tagged With: Kinda MathOverload
Read more