5 ms·

The value of a personal bug log

31 points by shedd 16y ago

yatsyk 16y ago
It would be nice to see few bugs and unit tests. It looks nice in blog post but I'm not sure that this approach could be used in real life.
wccrawford 16y ago
"never think about the bug again" That has never been part of the process. That may be what some people do, but it's not part of the process. Good programmers figure out how the bug happened -while- they are fixing it. There's no need to spend time afterwards. (Okay, there are a -few- cases that require research, but I've never met one of those that didn't require me to track it down anyhow.)
- darkxanthos 16y ago
  The reason he gives for spending time afterwards is at the end of the month to allow you to review a pattern across all of the bugs you worked on recently. That's difficult to do with just your memory. Just figuring out and understanding a bug is not enough. His point is to figure out and understand how bugs are manifesting themselves in your system... Slipping through all of your safeguards... showing your lack of control of your code.
StanDarsh 16y ago
I like to make video screen shots with CamStudio and record myself searching for the bug, finding the bug, explaining why the bug occurred, etc. I check the video into a log I have. It's worked very well, and it's much easier to watch the video instead of deciphering comments over, in some cases, multiple source files.
- ovi256 16y ago
  That's obviously a very powerful technique, and a close analog to dictation, which is widely used by scientists, investigators, writers, attorneys, and other people investigating complex issues. I use Google Notebook to take notes, and I just wish I could record dictation into my notes.
pmiller2 16y ago
"Because I screwed up" as a reason for a bug existing doesn't seem specific enough to me. My impression is if that's the best reason you can possibly come up with, then you should probably either think harder, or, if it seems to be just a case of "I made a typo," and it's one that's not similar to past mistakes you've made, you should forget about that particular bug unless a similar bug surfaces in the future.

31 points by shedd 16y ago

It would be nice to see few bugs and unit tests. It looks nice in blog post but I'm not sure that this approach could be used in real life.

wccrawford 16y ago

"never think about the bug again" That has never been part of the process. That may be what some people do, but it's not part of the process. Good programmers figure out how the bug happened -while- they are fixing it. There's no need to spend time afterwards. (Okay, there are a -few- cases that require research, but I've never met one of those that didn't require me to track it down anyhow.)

darkxanthos 16y ago
The reason he gives for spending time afterwards is at the end of the month to allow you to review a pattern across all of the bugs you worked on recently. That's difficult to do with just your memory. Just figuring out and understanding a bug is not enough. His point is to figure out and understand how bugs are manifesting themselves in your system... Slipping through all of your safeguards... showing your lack of control of your code.

StanDarsh 16y ago

I like to make video screen shots with CamStudio and record myself searching for the bug, finding the bug, explaining why the bug occurred, etc. I check the video into a log I have. It's worked very well, and it's much easier to watch the video instead of deciphering comments over, in some cases, multiple source files.

ovi256 16y ago
That's obviously a very powerful technique, and a close analog to dictation, which is widely used by scientists, investigators, writers, attorneys, and other people investigating complex issues. I use Google Notebook to take notes, and I just wish I could record dictation into my notes.

pmiller2 16y ago

"Because I screwed up" as a reason for a bug existing doesn't seem specific enough to me. My impression is if that's the best reason you can possibly come up with, then you should probably either think harder, or, if it seems to be just a case of "I made a typo," and it's one that's not similar to past mistakes you've made, you should forget about that particular bug unless a similar bug surfaces in the future.