[haiku-commits] Re: r41065 - haiku/trunk/src/apps/showimage

  • From: Stephan Aßmus <superstippi@xxxxxx>
  • To: haiku-commits@xxxxxxxxxxxxx
  • Date: Mon, 21 Mar 2011 21:18:57 +0100

Am 21.03.2011 13:14, schrieb Axel Dörfler:
superstippi@xxxxxx wrote:
  * Added appropriate work arounds for bugs in the Interface Kit
    concerning the hiding of tool-tips, and a bug in the app_server
    concerning the invalidation of unhidden views...

Is it really needed to waste your time like this? It makes much more sense to 
actually fix the bugs where they are, and those certainly look like one liners 
in the appropriate places (or at least not more code than what you added here).
Code like this should only ever be considered with release pressure, and in the 
release branch. It has no place in trunk, and only leads to poor code quality 
in the end.

Besides, actually having a test case at hand to reproduce the bug (or even be 
annoyed by it) greatly helps finding the time and motivation to fix bugs.

To be honest, I feel a bit upset about your mail. I hope it's ok to mention it and explain why: The basic problem is that I somehow cannot believe that you think I am not aware of the problems you mention. So as a result it feels like you don't trust my judgement of how I could make the best use of my time.

Obviously I was aware that the problems needed fixing. I reopened the DeskCalc bug, which several people including me commented that it was closed in error and there is a bug elsewhere. I logged two more tickets about the tool tip problems. All three tickets I assigned to myself, a sign that intend on fixing them eventually.

The tool tip problem is not in my own code, I have no idea how long it would have taken me to find the spot and a good fix. The app_server bug is in my own code. I am well aware that changing this code can have side effects, testing changes takes time. I simply assume that the app_server test environment build is broken as it always is when I haven't tried to build it in a while. Booting between Linux or testing in a VM is not giving quick turn arounds either.

At the moment, it feels like we are shortly before our next alpha release. I have no idea when I will have time next to look into either problem. Given all these circumstances, I decided it's better to fix the already commited tool bar in ShowImage with three lines of code after having understood the issues and logging the tickets with enough information.

Given all the above, I have to wonder why you think it is worth our time to have this conversation. Am I such a frequent offender and commiting work arounds and advertising laziness all the time that you think it is necessary to question my judgement of how I could make the best use of my time? You obviously have a valid point, it's better to fix the real problem. But was it really necessary to mention it to me all things considered? Having these conversations takes time as well, so one should trigger them carefully. That's not even talking about the demotivating effect of such comments. Please consider the implications of your mail and how it was supposed to make me feel.

Best regards,

Other related posts: