Comments for Old Dog, New Tricks http://mdenomy.wordpress.com Striving for better ways to write code Tue, 27 Oct 2009 01:13:34 +0000 http://wordpress.com/ hourly 1 Comment on New Team – First Retrospective by Ryan http://mdenomy.wordpress.com/2009/10/20/new-team-first-retrospective/#comment-79 Ryan Tue, 27 Oct 2009 01:13:34 +0000 http://mdenomy.wordpress.com/?p=84#comment-79 Inspect and adapt Inspect and adapt

]]> Comment on Project Retrospective by New Team – First Retrospective « Old Dog, New Tricks http://mdenomy.wordpress.com/2008/10/02/project-retrospective/#comment-78 New Team – First Retrospective « Old Dog, New Tricks Tue, 20 Oct 2009 20:54:43 +0000 http://mdenomy.wordpress.com/?p=35#comment-78 [...] of the software. It seemed like a good time for a retrospective. Of course, if I read my own posts, I would have remembered that the right time for the first retrospective may have been right when I [...] [...] of the software. It seemed like a good time for a retrospective. Of course, if I read my own posts, I would have remembered that the right time for the first retrospective may have been right when I [...]

]]> Comment on Using the Automatic Recovery Features of Windows Services by Per Lundberg http://mdenomy.wordpress.com/2008/02/28/using-the-automatic-recovery-features-of-windows-services/#comment-77 Per Lundberg Sat, 26 Sep 2009 19:23:17 +0000 http://mdenomy.wordpress.com/?p=21#comment-77 Hi! Thank you, this was "on target" to what I was trying to do. Writing a multithreaded Windows Service is obviously an elegant way to do it (when it needs to wake up at certain intervals and perform certain tasks), but... having it be restarted automatically by Windows is certainly an important part of it. I'm not entirely happy about this solution though, but I don't know of any better one. The ugly part of it is that this is what gets logged to the event log: "Failed to stop service. System.ApplicationException: Service failing because of previously logged exception" (Yeah, I chose a different exception type and text in my OnStop() handler, since I already write the actual exception text to the event log in the UnhandledException-handler) The ugly part of it is the "Failed to stop service"; it feels a bit misleading, at best. :-) If you (or anyone else) finds out an even better way to crash the service, please let me know! Best regards, Per Hi! Thank you, this was “on target” to what I was trying to do. Writing a multithreaded Windows Service is obviously an elegant way to do it (when it needs to wake up at certain intervals and perform certain tasks), but… having it be restarted automatically by Windows is certainly an important part of it.

I’m not entirely happy about this solution though, but I don’t know of any better one. The ugly part of it is that this is what gets logged to the event log: “Failed to stop service. System.ApplicationException: Service failing because of previously logged exception”

(Yeah, I chose a different exception type and text in my OnStop() handler, since I already write the actual exception text to the event log in the UnhandledException-handler)

The ugly part of it is the “Failed to stop service”; it feels a bit misleading, at best. :-)

If you (or anyone else) finds out an even better way to crash the service, please let me know!

Best regards,
Per

]]> Comment on Rethinking the C# using statement by Dave Amphlett http://mdenomy.wordpress.com/2008/02/11/rethinking-the-c-using-statement/#comment-53 Dave Amphlett Fri, 10 Jul 2009 10:33:35 +0000 http://mdenomy.wordpress.com/?p=18#comment-53 correction to above - I realised that a tiny class that does the attach in the *constructor*, and a detach in the Dispose() correction to above -
I realised that a tiny class that does the attach in the *constructor*, and a detach in the Dispose()

]]> Comment on Rethinking the C# using statement by Dave Amphlett http://mdenomy.wordpress.com/2008/02/11/rethinking-the-c-using-statement/#comment-52 Dave Amphlett Fri, 10 Jul 2009 10:32:16 +0000 http://mdenomy.wordpress.com/?p=18#comment-52 I came up with a use for this sort of anonymous 'using' and found this page whilst trying to prove to myself it was a reasonable thing to do. In my case I have various classes that can publish 'progress' of what they do, and a helper class that can show this progress on a progress bar. When I'm about to kick a class off to do some work I 'attached' it to the progress bar help class. When it had finished the work I 'detached' it. I realised that a tiny class that does the attach in the helper, and a detach in the Dispose() could make the code easier to read and use the simple 'using' statements 'final' functionality to ensure the detach is always performed. All very handy I came up with a use for this sort of anonymous ‘using’ and found this page whilst trying to prove to myself it was a reasonable thing to do. In my case I have various classes that can publish ‘progress’ of what they do, and a helper class that can show this progress on a progress bar. When I’m about to kick a class off to do some work I ‘attached’ it to the progress bar help class. When it had finished the work I ‘detached’ it.
I realised that a tiny class that does the attach in the helper, and a detach in the Dispose() could make the code easier to read and use the simple ‘using’ statements ‘final’ functionality to ensure the detach is always performed.

All very handy

]]> Comment on Project Retrospective by Kevin E. Schlabach http://mdenomy.wordpress.com/2008/10/02/project-retrospective/#comment-42 Kevin E. Schlabach Fri, 03 Oct 2008 16:10:22 +0000 http://mdenomy.wordpress.com/?p=35#comment-42 I have several posts on my blog related to <a href="http://agile-commentary.blogspot.com/search/label/retrospective" rel="nofollow">retrospectives</a> that might help you. Especially the one about <a href="http://agile-commentary.blogspot.com/2008/09/retrospective-accountability.html" rel="nofollow">accountability across retrospectives</a>. It's important in your first few retrospectives to insure you build trust by taking the raised ideas and following through with changing them! Good luck. I have several posts on my blog related to retrospectives that might help you. Especially the one about accountability across retrospectives.

It’s important in your first few retrospectives to insure you build trust by taking the raised ideas and following through with changing them!

Good luck.

]]> Comment on Energized Work and Personal Responsibility by Energized Work - Part II « Old Dog, New Tricks http://mdenomy.wordpress.com/2008/03/27/energized-work-and-personal-responsibility/#comment-40 Energized Work - Part II « Old Dog, New Tricks Thu, 02 Oct 2008 20:49:45 +0000 http://mdenomy.wordpress.com/?p=28#comment-40 [...] does someone go from blogging about Energized Work to being too busy to blog for 6 months. One day at a [...] [...] does someone go from blogging about Energized Work to being too busy to blog for 6 months. One day at a [...]

]]> Comment on Using the Automatic Recovery Features of Windows Services by Zaphod's Third Head http://mdenomy.wordpress.com/2008/02/28/using-the-automatic-recovery-features-of-windows-services/#comment-39 Zaphod's Third Head Mon, 30 Jun 2008 03:39:59 +0000 http://mdenomy.wordpress.com/?p=21#comment-39 This is helpful, but a stern warning about having JIT debuggers disabled on production servers would not be amiss. This is helpful, but a stern warning about having JIT debuggers disabled on production servers would not be amiss.

]]> Comment on Energized Work and Personal Responsibility by mdenomy http://mdenomy.wordpress.com/2008/03/27/energized-work-and-personal-responsibility/#comment-38 mdenomy Tue, 06 May 2008 00:31:32 +0000 http://mdenomy.wordpress.com/?p=28#comment-38 I guess it would depend on what skills you are interested in improving and whether they are applicable to the job you are working on. Sometimes I enjoy trying out some new technologies that I am not using in my current job. I find this can be a break mentally from "regular work" and keeps me up to date with the current state of technology for the next opportunity that might come up. I normally don't work on this type of "saw sharpening" at work because I can't, in good conscience, consider it part of my job. There have been exceptions to this, but only with my supervisor's knowledge and only then when there was no "real work" to do. So for me, that kind of learning has to happen on my time. The other part of your question seems to relate to better solutions for problems that are part of your job. For that, I think pairing is a great way to go, assuming there is someone who knows something about the technology. Even if there isn't someone with domain expertise, having two heads on the problem can keep someone from going down a rabbit hole (and never coming out). Finally in regards to your comment about "nobody ever adds time for ‘Learn a good approach to problem X"', I think it is up to experienced developers to identify bottlenecks and schedule time to adequately address them. That might require some research or a <a href="http://www.extremeprogramming.org/rules/spike.html" rel="nofollow">spike solution</a>, but to not plan for that time is a recipe for trouble. Thanks for the feedback I guess it would depend on what skills you are interested in improving and whether they are applicable to the job you are working on.

Sometimes I enjoy trying out some new technologies that I am not using in my current job. I find this can be a break mentally from “regular work” and keeps me up to date with the current state of technology for the next opportunity that might come up. I normally don’t work on this type of “saw sharpening” at work because I can’t, in good conscience, consider it part of my job. There have been exceptions to this, but only with my supervisor’s knowledge and only then when there was no “real work” to do. So for me, that kind of learning has to happen on my time.

The other part of your question seems to relate to better solutions for problems that are part of your job. For that, I think pairing is a great way to go, assuming there is someone who knows something about the technology. Even if there isn’t someone with domain expertise, having two heads on the problem can keep someone from going down a rabbit hole (and never coming out).

Finally in regards to your comment about “nobody ever adds time for ‘Learn a good approach to problem X”‘, I think it is up to experienced developers to identify bottlenecks and schedule time to adequately address them. That might require some research or a spike solution, but to not plan for that time is a recipe for trouble.

Thanks for the feedback

]]> Comment on Energized Work and Personal Responsibility by Ryan http://mdenomy.wordpress.com/2008/03/27/energized-work-and-personal-responsibility/#comment-37 Ryan Mon, 05 May 2008 15:35:06 +0000 http://mdenomy.wordpress.com/?p=28#comment-37 Granted, I'm a young bud with far less experience, but when you work an 8-9 hour day, and then close the books, when do you find time to improve your skill set? I've noticed that during task planning, nobody ever adds time for 'Learn a good approach to problem X,' it's always 'Implement X.' So you can spend all day implementing X, but I don't see where one gets the time to learn a better approach to implementing X if they don't use time in addition to what's been scheduled. As an example outside of software development, doctors see patients all day. If this is the only time they worked, it would only be a short time before their knowledge (skill set) became tired, old and rusty. I'm sure there are plenty of doctors like this, but if I had to choose, I wouldn't want them diagnosing me. However, maybe there is something to be said for all the experience you have, and it's no longer necessary to work so many extra hours since your base of knowledge is so vast. But I can't help feeling that the young buds still have to go through some learning pains to achieve the level of zen that you have. Granted, I’m a young bud with far less experience, but when you work an 8-9 hour day, and then close the books, when do you find time to improve your skill set?
I’ve noticed that during task planning, nobody ever adds time for ‘Learn a good approach to problem X,’ it’s always ‘Implement X.’ So you can spend all day implementing X, but I don’t see where one gets the time to learn a better approach to implementing X if they don’t use time in addition to what’s been scheduled.
As an example outside of software development, doctors see patients all day. If this is the only time they worked, it would only be a short time before their knowledge (skill set) became tired, old and rusty. I’m sure there are plenty of doctors like this, but if I had to choose, I wouldn’t want them diagnosing me.
However, maybe there is something to be said for all the experience you have, and it’s no longer necessary to work so many extra hours since your base of knowledge is so vast. But I can’t help feeling that the young buds still have to go through some learning pains to achieve the level of zen that you have.

]]>