Stories about Software


DaedTech Digest: Analysis Rules, Singletons, and Log Aggregation

Happy Friday, DaedTech readers!  Time for yet another installment of the DaedTech Digest.

I didn’t do one last week because of the US holiday.  For those of you not from the US, there’s this holiday called Black Friday that everybody celebrates by getting together and eating turkey the day before and then subsequently bludgeoning one another in retail stores the next day.  Out of respect to this noble tradition, I held off on making a post.

At any rate, it’s been a busy couple of weeks in my world.  We’ve migrated south for about 5 weeks and are living in a beach town on the Gulf of Mexico.  This photo below is a shot of where I’m working from today as I type this post.

So life is good.  With that in mind, let’s get to picks.


  • The jury isn’t totally back yet on this, but we’re trying out FreshBooks for managing Hit Subscribe‘s books.  And, so far, it’s great.  And it integrates with other favorite Gusto besides.
  • I’ve been listening to an Audiobook called How to Measure Anything, and enjoying it.  Any consultant worth his or her salt spends a lot of time figuring out how to quantify problems and solutions, and this is definitely helpful.
  • The folks at a site called Data Camp reached out to me about the study I’m doing over at the NDepend blog.  They seem to be developing a cool offering for teaching people about data science using Python and R.
  • I also pick time off.  I work very much on my own schedule and generally not full 8 hour days anymore.  But I also do tend to work 7 days a week.  Last weekend, Amanda and I played tourist and did nothing but explore the Gulf Coast, buy fresh seafood off of piers, dine out, explore new cities, and walk through state parks.  This was nicely restorative.

DaedTech Post Digest

  • In my consulting (and even long-ago salaried) travels, I’ve encountered a lot of myths about code reviews.  I wrote a post for the SubMain blog in which I looked at some of those.
  • I wrote a post for NDepend, in which I explored a topic probably not many have.  What’s the role of static analysis in your testing?   These things might seem orthogonal, but I think they actually have an interesting relationship.
  • Also for SubMain, I did a post in the CodeIt.Right Rules Explained series.  I examined why you should avoid single line if statements, why you shouldn’t base your enums on weird underlying types, and the problem with explicit rethrows.  All of this in C#, BTW.
  • In a post that went a little viral and predictably resulted in people calling me a know-nothing incompetent, I wrote about what the singleton design pattern costs you.  This was for the NDepend blog, again.  (But I’d have the last laugh later, when I would actually study it empirically and prove myself mostly right — stay tuned for that in a future digest.)
  • In a much less controversial post (because they don’t enable comments), I made the business case for unit testing on the ASPE blog.
  • For Scalyr, I wrote a post about log aggregation.  What is this, why would you want it, and how does it help you?

And, that does it for the digest round up.  Have a great weekend!