Another salary survey

The last time I wrote about salary surveys for test engineers was way back in 2009.  I saw this interesting salary survey a couple weeks ago and thought I'd dig into the topic again.

It looks like T&MW no longer does salary surveys.  I found this article referencing a 2010 survey, but when I tried to access that survey, this is what I got:

I tried searching on their website but found no other references to the 2010 survey, or a 2011 survey for that matter.  SO, have they switched to an every-other-year format, was it too expensive to continue publishing the survey, or did they get too many complaints from engineers using the survey data to ask for raises?  (I admit that I've used those surveys for exactly that purpose...)

I DID find this interesting "lifestyle" survey instead of the salary survey.  Supposedly they'll publish the results this month.

A salary survey for all engineers was done by Design News.  That's worth checking out.  But if you want a lot more detail, go to Salary.com.

Intro to DAQ

United Elecronic Industries has recently (again) put out a document called "Everything You Ever Wanted to Know about Data Acquisition."  This is apparently a three-part endeavor, and part 1 is available here.  I took a quick look at it, and it appears to be what it says it is - and introduction to DAQ.

Horrible LabVIEW

At my current company, I inherited some LabVIEW code that was written five or so years ago by a previous employee (who left at least three years ago).  The code works, so it has that going for it.

I would argue that there are some basic rules that any decent LabVIEW person should know.  I wrote a post many years ago about LV style, but what I'm talking about is more of a ground level set of standards.  This list includes things like:

1. Don't route wires underneath objects
2. Don't send wires right to left
3. Avoid complicated bends
4. Avoid stacked sequences
5. Avoid local variables
6. Comment the code
7. Utilize error trapping

The code I inherited breaks six out of seven of these.



I know that when I first learned LV (well over a decade ago), I wrote some sketchy code before I learned better. Maybe that's what this is - I really hope this guy's coding improved.  What he wrote back then is just painful for me now.

My new car

A few months ago I was in a car wreck (partly explaining my absence on this blog from April until July).  My poor Prius, with well over 200k miles on it, was totaled.  So I had to jump back into the car market.

For various reasons, I ended up buying another Ford hybrid vehicle, a 2012 Fusion.  While not reaching the 52mpg I got with the Prius, I've been getting a respectable 42mpg as well as more oomph when I accelerate.

As I mentioned last year when talking about my Ford Escape hybrid, this car has Sync in it.  But it has even more software and nifty tricks, including a ridiculously cool rain sensor windshield.  Can you just imagine how much testing all this stuff has to go through?

TestStand (part 3 - Is it useful?)

As part of a project I completed for my previous company, I  learned NI TestStand and implemented it for a test station.  I started writing a few posts about that experience and my impressions of TestStand a year ago.  I'm finally getting around to finishing it, and this is the 3rd and last post.

Is it useful?

That's a big question.  I've talked with three different people who've used it at bigger companies.  All three liked how it worked.  Personally, I can only answer that question for startups.

First of all, I had a few specific complaints about it:

• The database viewer was okay, but you couldn't look at stored procedures.  It seemed very "beta".
• I couldn't open more than one workspace at a time.
• The printing wasn't very friendly.  For example, I wanted to create a nice flowchart based off of the sequences/workspace.  I couldn't do it easily and needed to jump through hoops to "sort of" get what I wanted.
• There's no support for Tortoise SVN (of course, I have that complaint about LabVIEW as well).

Now that I've gotten that off my chest, let me address the broader question.  Yes, it is useful in startup companies.  At all the startups I've been at, you start out developing small programs to do specific tasks.  But eventually you get to a certain point where:

1. You hire technicians to run regular tests on your products.
2. You need a way to quickly make modifications to those existing tests.
3. You need to generate reports on those tests.

In other words, you get to the point where you need some sort of test executive.  A couple of startups I've worked at wrote their own test executive, either because they didn't know any better or didn't want to spend the money on a 3rd party solution.  To me, that's a waste of time and effort, and you spend the extra money anyway in effort.  Yes, TestStand isn't cheap, and yes there is a learning curve.  But it was definitely useful.

TestStand (part 2 - Learning how to use it)

As part of a project I completed for my previous company, I  learned NI TestStand and implemented it for a test station.  I started writing a few posts about that experience and my impressions of TestStand a year ago.  I'm finally getting around to finishing it, and this is the second of three posts.

Learning how to use it

I came into TestStand from the LabVIEW world, although I have a long background in text-based languages.  When my development group decided we wanted to evaluate TestStand, the first thing I did was download the evaluation software (free) and purchase the source materials for the training class (a couple hundred bucks).  After spending a week walking myself through the training and reading through the manuals, we decided to buy the software and implement it in our test systems.

Here's a couple of pointers for learning it yourself:

1. Download the manual and eval materials from NI (reference manual). 
2. Don't pay for the actual training if you don't have the cash.  Just buy the source materials for the training class.
3. This book is a collection of articles about using TestStand.  If you don't want to spend the $10, you can get most of it from here online.
4. Step through all of the examples you can find.

The God particle

It looks like the Higgs boson has been found.  Very cool.

As I've mentioned several times before, I spent time working in high energy physics, before the SSC was shut down.  Back in grad school we talked alot about how close we might be to discovering the Higgs particle and what it would mean.  It's taken a heck of a long time to finally get there, but sometimes research (and taking lots of data) is like that.