I often have people asking me about the free versions of software that are floating around the internet. And lately, I have been asked about a $500 version of EtherNet/IP that is available.
People ask me if this stuff is for real, if it works and what are the chances for success. I usually don’t spend a lot of time answering. Reason being that there isn’t really a good way of dissuading someone who perceives a short time benefit at no cost. If you have severe price pressures and your sales are eroding, people are going to take chances and assume risks that they would normally avoid. Sanity is something that some people have a really hard time to hold onto right now.
So what are those risks? Here are the major ones:
Free may mean that you spend a lot more time in development. You don’t get tools with free software and you don’t get a lot of documentation. So you have to figure it out. Sometimes you can and sometimes you can’t. If it works (or at least seems to in you lab), you’re a hero. You’ve saved the company untold thousands of dollars. If not, you’ve lost a few weeks or months but you can buy something that works.
Sometimes people confuse free with Open Source. The Free software stacks on the internet are unsupported. There is no body updating the code. No one finding bugs, making enhances, or making revisions to conform to the latest guidelines from the ODVA. Open Source is source that has a wide following like Linux. Groups of people all over the world maintain that software and enhance it. There are lots of forums to go to if you have a problem. Lots of help. You don’t get that with the free Industrial Stack.
Assuming you can get EtherNet/IP working in your lab. Your next stop is conformance testing where they send 4 or 5 hundred thousands messages to your device. At the ODVA, problems show up from time to time and the tester gives you a very cryptic series of messages showing the sequence of messages that led up to the conformance issue. Now what? You have to pull out the spec, all 6 inches of it, and start reading. You are now in a position where you have to get familiar enough with the technology to understand every last intricacy. Believe me, if the problem is deep in the stack, you have a great risk of breaking something else. As time Ticks away so does the development cost meter every hour costs around $195. The cost of free is adding up fast.
If you get past certification and ship you’re first unit you are still not out of the water. The risk actually increases greatly. You not only have to spend time on any problems that come up for your customer, you’ll probably have to tackle them on-site with a plant manager breathing down your neck. If resolution takes more than a few days, you could lose an order or worse an important customer. You are stuck and have no one to call for help. So you get the 6 inches of spec out, a big cup of coffee and go at it. The hours are adding up astronomically now. You’re boss is worried. You’re management is wondering who to blame for this mess. The cost of Free just went through the roof.
So what is the cost of free? It’s all risk. Time-to-market, Product Introduction, Company reputation, and your reputation are all at risk. Free is pretty costly when you consider all the extra hours you’ll have to put into the project and the risk you take on to save a few thousand dollars.