Not fixing the problem is an acceptable action by the company, but ONLY if an Errata is clearly provided (either in the part's data sheet or on their web site located where it is plainly visible when looking for documentation). Chip manufacturer's do this all the time and it's a reasonable solution to component problems found late in the development cycle.
On the other hand, sweeping the problem under the rug (and hoping for the best) is clearly unethical, at best. At worst it shows a casual disregard for their own reputaion and for their customer's time.
Not sharing this companies name only rewards this kind of behavior. It's venues like this that can have positive results in getting companies to do the right thing (but only if the light shines on them, specifically).
Charles, in this particular instance, fixing the firmware would have required rather extensive testing and documentation. The device was a safety relay. It would have likely affected the documentation filed with the responsible authority that approved the device (UL, TUV, CSA, etc.).
I really can see the manufacturer's point of view to not bother, since a new series was coming.
I only object to the fact that they continued to sell the defective product with no obvious warning.
The author is absolutely right: It's an integrity issue. For some reason, manufacturers sometimes ignore firmware problems, whereas they wouldn't ignore a faulty motor. Even though a firmware fix appears to cost less, it can waste massive amounts of time and labor.
The instructor at a BISSC training class, on the subject of food recalls, the instructor quoted a nameless CEO on the question "What is your company's name worth?" The CEO's answer was "Billions".
The context was, that if a company didn't get in front of a product recall, their company name would suffer greatly; the cost of the recall would always be less than not doing anything. Ideally, a food company would have no recalls, and an industrial automation company would make perfect parts.
This company does try to keep its customers notified. They send out product advisories when they discover a problem and know you have purchased that particular product, and they have an extensive knowledgebase. The information is there, if you know to look for it. I suppose in their eyes, they've done due diligence. In this case, I don't think the actions were enough.
I've also begun to wonder how extensive a problem this is as well.
I agree about a warnig being issued. On a broader level, one can infer from TJ's post that this kind of stuff is much more frequent than one supposes. Here, it came to light, but in many cases it doesn't. Do this "fix it" versus "cost" issue is something many manufacturers have to deal with, and I guess we can guess where the decision often ends up, if there's not a hard failure or solid safety issue involved. So this post raises some questions well worth thinking about.
I have to agree with your first inclination, TJ. There should have been an official warning issued and a red flag in the manufacturing system, as you suggested, so prospective buyers had the full skinny on the problem before committing to purchase the component. Just because the manufacturer responded on the up and up AFTER you found the problem doesn't absolve them of that responsibility--in my book, any way. You are right to call them on the carpet.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.