I had a contract job about 15 years ago to write diagnostic software for a product. When I asked where the spec was I was told there wasn't one, just go test it. I protested, that if I did not know what it was supposed to do, how could I tell if it was doing it right. After being given the brush-off by the manager and the managers boss, I quit, and told them why. A year later I was a newspaper report that that company had agreed to pay a fine of ONE HUNDRED MILLION DOLLARS for poor quality control. That sure made me feel good, sad as it was.
I'd have to disagree Bill, though it's difficult to call it unless we get some answers to the questions you posed. Sounds like "marketing" were the interface between engineering and the customer and didn't allow much direct contact between the two. To the detriment of the project, though you have to admit that Glen's self-imposed design criteria did the job, and did it well.
I had an interesting thought about future product development and marketing. Several times in the past, I have thought of ways to improve products or have foreseen better products way before management or marketing thought of them, but held back because they would say, "great, we want that tomorrow along with our requests". Since engineers may know the products capabilities better than management or marketing, it makes sense that engineers would likely know the next or better products to create. If this is the case, have any other engineers held back from management or marketing, product improvements and the reason why that was “necessary”?
This article is a good pitch for why engineers need to wear multiple hats, understand the application and system viewpoint -- and why they definitely need to interact with customers.
You will never get a perfect spec or enough time or enough resources. That's just reality.
I laughed when I read the part where your boss gave you a week to do the job, and I thought, well, at least he didn't say "I need it yesterday!"
I also had to wonder how many weeks had already gone by since the initial customer meeting. Sometimes the long pole in the schedule is the decision phase -- the time from initial customer contact to the time when the engineer is told "go!"
I can relate with you experience, thank you for sharing and providing your point of view which deserves recognition for your accomplishments. What I am trouble with after reading some of these postings, it is that everyone assumes that every company has a perfectly defined staff with well organized project management, which communicates perfectly well among groups, and schedules are well planned and perfectly timed and prioritized.
However, we have Marketing and Engineering groups always at opposite ends of the design spectrum, with competing priorities: time-to-market, lowest-cost, highest-quality / reliability It is well understood, that there needs to be a compromise to reach for all these design ideals.
From my point of view, what I have seen in many organizations is that we are lucky to get marching orders in a coherent product spec, in some reasonable schedule time line and sufficient resources to get it done. In most cases, in which I have participated in real product development, the product is pushed as hard as possible in each phase execution, and where costs are not too excessive, (a product development constraint) the accelerated steps are taken (faster turn on PCBs, machined prototypes, compliance testing, etc). However, when there are poorly defined specs, the delays and cost over-runs are second nature, in most product development. Success under those conditions is a rare event not the rule
Even when you take the time to understand your costumer and deliver the best product under the target goals, there is no guaranty that the Market will embrace your product, whole heartedly; there is always the cooker pressure to deliver an ideal product with the resources available. So we should not be surprised on poor outcomes, when there was not good planning or goals not properly reviewed and analyzed to make sure they align with the vision of timeline, performance, and costs.
I had an interesting challenge, I am not certain if it is right to call it marketing when there is only one item sold and more similar products are unlikely. But we started on the design, based on a verbal description, and when the description was completed the salesman then told us about all of the features that he had sold that we had not included. Suddenly the ability to do the job for the quoted price was gone. We did make the thing meet all specifications, but the profit was less than what we had hoped for.
Wow, a lot of complaints about marketing. I find it's all too easy to point the finger at the generic bad guy "marketing". Did you really mean product management? Perhaps there were no product managers at the time, so what person or group was responsible for providing the details you needed? Did the customers know what they wanted? Did you talk to the customers?
Unfortunately the article looks more like a rant and raises more questions about yourself that it answers.
Often Engineering and Marketing "cross swords", so to speak. My own experience is that the issues are largely a matter of mutual trust.
By this I mean that Engineering, thru repeated examples of good market information, can believe the requirements placed on them vis a vis specs, time frames, and cost constraints. e.g. thorough customer and market input.
Conversely, Marketing must be smart enough to understand Engineering's capabilities, and again with repeated experience, believe what Engineering tells them. Marketing must also accept that saying "make it so" does not automatically make it so.
Engineering also must clearly understand that time is a factor in most profitable projects, and must be able to be fairly accurate in their time estimates for a given project, assuming they have as much information as is possible to obtain regarding it.
Marketing must trust Engineering's abilities and timelines and not over commit to customers. Engineering must trust that Marketing will obtain any and all information that it can to support Engineering's work. Customers prefer accurate information, even if it is not always the schedule they want.
Most of all, Engineering and Marketing must clearly understand that their's is a partnership which must be nurtured for all projects. Failures in either party must be addressed and openly discussed and improved upon. No company will exist for long without this strong partnership.
Blog Doing Math in FPGAs Tom Burke 23 comments For a recent project, I explored doing "real" (that is, non-integer) math on a Spartan 3 FPGA. FPGAs, by their nature, do integer math. That is, there's no floating-point ...