A Loss Of Faith

[BenTC]

Not entirely correct. Engineers design things to be of use to society in an economical way.

Not entirely not entirely.

I think you're talking about where engineers respond to their pointy-haired managers who are run by accountants who take input from marketeers.

I mean - just look at how perfectly good websites are mauled by over-eager site developers until they barely work any more. Take ebay and hotmail as two examples.

We, the public, are given products based on a version of what the accountants and marketing people come up with between themselves. (Send them off on a B-ark and us engineers'll fix this planet good and proper, so long as we don't die from an unsanitised telephone.)

I think that quote stands - in general - but yes, the rest of my post referred to engineers run by PHBs, particularly in relation to the commodity goods which were Josh's example. There is a lot of other engineering where much more value is placed on quality and serviceability - for industry where downtime may cost 10s or 100s of thousands of dollars an hour. Its still the client's balance of economy of how many nth-redundant systems they want.

[Jccarlton]

I would like to see at least one "discover the networks" style report discussing the big oil link to AGW skeptics. So far all I see is various people and groups working independently, with no ties to any industry. And the repeated accusation. Where's the proof. I've been hearing this accusation for ten years now. Surely somebody must have proof. Or it's a lie. Just repeating the accusation doesn't make it true. Exxon Mobil doesn't send me checks to be a skeptic.

[MSimon]

The summabitches never even offered me a dime. Let alone sent me one.

[Jccarlton]

Engineers are lazy, their job security is in designing crap and force feeding it to the consumer. I have done enough work in the field to know that innovative, intelligently designed things are just an accident.

Note that this does apply to software engineers, too, even though they don't have the same level of accreditation (software engineers don't have to be licensed and jump through OSHA-style qualifications).

Take cars for example (but it goes to anything, really). Replace a component on a car, odds are you'll find one or two darn bolts that are just out of reach or require removing a bracket to get at.

Took a stove apart a while ago, was trying to replace the insulation on the oven because a mouse had nested in it. The design of that thing was clearly asinine. Bolts in difficult to reach places, different sized nuts, phillips screws, all without rhyme or reason. Finally got the thing apart and put back in half of the screws that I took out once I was done. The repair worked fine (the fiberglass insulation worked a charm, though it was not designed for the task).

Scientists discover new tech, engineers apply it.

Poorly.

Josh,
Before making snap judgments about people you might want to think a bit. One thing I have learned after being in the engineering field for a long time now is that engineering is the art of compromise. you are always having to compromise time, money, regulation, environmental effects, hardware requirements and a ton of other stuff.
Consider cars. One of the best jobs I have seen documenting how engineering works at an auto company is Arthur Hailey's "Wheels."
Haily was a hack writer, but he had a knack for distilling the essentials of how businesses operate:
http://books.google.com/books?id=nBjXmN ... fox-a&cd=1
Cars are not easy. This especially true because the styling department gets the fits whack and mechanical engineering has to come along and make all the power plant parts fit. Even when all the parts fit sometimes changes come along because of new government regulations that didn't exist at the beginning of the project or some unforseen problem crops up. What happens when a vibration problem manifests itself. There are a buch of ways to get rid of it, you have to consider a ton of issues. Can you solve it by figuring out where it occurs and making small changes. Sometimes you can, but as often as not you are too far down the design cycle. So you fix the problem with a bracket and move on. The exhaust system in my car had one of those brackets. when they replaced the exhaust I don't think they replaced the bracket and I have noticed that the exhaust is noisier. Enough to bother me, no. But that bracket is an example of the hack jobs you sometimes have no choice to create even if you would like a more elegant solution. Often as not the budget and time just isn't there.
As for your stove you have to consider one thing. It's not designed to be disassembled the way you did. Design for assembly does not mean design for disassembly. this is done for good reasons. For most cases, if you have to have to go to the lengths you did to repair the stove, it's not cost effective. Consider that most stove are serviced by service people and not the owners. Those service people are expensive, figure $100/hr. A new stove costs $500. So if the service call is more than 3 hours, you trash the stove. That's the price point the designers of your stove the engineers had to target. So they never intended for the screws to be accessible after assembly. This isn't bad design, just the compromises they had to make for a $500 dollar stove. You want different compromises, buy a more expensive car or more expensive stove. But you pay for it both in up front costs and in extra service cost.
Engineering isn't easy. Especially when you are working on the bleeding edge, where as often as not even the science hasn't caught up with you yet. I wish that I had time to be lazy. But I have a major new product sitting on my desk right now that I will have to be finished with in six months or so. I will simultaneously be doing cost down on the previous product. Plus all the little surprises that make my day really interesting. You want better designs. change the way things work so that companies don't feel they have have legal and tax departments bigger than the engineering departments.

[Jccarlton]

Making an oven:
http://www.youtube.com/watch?v=8cpwCtkt9oY

[MSimon]

[quote="Jccarlton"][/quote]

It always amazes me that so many know better how engineering should be done - never themselves having had to face the problems an engineer faces.

I remember doing a juke box and an emi problem cropped up from the fluorescent tubes that lit it up. We were allowed 3 days and a maximum of 20 cents in parts to fix the problem.

First we tried it the easy way. Capacitors. The all purpose universal fix for emi. No dice. The wire harnesses were already ordered so we couldn't change them. One brilliant engineer (not me) thought that maybe relocating the harness slightly within its limits might work. It did. He was a hero. No parts cost. And we didn't lose any schedule time.

The biggest problems engineers deal with are not design problems. They are problems caused by logistics. Once the flow starts you have to deal with things as they are. Not what you wish they might be. Lead time is on your mind 24/7.

==

Another war story. I was working on an automotive diagnostic tool. I told the company it was designed wrong. They insisted it was not and besides a redesign would cause a one month schedule slip - new PCBoards - reprogramming the auto insert machines, delaying cash flow, etc. - so they went with it. Against my advice. The first ten preproduction prototypes worked perfectly. That Simon is an idiot. He worries about nothing (a common engineering affliction). In production 90% failed on the line and 99% failed in the field. Well I guess they got lucky. Unfortunately. Ten million dollars in effort scrapped. Bad customer relations. There were other dumb parts of the design but that was the worst.

==

Josh really. You have no idea how the system works.

[MSimon]

In many ways it is like a military operation. Calculated risk.

[Josh Cryer]

Design for assembly does not mean design for disassembly.

In the software world we call this "designing for job security."

Consider that most stove are serviced by service people and not the owners. Those service people are expensive, figure $100/hr. A new stove costs $500. So if the service call is more than 3 hours, you trash the stove.

Or you spend $9 to buy fiberglass insulation, and do it yourself. And reserve the right to complain if the designers and the engineers made it so that it was easy to assemble by uneducated Mexican workers, but very annoying to disassemble by an intelligent person who wants to save money.

I'm an individualist, guys, I don't like crappy designs. I realize the collectivist world we live in, the "system" as MSimon calls it, doesn't operate that way.

That doesn't mean I am wrong in chastising many "lazy" engineering solutions.

[MSimon]

Design for assembly does not mean design for disassembly.

In the software world we call this "designing for job security."

A confusion of domains Josh. The requirements for hardware are different than software.

It is possible to make software do a simple task 100% perfectly.

In hardware there is always a probability of error due to noise and the limits of the hardware. And the limits of the hardware are constrained by price.

i.e. I could use logic levels of +100 volts and zero volts along with currents of +100 mA and - 100 mA. It would lower switching speed and increase power but the chance of random noise affecting the circuit goes from once every 10,000 years to 10,000X longer than the expected life of the universe. But it would be a power hog and a 10 gate logic element might be the best you could get.

So what do you want Josh? A $500 stove that is very hard to fix in some cases or a $2500 stove that makes every thing easy. There are trade offs.

i.e. you want a stove that 95% of the population can afford (and .1% loses its investment for some failures) Or do you want a stove 5% can afford and no one loses their investment due to unfixable. There is a market for both. You pays your money and takes your choice.

You know what the proper compromises are called? A successful design.

Software only guys are head in the clouds thinkers.

[MSimon]

I'm an individualist, guys, I don't like crappy designs.

Then pay for good ones.

Revised for clarity.

[Jccarlton]

Design for assembly does not mean design for disassembly.

In the software world we call this "designing for job security."

Consider that most stove are serviced by service people and not the owners. Those service people are expensive, figure $100/hr. A new stove costs $500. So if the service call is more than 3 hours, you trash the stove.

Or you spend $9 to buy fiberglass insulation, and do it yourself. And reserve the right to complain if the designers and the engineers made it so that it was easy to assemble by uneducated Mexican workers, but very annoying to disassemble by an intelligent person who wants to save money.

I'm an individualist, guys, I don't like crappy designs. I realize the collectivist world we live in, the "system" as MSimon calls it, doesn't operate that way.

That doesn't mean I am wrong in chastising many "lazy" engineering solutions.

Having just been wrung through the wringer in a design review, I have to wonder if you perhaps should not go around insulting everybody. In my job I can't do everything by myself. There just isn't time and there are parts of the job that are not my specialty. why should I, as a mechanical engineer, learn to how all the electronics work? I have to know enough to understand how to interface with the electronics, but knowing how to use IC's and create more than the simplest PC boards, no I don't need to do that. The same goes for software. The average product has all of the various engineering disciplines working together. In fact I think of my role as senior designer as a facilitator and communicator as much as somebody who puts things together.
The problem with your way of thinking is that you are a distinct minority. I can't design for that minority. Especially as doing that means that the price point goes out the window. The fact of the matter is that if you want a stove that behaves the way you want you can get a commercial grade stove. But it's going to coast you over $5k. But you will be able to tear it down and fix it any way you want.

How long did it take you to dismantle the stove, go down to Home Despot, buy the insulation, fit the insulation and put the whole thing back together again. A whole day is my guess. Time is money and it's valuable. That 100/hr I quoted is about average for a service call. The fact is for an engineer, replacing the insulation jacket is a very low logistics item. For most stoves the MTBF for the fiberglass is beyond the expected life of the stove. The fact is that the parts that do need to be replaced, like the heating elements are easily removed, often without screws. Some of the other parts, like the timer, actually are designed to come out as easily replaces units. but major structural element just don't fail that often. Which means they don't have to be accessible or removable.
As for those uneducated Mexican or other immigrant workers, did you know that if you fly you entrust your life to people like that. In the company that I worked for all of the aircraft engine harnesses were made by immigrant labor that probably didn't have a PHD. But they do work hard and pay attention to what they are doing. In fact I would trust them more than some American workers. You don't want to know the stories I've heard about drunkeness on the job and working on aircraft that I have heard over the years.

[Betruger]

Bolden's press conference says (unless I heard wrong) that climate data from near future NASA sats will be fully disclosed.

[Josh Cryer]

MSimon,

So what do you want Josh? A $500 stove that is very hard to fix in some cases or a $2500 stove that makes every thing easy. There are trade offs.

They used to say that about computers, now 12 years olds are building their own computers (from parts). In fact, I would wager that the more expensive stove is even harder to fix, with replacement parts that come from the manufacturer only sold to licensed repairmen and such. I know for a fact that when I replaced the internal fan assembly in my laptop the fan from the manufacturer was 10x the cost from a third party guy who had OEM parts sitting on a shelf somewhere (I am still shocked that he sold the assembly to me for $15 plus shipping when HP was selling it for $150! What a guy! Guys like that are *increasingly rare*).

15-20 years ago taking a computer case apart was a flipping nightmare (at least *they* used similar screws). Now-a-days it is standard for two-three hand screws to remove the whole case side. Bicycles used to be a nightmare to repair (I remember walking to the dollar store as a kid to get a patch kit and fixing my bike, it was no easy task), now you have quick release wheels.

The reason ovens are not designed for the consumer to fix or repair is because there is no demand for it. Plain and freaking simple.



Jccarlton, your "time is money" bullshit is just that, bullshit. It took me 10 minutes at Home Depot to get the insulation (because I work in construction I knew exactly what to get), and about an hour to do the repair. It would have taken me 15 minutes if they knew how to design the thing.

Or it would not have even been necessary if they designed it so that a mouse could not get in to the insulation. It was a joke to the Nth degree. Since I moved out of that "rat hole" I never got around to actually fixing the defficiency in the design by wiring pieces of screening to the various entry locations.



Betruger, NASA data products must all be disclosed to the Planetary Data System within 6 months. All. Period. It is a precondition on the grant process. I just hope it speeds up the development of CLARREO so we can get an answer and shut everyone up once and for all.

[Betruger]

You still have to model, once the data's pinned down. So it probably won't shut everyone up.

[BenTC]

In fact, I would wager that the more expensive stove is even harder to fix, with replacement parts that come from the manufacturer only sold to licensed repairmen and such. I know for a fact that when I replaced the internal fan assembly in my laptop the fan from the manufacturer was 10x the cost from a third party guy who had OEM parts sitting on a shelf somewhere.

You are mixing your arguments. That is a defficiency in pricing, not a defficiency in design.

15-20 years ago taking a computer case apart was a flipping nightmare (at least *they* used similar screws). Now-a-days it is standard for two-three hand screws to remove the whole case side.

That improved case design is for the benefit of Business IT Department decision makers. When managing thousands of PCs the small failure rate causes enough trouble that simpler hardware replacement is attractive, which affects bulk buying decisions and hence engineering design objectives. Home users do not figure in this.

Bicycles used to be a nightmare to repair (I remember walking to the dollar store as a kid to get a patch kit and fixing my bike, it was no easy task), now you have quick release wheels.

Cyclists get punctures all the time. Tyres have always been user replaceable, long before quick release. Making tyres easier to change is a selling point and hence becomes an engineering design objective. User replaceable insulation is not a selling feature that will ever appear on a consumer brochure - hence for consumer ovens, not an engineering design objective.

Since I moved out of that "rat hole" I never got around to actually fixing the defficiency in the design by wiring pieces of screening to the various entry locations.

Why not take the TIME to do it right? Sounds fairly straight forward. It is almost like you had something better to do - time-risk-reward compromise, you might call it, aka time-is-money-buillshit.

The reason ovens are not designed for the consumer to fix or repair is because there is no demand for it. Plain and freaking simple.

Correct. Hence simple-consumer-repair is NOT a design objective.