Curse you, Steve Jobs!

[packbat]

Zee laptop, eet is kaput! All kaput!

You know, it could be my carting it about in the backpack every day, but I think Apple has lost quality since the old Mac Plus days. I have taken this in to replace the hard drive, to replace the monitor, to replace the logic board ... this stinks.

(Of course, the computer I want to replace it is a MacBook....)

Comments

[baxil.livejournal.com]

I'm not certain it's Apple so much as it is the march of technology. You see this across the tech sector.

To name one of a great many examples in recent memory, I just found out when I bought my latest external hard drive that Maxtor has reduced their warranty from 5 years to one.

[packbat.livejournal.com]

Y'know, I just thought of something.

One of the latest trends in American manufacturing (and I'm studying mechanical engineering right now, they're telling me this is a recent trend) is an increasing modernization of the product design process, including such things as Quality Function Deployment. (If you're interested in history, this was a technique which was invented here, but never caught on in the US until after Japanese firms – who the inventor taught it to during post-WWII reconstruction – began using it to produce cheaper products than American companies could. So, kinda like an industrial analogue of Blitzkrieg.) Among the steps involved in the process are judging the customer requirements, which includes determining the numbers for how long the product will be used.

In one of my other classes, the teacher was just saying last Thursday (quoted from memory), "We don't design for infinite life any more. We design for exactly the number of cycles we need."

[roaminrob.livejournal.com]

Browsing your journal now as a result of Bax making his BaxilDragon M:TG card based off your M:TG card...

I have a bit of experience with this, as I just got out of a manufacturing job. I was an output smoothness tech for a major potentiometer manufacturer; a lot of our stuff ended up as critical controls in tanks, missiles, commercial aircraft, etc. Based on my experience there, I'd say your teacher is exactly correct. I'd even take it a step further: we don't manufacture for infinite life.

In the parts we built, it was possible for a part to have a manufacturing defect which made the part unreliable or shortened its lifespan, but in such a way that the part met the customer's specifications anyway. One excellent example is lube contamination: before housed units have their covers epoxied on, they get a light coating of lubricant. Nothing in the facility was clean, there was often dust and occasionally conductive dust (from machining) in the air. This dust would end up in the lube before the cover was epoxied, then the part would be sealed and come to me.

At my station, a common spec was an allowed voltage fluctuation of 10mV over 150 milliseconds, and I would see out-of-spec fluctuations come and go as the part's rotor was turned. The official instructions from the general manager were to continue turning the rotor until I got a clean sweep, and then pass the part on to shipping. (!)

I've been so frustrated by crappy products that I often rejected stuff that I knew was bad (and occasionally got in a little trouble for it).

If most manufacturers are anything like the place I worked at, then their highest priority is their gross monthly (and quarterly) production in dollars. The more crap they can poop out, the better. If that means cutting corners and shipping more borderline product, then that's what they'll do.

[packbat.livejournal.com]

It makes me think of something I read in a history class. Have you ever seen the ASME Code of Ethics (http://files.asme.org/ASMEORG/Governance/3675.pdf) for engineers? Historically, this code, and many like it, were born of the movement by Morris Cooke and the like, sometimes described as the "revolt of the engineers," to insist that the first responsibility of the engineers was to society, not to their bosses.

Morris Cooke lost. The revolt of the engineers only lasted through the first couple decades of the twentieth century. By 1928, engineering schools were already talking about producing "a product which is saleable" to future employers.

Source: "American Genesis: A Century of Invention and Technological Enthusiasm, 1870 - 1970", Thomas P. Hughes, University of Chicago Press. And technically, he didn't state it as baldly as I did.

[roaminrob.livejournal.com]

I hadn't heard of any of that. Thank you.

My guess is that most engineers approach everything from a point of reason, and it's relatively easy to argue against designing things for very long lifespans. You might say that it's not cost-effective, or that it would outlive its thpractical application, and the engineer would be forced to say, "Well, yeah..."

Even given that admittedly lame excuse, I can't figure out why the movement would have died out.

I think engineers need their own version of the Hippocratic Oath just as badly as doctors. There is significant pressure on engineers to design products for the lowest possible cost, even in applications where lives are at stake.

[packbat.livejournal.com]

I think "theoretical" would actually be correct, there – just from my own experience, a lot of people tend to keep things running much longer than that. ;)

Well, I don't think it was as much a matter of 'Morris Cooke et al. gave up' as 'Morris Cooke et al. failed to convince universities to teach the next generation that their first responsibility was to society, not to their employers'. I mean, we still get about one lecture's worth of 'engineering ethics' in the senior design class, and we still read ASME's code of ethics, but that's barely more time than we spent on, say, scheduling.

I think engineers need their own version of the Hippocratic Oath just as badly as doctors. There is significant pressure on engineers to design products for the lowest possible cost, even in applications where lives are at stake.

Well, I don't think it's as bad as all that. If an engineer screws up on a job that gets someone killed, even if the failure couldn't be predicted by theory ... well, ex-engineer is only the beginning of the possible consequences. When there's a risk of direct harm, engineers are extremely careful for a good reason.

That doesn't stop them from shipping the labor out to Chinese sweatshops to save a nickel a part, of course. (Hey, that nickel adds up to $50,000 if you sell just a million of 'em!)

[roaminrob.livejournal.com]

No, no, haven't you heard of "thpractical" applications? <G>

I think that responsibility to society versus responsibility to one's supervisor is something that could become a part of the engineering culture; I wouldn't expect a university to adopt such an attitude officially because in the short run it would serve to make their students somewhat less employable than otherwise. So, the solution is to take those attitudes outside of the university; it's something that I would expect to see adopted in various journals, for example.

I'm somewhat more cynical about the effects of lowest-cost design and production. From what I've seen, design failures rarely publicly get traced back to the engineer, usually because it comes out that it was a management-level decision to do things that way, so the executive goes down instead. And, I'm not at all arguing that it shouldn't be that way, I'm just saying that engineers really aren't subject to the most severe professional consequences.

There may be relatively few circumstances where a product is actually life-critical, but then we should begin to look at the costs and benefits to society at large in robust design versus cheap design. Hell, look at the costs of replacement and disposal alone for everything in the consumer market.

Addressing the design of a thing may not keep it from being ruined during a production, but at least it's a start.

[packbat.livejournal.com]

Come to think of it, we did discuss "Design for Environment" for about ninety seconds spread out between two or three lectures... ;)

But you're right – the universities have no incentives to try to instill these ideals. Although ... huh. Thinking about it, there's a lot of stuff offering implicit support for a societal ethic all through education – Silent Spring on the reading lists, for example – but it doesn't exactly get emphasized. It makes me think of The Gate to Women's Country.

Dang. Every time I think about it, the methods of education start looking cleverer and cleverer. :)