Hacker News new | past | comments | ask | show | jobs | submit login

I had a similar experience in a mechanical engineering class, our professor was discussing MTBF (mean time between failure) and how calculating this more accurately allowed companies to better engineer their product for obsolescence. For a company, a product that is too reliable is unprofitable in the long term and so they adjust by making sure that the product works beyond the terms of the warranty but not much more.

My professor was himself horrified by that (but he was a teacher because he believed in not working for corporations so that was not very surprising)

I do know one company that has a reputation for not doing this. Miele in Europe but then their appliances are double to triple the price of typical companies.

EDIT: corrected Mean Time Between Failure instead of Mean Time Before Failure. Thanks slim




It's obvious that companies do this sort of thing. The main thing that prevented it in the past was that parts and labour were actually a significant cost, so things would be more expensive (in real terms) than today. When a TV or washing machine could be as much as 100% of a month's salary for a family, people couldn't afford a new one every 4 years. Companies realised this, so they made stuff that lasted longer.

But today, when everything is made in Asia by people who earn orders of magnitude less than the people buying the products, that TV or washing machine is ~10% of a months salary, and you end up with "fixing it is more expensive than buying a new", because the labor costs of fixing stuff (locally) is orders of magnitudes larger than labor costs for manufacture (in Asia).

Corollary: if global salaries become more equal in the future, we will get quality long-lasting stuff again.


I'm afraid your corollary won't apply; the lowest wages will just move elsewhere. There already was a company in China that moved production to the US because of wages there (plus transport costs, etc). I could see countries in Africa become the next manufacturing powerhouse(s), direct access to major seaways to both the US, Europe and Asia, etc. I don't know if they have the natural resources though, or the political stability for that matter.


This exploitation of cheap foreign labor is a big reason why rural America is doing so poorly. Corporations used to go to rural America for cheap labor, but because of our environmental and labor laws as well as our higher standard of living, it's not viable. I honestly don't see how this trend can continue without destabilizing our country.


If you look at a typical smartphone, the retail/supply chain labor costs in the US are at least twice the production and logistical costs in the country of origin.

Apple could easily increase the wages of all their Chinese staff to $10/hour and they would still be making 30-40% margins on every phone.

I suspect the real drivers environmental, China is willing to pour megatons of waste into their wilderness in a way that we would never consider here in the US.


It makes me wonder if the people at the top could handle expecting less, could trends reverse themselves and more people benefit, or if it's beyond them with too many other factors to consider. Does it really all come down to "We have to grow every quarter or die"?


In any competitive market if one manufacturer decides to stray far from optimal behavior for such reasons, it won't even last a decade before their competitors overtake them (no matter if simply taking their marketshare, or with literal buyouts, or by taking over their assets in a bankruptcy/restructuring sale) and reverse those practices.

If the industry margin is e.g. 10%, then you might assume more beneficial practices that cost up to 10% if the company is privately held. Not more, and not anything significant if you're public - since if you do so, then it would be trivial for anyone with big resources (e.g. an investment bank or hedge fund) to buy your stock to gain a significant voting percentage, replace management with literally anyone else, and sell stock that immediately becomes so much more valuable.


I don't know why you're being downvoted. The board of directors exists to maximize profits, not create a stable society. As long as greater profits can be achieved by gutting the lower and middle classes, it will happen. To create change, the government must step in, which is almost impossible since the major corporations of this country have lobbied so much and spread so much propaganda.


It already has; see the 2016 election. Factories will come back, but they won't require labor anymore, so in the long term something fundamental needs to change.


> if global salaries become more equal in the future,

> the lowest wages will just move elsewhere.

Uhh.


The richest capitalists are planning a move to Mars, and presumably instituting (effective) slave labor or indentured servitude there.


Yeah but if you finish your contract on mars you get the anti-aging treatment for free.


Ever notice how the high cost of wages in the USA combined with RORO car carriers (https://en.wikipedia.org/wiki/Roll-on/roll-off) to lose 85% of the UAW jobs in the USA? They allowed very low cost sea transport to the USA. The UAW constant striking for wages, perks and pensions (now close to $75 per hour)and other labor restrictive rules threw away 85% of their members jobs. This was an unintended consequence of greed and management parallel avarice who just upped the cost of cars by passing it to us. Now most people can own only used cars. The net effect also increased all other US wages - all of which now come home in spades. Furniture and matresses, locally built items can still be US made.

Remember - a bad job has low wages. I high wage job can also be bad - to the economy.


Yep, there is absolutely no way whatsoever policy could have prevented that from happening. It's all the fault of workers and there is no choice but for all of us to reduce our standard of living to compete with the world's most destitute places.


I think automation is going to do the very opposite.


Ironically, I work for a manufacturer who tries to support their product for at least 20-30 years, but obsolescence is an incredible challenge due to use of a lot of commercial of the shelf electronics which themselves go obsolete at a ridiculous pace. What I'm saying is, even if you try to engineer a product to last, if the supply chain supporting you isn't to doing the same, the problem often compounds.

In my case, obsolescence is a problem to be addressed to attempt to support a long product lifecycle. In most companies, planned obsolescence is a tool to advance tech, decrease support and manufacturing costs over the long run, and increase profit.


> What I'm saying is, even if you try to engineer a product to last, if the supply chain supporting you isn't to doing the same, the problem often compounds.

This makes sense to me. If you want to engineer and support a long-lasting product, you have to own more of the supply chain, make more of your own component parts, or limit yourself to components that are "standard parts" and likely to remain available.


What if your company decided to sacrifice features for maintainability and kept the electronics in the product to a minimum? There might be a strong market for something that does only 1-2 things but can be maintained indefinitely with simple materials and tools.


Yes, actually this is a strategy for sure.

Historically, the company I work for has been a market leader in the niche they serve, and as such, their products were initially incredibly complex as they were first to market and invented a lot of the technology. Over time, those products have been simplified and cut down, with only the most important features kept (and a few "whiz-bang" features).

Interesting issue with this, though - while I completely agree with your suggestion, some of the pushback for taking that direction involves the perception that the inherent complexity of the products is a strategy in and of itself. If the product is simplified and streamlined then it is easier to copy and harder to protect, according to some who think that way.

Personally, the competition is going to be there if it really is that much simpler to make the same product with less components/features and easier to maintain. Patents, etc. protect a business' position to an extent but, ultimately, competition still often points the way to more efficiency that benefits everyone.

I think much like companies use "planned obsolescence" to try to control their market position, they also use "intentional complexity" to try to do so as well. Both strategies have their pros and cons.

I think really strong companies maintain market dominance with simply constructed, simply understood offerings, done well, at a fair market price, with the related mind-share branding that goes with it. But that definitely sounds a lot easier than it is in reality.


Then you can't offer the features and efficiency that the competitors' models do. Replacing the simple mechanisms of yesteryear with microcontrollers and sensors is how our appliances have become so much more efficient, just like humans have become the dominant species on this planet by having the most complex brain, instead of trying to get by with the mental abilities of an amoeba.


Miele did have parts for a 30 years old vacuum cleaner a couple of years ago when I needed them.


>I do know one company that has a reputation for not doing this. Miele in Europe but then their appliances are double to triple the price of typical companies.

I think Miele serves as an excellent example for why the market has moved away from designing products with a 30+ year expected service life. If the average modern appliance lasts for 5-10 years as this article claims (and has proven true time and time again in my experience) but the reliable brand costs 2-3x more, you would have to be absolutely certain that the "reliable" product will actually last at least 20-30 years if you are to ever make up for the initial cost premium.

The problem is that even the most well built appliances will have random part failures and the chance of a failure happening (25-50% IME) over a span of 30 years is high enough that it has to be treated as an inevitability. Even if the parts are available that far in the future, the cost of having a technician diagnose the failure ensures the repair will cost at least $150 in labor alone and the average cost for any appliance repair with parts included is somewhere between $300 and $400 in my experience.

My point is that the "cheap and unreliable" appliance is almost always a financially sound choice even though it leads to a graveyard of dead appliances over time. The real problem is that graveyard of broken appliances we leave behind by making the financially sound decision to buy an appliance we know will only last ~10 years at best.


There's another factor to this - the US homeownership has gone down over the years, and will likely go lower.

More renters = even less incentive for reliable products, as the cheapest install that works is sufficient to rent.


Not necessarily. The problem here is that many appliances are the property of the landlord, and it's their responsibility to repair or replace them. Landlords who buy cheap junk appliances that need quick replacement are going to see their profits wiped out.

I really don't see how home ownership is a factor here at all. This is just a symptom of modern society where people don't expect things to last and don't penalize mfgrs who sell them short-lived junk, and instead look for generally the cheapest stuff.


> Landlords who buy cheap junk appliances that need quick replacement are going to see their profits wiped out.

Have you rented recently? I can tell you from my own experience (US/Ca and Ut) as well as family/friends who don't own - that often landlords (or property management companies hired to manage the rental) much prefer "new,shiny,cheap".


> For a company, a product that is too reliable is unprofitable in the long term and so they adjust by making sure that the product works beyond the terms of the warranty but not much more.

My 2002 Volvo V70, which I bought second hand a couple of years ago, is so reliable that I consider buying a Volvo again.


>(W)hich I bought second hand a couple of years ago

Would you consider buying a NEW Volvo? If not, your purchase exerts little to no pressure on the only market which Volvo is concerned with, namely the market for new cars.

Yes, a reputation for reliability increases the demand for used Volvos thereby increasing the resale value which positively factors into the decision to purchase a new Volvo. However, the increase in "value" to the buyer of the new Volvo is so marginal that it is overwhelmed by any number of other factors involved in the final judgement of value in the decision to purchase a new car. A Volvo V70 of similar vintage to yours can be had for anywhere between $1,600 and $2,500 depending on condition according to my local Craigslist. If their reliability is so well regarded that a Volvo V70 is hypothetically worth 50% more than an equivalent Subaru wagon, resale premium afforded to the original buyer is just $1,250. That premium is so small that a manufacturer's sale incentive can wipe it out entirely, nevermind the fact that many of Volvo's competitors are less expensive by a far wider margin.

Basically, my point is that long term reliability doesn't really matter to manufacturer's beyond a certain point. As long as the vehicle meets the expectations of the original buyer and remains useful enough to keep the resale value out of the gutter (which doesn't appear to be the case with Volvos, at least in my area), the manufacturer has no incentive to further improve the reliability of their products.


Resale value definitely affects new car purchasing.

For one, lease prices are partly determined by the expected resale value, and a lot of new car drivers get them by leasing them


> Would you consider buying a NEW Volvo?

Yes - sorry I forgot to add that bit :)


Plus 1 - my 2002 v70 had quarter a million on the clock and was very happy. 2.4 diesel was built to last. Not sure how the newer greener engines would be, all diesels now need a DPF that needs changing regularly, or a similar device to keep emissions down. I bought another (newers and sadly non Volvo) diesel, and I'm regretting it on a longevity POV - it's a great car now, but without pouring money into I'm not sure how long it will last - if I get 3 years I'll be happy (often less that 7k miles per year, so I don't tax my cars - just irregular long motorway trips with the rest of the time the car barely used).


Any diesel engine that has SCR and isn't miserly on the urea should be good to go. It's the ones that cheap out and rely excessively on EGR that you should watch out for.


Mine is a citroen so uses a urea system - hopefully good then... It's a lovely family car and easy 53mpg - though as a cyclist and parent I'm now feeling guilty about particulates, but as I say I drive irregularly and mainly motorway miles.


The urea system allows the engine to run lean and hot without excess NOx emissions, which reduces particulate emissions. The particulate filter should take care of the rest.


Great to hear! Pleased that they haven't been drawn into the emissions scandal (so far?) - really hopeful that my next car can be a electric or hybrid - but just have to see how the used car market pans out.


> their appliances are double to triple the price of typical companies

This seems like the most important detail by far.


Mean Time Between Failures


That's an interesting contrast. I mean, you're correct, but why is the acronym "between" instead of "before"?

As the article describes, when replacing a part costs a comparable amount to the price of a brand-new appliance, people buy new appliances. With this mindset, the correct term is no longer "between" failures - when you might replace a part and average the described time until the next failure - but "before" failure - at which point you replace the device and buy a new one.


That's a good point. I'm not sure why I misremembered the term but in this case it might actually be more accurate.


The other one is Mean Time To Failure.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: