Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why is it hard to buy things that work well?

danluu.com

Post reply on HN
Print view View original

451–460 of 542 posts

Re: Why is it hard to buy things that work well?

#451

Post by derefr »

pclmulqdq wrote: post #407
ownagefool wrote: post #318

Earlier quoted context omitted.

What if the saw breaks, what if the person who knows how to use the saw leaves? There are good answers to these problems, but I find folks who find comfort in outsourcing assume it's easier for the downstream org solve them. Personal experience suggests it's often a false sense of security, and you often have to have your own experts anyway. Saw the same in software. We won't get budget to patch so let's outsource. T…

A lot of outsourcing decisions seem to be made on the premise of "what if it breaks?" It's a terrible fallacy. Your wafer saw (or server) that is utilized 1% of the time is going to need a lot less maintenance than a shared wafer saw that is used 16 hours a day. You will actually experience a lot less downtime by investing in your own wafer saw (or your own servers) rather than outsourcing (to a foundry or a cloud pr…

> You will actually experience a lot less downtime by investing in your own wafer saw (or your own servers) rather than outsourcing (to a foundry or a cloud provider).

Analogy doesn't work; "in-sourcing" of previously-IaaS resources as a cost-cutting measure, is used almost exclusively for "base load" (vs. elastic load, which is the comparative advantage of IaaSes); and so your in-house servers are usually running pinned at 100%. So the hardware components will wear out just as fast, if not faster, than the IaaS's servers will; and when they do, your ops team won't have as many spare parts on hand as the IaaS does; nor the ability to live-migrate the enforced VM to another exactly-equivalent substrate host in the fleet (which was already warm) to avoid downtime altogether.

Also, apart from that, there are economies of scale in reliability. Your in-house backups are probably just a ZFS pool or a RAID5+1 array or something. They're certainly not hosted within a 17+-copies Dynamo-ish system, the way most IaaS object storage (incl. archival storage) is, because the costs of doing that in-house are ridiculous.

Re: Why is it hard to buy things that work well?

#452

Post by JohnBooty »

jader201 wrote: post #215
Strilanc wrote: post #88

It's pretty depressing how many of the comments are purely answering the title and not discussing the content of the article. In particular, the comments seem mostly focused on consumer product build quality whereas the article is more focused on e.g. businesses trying to outsource something (like same day delivery) and getting a worse result even when they pay more than doing it internally. And the cultural obstacle…

> It's pretty depressing how many of the comments are purely answering the title and not discussing the content of the article. I’ve said this before on other threads, and it seems to resonate with others, so I’ll say it here too: I personally don’t care near as much about a single person’s opinion of a particular topic, as much as I care about the discussion of many people around a particular topic. And by “topic”,…

    I often wish HN would just allow titles to get 
    posted and HN folks just have a discussion around 
    that. (Similar to Ask HN, but not necessarily in the 
    form of a question.)
Sounds like a real recipe for hivemind to me. I mean, any community struggles with hivemind a bit. But, at least with discussions centered around articles that exist outside HN, we get a little fresh air.

    especially if I don’t even know whether 
    this person is an expert on said topic.
He's a pretty respected voice, or at least popular on HN.

I find it very valuable to look into the authors of linked articles, see which authors pop up on HN frequently, etc.

Gives you an idea of who the leading voices in the industry are, helps you to know which way the winds are blowing.

Especially for somebody with a slower reading speed this may be crucial; helps to understand which article may actually be worth the time investment.

Re: Why is it hard to buy things that work well?

#453

Post by ebruchez »

WrtCdEvrydy wrote: post #390
bgentry wrote: post #382

Earlier quoted context omitted.

After purchasing the top two Wirecutter picks for electric kettles (Cuisinart and some gooseneck kettle) both died within a year. The gooseneck one was rusted on arrival, clearly awful build quality. I decided to try paying much more for a Fellow Stagg EKG, and it was a great decision. It’s lasted over 3 years and has been an absolute joy to use compared to the prior mass market garbage. I often wish for a Wirecutter…

Honestly, I had spent 20 years in the US and we consistently bought the cheapest appliances ever. When I bought my house I finally said "screw it, let's see what decent appliances look like". Japanese rice cooker set me back $95 and I thought I would never hear the end of it, and after 4 years, it had already paid itself off (we were doing $14 rice cookers every 6 months). Air fryer was $70 but the previous $40 only…

> one of our friends suggested bosch because we wanted a quiet appliance

We've been very happy with our Bosch. Don't ever buy a cheap dishwasher.

Re: Why is it hard to buy things that work well?

#454

Post by Mezzie »

kingcharles wrote: post #231
ncmncm wrote: post #216

When I interview, I like to look around and see how many short people work there. A place that picks up the short people other companies don't must be recognizing their value. I would like to favor places with more women in tech jobs, but generally just don't find any at all, anywhere. Everywhere I have worked they were always trying to hire more women but never got any applicants, or couldn't hire the ones they foun…

Huh, interesting. I would second MtF transgender. I've worked at places that will interview so they can appear what you might now call "woke", but then not hire based on their dislike, not skills or personality. And there are so many MtFs in tech. I'd love to see statistics, but it feels skewed.

It is skewed.

This is one of those things I Can't Say in most places, but (unless you transitioned in elementary school) the MtF experience in tech and the cis female experience are totally different, and they (the MtFs) know it, because so much of our work is done online where you can be whatever gender/sex you want. I know how people reacted to me when they thought I was male vs. female.

Age plays into it, too. Everything was fine before I went through puberty, but a lot of young techie men have a really hard time with their sexuality and take that out on their female colleagues in various ways. (I've also seen some gross reversals in female-dominated professions re: issues with men, so...) Now that I'm in my 30s and the men in that insecure/figuring things out age range aren't interested, it's better again.

Re: Why is it hard to buy things that work well?

#455

Post by mkr-hn »

dotancohen wrote: post #302
JoachimSchipper wrote: post #275

Earlier quoted context omitted.

Ehm, doesn’t Tesla have famously bad quality control in several areas, e.g. fit of body panels? EVs, as a category, need less maintenance - but Tesla’s are not an example of a trouble-free product! (They are legitimately cool though!)

Sure, the early Model 3's had body panel fit problems. I actually bought one last week, and the fit is amazing, both interior and exterior. I don't know why that point keeps coming up. For what it's worth, I used to be a tech at a Ford dealer, so I know what bad fitment looks like!

Initial news always spreads further and faster than updates. This is normal and should have been expected by the company. They took the risk of rushing and get to suffer the consequences.

Re: Why is it hard to buy things that work well?

#456

Post by blablabla123 »

yrral wrote: post #16

I think it's that you don't notice things that work well. House foundations, light switches, filesystems, silicon manufacturing, water delivery, grocery store logistics, etc etc etc. These are all things that most people never notice because they just work. It doesn't even occur to people day-to-day that these things can fail.

Just to give one example, CI pipelines seem to fail all the time. For closed source and open source project. Just like this, it worked in the last commit and in the new commit it fails despite the test suite passing. The ultimate reason is routine tasks pulling in a ton of complexity of which only a tiny fraction is being used.

At workplaces this creates a lot of absurd situations that eat up insane amounts of productivity.

Or another example, it's pretty common that water pipes don't work as expected. (Congestion, low pressure, undesired backflow, tricky to get water at body temperature...) Nobody really complains, everybody lives with it and learns to completely ignore it. I'm not saying these problems occur everywhere 100% of the time but often enough to show there's something structurally not working

Re: Why is it hard to buy things that work well?

#457

Post by josh2600 »

ownagefool wrote: post #318
kccqzy wrote: post #242

> An example from another industry: when I worked at a small chip startup, we had in-house capability to do end-to-end chip processing (with the exception of having its own fabs), which is unusual for a small chip startup. When the first wafer of a new design came off of a fab, we'd have the wafer flown to us on a commercial flight, at which point someone would use a wafer saw to cut the wafer into individual chips s…

What if the saw breaks, what if the person who knows how to use the saw leaves? There are good answers to these problems, but I find folks who find comfort in outsourcing assume it's easier for the downstream org solve them. Personal experience suggests it's often a false sense of security, and you often have to have your own experts anyway. Saw the same in software. We won't get budget to patch so let's outsource. T…

As someone employing a lot of engineers right now, one of the problems is that engineering will just keep requesting resources ad infinitum if there’s no budgetary limit. The constraints of the business are actually always financial, and without proper financial modeling you’ll cease to have a business.

This is true in process scaling no matter the process.

Re: Why is it hard to buy things that work well?

#458

Post by marcusverus »

cosmojg wrote: post #438
impalallama wrote: post #419

> [It] involve companies being efficient and/or products being basically as good as possible because, if it were possible for them to be better, someone would've outcompeted them and done it already. Must be a generational thing but I don't think anyone i know (ie under 30) genuinely believes this anymore.

I think the more correct statement is that products are as good as profitably possible and no better. If the margins on a better product aren't large enough to justify spinning up a new business just to sell it, the existing product will remain as it is indefinitely. To top it all off, consumer expectations can be surprisingly low. It's often the case that no one even wants or knows to want a better product, let alon…

I think you're right--particularly w/ regards to consumer expectations being low. I'm a part of this problem as well--when making purchases, price is almost always the deciding factor.

It's also worth noting that "as good as possible" and "as good as profitably possible" aren't just two flavors of the same principle w/ regards to quality--they're directly at odds. I'll prove this (with scientific finality) via an anecdote about toys.

My kiddo was gifted a little Fisher-Price school bus that was made in 2005.[0] It's pretty neat! As you roll the bus forward, some internal mechanism causes the driver to turn left and right (making it look like they're turning the steering wheel back and forth), and the passenger seats to move up and down (a dope allusion to the classic kid's song, 'The Wheels on the Bus', in which, as the wheels of the bus go 'round and 'round, the people on the bus go up and down). It's clear that this product was lovingly designed, with playful curves and playful details, like the face of the bus. The wheels not only have hub caps, but the shape of the tires even mimic a pneumatic bulge! The space inside of the bus is even big enough to accommodate human hands!

More recently, she was gifted a newer version of the same toy.[1] It's smaller in every dimension, and flimsier. The slick design is gone, replaced by simple, boxy one. The little things that I loved about the old bus are gone--the wheels are now a single piece of plain molded plastic. While the wheels of the bus still go round and round, the people on the bus, alas, do not go up and down.

It's fine, as far as toys go--the kid probably doesn't notice a difference. But it's clear that the guiding principle in the evolution of this product is not only not quality, but something that is directly in conflict with quality.

[0] https://www.ebay.com/itm/304338207633 [1] https://www.walmart.com/ip/Little-People-Sit-With-Me-School-...

Re: Why is it hard to buy things that work well?

#459

Post by WrtCdEvrydy »

JohnBooty wrote: post #443
WrtCdEvrydy wrote: post #390

Earlier quoted context omitted.

Honestly, I had spent 20 years in the US and we consistently bought the cheapest appliances ever. When I bought my house I finally said "screw it, let's see what decent appliances look like". Japanese rice cooker set me back $95 and I thought I would never hear the end of it, and after 4 years, it had already paid itself off (we were doing $14 rice cookers every 6 months). Air fryer was $70 but the previous $40 only…

There's also just a... man, I don't know how to describe it. Kind of a mental benefit to using slightly nicer things. When I was young, almost everything I owned was the cheapest possible version of that thing. Everything just kind of sucked, brutally cost-optimized to the point of being somewhat nasty to use and barely functional. I was still very fortunate: I had food to eat, clothes, etc. A lot of kids in the worl…

I'm not a super stingy guy and we're a Cuban family so rice is an every day dish.

It's not super fancy or anything but it fills that rice craving and is a multi-use device.

Re: Why is it hard to buy things that work well?

#460

Post by amanaplanacanal »

kingcharles wrote: post #231
ncmncm wrote: post #216

When I interview, I like to look around and see how many short people work there. A place that picks up the short people other companies don't must be recognizing their value. I would like to favor places with more women in tech jobs, but generally just don't find any at all, anywhere. Everywhere I have worked they were always trying to hire more women but never got any applicants, or couldn't hire the ones they foun…

Huh, interesting. I would second MtF transgender. I've worked at places that will interview so they can appear what you might now call "woke", but then not hire based on their dislike, not skills or personality. And there are so many MtFs in tech. I'd love to see statistics, but it feels skewed.

I don't have statistics at hand, but folks on the autism spectrum are more likely to be queer or otherwise gender non-conforming. So if there are more autistic people in tech this certainly makes sense.
Post reply on HN