TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Why is it so hard to buy things that work well? (2022)

693 点作者 janandonly5 个月前

77 条评论

mattgreenrocks5 个月前
This post is so interesting to me, esp. the build-vs-buy spectrum.<p>As Dan notes, a lot of software is just...not very good. It either isn&#x27;t upfront with flaws (as in the case of the Postgres -&gt; Snowflake tool), has too much scope, or is abstracted poorly. Finding things to buy&#x2F;use (as in the case of open source) can often eat a lot more time than you anticipate.<p>I&#x27;ve been dipping my toes into the JS ecosystem, and I keep bumping into the fact that using mentally cheap signals of quality (such as stars or DL counts) almost never indicates the quality of the thing itself. Winners seem to be randomly chosen, almost! The only way to assess is to read the code and try integrating it in.<p>I&#x27;d go farther to argue that the larger an ecosystem&#x2F;market is, the more untrustworthy it behaves as a whole, simply due to the size, and the types of people attracted to it who want to get influence&#x2F;money. See also: appliances that everyone needs.
评论 #42432498 未加载
评论 #42432141 未加载
评论 #42432096 未加载
评论 #42435093 未加载
评论 #42432412 未加载
评论 #42432183 未加载
评论 #42431876 未加载
评论 #42431953 未加载
评论 #42434485 未加载
评论 #42438603 未加载
评论 #42432014 未加载
评论 #42434675 未加载
评论 #42434146 未加载
georgewfraser5 个月前
I have some insight into this because this claim is about my company Fivetran:<p>“…relies on the data source being able to seek backwards on its changelog. But Postgres throws changelogs away once they&#x27;re consumed, so the Postgres data source can&#x27;t support this operation”<p>Dan’s understanding is incorrect, Postgres logical replication allows each consumer to maintain a bookmark in the WAL, and it will retain the WAL until you acknowledge receipt of a portion and advance the bookmark. Evidently, he tried our product briefly, had an issue or <i>thought</i> he had an issue, investigated the issue briefly and came to the conclusion that he understood the technology better than people who have spent years working on it.<p>Don’t get me wrong, it is absolutely possible for the experts to be wrong and one smart guy to be right. But at least part of what’s going on in this post is an arrogant guy who thinks he knows better than everyone, coming to snap conclusions that other people’s work is broken.
评论 #42435279 未加载
评论 #42434733 未加载
评论 #42437720 未加载
评论 #42435811 未加载
评论 #42435278 未加载
评论 #42434708 未加载
评论 #42436442 未加载
评论 #42440749 未加载
评论 #42437189 未加载
评论 #42434627 未加载
评论 #42437317 未加载
评论 #42436964 未加载
评论 #42435744 未加载
评论 #42437930 未加载
评论 #42436443 未加载
asveikau5 个月前
&gt; markets enforce efficiency, so it&#x27;s not possible that a company can have some major inefficiency and survive<p>This just seems totally false on its face. If you&#x27;ve worked at the big guys you know they aren&#x27;t magically smarter, they do very inefficient things frequently.<p>It&#x27;s so intuitively false that I&#x27;d have to wonder about someone who thinks it&#x27;s true.
评论 #42432071 未加载
评论 #42433033 未加载
评论 #42433133 未加载
评论 #42432511 未加载
评论 #42436274 未加载
评论 #42433091 未加载
评论 #42434614 未加载
评论 #42435557 未加载
评论 #42435613 未加载
评论 #42436967 未加载
评论 #42433884 未加载
评论 #42437196 未加载
评论 #42432527 未加载
评论 #42439097 未加载
评论 #42432112 未加载
评论 #42435421 未加载
评论 #42434482 未加载
评论 #42432250 未加载
评论 #42434453 未加载
Havoc5 个月前
For a lot of fields the answer is much simpler. The buyer is simply not equipped to evaluate it at all.<p>eg I bet north of 50% of people judge their tax accountants by the size of refund not technical and legal accuracy. But they’re still liable so accuracy is an important measure of „good“<p>Same with say dentist. If he says he needs to do procedure x what am I going to do except ask some layman questions. Or doctor. Even trades often seem simple but have significant accumulated practical learnings that are not obvious to laymen.<p>It’s tempting to boil everything down to an optimization question &amp; just finding the right metrics, especially for those STEM minded but often that’s not how reality works
评论 #42432925 未加载
评论 #42432203 未加载
评论 #42433119 未加载
评论 #42442199 未加载
评论 #42438743 未加载
评论 #42488406 未加载
评论 #42432131 未加载
评论 #42434180 未加载
评论 #42435231 未加载
criddell5 个月前
I&#x27;ve been thinking about good design lately. Things have to work well, but life is so much better when they are also beautiful. I think Don Norman&#x27;s essay <i>Emotion and Design</i> started me down this path.<p>The Conan OBrien and Jordan Schlansky podcast talked about this in the context of nose hair trimmers. It was very funny, but it really resonated with me too. Schlansky starts with:<p>&gt; I believe that we can live minimally. But the products that I do buy, I want them to be of a very high quality. I want them to have something special about them, and then I have to buy fewer products going forward because they last longer.<p>A little later he says:<p>&gt; We define ourselves by the objects we interact with every day. I surround myself with beauty, with high levels of aesthetic pleasure, and it&#x27;s not only putting on beautiful clothes. It&#x27;s also using a beautiful nose hair trimmer.<p>I&#x27;m going to get a trimmer, so I want a thoughtfully designed and well made version of that. This has been my mission around my home since the start of the COVID pandemic. Upgrade all the little things around my home that annoy me or that would make my day a little bit better if they were upgraded.
评论 #42435423 未加载
评论 #42432734 未加载
评论 #42434276 未加载
评论 #42432966 未加载
评论 #42454895 未加载
评论 #42434544 未加载
评论 #42434229 未加载
cs7025 个月前
This is a thought-provoking essay by Dan Luu, whose essays I always find thought-provoking.<p>I&#x27;m surprised Dan didn&#x27;t make the connection that the webs of mistrust between fiefdoms that form inside organizations as they grow are... <i>Nash equilibria</i>.[a]<p>Organizational webs of mistrust are nothing more than complicated versions of the Prisoner&#x27;s Dilemma.[b]<p>Unless you have a CEO actively enforcing trust and collaboration, different fiefdoms naturally evolve behaviors that ensure they can survive and thrive in the face of possible betrayal by any other untrustworthy fiefdoms in the organization. We see similar behavior in natural ecosystems, which tend to evolve toward suboptimal equilibria that is robust to betrayal between groups, instead of global optima that requires perpetual honesty between them.[c] In many settings, robustness against betrayal is an evolutionary advantage.<p>---<p>[a] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Nash_equilibrium" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Nash_equilibrium</a><p>[b] <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Prisoner&#x27;s_dilemma" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Prisoner&#x27;s_dilemma</a>
评论 #42433202 未加载
jefftk5 个月前
2022, per <a href="https:&#x2F;&#x2F;x.com&#x2F;danluu&#x2F;status&#x2F;1503512394126938120" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;danluu&#x2F;status&#x2F;1503512394126938120</a><p>I wish Dan would put dates on his posts!
评论 #42432309 未加载
mnky9800n5 个月前
For some reason this makes me think about cooking stock. Like the liquid you use to make gravy, soups, sauces, etc. most people I know will use either a box or jar of stock or a bouillon cube. There’s nothing wrong with this. It will produce a satisfactory meal. However you can easily make stock if you cook for yourself. Simply take all the parts of vegetables and meats you would throw away (think bones and onion paper) and cover it with water in a pot and add salt and pepper if you want and boil for an hour or so. You can do it while you cook other stuff. This stock can be frozen for later. But also, it’s obviously better than the stuff from a jar or a cube. Maybe it’s because you put some effort into it, maybe it’s because there’s a complexity to all the flavors that’s removed by industrialization. But it’s a stock that works very well. And I think it’s simply it requires time. Time we often think we don’t have. Perhaps this is all an unrelated thought but I do recommend that if you enjoy cooking at home making your own stock.
chias5 个月前
&gt; if it&#x27;s so obvious, someone at the company would have fixed the issue or another company would&#x27;ve come along and won based on being more efficient or better<p>I think there&#x27;s two kind of people at any tech company. There&#x27;s the kind of person who has removed an unused piece of cloud infrastructure that nobody remembered existed but whose spend was north of $50k &#x2F; month, and there&#x27;s the kind of person who does not believe such a thing could possibly happen.
imgabe5 个月前
I think there is a fundamental misunderstanding about both evolution and markets, which is that either one &quot;optimizes&quot; something. Neither needs to create a perfect outcome. An organism doesn&#x27;t have to be the best, most efficient possible at getting food, it only has to be good enough to survive long enough to reproduce. A cheetah does not need to be the fastest cheetah on the plains to survive, it only has to be faster than the slowest antelope. A company does not need to make a perfect product, it only has to be just good enough to turn a profit, any profit, in order to survive.<p>Once you realize this everything starts to make sense. Why can accountant have errors? Because they aren&#x27;t bad enough to lose customers. Why can a product be made shoddily? Because it isn&#x27;t quite shoddy enough that people stop buying it. Indeed, the most optimal thing for a company, profit-wise, is to be just as shitty as possible without going out of business. Any additional resources spent on increasing quality that is not demanded by the market are, in a sense, wasted.<p>This hurts from an engineering perspective because we want to make things that are good, but too often customers don&#x27;t actually care about how good it is, just that it is cheap and just barely good enough to be worth purchasing.
not_your_vase5 个月前
I am so interested in this submission (based on the title), but that&#x27;s such a big wall of text, even the LOTR trilogy looks more manageable<p>If you have read it, do you regret the time you spent on it? Just trying to get some &quot;goodreads&quot; reviews, with or without spoilers, before committing to it...
评论 #42431591 未加载
评论 #42432306 未加载
评论 #42431340 未加载
评论 #42431638 未加载
评论 #42432423 未加载
评论 #42431841 未加载
评论 #42431569 未加载
评论 #42434458 未加载
评论 #42434943 未加载
评论 #42433125 未加载
评论 #42431227 未加载
评论 #42431857 未加载
评论 #42432028 未加载
评论 #42432276 未加载
评论 #42431541 未加载
评论 #42432319 未加载
评论 #42439107 未加载
评论 #42430704 未加载
评论 #42431675 未加载
评论 #42431720 未加载
评论 #42433161 未加载
评论 #42431239 未加载
评论 #42431526 未加载
bombcar5 个月前
One thing not mentioned, but I&#x27;ve seen time and time again, is if you <i>buy</i> something, you buy what you think you need&#x2F;want. You make a decision and live with it.<p>But if you <i>build</i> it, you can learn that what you wanted wasn&#x27;t what you thought you needed, and the build can adapt and change to meet that actual need. A bought item&#x2F;tool will do what it is designed to do, and can sometimes be shoehorned into what you needed, but often what you needed wasn&#x27;t what you thought you needed.
评论 #42435602 未加载
markus_zhang5 个月前
I have reached the age (40+) and I need much fewer than I thought I&#x27;d need. I think the best way to reduce the number of lower quality buys is just to buy less, way less, while keep a large amount of $$ float in the market. Yes the CAD or USD also de-value in time but IMO they devalue a lot slower than most of the products on the market.<p>And when I buy I only buy second hand stuffs with a deep discount. I&#x27;m glad that a used workstation lives for 2+ years (in fact, all of my second handed workstations live for 2+ years for just $500 and less).
评论 #42433788 未加载
jpm_sd5 个月前
I thought this was the most interesting point in the article:<p>&gt;&gt;&gt;<p>People often think that having a high degree of internal distrust is inevitable as a company scales, but people I&#x27;ve talked to who were in upper management or fairly close to the top of Intel and Google said that the companies had an extended time period where leadership enforced trustworthiness and that stamping out dishonesty and &quot;bad politics&quot; was a major reason the company was so successful, under Andy Grove and Eric Schmidt, respectively. When the person at the top changed and a new person who didn&#x27;t enforce honesty came in, the standard cultural norms that you see at the upper levels of most bi
mrngm5 个月前
Earlier notable thread in 2022, 518 comments: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=30679935">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=30679935</a><p>The title could use a [2022], as that seems the year it was originally posted.
backtoyoujim5 个月前
People with large swathes of equity in a company want people with near zero equity to act like they will be getting C-level rich off of their wage&#x2F;salary if the company&#x27;s business plan hits.<p>Founders are not desperate for talent. They are desperate for labor will shoulder vastly unequal labor for vastly underpaid compensation.<p>Capital always pushes for slaves. Labor can only push for using their smart phone during their bathroom break and not giving a shit about the final product as its success does not offer labor truly valued compensation.<p>Minimum wage is a thing but maximum wage is not in a civilization run by capital.
secabeen5 个月前
&gt; The good accountants are typically somewhat expensive, but they&#x27;re generally not charging the highest rates and only a small percentage of somewhat expensive accountants are good.<p>This is an interesting observation that I see a lot of value in. Not everyone is money-driven in the same way, and there&#x27;s a lot of talent that is willing to turn down additional money or additional work because they are happy enough. They have reached a local maxima, and while there may be other maxima, those are not guaranteed, and going for them could result in overall loss.
madrox5 个月前
Market efficiency follows a lot of the same laws as evolution, and I think the author identifies the same fallacy many do about evolution: that there is some implied natural progression to perfection or negentropy.<p>You don&#x27;t need to be perfectly efficient to survive; you don&#x27;t even need to be the most efficient. You just need to be efficient enough in your niche to make it another day. As environments&#x2F;markets change, the measure for efficiency does as well. Adapt or die.<p>The problem with build vs buy in general is that evaluating software is <i>hard</i> and there are a ton of variables whose values are often only clear in hindsight. Not only does the thing have to work, it has to work in a way that will scale with how you use it. Also, when comparing it to build, you have to know whether you have the right expertise (and interest) in-house to do what you want. There&#x27;s also straight up timing considerations. There&#x27;s really no formula to this, but at the end of the day you have to acknowledge that there&#x27;s always going to be a slight misalignment of incentives between you and your vendor. You need a working product; the vendor wants your money. People want to believe these end up being the same thing, but especially for startups that aren&#x27;t thinking past their next sales cycle this simply isn&#x27;t true.
eviks5 个月前
&gt; Talking purely abstractly, it&#x27;s hard to settle the debate,<p>Hm, the mistake in the abstract is pretty obvious and is acknowledge in the theoretical discussions of all the &quot;perfect efficiencies&quot; theories - real life is never perfect! So sure, at the superficial&#x2F;cocktail party level, you&#x27;ll always have a lot of folks ignoring the obvious and repeating this silly mistakes. But then it&#x27;s just as easy to do the same when looking at the specifics. I mean, the original example came from defending a specific obvious error?<p>It&#x27;s also not hard to find flaws in the specifics: like with the air conditioner friend &quot;missing expertise&quot;, you don&#x27;t need to have any expertise to look for such a directly measurable (even wihout instruments) stuff such as noise level! I mean, your friend could&#x27;ve gone and listened to different units working if this information is not covered anywhere online.<p>Even though the fundamental issue of the challenge to find relevant reliable information is indeed very hard, there are so many force working against a poor guy at any level of actual expertise of said guy (so the post is right on the money here)!
throwway1203855 个月前
I think on the point of build vs buy, what a lot of &quot;buy&quot; people neglect is that you can often cut a ton of use cases out of the product that you&#x27;re paying for, and when you cut those use cases out then build becomes a really cheap proposition versus buying. Plus, when you buy you still have to maintain internal staff to manage the product. If you build, those people can also work on other things that advance the plot for whatever you&#x27;re building elsewhere in the company.<p>If you have to make payroll and you&#x27;re not interested in handling payroll business cases in your core business, absolutely buy payroll software. But if you need to spin up containers incidentally to your core business, maybe consider wrapping the same stuff Podman or Docker wraps if it makes sense. Going down a level of abstraction in this way can allow you to work more efficiently than if you just used those tools off-the-shelf.
UniverseHacker5 个月前
I&#x27;m very much into buying things that work well... to me the trick is to ruthlessly research what actually works well, and not be biased by what is new or more expensive. Most often these end up being older, cheaper, things built with a simple timeless design and no extra complexity that leads to unreliability.<p>Some examples:<p>Maytag 575 commercial grade residential washer and dryer<p>1980-early 90s Mercedes W124 or W201 diesels<p>1980s Volvos (mentioned in article)<p>IBM Model M keyboard<p>Military&#x2F;commercial security marketed heavy nylon and wool clothing<p>Vintage Yamaha solid state electrical devices (or anything 1970s ish with real hardwood cases, and discrete components you can test and replace)<p>1970s professional grade Pioneer studio monitor headphones (e.g. Monitor 10s, etc.)<p>I&#x27;ve often had people accuse me of being a &#x27;hipster&#x27; into vintage stuff that is impractical because of the aesthetic. In such cases I usually could not convince them that I just had what worked best- I didn&#x27;t care about the age or aesthetic, but some good stuff happens to be old.<p>For services, companies, and products I recommend reading every 1 star review you can before choosing. For actual good stuff, the 1 star reviews will all be petty: they got what they paid for and it was perfect but are mad or confused about something else, or they just had random bad luck. I keep forgetting to do this, and everytime I have an awful experience, someone else had it before me and warned everyone but I didn&#x27;t remember to look.
评论 #42436659 未加载
评论 #42434315 未加载
评论 #42435993 未加载
buryat5 个月前
We need a system based on verifiable proofs that will help the society to streamline selection of long running working products that will push the businesses to optimize their processes and fight for the economy share. Long working products will signal that producers can not just sell the same inferior products and reap profits. As a cause-effect it will make available many elements of the society to explore new economic models and embark on a journey in search of their application. And since everyone wants to keep consuming it will give a healthy economically-based incentive for the majority of the population to start producing tangible value.
tempestn5 个月前
&gt; For example, in my social circles, there have been two waves of people migrating from iPhones to Android phones over the past few years. Both waves happened due to Apple PR snafus which caused a lot of people to think that iPhones were terrible at something when, in fact, they were better at that thing than Android phones. Luckily, iPhones aren&#x27;t strictly superior to Android phones and many people who switched got a device that was better for them because they were previously using an iPhone due to good Apple PR, causing their errors to cancel out.<p>Uncomfortably reminiscent of politics.
评论 #42440757 未加载
1970-01-015 个月前
Shopping without researching goes hand in hand with a fool and his money. For the typical consumer, visiting consumerreports.com and picking one of their top recommendations would solve a majority of these problems.
评论 #42433776 未加载
评论 #42432173 未加载
评论 #42432479 未加载
K0nserv5 个月前
Core properties of well functioning markets is information symmetry and transparency. I don&#x27;t think this describes a lot of consumer goods. Even consumers who want to do their research will find that there isn&#x27;t enough information available.<p>Personally, I feel the responsibility of government in a capitalist system, is ensuring the free market is actually a free market. I&#x27;d love it if there was a government agency charged with improving the above mentioned problem of information asymmetry and lacking transparency. For example, when I buy a fridge I&#x27;m happy to spend more now, if it means I spend less over the lifetime of the item, but determining the cost per year for every fridge on the market is a huge task for an individual.
评论 #42434141 未加载
peterlada5 个月前
The most expensive thing in this work is human attention. Ie somebody has to give a shit so the thing works well.
uludag5 个月前
Reading this, I realized a reason why I try to move as much of my computing life into Emacs as possible: because having full control over a, perhaps inferior, Elisp application does result in a better experience, as I have full control over everything.<p>At times I would think that my motive to do this was purely superficial (for fun, masochism, idealistic purity, etc.), but perhaps the software world is such a market for lemons that the best one can do is put all the watermELons they can get ahold of in their Emacs cart and go their way.
spiritplumber5 个月前
Because most people will buy the $100 thing over the $130 thing even though the $130 thing is better made.
评论 #42431803 未加载
评论 #42431831 未加载
评论 #42431929 未加载
评论 #42432901 未加载
评论 #42431885 未加载
评论 #42436652 未加载
评论 #42431844 未加载
评论 #42432483 未加载
creer5 个月前
Customer reviews and professional advice are still not solved - and still a huge opportunity in my mind.<p>1) Customer reviews would in theory help people choose better and fast. There will always be the issue of difference of opinions but we are far from a point where this would be the only problem.<p>2) Professional advice is great but as anyone who has been around the hiring of consultants can tell you, if you are not competent in a field then it&#x27;s hard to choose someone who is. And beyond that, diligence and listening are not uniformly distributed.<p>Until these are solved, the goal of the firm is to &quot;sell something that people buy&quot; instead of &quot;sell something great&quot;.<p>It&#x27;s interesting that in a large part, Amazon makes its distribution business out of this. If you watch what you buy and for the most part, you can reliably and easily return duds. There is some, poor customer feedback on the site, AND there is very easy returns. And that still does not solve the issue of items that are difficult to test good or dud such as a battery or an electronic device where it might be a while until the problem appears. Costco is also in this strategy with a very long return window that mostly will give you time to notice whether the item does the job or not.
sl-15 个月前
To me this brings to mind Eliezer&#x27;s Inadequate Equilibria(<a href="https:&#x2F;&#x2F;intelligence.org&#x2F;equilibriabook&#x2F;" rel="nofollow">https:&#x2F;&#x2F;intelligence.org&#x2F;equilibriabook&#x2F;</a> ), which goes on explaining how these non-optimal, but stable situations arise all around us in societies.<p>Especially the part how everyone can see the inefficiencies, but there is no (easy, non-coordinated) way to fix them.
milesward5 个月前
It&#x27;s because buying things allows you to see the price of the thing, which means that will be the first and last thing you evaluate: quality is radically harder to quantify, so we resort to evaluating on the thing that comes in an easy quantity. Sellers are thus always incentivized to figure out how to make the price low, since basically all other factors can&#x27;t be rigorously evaluated until after the sale...
wtcactus5 个月前
The idea that competition alone solves this problem ignores that this is an incomplete information system.<p>I really, really want to buy products that work well, hassle-free and last a very long time, and I&#x27;m willing to pay premium for it. But how? Where do you even get that information? All you get is anecdotal experiences &quot;buy this brand, I have a washing machine from them for 20 years going fine&quot; that just don&#x27;t apply to your problem.<p>Sometimes I get lucky. I have a Mazda 2 with 16 years now. Very, very little issues, extremely reliable. But if I went to buy a new Mazda 2 model, other than my past good experience, would I even get something similar?<p>You can&#x27;t trust online reviews. 1st because they are played by the companies, and 2nd, I&#x27;ve noticed that people came to expect so little that they give 5 stars to something even when they specifically write that something was a little off. It&#x27;s like the all society has accepted that you will only be able to use what you bought for 2 years, and even so, you are expected to face some problems with it.
noelwelsh5 个月前
By definition most things will be average. You just have to look around to see, for example, poorly designed websites that don&#x27;t have even a minimum of CSS that would greatly improve readability. That&#x27;s fine for most things. Most websites don&#x27;t need to have great design, even if improving the design would be a trivial exercise.<p>Only a few industries are power law industries, where only the best dominate, but even in those industries there are different definitions of what &quot;best&quot; is. Take entertainment, for example. It&#x27;s a power law market. Taylor Swift and K-pop band de jure make magnitudes more money than the tenth or hundredth best artist. Yet there are people who have never listened to either but will queue around the block for the latest Aphex Twin release. Similarly in written work, if you adopt a particular style, such as aesthetic of an unstyled web site, there are certain people that will resonate with (in this case, people who started using the Internet in the 90s or 2000s) and they will perhaps appreciate your work more for it.
reginald785 个月前
Things that work well are hard to identify, it takes time and effort. Worse, companies are constantly changing products (either they stop making models or change the way they are made) so even if you successfully identify one thing the knowledge will rapidly become obsolete. So rapidly that you&#x27;ll often not even have an opportunity to reuse that knowledge a single time.
notShabu5 个月前
In trading there is an idea called the &quot;hurdle&quot; which is a measure of inefficiency that any trade has to be above in order to actually make a profit. E.g. if the bid&#x2F;ask spread for gold chains or options is 2% then any arbitrage has to be better than 2%.<p>For companies this hurdle rate is from political friction. Any improvement that doesn&#x27;t provide benefits in excess of the political friction generated doesn&#x27;t get implemented.<p>And in large institutions this friction can be very high. E.g. planes not crashing for Boeing or kids-not-being-shot for the U.S. seem like obvious things-to-fix.<p>But they generate so much destabilizing political friction that they just can&#x27;t.
dsego5 个月前
&gt; where the notice falsely indicates that the person wasn&#x27;t home and correctly indicates that, to get the package, the person has to go to some pick-up location to get the package.<p>Didn&#x27;t know this was a universal experience with package delivery and the post office. I always thought it was just my national postal service that does this.
评论 #42433733 未加载
everdrive5 个月前
One thing I’ll say is that I’ve _never_ seen a company properly understand a product, properly staff a product they purchase, or spend much time ensuring that their product is properly configured. But despite this, companies are shocked over and over again when the services or softwares they purchase do not live up to expectations.
1970-01-015 个月前
A massive unmentioned bullet point in this rant is <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Planned_obsolescence" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Planned_obsolescence</a><p>Stuff is literally designed to stop working while also being financially unrealistic to even salvage and sell for parts.
评论 #42433358 未加载
评论 #42442357 未加载
评论 #42433195 未加载
dsjoerg5 个月前
Coase&#x27;s &quot;Theory of the Firm&quot; is the more serious, less straw-man exposition of these issues.<p>===<p>What’s New (If Anything)? While the article restates familiar Coasian concepts, it adds:<p>Modern Examples: It applies Coase’s framework to contemporary issues like software outsourcing, tech hiring, and cultural differences in trust and service delivery. Focus on Culture and &quot;Unreasonable&quot; Actors: It emphasizes how outliers (individuals or firms) push against systemic inefficiencies, which isn’t central to Coase’s theory. Bottom Line: The article revisits Coase’s foundational ideas without explicitly connecting to them. It’s less a groundbreaking insight and more a modern illustration of why transaction costs and organizational trust matter, consistent with Coase’s theory of the firm.
mojuba5 个月前
tl;dr information asymmetry. Didn&#x27;t read the entire text, just skimmed through.<p>I think fundamentally the reason behind info asymmetry in our day and age is that products and services have become so much more complex that there&#x27;s simply no time to independently assess the qualities such as reliability, durability, and a myriad of other variables in each particular case.<p>I think it&#x27;s the complexity that is becoming our enemy number one, and too many variables when choosing a product as a consequence of that. Is the air conditioner too noisy? Does it require a WiFi connection and even a mobile app to function? People may omit some of these things even in their Amazon reviews.<p>And then some of the insanely complex products like mobile phones are practically impossible to evaluate objectively. I once stumbled upon an article that explains why Android requires roughly 2x RAM and a slightly larger battery compared to an equivalent iPhone in order to have the same efficiency and performance, supported by some benchmarks (blame garbage collector I guess?). How many technical people or experts are even aware of this?<p>My process of purchasing stuff comes down to two principles: (1) devices that Apple makes are generally OK to buy, they are less likely to disappoint; (2) for everything else: research, read reviews; the time spent on a product is proportionate to the price of the product.
评论 #42448269 未加载
评论 #42441530 未加载
swayvil5 个月前
Can-openers. 99% of them are utter trash. The one that works is a copy of a 60 year old model.
评论 #42433484 未加载
评论 #42432697 未加载
myflash135 个月前
This is not just about information asymmetry. If you want to do something well, you have to find someone that truly cares about doing it well. Price is almost irrelevant, in fact, it may actually be a negative signal, because the best performers are often not motivated by money alone.<p>&gt; So, in Korea, there&#x27;s some service like Amazon where you can order an item and, an hour or two later, you&#x27;ll hear a knock at your door. When you get to the door, you&#x27;ll see an unlabeled box or bag and the item is in the unlabeled container. If you want to return the item, you &quot;tell&quot; the app that you want to return the item, put it back into its container, put it in front of your door, and they&#x27;ll take it back. After seeing this shipping setup, which is wildly different from what you see in the U.S., he asked someone &quot;how is it possible that they don&#x27;t lose track of which box is which?&quot;. The answer he got was, &quot;why would they lose track of which box is which?&quot;. His other stories have a similar feel, where he describes something quite alien, asks a local how things can work in this alien way, who can&#x27;t imagine things working any other way and response with &quot;why would X not work?&quot;<p>This gem near the end of the article ties in with Marc Andreeson’s quote at the beginning of the article. The fact that he thinks hiring can’t be more efficient sounds to me like typical American head-in-sand insular thinking. He’s stuck in a cognitive frame, and in a particular system where people can’t imagine doing things differently, because they simply haven’t seen anything else. Try living somewhere else for an extended period of time, and you’ll realize that a lot of things you didn’t even think about could actually be better. The author gives the example of deliveries in Korea. I say the same about many things in Eastern Europe, including deliveries, payment systems, and just everyday services like dentists - a LOT of things work better than they do in the West, not because of monetary incentives, but because things are simply done differently here, and some things are done better.
评论 #42432565 未加载
sneed_chucker5 个月前
Market for lemons. Read it, explains everything.
Sharlin5 个月前
Does this long-winded article manage to say anything more interesting than &quot;maybe markets aren&#x27;t <i>actually</i> 100% efficient, just like cows aren&#x27;t actually spherical and of uniform density&quot;?
评论 #42434234 未加载
throwaway_203575 个月前
A lot of it comes down to the attitude that product quality (at least when it comes to a product&#x27;s longevity aspect) is less and less valued because the low quality alternatives became so cheap. Or, as someone put it here once: &quot;I don&#x27;t want to have to think really hard about which trivial household item I want to spend the rest of my life with. I&#x27;d rather pick a random cheap one and then not feel too bad about replacing it, if necessary.&quot; It seems the negative externalities of this trend are not effectively priced at the moment.
评论 #42433509 未加载
zombiwoof5 个月前
Because people like Mark Andreeson only care about money. That guy is a one trick pony who just has money and part of the problem. He’s a pump and dump bozo no better than some banker on Wall Street
micromacrofoot5 个月前
Because it&#x27;s easy to lie about things that don&#x27;t work well<p>half the time after we buy something, we&#x27;re willing to lie about how well it works to ourselves
switch0075 个月前
I swear a lot of people buy the cheapest thing knowing it won&#x27;t last because they are addicted to shopping
评论 #42435224 未加载
dehrmann5 个月前
&gt; every social media company has kernel expertise as a core competency<p>It gets even more interesting when you&#x27;re someone like Google who has both search and Android. There&#x27;s a case for having <i>two</i> kernel teams so server teams aren&#x27;t directly competing with mobile teams for resources.
RicoElectrico5 个月前
Oh, the appendix part is such a poignant description of the US society. It is perfectly acceptable to be enriched at someone&#x27;s cost because <i>they should have known better</i><p>If it&#x27;s not both parties benefitting from a transaction then let&#x27;s name it straight - it&#x27;s a scam.
fatbird5 个月前
Markets are only as efficient as their information is good. At this point, we should all have a very healthy appreciation for how warped the infosphere is for everyone, and that will of course have deep applications to the &quot;efficiency&quot; of the market.
camgunz5 个月前
Super enjoyed reading this post as usual.<p>I want to make a small leap. It takes extraordinary effort and resources to gather data, whether that&#x27;s on your ETL SaaS, your health insurance, your congressional candidates&#x27; policies, or who&#x27;s right about vaccines on Twitter. If &quot;a market for lemons&quot; is one &quot;where information asymmetry means that buyers can&#x27;t tell the difference between good products and &#x27;lemons&#x27;&quot;, literally every market is a lemon market. Since we know the incentives there reward marketing over performance, we have to have some kind of countervailing force if we don&#x27;t want that (I don&#x27;t). Luu cites &quot;unusually unresonable&quot; people for companies here, in traditional markets those are regulators, on social media those are content moderators, in politics those are political parties and the press, etc.<p>Jump from here to wherever you want:<p>- unregulated markets quickly become back alleys stuffed with lemons<p>- (therefore) regulation greatly increases market efficiency<p>- competing definitions of &quot;lemon&quot; undermine the entire system<p>- regulator corruption is inevitable--successful systems account for this<p>- regulation does stack--intuitively one would think no one will buy &quot;sure, the car regulators are corrupt, but the car regulator regulators are on the up and up, so it&#x27;s fine&quot;, but synonyms for &quot;car regulator regulators&quot; include &quot;FBI&quot; and &quot;journalists&quot;<p>- if you wanted to succeed as a person of style over substance you have a strong incentive to assail the regulators
pphysch5 个月前
If a product is straightforward, works well, has no lock-in, then customers don&#x27;t spend on support or switch to a competitor. So don&#x27;t make good products. That&#x27;s the message for most of the SaaS world.
aidenn05 个月前
I no longer trust Amazon with near certainty to send me the product I ordered, so the efficiency of me obtaining things for which it is very important I get the product I ordered has gone way down.
tonymet5 个月前
1. people lack good taste . (aka bad aesthetics)<p>2. it&#x27;s easier to agree (follow consensus) than disagree<p>3. (corollary to #2) Products are &quot;sticky&quot; and good selling products tend to sell well, despite their faults.
评论 #42434279 未加载
__MatrixMan__5 个月前
Because making things that work well is not what modern markets are optimizing for. They&#x27;re efficiently pursuing other agendas because they can get away with making things that work sorta.
pixelpoet5 个月前
Fabian Giesen&#x27;s name jumped out at me on the page, not sure why but that was pretty surprising to me (he&#x27;s normally writing about compression &#x2F; graphics &#x2F; maths).
spankalee5 个月前
Physician, heal thyself<p>This page is almost impossible to read at default browser settings. A couple of lines of CSS would go a long way to make the site work well.
评论 #42458703 未加载
larsrc5 个月前
And here I thought here would take about how new models come out so frequently that any information about long-term quality is long obsolete.
numitus5 个月前
in my opinion, the problem here is optimization. It is not difficult to make a courier service that will work like clockwork. It is difficult to make one where illiterate couriers will deliver parcels on time for a minimum wage, and the price will be comparable with competitors. The processes are simplified until the buyers of your product or service start suing you.
increscent5 个月前
A notable exception to the trend of poor&#x2F;unknown quality in consumer products is the outdoor product market. I buy a lot of gear for skiing, canyoneering, climbing, cycling, backpacking, and other sports. In these sports, the safety and functionality of gear is critical. And you spend enough time using and evaluating the gear that many (if not most) users become experts. Perhaps this type of market is the exception that proves the rule.
Funes-5 个月前
Because of &quot;profitabilism&quot;, that&#x27;s what. Profit margin over all--including, of course, product quality.
t0bia_s5 个月前
Why is it so hard to format text that is readable well?<p>Reading it on UHD monitor is almost impossible with full width of paragraphs.
评论 #42434700 未加载
评论 #42436462 未加载
评论 #42434387 未加载
scotty795 个月前
Quality is multidimensional. Local maxima exist in any non-trivial optimization process.
photochemsyn5 个月前
Generally speaking, if a corporation is controlled by the finance and sales people, instead of buy the engineering people, the products it makes will suffer from things like planned obsolescence. Monopolistic control of markets makes this worse, as there&#x27;s little if any competition to turn to. Boeing is the poster child for this problem in the USA - Boeing relied on monopolistic control of the airplane market, and cut R &amp; D to increase shareholder and executive profits, and was then blindsided by Airbus developments in fuel efficiency, and their response was to try to rush a low-cost development project with disastrous results.<p>Investment capitalism naturally gravitates towards such behavior; this is why the most disastrous thing SpaceX (compare to Boeing) could do is go public and let big shareholder conglomerates like Vanguard&#x2F;Blackrock&#x2F;StateStreet&#x2F;FMR etc. control the makeup of the corporate board and thus, engineering decisions.
zombiwoof5 个月前
Better recurring revenue if you need to say buy a new iPhone every year
评论 #42436956 未加载
dukeofdoom5 个月前
One strategy is after you buy reinforce possible failure points
eddyzh5 个月前
What an excellent written article. Thanks Dan Luu!
ed_balls5 个月前
&gt; “Find the dependencies — and eliminate them.” When you&#x27;re working on a really, really good team with great programmers, everybody else&#x27;s code, frankly, is bug-infested garbage, and nobody else knows how to ship on time.<p>Does anyone have a feeling that this has never been further from the truth?
deeves5 个月前
Cause it’s even harder to build them
sylware5 个月前
no bugs or planned obsolescence --&gt; &quot;no business&quot;
yapyap5 个月前
a guess before having read it: corporate greed?
kransky5 个月前
Because it’s all made in china and the Chinese just copy crap and don’t work it. Such is the communist economy. Suck it, libs, that’s your utopia.
pluc5 个月前
Doesn&#x27;t everyone know and acknowledge that in a capitalist system, there is more money to be made when products have a finite operational window? It&#x27;s a lot more profitable to sell you a phone&#x2F;fridge&#x2F;car every few years than it is to make one that lasts 10, 15, 20 years...
fHr5 个月前
we need to enshitify our supply chains with grifters
DataDive5 个月前
Reply: things don&#x27;t work well for the same reason this very post is so hard to read wall of text.<p>I find it deeply ironic that the author can&#x27;t even produce readable text, while complaining about other, much more complex tasks not working.<p>The author refuses to take minimal, common-sense measures to make the text more readable.
评论 #42432259 未加载
评论 #42432249 未加载
code-faster5 个月前
Will AI change this? If AI can assess who is good for us, then we can get somewhere
评论 #42431268 未加载
评论 #42431804 未加载
评论 #42431690 未加载