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。

Fashionable Problems

365 点作者 hackerews超过 5 年前

36 条评论

rgbrenner超过 5 年前
I think pg understates the problem. It isn&#x27;t just fashionable. Think about how the solutions we use today evolved.<p>Someone had an idea originally and certain decisions were made about that approach.. those that best adapted to the conditions of the time were successful... rinse and repeat over several decades.<p>Those ideas that preserved the past were more likely to succeed because they preserved the ecosystem that already existed. Ideas that diverged too much from the existing successful tech had a huge obstacle in their way: they had to recreate all of the existing solutions in their new model. So even if they would lead to a better solution over time, they may never get past that initial roadblock. And the longer we stay on the path, the bigger that roadblock becomes.<p>If we are all thinking of ideas around the existing model and assuming all of the existing assumptions, then we end up with similar solutions (fashionable solutions). We&#x27;ve reduced the solution space, leading to a limited number of solutions.<p>Perhaps groundbreaking&#x2F;valuable tech can be found by questioning those existing assumptions; identifying where tech is still built on assumptions that are no longer true; or reexamining past solutions to see if they can be solved in better ways today with what we&#x27;ve learned since then...
评论 #21859684 未加载
评论 #21859439 未加载
评论 #21859680 未加载
评论 #21859505 未加载
评论 #21861473 未加载
评论 #21860141 未加载
评论 #21860187 未加载
评论 #21860588 未加载
评论 #21860646 未加载
majos超过 5 年前
This post sorely lacks evidence for its big first claim:<p>&gt; I&#x27;ve seen a similar pattern in many different fields: even though lots of people have worked hard in the field, only a small fraction of the space of possibilities has been explored, because they&#x27;ve all worked on similar things.<p>Anyone want to step in with some examples? Without them, the thrust of the essay seems to be: &quot;If only other people understood what problems are worth working on! Especially in the well-studied areas of essays, Lisp, and venture funding! Too bad they do not. Well, goodbye.&quot;
评论 #21860103 未加载
评论 #21859602 未加载
评论 #21859518 未加载
评论 #21860324 未加载
评论 #21859899 未加载
评论 #21864043 未加载
评论 #21861579 未加载
评论 #21859583 未加载
评论 #21859958 未加载
bawolff超过 5 年前
Seems like a lot of weirdly defensive people in this thread.<p>But the essay doesnt really seem that revolutionary more common sense. If you want to make an impact, dont work in an oversaturated field. The low hanging fruit is probably already picked and other people will probably get there before you do. But you also dont want to work in a field nobody cares about as noone will care. Working on a problem with proven demand, but seems &quot;boring&quot; and hasn&#x27;t changed much recently is a good bet, as there is probably new insights you can apply and new contexts that have appeared since last time there was a frenzy for that field.
评论 #21859672 未加载
评论 #21860268 未加载
MaysonL超过 5 年前
Just listened to a podcast with a VC who applied this pattern to nuclear power. The big problem, which almost nobody was investing to solve, was nuclear waste. He went out, attended a number of nuclear power events, and found some people who thought they had a productive attack. They did, and after he threw some money at them, ended up with a 40 or so X return.<p><a href="https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;audio&#x2F;2019-08-16&#x2F;josh-wolfe-discusses-innovative-investments-podcast" rel="nofollow">https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;audio&#x2F;2019-08-16&#x2F;josh-wolfe-d...</a>
est31超过 5 年前
&gt; Even the smartest, most imaginative people are surprisingly conservative when deciding what to work on. People who would never dream of being fashionable in any other way get sucked into working on fashionable problems.<p>This makes no sense. The concepts of &quot;conservative&quot; and &quot;fashionable&quot; are almost polar opposites. Becoming a musician or an actor is fashionable and the dream of many. But it won&#x27;t bring any money to most people who choose that career. It&#x27;s no conservative choice. Instead, it&#x27;s conservative to go into STEM, law or finance. But those fields are &quot;boring&quot;. The pattern repeats inside a field as well. It&#x27;s conservative to be a cobol coder or a DOS expert, and you&#x27;ll certainly make money. But it&#x27;s not fashionable. How come you aren&#x27;t building a cryptocurrency using self driving car that has a drone port on the roof! It results in the woke fields being overrun with very smart and capable people and capital, while tons of fields that use slightly outdated stuff are ripe for harvest but nobody is around to do it.
评论 #21859504 未加载
评论 #21859513 未加载
blast超过 5 年前
Is this the first PG essay that was composed on Twitter?<p><a href="https:&#x2F;&#x2F;twitter.com&#x2F;paulg&#x2F;status&#x2F;1183686114844069888" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;paulg&#x2F;status&#x2F;1183686114844069888</a>
评论 #21859347 未加载
lquist超过 5 年前
<i>If you want to try working on unfashionable problems, one of the best places to look is fields that people think have already been fully explored: essays, Lisp, venture funding – you may notice a pattern here.</i><p>Do people really think the fields of essays, LISP or venture funding are fully explored?
评论 #21859042 未加载
评论 #21859047 未加载
评论 #21859059 未加载
评论 #21864903 未加载
fnord77超过 5 年前
&gt; If you can find a new approach into a big but apparently played out field, the value of whatever you discover will be multiplied by its enormous surface area.<p>If you found something in a big, fashionable field with a huge surface area (cough ai&#x2F;ml cough), wouldn&#x27;t this multiplier apply to that fashionable field, too?
评论 #21859167 未加载
评论 #21859136 未加载
MAXPOOL超过 5 年前
Humanity as a whole as problem solving algorithm might be similar to particle swarm in very complex neighbourhoods and topologies.<p>Particles (individual humans or small groups) move around in problem space as particles with position and velocity. Each individual&#x27;s movement is influenced by its best known position locally but also the the best known global positions in the search-space. When better positions are found by others, individual change their course (take hints) and move towards them.<p>The problems would be the same. Too much randomness and it&#x27;s just a random search. Too much convergence leads to local optimums.
starpilot超过 5 年前
This advice is just so... airy. Like most of his advice. &quot;Build something people love&quot; Got it, now what? What do I build, exactly? Like the incredible Onion talk on startup: &quot;Step 1: come with up an idea. Step 2: Build it. We&#x27;re at step 2 - we&#x27;re half way there&quot;
savrajsingh超过 5 年前
I think there are still a few major breakthroughs left in our understanding and use of electricity.
nostromo超过 5 年前
I love this mini-essay.<p>The counter to this is that it&#x27;ll be harder to raise capital for un-fashionable problems.<p>If you pitched &quot;ML for sandwich makers&quot; right now you could raise a million bucks because so many VCs are making fashionable bets on ML.
评论 #21859138 未加载
bobbyi_settv超过 5 年前
Fashionable problems are the ones for which you&#x27;ll have the easiest time recruiting employees, raising funding and generating press coverage, even taking into account that their fashionability leads others to pursue them.
galaxyLogic超过 5 年前
Well that&#x27;s the whole paradox of Fashion isn&#x27;t it? You want to follow fashion but be almost at the top of it. If you are too much next year&#x27;s style nobody thinks you fashionable but just crazy. But if you follow last year&#x27;s trend you are also unfashionable.<p>The paradox is that nobody really knows what will be the next fashion and similarly nobody knows what&#x27;s the next worthwhile problem-area to work with.<p>There&#x27;s a good reason why fashionable problem-areas are well-researched it is because the results so far have been useful and promising.
echelon超过 5 年前
My worry in solving an existing problem in a novel way is that the incumbents can catch up faster than you can scale.<p>If I were to take on Netflix&#x2F;Disney&#x2F;Twitch with some new kind of video entertainment product, they&#x27;d have deep pockets to fund a competing offering.<p>The lever of equity might work to attract better talent, but only if you succeed. There&#x27;s a lot of risk.<p>Scaling rapidly also means giving up control as you seek capital. It&#x27;d be hard to organically grow and go unnoticed.
评论 #21863589 未加载
diego超过 5 年前
Well, I have a hard time thinking of a field that I believe is fully explored. History has shown time and again that it&#x27;s really easy to be fooled in that respect.
mark_l_watson超过 5 年前
Agree. It is also a good reason for having multidisciplinary interests so we might have different ideas for looking for very different solutions to problems.
shrubble超过 5 年前
In this thread a guy talks about using the Q language and then someone else jumps in and says &#x27;it&#x27;s not scalable etc.&#x27;<p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=21854793" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=21854793</a><p>What if all this cloud&#x2F;k8s &#x2F; serverless stuff is really all piffle? What if running stuff on dedicated hardware ends up a better solution in some fashion?
评论 #21859493 未加载
评论 #21860644 未加载
smitty1e超过 5 年前
Let&#x27;s write &quot;The Online Packaging System To Ruminate About Them All&quot; (TOPSTRATA).<p>Are we not eternally one packaging system short of Nirvana?
kick超过 5 年前
Is this a repost? I feel like Paul has already written this essay.<p>Edit: Oh, it was originally a tweet: <a href="https:&#x2F;&#x2F;twitter.com&#x2F;paulg&#x2F;status&#x2F;1183687634763309056?s=20" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;paulg&#x2F;status&#x2F;1183687634763309056?s=20</a>
mmaunder超过 5 年前
I&#x27;d suggest focusing on people and the problems they have rather than industry, your space, the news and your colleagues. That way you&#x27;ll serve real needs rather than getting a ton of street cred as you dissapear down an academic rabbit hole.
mmahemoff超过 5 年前
&quot;Fashionable solutions&quot; are a related phenomenon to fashionable problems. Solving a problem the same way as everyone else (and often expecting a better result).<p>Both are widespread.
评论 #21859490 未加载
lazyjones超过 5 年前
People are different, but I cannot imagine how someone could seriously work on (as in: devote their whole working time to, by their own choice, as an entrepreneur, researcher or hobbyist rather than an employee or student) problems they don&#x27;t genuinely love. Sure, some do it for the money, but then they probably just love money and found a promising opportunity to earn it.
jonplackett超过 5 年前
It&#x27;s not just people&#x27;s desires and people following what is &#x27;fashionable&#x27;. I read a while back that in Physics it was impossible to get funding for anything outside of what was fashionable, String Theory was their example.
akr1超过 5 年前
Funny, we&#x27;ve just launched EssayMash.com yesterday trying to accomplish what PG writes about: further exploration into essays. We host monthly essay competitions on an important topic with $300+ in cash prizes.
galfarragem超过 5 年前
What I understand from PG words is that people, even the creative ones, by lack of courage or macro vision are not a &#x27;Elon Musk&#x27; and go with the herd building yet another crypto or yet another SAAS.
rdiddly超过 5 年前
Nice to see the recent burst of writing activity from PG. Also nice to see people here putting it through the paces. Would expect nothing less.
blondin超过 5 年前
yeah but i am afraid pg is not relatable anymore (to most of us anyway).<p>devoting more than 20 percent of your time to problems that are unfashionable but dear to you is ill-advised because: a. they don&#x27;t pay the bill b. they take time and in the grand scheme of things spending time with others on things you all understand is better than being happy alone.<p>but of course there are exceptions...
评论 #21863600 未加载
blueboo超过 5 年前
Paul is confusing ignorance for insight here. It&#x27;s the same phenomenon whereby a weekend visit to Paris has your uncle explaining the European soul but your year in Kyoto leaves you barely able to generalize at all from your (actual) knowledge of nuance, complexity and diversity of another culture.<p>Everyone is trying to break the mold at various scopes.<p>Meanwhile, you&#x27;re welcome to reproduce your late-90s success at any moment, Paul. We&#x27;ll wait.
评论 #21860042 未加载
mc3超过 5 年前
Sounds grand. There are a lot of us who just want to pay the rent. If that means a React job, so be it.
notkid超过 5 年前
The heuristic of trying to work on what you genuinely love is not helpful or practical for most people. It sounds good, but it is just a platitude. Genuine love and fake love feel and look pretty similar. Most people naturally start loving the life they live in, if it is generally positive. Then, they make up a self-affirming, coherent narrative that justifies their emotions, decisions, and interests. If you do an AI startup, life goes well for you, and you embrace that decision and life, how can you differentiate whether it was really a genuine interest or not?
评论 #21860138 未加载
评论 #21860142 未加载
ismail超过 5 年前
I think it goes much deeper than people choosing to work on fashionable problems. Here are my thoughts<p>1. The more defined and mature the problem space is, the more the assumptions that underpin that field are taken as a given.<p>2. These assumptions may become so deeply ingrained that people become effectively blind to the entire range of possibilities.<p>3. These assumptions frame how the problem&#x2F;solution space is looked at. Therefore the solution space is constrained by the set of assumptions (about what the problem is , how to solve it, what to do)<p>4. These assumptions are recursive, in that they are contained within other assumptions. It is turtles all the way down. At some level, someone working in the space may not even understand what the core assumptions are. We have to have these assumptions though. See the next point.<p>5. The interesting thing about this is: The constraining of the problem&#x2F;solution space is actually a <i>positive</i>. It enables co-ordination and incremental improvements and refinement. It allows people new to the domain to quickly get productive.<p>I like to think about it this way:<p>Picture yourself in a massive area that is pitch black. You are grasping around and can not see much. Someone figures out how to get a tiny fire started. With this tiny fire you get to see a little bit. Using this you can build a bigger fire illuminating more of the area (but still leaving the entire space unexplored). This eventual results in the ability to build a permanent light illuminating a specific corner of this space.<p>This specific space with light illuminating it becomes highly productive, people can do all sort of things. Like read etc. Yet, there are still areas left unexplored. The light cannot simply be taken across. It takes work, and it takes turning your back towards the current &quot;lit&quot; up space, and taking a step back into the dark. A scary thought for some.<p>5. Importantly: These assumptions have been inherited from the past. So they existed and were relevant at a specific point in time. They may or may not be relevant as of today. We would have to peel several layers to get to the core.<p>6. While 4 is a positive, it is also a negative. The idea&#x2F;areas greatest strength (maturity, constant improvements, efficiencies) is also its greatest weakness (constraining the search space)<p>To take a step into the dark, is to turn your back to the lit up parts. You have to question the underlying assumptions and see if they are still relevant. If you discover an assumption about the world that is no longer accurate, then you found a new space to illuminate.<p>To put it in another way, to explore the dark is to shift your perspective on the problem&#x2F;solution. It is to see with &quot;new eyes&quot;. Initially it may be dark, but slowly with diligent work, and passion you could light up a completely new and novel area.
bdotdub超过 5 年前
Area man write blog post saying things are great and just so happens to overlap exactly with what he does
gist超过 5 年前
&gt; Even the smartest, most imaginative people are surprisingly conservative when deciding what to work on. People who would never dream of being fashionable in any other way get sucked into working on fashionable problems.<p>How can a statement like that be made? Is there some kind of authoritative directory of &#x27;the smartest, most imaginative people&#x27; being &#x27;surprisingly conservative when deciding what to work on&#x27;.<p>Implied I guess Paul means &#x27;who I&#x27;ve met or who I know of&#x27;. So then say that.<p>It&#x27;s a big world out there. Who knows what anyone is working on or what they are thinking or have tried and why they haven&#x27;t pursued it.<p>This is a bit like saying &#x27;people love their dogs and will do anything for them if they are sick&#x27;. Just a general statement of opinion by one person (and generally accepted as being correct) but based on not anything even close to being scientific and&#x2F;or backed up by any actual data. That part is fine. But if that is the case state it as such and not some absolute. Why does this matter? Because when someone like Paul writes something it will be taken by others to be some kind of important thought or fact.
评论 #21859488 未加载
ronilan超过 5 年前
Silicon Valley’s biggest problem is that is tends to see problems as opportunities and opportunities as problems. There is no opportunity in that problem.
sillysaurusx超过 5 年前
As a long-time pg supporter, it pains me to say this: I think at this point pg could write anything and it would show up immediately with critical acclaim.<p>It was more charming when he had to work hard to make his points known.<p>But hey, fame, right? Just famous people things.<p>There&#x27;s so much more to say in this case, though! <i>How</i> do you avoid the traps? Waving a wand like &quot;Just love something&quot; leaves far too much to the imagination. Pointing at a prior essay at loving your work is helpful, but different.<p>Often, you have to actively offend people in order to find good problems to work on. The idea that people have devoted their lives to the wrong thing is inherently offensive to them. That&#x27;s a point not covered here.<p>For example, I imagine that a lot of people who&#x27;ve studied 3D rendering for their entire lives are about to feel very outdated the moment neural network renderers displace them. And that&#x27;s also a good counterexample to the point that &quot;Often, the best place to search for new ideas is a place thought fully explored.&quot; It might often be true, but it&#x27;s not always true.<p>And then there are the in-betweens. Bitcoin was in a field both thought fully explored (crypto + finance) and also unexplored, in a certain sense.
评论 #21859175 未加载
评论 #21859297 未加载
评论 #21859222 未加载
评论 #21859293 未加载
评论 #21859225 未加载
评论 #21859253 未加载