to be completely free, uninhibited, released from guilt and predjudice. to just think, play, laugh, fart, and yawn. boring days spent looking out the window wondering how crows fart, or what makes zebras sassy.
the only limit is our own adult psychosis. we are prisoners of our own minds, locked and shackled by constructs of our own design. too focused on the failures of the past and anxieties of the present to take a moment and b-r-e-a-t-h-e. like who fucking cares about businesses and startups and shit. saas my ass. when you actually stop trying, give up, and stop caring, and truly let go (like in the movie frozen) you can actually go in and through and back out the other side.
just learn to get out the way of yourself. that's what im working on, but it's really hard. you think you're supposed to be there and generate brilliance 24/7 but it's not possible.
go outside and find a patch of earth where things are growing. get really close to the ground and start looking at it real hard. there's honestly so much stuff there, it's wild. like who knew? wow thats nuts. acorns, rocks and shit. touch the dirt and rub it on your face. mushrooms are literally everywhere. the water in my body was once in a bowl of rice that your great grandma ate at becky's 27th birthday party.
just give up just give up stop caring be free
People obviously do do dumb things, and some stuff is obviously crap. But if you change perspectives a lot of stuff actually….makes sense?
Like, looking at the world and seeing a bunch of bullshit is actually a flaw in your own thinking, not the world.
Take fast food for example. Widely derided, clearly poisons the body, not the best taste, produces a lot of waste. More expensive than cooking.
But, flip it around:
* Standard experience —> you know exactly what you’re getting
* Much cheaper than restaurant food
* Much faster than restaurant or home cooked food
* Tasty
* Easy
It’s something with obvious downsides but also obvious upsides.
It isn’t obviously dumb. It is only obviously dumb if you decide the downsides outweigh the upsides and then ignore all the upsides. Which is flawed thinking.
You can repeat this analysis and find that on their own terms most people mostly aren’t doing things that are totally irrational.
Even something like a lottery ticket gives someone hope despite the clear negative EV. You can take that too far, but….
The point is if you think everything is bullshit, your analysis is probably crap. And merely allowing yourself to use the term bullshit rewards this sort of lazy, negative analysis.
Whatever you sell doesn’t have to be the absolute best of all possible things. But it has to meet a need, and the vast bulk of stuff which sells does exactly that.
As an example, audiophiles might sneer at consumers buying "cheap" "digital" "crap" instead of having the good sense to appreciate what a high end speaker or vinyl system is capable of, but those consumers aren't upstream, they're trying to solve the issue of needing a sound system. People in the industry are for the most part audiophiles, so they're going to have an appreciation for things their consumer base is basically illiterate in. Still, the business is run on solving problems, and the problems of most are not the problems of the high-end user.
we're solving problems, or solving really hard problems, but either way can't deceive ourselves if we're just trying to solve the problem of making ourselves money, or actually of providing value to a variable population of problem-havers.
anything in that game requires a lot of planning to make something livable that can be sold at many groups over many years so you can finally get enough leverage to move the minds along some value axis
nerf was last minute class project.
If you want to really blow your mind, start looking into soil biology. Entire ecosystems of bacteria, fungi, multicellular creatures, plants, minerals, water....
And ALL OF IT IS CRITICAL TO LIFE.
People just think "oh, it's just dirt", and move on, when nothing could be further from reality
Old people have been bitten by the “A 95% inference will be wrong 1 in 20 times” thing a lot more, and also have been exposed to a wider array of experience where they might often judge an inference as being less certain than someone with a narrower range of experience would.
I don't have a specific reference but if you google it, you should find what I mean.
A 40 year old mature student was last in school more than 20 years ago. Younger students were last in school as recently as 1 year ago. Learning is a skill and it's perfectly possible to be out of practice. There may also be some knowledge gaps because recent practices and knowledge are different to common knowledge 20 years ago.
It's also possible that mature students are more assertive about asking questions.
And also - what some of the students in a class consider "obvious" is not at all obvious to all of the class.
Of course if someone is showing clear evidence of dementia, that's a different issue.
edit: needs of the many refinements out weight the needs of the few refinements
1. "Mum, when I grow up I am going to invent a telephone with no wires that's also a flashlight, a compass, a map, a jukebox, a movie theater, and even an ENTIRE computer!"
2. "Dad, when I grow up I am going to invent a table saw that will JUST KNOW when it touches your finger and it will disappear so fast that you'll only get a little cut next time!"
3. "Sis, when I grow up I am going to figure out some way to make it possible for even six year olds like you to have their own TV channel and put whatever you want on it! You can show everyone your toys and dresses. We can call it Roxanne-TV!"
https://inventionland.com/blog/ten-kid-inventors-that-change...
For example, laptops and phones produced 20 years of incremental improvements in li-ion batteries, and suddenly electric cars are "unlocked".
I think to some extent we are seeing the same thing with the latest AI models, where if you just throw enough parameters and GPU's at a problem, you can get pretty good results.
In a way, I see some parallels to aircraft, in that it was important to study birds as the best example of heavier than air flyers early on, but understanding the nuances of the biomechanics of flapping wasn't necessary to surpass them.
Model architecture cleverness will eventually outpace more GPU time, but for now the roi on GPU time is higher than clever design.
Streaming gaming is something I am still laser-focused on. Our networks are only getting faster. I feel there is a gigantic opportunity here that is being passed up by efforts like Stadia.
To do this thing correctly, the game engine itself needs to be developed with these concerns in mind. Expectations of things like having GPUs available on the same machine that is servicing player input events should be eliminated. The simulation itself may benefit from being distributed across multiple nodes connected via a low-latency network.
The future of shard-less, million player MMO worlds will not be feasible if every player's computer will be required to replicate and render all of that game state.
Sharding would have to happen to handle geographic distance between people, wouldn't it?
I’m curious to hear your thoughts.
If a machine in 5-6 years can leverage AI to render an entire unique universe of AI driven characters why sit in an MMO full of gold farmers, normalized, repetitive meme spammers aka other people who browse the same normalized social media?
Y’all have a good life, but I don’t need to know you all personally, nor did I ask you to solve these “problems” looking to be sold for coin.
I think the problem here isn't render, on that scale you'll have to shard on the server end anyway.
My day job is to work on software working with self-organising maps, which were last popular in 1980. As far as I know, they were never really used for anything significant then. The reason we can be so successful using them is that we discovered a relatively new environment in which they can improve profits fairly immediately: e-commerce.
Seems like an expensive way to structure a "this is like" or "people also bought" compared to an SVD.
This is not the way to look at it.
'Ideas' are too early to be judged as good or bad.
Ideas are fodder, they are there to be moulded into things that are good or bad.
Often ideas are just foils to help you think.
They are so contextual as well: 'full screen smartphone' was maybe only possible by Jobs, with that much direct and coercive power over engineers. Motorola CEO was not going to pull it off. But Apple did work with motorola to make a 'music phone' as an experiment.
I think your note about 'yesterday's ideas' is bang on.
My backlog of games/puzzles/stories I'd eventually like to get out there might be >100 at this point.
Focusing mostly on a bit more of an ambitious game right now, and it's probably still 6 months before I get it into Steam early access (coding in my spare time and energy, which is sparse lately), so everything else is just waiting at this point.
But sometimes I get an idea that I need to at least get some playable prototype with index cards, pen, paper, and tokens, at least. Just had another concept pop into my head a couple days ago for a roll-and-write I should at least make a board and write up some rules for before I forget.
Waiting for the day when something like ChatGPT can code up a basic but working game based on my description of its mechanisms. Even if it can only handle abstract games on a sqauare or hex grid, that'd still be useful for me.
go e-ink & vr and save the $$ not spent on consumable resources for venture capital.
One of the surest ways to come up with new ideas is to pay attention to what doesn't fit in. We're used to seeing things in a certain way, and often times it's the things that don't quite fit into our preconceived notions that can provide the best insights. The most obvious place to look for these anomalies is at the frontiers of knowledge.
Knowledge grows like a fractal; from a distance its edges may appear smooth, but when you get closer you begin to see the jagged gaps and spaces. These gaps can be surprisingly obvious; it can seem strange that no one has thought to ask a certain question or investigate a certain problem. When you explore them, you can often discover entirely new areas of knowledge that have been previously untouched.
That's why it's useful to investigate these gaps in knowledge. It may feel uncomfortable at first; we don't always like to challenge our preconceptions and explore things beyond our comfort zone. But often times this can be the source of some of our best ideas.
You don't even have to look far for these anomalies; they can be found in everyday life. Much of stand-up comedy is based on this, by finding the oddities and quirks about our daily lives that we normally take for granted.
So don't be afraid to take a closer look at what seems strange or missing or broken. It could lead you to unexpected discovery and perhaps even uncover some of your best ideas. This may be challenging and uncomfortable, but it's often worth it.
If he were to make it longer some examples might be interesting to read about and drive home the message, he no doubt has examples in his mind, and his readers would no doubt be curious as to what they are
Maybe he’s trying out if he can use his blog for short form too, since Twitter is going down a new / strange? path.
wouldnt that require a 'back to the future' tie-in?
Patio11 tweeted about using ChatGPT to immitate social signaling by rewriting content. I thought that was fun so I started doing it. That grew into Spencer Westford: https://vc.blankenship.io
That grew into a product: https://persona.ink
That keeps snowballing. Ever since building for myself my notebook of ideas is growing faster than I can keep up.
I've been doing this for years and all it has done is drill me further down into super-niche rabbit holes that the majority of people don't care about.
I mention this because I see this kind of advice all the time on here, but it so often comes bundled with the unwritten assumption that the kinds of things the reader will build won't stray too far from the mainstream.
After more than a decade of "building", I can state with confidence that profitable ideas don't spring up from countless hours invested in building e.g. a better comb filter for NTSC.
The reality of the situation is that if your interests are things like "doing cool stuff with AI" or "SaaSify yet another thing" or "shoehorn Merkle trees into yet another square hole", then this advice will likely work for you.
If you're the kind of dude (not me) to write ld-decode¹, then it will likely not.
I mean, shit, Donald Graft² has been "building" far longer than I have and yet none of his "building" has led to a business AFAICT.
If Paul Graham wants to say something, he should say it directly and not hide behind a chat gpt layer.
I'm wondering what ways/technologies we'll create to tell if someone actually generates text themselves. Right now, I can feel a rising desire to avoid text-based communication with people, almost a fear of getting catfished by text.
That 5% is pretty critical however, as without it you're just continuing to redo what's already been done.
It is like saying that to lose weight 90% is about nutrition and 10% is about exercise. 90-10% of what, exactly? And what would change in my approach to diet/exercise if it was 80-20% or 70-30%?
I think vague quantification does more damage along the way to achieving goals than we suspect. Maybe 10% more.
[1] "Robert Recorde invented the equals sign (=) and also introduced the pre-existing plus sign (+) to English speakers in 1557." https://en.wikipedia.org/wiki/Robert_Recorde
Can I have my big break and retire early? Sure, but the chances of that happening are not that great. Many serial entrepreneurs fail multiple times with nothing to show for it.
All I think about now is - when I am on my deathbed, and I look back at my life, will I regret not living it, or that I didn't spend enough time writing code in front of my computer on an off-chance that my idea works out?
TBH, at this point, if there is a problem that hasn't been solved, its because some people tried and failed and found that the problem is not solvable currently. It may be more fruitful to start with something promisingly popular, and improve upon it (e.g. tiktok)
pre or post order traversal of copyright/trademark?
perhaps switch heap stacks?
How to get new ideas? How do I stop them coming?! My method is to scribble them on a piece of card and chuck that into a tub to get them out the way for now.
I know that tub is being emptied with nothing in there seeing the light of day. Simple fact is if I work on my own stuff I don't get to pay rent or eat this month.
One day. In the next lifetime if not this.
The bottleneck here is to have the domain knowledge and experience to judge which generated ideas are good.
> The bottleneck here is to have the domain knowledge and experience to judge which generated ideas are good.
This comment reminds me of my grandpa.
I occasionally call him when I have a "brilliant" idea, and he dismembers the idea with simple questions.
It's like magic. Even if my idea sits outside his domain knowledge, he has an incredible gift for finding fatal flaws in things.
This gift has proven to be lucrative over his lifetime, and I suspect it makes him deeply unhappy.
I'd like to acquire this power too, one day.
downside of former is lack of sources, downside of latter is context constraint
{"prompt": "<prompt text>", "completion": "<ideal generated text>"}
So I'm not sure how you would feed it PG's essays in this mechanism
would tying infinate google search feeds to chatgp crash the internet or just reduce down to 42?
How can I reliably find "Frontiers of Knowledge"? Especially in fields I'm not savvy in?
currently, search engines/chatgpt seem to fill that roll vs more historical library/university.
aka search term "frontiers of knowledge" <replace with subject area field(s) unfamiliar with>
compared with traditional library/university book writer/teaching assistant
note: sed & awk not required to do text replacement, but good editor certainly helps
But if patient enough to wait, will eventually show up on internet as a video(s) way before the random monkeys typing away at a type writer get to it [1]
[1] : https://en.m.wikipedia.org/wiki/Infinite_monkey_theorem
Similar to "look for anomalies" my general rule of thumb is to solve some sort of problem, even if it's rather small.
So for example:
1. I pick a topic/product. Let's say: Gmail
2. I look for things that I'd like to be able to do, wish I could do, or consider frustrations I've experienced in the past.
- I want to review how many emails I get a day on average, where are my "inbox metrics"?
- I want to see threaded messages more clearly, why is the design so opposite a normal chat message?
- I want to create emails faster, why do I have such an elaborate editor and format for something that I typically want to use informally?
3. I consider ways to solve said problems
- Where are my "inbox metrics"? - Chrome extension to show average emails per day, and graph on a timeline which types of emails I get on certain days. Huge personal data opportunity here.
- I want to see threaded messages more clearly - An app that connects to my gmail that intelligently restructures my emails to make them appear like chat messages. It handles all the noise, I just message as I would in any instant messenger.
- I want to create emails faster - predictive text templates, quick starters, GPT powered replies, change the "bulky editor" out for a simple text box + clever replies.
4. I then research if there are existing solutions to said problems, and if there are, I either ignore and move on, or consider ways that I could quickly one up their product to make mine the better product in the marketplace.
For my specific use-case, I'm trying to build a product per day, so I'll look for speedy ways to solve the problem that I've selected in a novel and agency increasing way. But I think this is a great structure for anyone looking to build a startup or a small side hustle.
I'm really curious about how that's going. I can't quite suss out what your biggest successes are on your website.
In a year I hope to have 2-6 that are decently profitable, and ideally more that are valued in the open source community.
I think a few of the most successful ones so far are:
https://ghostlystock.com/ https://www.barbellformac.com/ https://www.heftysearch.com/ https://www.oldestsearch.com/
combined web traffic is around 5-10k views per month (across the pages you see on the main site)
Side note: I'm improving that page to make finding popular ones easier. Some aren't worth more than a quick laugh/glance, while others are really worth using.
best done on one's own terms vs. letting life exclusively dictate when the wierd/strange/unknown happens.
edit:' does result in lot more blue screen issues.
Going out and “pursuing” ideas often leads to some confirmation bias as we’re determined to “find” an idea, rather than sitting back and letting the lightbulb go off when you’re not actively trying.
To each their own though.
- A man who built a SaaS for local administrations and has about a ~50% share of the market. He spent 30 years working for the Spanish IRS in different positions. I don't know the insider story, but looking at the website he just made a SaaS solution with good UX in a space where UX tends to be atrocious.
- RatedPower's CTO. RatedPower makes a software that automates the design of renewable power plants. This guy had been drawing the plans by hand as a renewables engineer, realized a computer would be perfectly equipped to do it, and the competition in the space had grown complacent and barely innovated in years, so he did it.
The older I grow the more I get excited by boring ideas. People pitch ideas to me frequently and most of the time I literally end up thinking "not boring enough"!
Plus, a lot of low hanging fruit stays there because no one looks at it. Everyone is looking at x, y, and z, and there are a thousand variations of those, and they ignore a, b, and c because no one brought it up.
1) aka two see m is to b without [need for financial] backing.
2) sed e/two/to/ | tr '[a-z]' '[A-Z]'
sed e/m/money/ | tr '[a-z]' '[A-Z]'
sed e/b/be/ | tr '[a-z]' '[A-Z]'
3) crap -- type face doesnt do this justice --- gotta go re-invent tex for ycombinator
4) ha ha ACME; errr.. sed e/ol/al/
sed g/ol$/lol/
sed g/al$/ai/
sed g/capit/car(cdr(ulate)) /
^Gli movie ref[1]
^G ed ^D
So would be nice to have a PG bot without all of that.
Short and concise, this combines two of my most favorite mindshares.
1. "Hackers and Painters" – I stumbled across in late 2004 at the Technical University of Vienna's (excellent) bookstore. Reading it changed my life.
2. "Ideas are the Enemy of Observations (2012)" – Just Three months ago yamrzou shared this gem of a text. For me, it has become the most meaningful post ever shared on HN.
"How to get new ideas" seems to fit snuggly right in the middle :)
The essay writer's main problem is that they are attempting to infer causality from observation or to find an explanation of the data that justifies or supports their worldview. After all, "why?", "how curious", "I wonder how that happens" are powerful questions and statement that have led to countless inventions, investigations, novels, and pleasurable, although at times frustrating, time.
I would recommend something more scientific: observe the world (data/observations), develop a hypothesis that is hard in Platt's sense (meaning it should be discriminating) and test the hypothesis on new data and observations, where new means not used to develop the hypothesis to be tested. And they will see that they should not only collect observations that describe the world, but develop and test hypotheses that can test the mechanisms that lead to those observed facts.
Sometimes this is possible, sometimes it is not. But it is always entertaining.
Try it! Generate typical HN comments to this PG post [ … ]. - Include both positive and negative comments. - add replies that focus on annoying details that draw attention to the commenter - include fan boy defence of original post - point out obvious exceptions
* Overture, Picasa and many more OGs.
It starts with finding the right problem/need and before jumping into to solutioning, the smart person would validate and pitch the problem to as many relevant target customers as possible. Once the problem is validated and tangible root cause is flushed out then it's time to explore the solution and pitching it to the same customer research group for validation before going into building mode.
Finding ideas is not hard, but the process is gruelling, frustrating and potentially rewarding. Good luck!
How to fill those gaps? Write.
Here it is:
How to Get New Ideas
January 2023
(Someone fed my essays into GPT to make something that could answer questions based on them, then asked it where good ideas come from. The answer was ok, but not what I would have said. So I asked: What would I have said?)
The way to get new ideas is to notice anomalies: what seems strange, or missing, or broken? You can see anomalies in everyday life (much of standup comedy is based on this), but the best place to look for them is at the frontiers of knowledge. Knowledge grows fractally. From a distance its edges look smooth, but when you learn enough to get close to one, you'll notice it's full of gaps. These gaps will seem obvious; it will seem inexplicable that no one has tried x or wondered about y. In the best case, exploring such gaps yields whole new fractal buds.
so, 'through the looking glass' by another name?
edit: at time of writing, in wrong hemisphere for summer's day.
If I'm correct that PR review is currently much less efficient than it will be soon, it won't be because I'm smart or this is a new idea. It would just be because our company has spent last five years building code review tools, right place right time. Eventually there was enough infrastructure accumulated (and ambient events unfolding, ie OpenAI) that it became a small step to pass the edge of what PR review meant circa 2022.
My approach to finding ideas is to think, write and use words I like to explore creatively and join arbitrary ideas together. Writing and talking IS thinking itself!
I even journal all my ideas in the open on GitHub! I share all my ideas with everyone, including startup ideas. Links are on my profile. I LOVE SHARING IDEAS.
I enjoy reading people's problems to encounter inspiration for problem solving. I would read Slashdot when growing up to learn about computer issues people were solving.
Some words I enjoy are "mesh", "parallel", "multiplex", "multi", "schedule", "marketplace", "tree", "additive", "auto", "query", "traversal", "graph".
Also recongize that inventors would be excellent in story writting when asked to produce a recipe for creativity. So the step by step instructions may be nonsense.
Creativity, the far reaching one, not the incremental one, is topological adjunct to insanity. Which is why the radical new things, only get funded in times of desperation. If I were a inventor, i would be heading for ukraine or russia right now.
During the apocalyse, it got really bad. Probably due to isolation and so I didn't have my usual inhibitors. ("Being polite company" turn upkeep.)
--
The one strategy I have "to let it go" for my idea factory is to journal. Write however much of the notion down in notebooks that emotionally allows me to mentally stop chewing on it. Sometimes just a one liner. Sometimes long white papers.
So I'm not hyped about ChatGPT at all
> much of standup comedy is based on this (anomalies)
I think it's more about capitalize popular rants though. Not sure most of them are even feasible things to fix.
It's an ok essay .. though nothing sparks new idea of thinking about new idea.
It's not entirely clear whether or not the essay was written by GPT because it's not clear who PG asked. Himself or GPT?
I'm leaning toward the former, but can't rule out the latter. Maybe that's the point.
When you start from that, only the ideas that most people would never consider float to the top.
It's worked for me and my business.
edit: perhaps a stippling[1] to stipulation [2] ratio tuppled with relevant/appropriate pruning technique(s)
[1] stipplism vs pointillism : https://www.davidaccurso.com/the-art
[2] document of concurrence : https://www.thesaurus.com/browse/stipulation