It really fascinates me how the "other" clientele of HN operates. The world of networking, marketing yourself, creating a resume of sorts through your blogging. I'm just some dude with a decent job occasionally writing blog posts because it amuses me. I'd be pleasantly surprised if a dozen people read them before I die. AI doesn't really factor into the calculus of whether or not I want to continue doing it at all.
Yes. I don’t really understand the “you will be replaced” mentality. I do things because I like doing them not because they have some optimum economic value.
Like read the ad copy of old apple or HP or xerox and see how they wanted to help make tools that help you think. Today everyone wants to make tools that help you make money. I just don’t see why my life needs to be optimised for making money compared to engaging in interesting ideas and creating peace and happiness in the space around me.
I love designing posters for thé love of it. AI helps me learn new shortcuts or suggest tools, helps me delete this or that, but essentially - my appreciation for the process comes from actually moving things around myself and having a say myself.
With generative art/writing, well honestly - the author is very very limited in what he can command the NN do and particularly if the author has a string vision.
Same for music - I wouldn’t let AI arrange my Renoise tracks, I’d love to do it myself. Is love to create the VCV patches by carefully tuning each knob. And particularly when I actually know/imagine/envision how it should sound.
It takes understanding of underlying tech and love for the work. Love - very important.
For people lacking vision and creativity, those in a hurry, those under pressure to do shitjobs they hate - it idées may be better to use AI. But not for the real artist.
What are your favorite design posters from "the good old days"? haha. Can you share the posters you have made? It's cool that you make such things.
```
I just don’t see why my life needs to be optimised for making money compared to engaging in interesting ideas and creating peace and happiness in the space around me.
```
This hit me hard! Thanks a lot mate for this comment.
External validation and being useful to others matter. A world of everyone playing violin in a closet for nobody is a dystopia.
This view that you can't just do an activity for your own joy, you have to do it for the approval of other, is the actual dystopia. I wonder how old you are because this seems very much like the attitude of someone who was born in the age of social media.
That would be sad. Did you mean to respond to some post that was about that?
Yes, I responded to yours.
Weird, since that notion appears nowhere in my post.
Given that I'm not the only reply you received reading it that way, perhaps you should consider that you did not communicate what you think you did.
Please point to any part of my post that reads best as "you can't just do an activity for your own joy".
You can just admit you made a mistake. It's fine.
> a dystopia
i think this part gave people the impression that you meant that doing something "for nobody" (eg. for your own joy) was somehow bad/malevolent
See the comment below.
You can just admit you communicated poorly. It's fine.
My niece would disagree. She loves playing violin just for the playing of it.
It's not that nobody enjoys doing anything just for its own sake—it's the destruction of ways to provide for others, socially. Humans really, really like doing that. The two things may as well be completely different activities, scratching totally different itches. No amount of solo activity will make us feel whole in the same way feeling that our abilities and passions are helpful to and desired by—not just humored by—others we care about will.
You cannot share happiness with others if you don’t create it for yourself first as you will have no happiness to give away.
only inasmuch as a world where everyone playing violin in the streets is a dystopia, too...
A gentleman is someone who knows how to play the bagpipes but chooses not to. May also apply to violins.
Seeing value only measured by utility seems to sidestep the reality that not everything that counts, can be counted.
In startup land, the world of innovation accounting for organizations is one attempt to better understand that.
> I just don’t see why my life needs to be optimised for making money compared to engaging in interesting ideas and creating peace and happiness in the space around me.
I feel the same way. But I think this is kind of forced on us nowadays. For example, I might want to write a blog post to share my ideas/work/whatever. But it will be used to make money. It will be used to feed my competition. Maybe not necessarily my content, or not necessarily directly compete with me. But as a class of people, by creating more content we are feeding our competition and some minority will make a lot of money out of it.
Maybe I was not doing it for the money in the first place, maybe it was being read by 5 humans in total anyways. But still, I have no escape and I personally find the whole idea disgusting. I will take no part in it.
Yes I agree. But also I just don’t know what to do about it. I mostly try to make things more private. Like I write a newsletter about my thoughts on ai/ml, physics, etc that’s only for my research institute sent through the email listserv. Maybe nobody reads it but also nobody monetises it. And I get an outlet that is hopefully useful to others I’m directly connected to.
> I just don’t see why my life needs to be optimised for making money compared to engaging in interesting ideas and creating peace and happiness in the space around me.
This means your are rich, and you can afford the luxury of such mindset.
> This means your are rich, and you can afford the luxury of such mindset.
The people I know that are focusing on "engaging in interesting ideas and creating peace and happiness in the space around [them]" are the "poorest" I know of.
They chose to have a part-time job (if they have one at all), they chose to live a simple life (from Western standards)... to do what they care about.
The only luxury they have is living in a society with enough solidarity to allow them to live this way.
I'm not rich and I have this mindset. I make far less than most people on HN do, less than industry average too, but I've been able to live an okay life regardless.
The goal was never to get rich as I realized early on that was improbable and I focused on other things.
I have met a lot of people who live in poverty. They're not all that interested in making money (although they are not necessarily averse but it's definitely never an all-consuming priority). By and large their focus is on human connection.
It's not the default human condition to be consumed by greed any more than it is any of the other deadly sins.
Exactly. People are trying to make ends meet. They aren't trying to optimize for money making for fun We're trying to eat today, and maybe someday retire, if we're lucky. It's harder than it used to be, so people are looking for strategies.
It's not some morally superior attitude to "engage in peace and happiness" blah blah you're just rich congrats
There is a very large gap between being "rich" and having to turn every moment into money so that you can eat beans and rice. The general point upthread is that for many/most people, not everything they do has to be a "side hustle."
(And there are probably quite a few people here who are pretty comfortable who still look at a lot of activities through the side hustle lens.)
People who are just trying ro make end meets with no hope of improvement do often adopt this thinking. It is that or alcoholism, drug use and gambling.
People need meaning in their life and work that mistreat you is not that.
I suppose I rather appreciate the lot of you who believe I am wealthy and therefore able to find happiness in life. I suppose I’ll have to disappoint you a bit and tell you that I am not wealthy and certainly not wealthy by most standards on HN. I simply accept that it’s my job to create my happiness. Nobody else will do it for me.
I might be the kind of person that requires external validation, what I noticed is I simply stopped writing online, for the most part. It just feels like writing/tweeting/etc into a cacophony, makes me slightly uncomfortable adding to the noise and discouraged that nobody (no human at least) will read anyway. So I just write what would be blog posts on Notes app and don’t post anywhere…
I read this comment.
That said I have also found myself writing less, but not because I think it’s less worthwhile than it was 15 years ago when I started, it’s that the internet feels like it has become less deep. Probably not because the internet has changed but because I’ve learned more so fewer and fewer things tickle that curious part of the brain and feel worth writing about. The things that do feel worth it are so off the deep end that there are fewer and fewer readers who are interested.
Writing on the internet sometimes feels like being a blob of uranium in a reactor where control rods are slowly being inserted at the same time as the reaction chamber is being expanded. There is factually more fissile material out there, but it's no longer as close to critical mass.
I am not a blogger, but was very active on StackOverflow for years -- mostly asking, but sometimes answering. I used it to learn and to create digital footprints, similar to blogging. I blame most of my slow down on experience and age (same-same, but different). After a while, my questions for SO became so hard, they would only get a few views and no answers. I'm no genius, but with enough experience, you will stumble across some never-before-answered problem.
So you actually enjoy writing, without getting anything in return. Sounds like it'd be the good kind of blog (as opposed to content marketing, personal branding and all that hustling). If you were to publish it, maybe one person will come across it and get something out of it.
There's this Spider Man quote I like: "If you help someone, you help everyone". I think it's not comic canon but just from the PS4 game. So getting meta here, that random line a writer for a random video game came up with had a big impression on me.
It's a typical programmer fallacy to avoid redundancy. If somebody else already wrote about something, why would you? Yet in communication, redundancy and repetition is actually quite key. We need to hear ideas multiple times and from different angles before they land.
I used to blog and write a lot and I never cared how many — if any — read it. Purposefully avoided analytics tools etc.
But after the last few years with the proliferation of “AI” tools and the increasing amount of noise on every level I just don’t like feeding the “grey goo of information”. It might be unreasonable but I’ve felt it for over a year now and it’s not going away. Instead I value interpersonal conversations a lot more again. I hang out in discord voice chats with a few people at a time. Text communication feels soulless and low signal to noise in general now.
Anecdotally almost every text chat server I’m on has less active users writing than I’ve ever seen in 25+ years of using the internet. Might be a coincidence but I wouldn’t be surprised if people’s behaviours are changing. Just like knowing you’re being watched changes your behaviour, knowing text content may or may not be fed to or generated by a slop machine algorithm probably changes how you view text as well.
"I just don’t like feeding the “grey goo of information”"
If your voice isn't spoken then it will go unheard. It is of course up to you but I think that blogs and websites run by real people are invaluable (and I'd love to see what AI makes of that word!)
This is the bigger issue at the moment: https://lwn.net/SubscriberLink/1008897/bbe01d846ec5e9cd/
I used to write technical things. I would not be heard anyway.
I don't feel like giving free inputs to an AI company. They will use it and once they have domination they will abuse their position.
Scraping is a short term anomaly. The hype cycle is already dying which quickly reduces the amount of resources devoted to such efforts.
Where I think that post failed is poisoning the well of anyone ignoring Robots.txt is arguably a moral imperative. That post was concerned with wasting kWh, but ensuring that bad actors don’t profit discourages people from being bad actors.
My «voice» isn’t heard from a trained LLM anyway. Only a vague, blended and remixed imitation at best. As a human I have no interest in feeding that beast.
I can hear your voice, right here, right now!
You wrote your comment in English but used French style quotation marks and made a few minor grammatical errors in your first sentence (won't instead of isn't, by instead of from). Your second sentence beautifully paints a gloomy image with just nine words and your final flourish has just the right level of "fuck it".
If you can deliver that amount of character in a single line in a second language then please do feed the beast. We are all the product of what we eat. The beast needs to eat its greens and suck up some vitamins.
15 years ago felt very different to me as well. Many of us lucked into the conditions that made many online interactions feel worthwhile. The loss of these settings sucks.* But such a creative environment is still attainable and can be found in pockets.
* To be fair, nostalgia around many off these communities can be misleading. Many were never designed to sustain realistic growth nor the inevitable pressures to monetize them. They were naive and ignored human nature, a product of the mentality of their creators or perhaps their participants or both. We know better now. Community building is tricky and worthwhile, rarely a matter of formulaic scaling.
All of my best writing has largely gone unwritten.
It's the few highlights in the cacophony called life, that makes life worth living.
Me too. At least i can be wrong in private that way.
I have actually cut back on blogging (partly) because I don't want my hard work to be slurped up and regurgitated by an AI. I write for other people. Not for AI
People had been stealing ideas long before computers started doing it.
> People had been stealing ideas long before computers started doing it.
So it’s not ok to be bothered by it because it’s been happening a long time?
you're not allowed to make fire. get inspired by music and make it yourself. write anything in any known written language.
ideas are there to inspire and to freely go from one to the next and evolve.
this ownership fantasy people have of knowledge and ideas is a deteriment to human progress.
its a new thing. not an old thing. and its only there for people to take money of others.
do you realy think its a good thing? its a large part of what makes so many people poor and hungry in the world.
Operative word being "you", not "a blender that predicts what a perfectly average person would create". AI is a blight on creative works.
Do you think you might be being a touch reductive there? Perhaps ignoring factors like context, medium, the economic realities of who makes this “data” and who ends up making money off of it at their expense, the difference between mechanically harvesting data because it’s there and internalizing things through the lens of your mind and existence and making it your own… you know… pretty much everything about it except the razor thin slice of the situation you’re choosing to take into consideration? Any complex and nuanced situation can become cut-and-dried if you ignore enough.
I can't imagine how culture could exist without ideas being taken without permission.
Straw man. I can’t imagine how culture could exist without some people being thrown in prison. That doesn’t mean that criticizing a law or law enforcement practice is equivalent to arguing for abolishing prisons.
We can agree that prison and appropriation are both necessary for culture, regardless of their negative effects.
Your problem is the scale at which knowledge and ideas are being appropriated, but my point is that it was already happening a lot but it was far more implicit; now it's just explicit that it's happening because we see the process laid bare.
No, that is not “my problem” with it. It is one facet of it, there are lots of sucky ways to take ideas that have nothing to do with scale, and there are lots of ways that fundamentally mechanically harvesting “ideas” as data is different than just learning quickly, but I’m not interested in re-arguing any of this for the 500th time on HN.
It really is getting tiresome. I guess a lot of HN commenters are directly or indirectly heavily invested in the AI bubble and so cannot / will not argue in good faith because they’re barred by personal interests.
I saw a hint of it with crypto and NFT hype cycles, but this is on another level.
It's rude to subtly accuse the GP of not arguing in good faith due to personal conflict of interest without any evidence.
I was enjoying the discussion until the parent broke it of with a comment that sounds like a SJW refusing to "educate" us.
Sounds like that contemptible SJW needs to stop being so judgmental. And it’s definitely rude to imply motives that someone doesn’t directly state, isn’t it?
exactly this lol
I consider it a honor if people are inspired by my ideas or thoughts.
This could in theory still apply to machines but I would have to think about it first. It might require a different format.
I welcome people to steal my ideas. I don't welcome the AI companies. They claim they want me to lose my job. Most people online don't want that for me.
Yeah, and the movie and music industries moved fast to have laws passed protecting their "ideas". The average person has no equivalent protection nor the means to pursue legal action. So opting out of the charade is a perfectly legitimate move.
And in clinging so hard to their 'ideas', the substance has slipped through their fingers and they've mostly reduced themselves to cloning their own movies.
I can't see how it's protecting creative risk; it protects the infrastructure for some to take risks, but that infrastructure isn't being used anywhere near its potential.
I'm no fan of the music/movie industry, but they abide by a set of common rules. If you take a piece of music and use it verbatim or close to it in your song, the original composer must be credited and the royalties shared.
It's still too much of a reach to expect regular people to want to contribute personal knowledge for free, into an opaque corpus that rewrites your words, sends them to someone else and arbitrarily decides whether you are to be credited with them.
I am a person (allegedly) and I would benefit from AI regurgitating your content.
Much the same way I benefit from Google indexing the internet, and summarizing news articles.
Reading and writing are both intimate activities. The reader holds the writer's thoughts in mind, and the writer knows this and acts accordingly. Personally, I don't particularly enjoy reading material that was made by an LLM. The fact that so little effort was applied suggests that there is not much reason for this to exist beyond a chance to serve a few quick links. Since the llm is also running this as a business, I would also point out the social connection between reader and writer does come with some expectation of a reward. Whether it is to be paid in cash or respect is beside the point. People often expect some reward for their efforts and they are not wrong to want that. People are often uncomfortable to put it so bluntly because that would compromise the quality of the relationship, but upending this relationship is really a perversion of logic carried out for purely selfish reasons ("I want what you are making, and I will give nothing in return").
I agree, but its not the only way to use an LLM. I tend to focus my time with ChatGPT on getting it to prompt me on good ways to do things. Like playing cooperative writing games, or coming up with coding training exercises.
>I would also point out the social connection between reader and writer does come with some expectation of a reward. Whether it is to be paid in cash or respect is beside the point. People often expect some reward for their efforts and they are not wrong to want that. People are often uncomfortable to put it so bluntly because that would compromise the quality of the relationship, but upending this relationship is really a perversion of logic carried out for purely selfish reasons ("I want what you are making, and I will give nothing in return").
I often ask Chat GPT to cite anything it comes up with that I want to repeat, and if it cant cite it I ignore it.
I have been reading blogs, books or whatever for free for years often (99%+) of the time without payment or attribution. I have never considered myself to have a relationship with an author, and often wonder about people who have parasocial relationships.
I dont understand how a new tool that can chew on the data for me before presenting it hurts peoples feelings tbh. I would be honored if anything I have ever done was worthy of inclusion.
>Much the same way I benefit from Google indexing the internet, and summarizing news articles.
wow there. You do note the big difference between AI regurgitating someone's content and Google indexing it?
If I put effort into creating something digital and people find it through Google's massive index of the internet, at least they'll see who its by, enjoy it within its context and have the whole thing before them. This is way different from some algorithm mashing what it scraped from a creator's work into its own rehash of a bunch of content.
You can understand why someone might detest the one and appreciate the other?
If I create, I might want you to benefit from it, but I might also want you and others to know it came from me. Not just consume it as an undifferentiated part of some parasitic corporation's AI slurry.
Google gives attribution and maybe provenance, while AI gives you smoke and mirrors. I guess we'll decide if copyright has any legs left to stand on in the modern world, or if it falls as collateral. It's so sad that commercial piracy has hit such an incredible tipping point that even I feel bad for creative people and their bleak economically dead future ahead.
I want to be optimistic, but I suspect what we'll end up with is a legal interpretation which allows LLMs to steal content from creators without the cool side benefit of being permitted to steal content ourselves. It will become legal for an LLM to turn your blog into a part of its library, but you won't be able to apply the same logic to read out-of-print books.
Sure, but I am the creator of the content. And I'm fine with Google indexing the content. I'm not fine with AI doing it. Especially since many of the AI's monetization goals are to make me, specifically, jobless
Makes me wonder, how about changing your blog to a mailing list?
That's no guarantee it won't get slurped up by an AI at some point. Anything that goes into, say, GMail is ripe for plucking. And there's always a good chance your newsletters will get publicly archived on some web page somewhere, whether intentional or not.
I had the same thought. Still, perfect or not, I bet it'll be an attractive option for some.
I guess our gmail content has been fed into an AI of sorts since many years ago. I would surely hope, however, that Google would not use it for any sort of non-private LLM training data.
Yeah maybe that's the move. I don't really have a following though, lol
I don’t think there has to be some practical economic business justification to a) feel bummed out by your creative output getting munged up into something that, for all its better uses, will feed the great fire hoses spraying trillions of gallons of bullshit all over our information landscape, or b) reduce your creative output because of it. It’s weird how entitled people feel to other people’s creative work and get mad when people don’t freely create for and share with them, while simultaneously minimizing the value that work and its authors bring to our society. Despite what many say, the way people receive and interact with your work mentally/emotionally is really important, and all your work being sucked up into these models— often to create commercial products that are openly antagonistic to the people that created the work that made it possible— changes that. It’s sad that AI has devalued creative processes even to the creators themselves.
You could hack the email addresses out of a website, make disposable email addresses and send your writings to everyone including the replies from the previous newsletter with your response to them.
AI doesn’t care, the people using AI care. If you really write for other people, I’d recommend you reconsider blogging again.
Even if you write primarily for yourself (vanity, marketing, client acquisition, there is nothing wrong with that) and not for other people, I’d still recommend you publish your stuff. Not publishing will have always <= effect than publishing, even if AI slurps it up.
JD Salinger took a lot of heat over his reticence to publish any more of his writings after Catcher In the Rye and the Glass family stories. In 1974, he responded to the NY Times, "There is a marvelous peace in not publishing. It's peaceful. Still. Publishing is a terrible invasion of my privacy. I like to write. I live to write. But I write just for myself and my own pleasure. I don't necessarily intend to publish posthumously, but I do like to write for myself. I pay for this kind of attitude. I'm known as a strange, aloof kind of man. But all I'm doing is trying to protect myself and my work."
Isn't there a robots.txt equivalent for AI?
You'd probably have a better time using something like Cloudflare to block known scrapers than trust companies that pirate terabytes of books to train their models to obey a "please don't" text file.
https://blog.cloudflare.com/declaring-your-aindependence-blo...
It’s not respected, there’s a lot of posts on HN about this.
Just one quick reference: https://www.tomshardware.com/tech-industry/artificial-intell...
What about the idea that writing for the AI is your best opportunity to help contribute to it's direction?
Where's my cut of the AI profits?
Nice retort, but serious question: Except for NVidia (and maybe AMD's GPU division), is anyone making any money from AI/LLMs yet? I think no.
SV companies have that cycle where they give things for free until peoppe get dependent or until they killed competition. Then they switch to using every trick in a book to make you pay.
what will happen if it was? some kind of end of the universe event?
i really dont think its such a big deal. If you enjoy blogging, then enjoy it. Maybe its an excuse, AI, for something that was already in the making.
I’m with you, my blog is a static site with no JS hosted on GitHub pages. Unless I ever see anyone discussing it or linking to it I have no way of even knowing if it’s being read or not. I write for me, mostly.
> I'm just some dude with a decent job occasionally writing blog posts because it amuses me. I'd be pleasantly surprised if a dozen people read them before I die.
That is exactly the right approach. Most of the posts on my blog have low 3 figure hit rates after being up for a decade (the long tail is a joke) - that averages to maybe 20 hits a year. But that is not that point. I enjoyed writing the posts and just maybe somebody, somewhere enjoyed reading my hot take on The Last Jedi or whatever.
I don't really understand the people who blog for "exposure" or money - it seems like such an effort for very little material gain.
I guess I am in the same boat. I blog because I have this creative urge to write about subjects that interest me, and put it on the web because why not. AI would only accomplish taking away the part that I find "fun" about it, the reason to do it in the first place.
The fact that you see it this way, actually makes your writing so much more interesting. It's genuine.
this is the spirit :). i dont mind ppl using AI or not actually. for some it really helps their writing. its what it exists for imho. but i do appreciate a good writer. other then that, blogs are about the topic, not it being written in some uber writing style.
for the writer, id hope like you it gives them joy, because that usually is something which gives it some spirit and joy to read also. its not important at all how things are worded if there was fun to be had!
Yeah, it's very interesting. To me, this is like saying "why write a journal when AI can generate years worth of journal entries?"
Not even coherent, really. That's not why I write at all. It's also not why I read.
Honestly, that's probably the best approach: writing because you enjoy it, not because of algorithms, engagement metrics, or personal branding
Is your job super-safe? If so, that's awesome :-) The whole marketing thing only becomes important if you have to get a new one, and then it can become important very quickly.
That's where it turned for me. Originally, I had started a small tech-topic blog with the idea that it would be my portfolio because I really wanted to write for a tech publication, most because I thought I had the chops for it and I want a job where I can travel and work.
Things started off okay, me writing about my projects, etc on a small self-hosted site with zero analytics, keeping things small and manageable in my free time. But the lack of feedback sort of left me I limbo. Was I writing in an engaging way? Were my subjects interesting to more than just me? I had no idea. Eventually, that iteration of the blog got deleted.
And I made another. And another. And so on.
Til I landed on the current version, which is basically me just faffing about with a editorials about tech for fun since I have little time for actual projects anymore, let alone the accompanying writeup.
I still want that writing job, but I realize how much of a pipe dream it is, now. Tech bloggers were already a dime a dozen before I showed up and genAI only saturated that market even further. That, and I still have no interest in working for or hosting a site that is hostile to my reader by being a bloated sludge of scripts and sloppy use of frameworks, which limits my market for a writing career in disappointing and obvious ways.
When I see discussions like this pop up about writing online in today's landscape, it seems to always come down to "write what you find interesting or fun, but keep your eyes expectations near zero" which seems so self-defeating considering how much work it often takes to maintain a blog while you also have to tend to real life. As much as I loath places like Medium or Substack for asking for money up front, I do understand why those writers choose to go there instead of walking my lonely path.
Writing is thinking. So is drawing.
To think clearly, come up with new ideas, make and truly understand things, we need to put marks on the blank page ourselves, and not just repeat what teachers or textbooks tell us like the majority of students Richard Feynman had during his time in Brazil — https://v.cx/2010/04/feynman-brazil-education
LLMs/AIs are useful to help us get farther, faster, like witty, skilled, intelligent friends who sometimes take too many magic mushrooms during conversations.
Forgetting about our own agency and individuality is bad for us, and dangerous for society.
“I am no bird; and no net ensnares me; I am a free human being, with an independent will.” —Charlotte Brontë, Jane Eyre
“Monsters exist, but they are too few in number to be truly dangerous. More dangerous are the common men, the functionaries ready to believe and to act without asking questions.” ―Primo Levi, If This is A Man
To create and be free like an animal outside a cage, ask, write, and draw your own questions. Look, and find out for yourself, rather than blindly believing what others tell you.
Two useful books:
Drawing on the Right Side of the Brain by Betty Edwards https://archive.org/details/DRAWINGONTHERIGHTSIDEOFTHEBRAINH...
The Hand - How Its Use Shapes the Brain, Language, and Human Culture by Frank R. Wilson https://www.penguinrandomhouse.com/books/191866/the-hand-by-...
For anyone put off by the dubiousness of the left/right brain thing, rest assured that Drawing On the Right Side of The Brain is no less useful as an introduction to abstract visual creativity because of it.
This comes back to my initial thought on the effect of AI on the "producer" rather than the "consumer".
I do a lot of things badly because it helps me develop skills and makes me happy. I wouldn't outsource this to an AI even if it did all these things better and the world benefited more from it. This is for me.
There's something deeply human about putting thoughts into words (or images) and shaping them into something tangible. AI might help speed things up or spark ideas, but if we rely on it too much, we risk losing that process of real discovery.
feynman-brazil-education was an amazing read. Thank you.
I am from India, and I have a similar experience with my education — one that forces you to memorize, never experiment, and never connect the dots. It felt like reading about my own past and realizing just how bad it was.
I have seen the same in Sri Lanka. It was not as bad in my day and I was mostly educated in Britain.
Sadly, now the UK is becoming more like that. Schools are judged by league tables based on exam and test results. My daughter's sixth form college (school for 16 to 18 year olds) has deteriorated quite a bit since her older sister went there five years ago and its pretty clear to me that is the cause. A lot of other schools seem to be the same or worse.
>Writing is thinking. So is drawing.
As a general rule, any kind of explaining done with the intent of making the recipient understand the _concept_ will require oneself to have though and understood the concept. Explaining could be in the form of writing, drawing (as you eluded to), verbal etc.
Drawing on the Right Side of the Brain is discredited pseudoscience bullshit. The exercises will make you a better drawer but not for the very stupid reasons the author claims, but because they are drilled exercises.
> discredited pseudoscience bullshit
Care to substantiate? If the contention is on specifically which hemisphere of the brain is responsible for drawing ability, this is besides the point. The author even says
> Since the late 1970s, I have used the terms L-mode and R-mode to try to avoid the location controversy. The terms are intended to differ- entiate the major modes of cognition, regardless of where they are located in the individual brain.
The entire left brain/right brain premise that the book is based in is totally bogus. The author constantly refers back to this nonsense theory throughout the book (yes I have read it). What she refers to (constantly) as "R-Mode thinking" simply isn't real. The book works as a learning aide because she largely takes regular learning techniques and drills, and dresses them up with nonsense scientism.
https://www.smartermarx.com/t/regarding-betty-edwards-drawin...
> You're building up a portfolio of writing about topics that interest you.
This reason resonates with me immensely.
You're not just writing about what you've figured out, sometimes you're actually deepening your understanding as you write! Writing is the thinking process: https://news.ycombinator.com/item?id=32628196
I have been writing every day at my blog for three years now, and it's been very rewarding for me to figure out what I actually care about and seeing patterns.
I like thinking about it like a bunch of skateboarders lugging the video camera around to capture the moment. (They did this before social media!)
P.S., You may also enjoy the similar sentiment in this post: https://news.ycombinator.com/item?id=42992159
> I like thinking about it like a bunch of skateboarders lugging the video camera around to capture the moment. (They did this before social media!)
We sure did! I’ll never forget my first vx1000 with a death lens.
VX1000 was peak camera in skateboarding. The one modern camera that seems somewhat close has been the whatever Lumix that the dudes at April Skateboards use. It's hd but has a 4:3 ratio and looks VX like in my opinion
https://www.youtube.com/watch?v=GuXqrq7aMO4
PS: I run skatevideosite.com, if anyone is interested in helping dev the site and skates hit me up!
For me, the #1 reason I don't blog more, especially about tech topics, is that they take too long. Maybe you can bang out a useful blog post in 20 minutes. For me it's more like 4 to 8 hours.
I have to make samples. Since I do mostly web tech I want the samples to actually work, no "here's some code, trust me". I also need diagrams. And, I have to proofread since I'm terrible at getting it right in one or even 5 checks. I write once, add samples, write some more, add images, write some more. Every time I write I add errors, so it always takes multiple passes.
Proofreading is one place that AI can actually be a friend rather than a foe. If you give Claude your draft and tell it explicitly to call out misspellings and grammatical errors only, it does a really good job.
Much like (say) MS Word does? In real time.
AI is a lot more powerful in this aspect.
MS Word would find no qualms in this sentence:
"I started the car and went for a drive on the highway. There were many other cats on the road but it was nevertheless agitating."
Given the correct prompt (that avoids changing your literary style altogether), AI can quickly suggest cats -> cars and agitating -> peaceful, since it's much better at contextualizing.
Proofreading is easily done with an editor. I think AI is much more useful for giving critic and advice on how you write your sentences. Setting the tone, refining the main idea, pointing out redundancies are some of the things that I find very useful.
Sometimes, but you have to be careful. IME Claude and (to my surprise) Grok 3 are really good at understanding your style and adapting their suggestions to match. ChatGPT, by contrast, tries to change everything into some kind of corporate drone.
Another thing is to know when to stop. AI can endlessly refine your sentences if you let it.
My drafts usually live for days, weeks, or even months before I publish them. (And sometimes I throw them away after having worked on them for many hours.)
My advice is to "chill": focus on the process instead of the result and let the posta take the time they take.
I like writing too. Funnily enough, the age of LLMs makes this even better. I wrote a little MCP server (this is trivial with Claude) that interfaces with my blog so it can full-text search for articles and look up articles and look at recent articles and stuff like that and it is pretty good at finding references in what I've written to thoughts I've had. It's a bit trigger-happy when looking up my blog posts (I have to put more in the assistant prompt in the Claude app to get it to stop defaulting there).
The other thing that's nice is that LLMs make the process of writing better. When I cite stuff I can just screenshot the website and ask ChatGPT to write the citation and then check it. Things like that are more painful to write than to check and LLMs shine there.
Blogging pushes me to explore things I probably wouldn’t otherwise. That’s been the main reason I’ve stuck with it pretty consistently[1] for the past five years.
Getting attention was never the goal, so the rise of LLMs has mostly been background noise to me. There have been plenty of times when I’ve searched for something on Google, only to land on my own page.
Over the years, though, things picked up. Now, I’m seeing around 30k monthly readers—way more than I ever expected. More than once, I’ve written about something I barely understood, only for the post to hit the front page. Then people corrected me, and I learned a ton in the process. That’s something I wouldn’t trade for anything.
[1]: https://rednafi.com/
Me too! These last two posts blogging about blogging are unusual for me. I'm working through a book (Sebastian Raschka's "Build an LLM from scratch") and posting about that at the moment. It's likely not a coincidence that I'm procrastination-posting before going through the trickiest bit...
I love reading meta-writings at times, as long as there’s a real human behind the keyboard. This was a fun, quick read.
Really enjoyed reading about your blog stack, motivating me to get my own up and running.
Love the blog!
Thank you. One reason I wrote it was to demonstrate how easy it is to spin up a blog where everything is automated and you never have to worry about the infra.
I’ve tried building a static react blog and hosting it on vercel and while it was easy to set up there was just too much styling and configuration to sort through that by the time it came to writing i was pretty much unmotivated. Markdown seems to be the key here, going to try spinning one up tomorrow. Cheers.
Yeah, landing on your own blog post during a search is always a funny (and slightly surreal) moment!
Funny indeed. I recently needed to set up GNU stow to manage my dotfiles on a new machine.
So I searched google and lo and behold, my own stuff was among the top 10 results.
https://rednafi.com/misc/dotfile_stewardship_for_the_indolen...
I think that as more people offload understanding to LLMs, being able to deeply understand a topic will make you stand out more and more. Doing things and explaining them are two of the best ways to get that deep understanding.
When I write about a technical topic, I open a new markdown doc and just go. You quickly run up against the limits of your own understanding, which is a valuable exercise.
Exactly, reading and consuming information is one thing but teaching it to someone else is something else entirely. If you're not writing with the goal of self-marketing and content-farming, it's still worth it.
You're conducting multiple parts of a dialogue when writing, but discussing what you've learnt can also be quite a good way of solidifying learning and encouraging further thought.
I started blogging to record the history of projects I've worked on (e.g. Microsoft Band https://meanderingthoughts.hashnode.dev/history-of-microsoft... Launching HBO Max https://meanderingthoughts.hashnode.dev/only-300-software-en...) but, perhaps not surprisingly, my only "successful" article has been the one controversial article I wrote cautioning against using SSR for everything.
That said I believe documenting history is important so I'll keep sporadically writing down notable events I've been involved in!
I also blog a fair bit about AI, and there is no hope getting views there without playing the game.
I think there is a more important reason to blog besides the 3 reasons listed: to force yourself to slow down, organize your thoughts, fill in the holes, and articulate your points.
"Writing is understanding."
"Writing is nature’s way of letting you know how sloppy your thinking is" - Guindon
Though admittedly I first heard it from Leslie Lamport and that's who I associate it with.
That's what I was trying to cover with the "make your newly-acquired knowledge concrete" bit, and was my focus in the previous post. This time around I wanted to look into the aspects that might be impacted by AI (and why I didn't think they would be).
You don't really need to blog for that though? You could just write to a text file on your PC and not throw it online.
Socrates would like a word with you.
I find that thing where people say "I'm not going to publish anything creative ever again, it'll just be used to train AI" so depressing.
It feels like such a dismal excuse to avoid andding any value to the world.
You calling it a "dismal excuse" is an emotional hand-wavy dismissal that doesn't actually answer the fear or solve the problem.
Moreover, "the world" isn't a thing that you can add value to. You can add value to other people by sharing works with friends, or you can add value to AIs that are going to be used to replace you without getting compensated for it. One of those should be obviously bad.
The world is already flooded with more "artistic value" than you can ever experience in your life. This market has been over-saturated forever. And it's all free of charge because "building a portfolio" for a chance of getting paid in the future is the go-to strategy. It was only a matter of time for a new predator (AI data collection) to arrive and exploit this situation. A handful of artists trying to react to this shouldn't be alarming. Good for them. Not because more free publishing would hurt them, but because they can invest their time better than into whatever the marginal benefit of posting the 50th free artwork might be.
It's not an excuse, it is a reality. Why spend your personal time and effort for someone else with a deeper pocket to automatically extract value from your work.
There is certainly a line where if you're popular enough and have significant google juice you'll still get organic traffic, many small bloggers can go their entire posting history without getting more than a smattering of hits and now chatgpt is taking away that.
"Why spend your personal time and effort for someone else with a deeper pocket to automatically extract value from your work."
That's the exact attitude I'm talking about.
Because creating things is good! Because it's good to put value out there into the world, even if someone else might also use it.
I think it can significantly change the harm-benefit calculus. (But I'd love to be wrong.)
In the past, I could be fairly confident that if someone else uses my work (and I want them to do that! that's the point of sharing!) the good that it causes will outweigh the bad that it causes. It's not like I'm helping people make missiles.
But now it's entirely possible (especially if my content is unpopular, such that LLMs make a larger proportion of its readers) that the bad outweighs the good, given the negative effects that LLMs have had and continue to have on our world.
Missiles save lives too.
Why spend your personal time and effort for someone else with a deeper pocket to automatically extract value from your work.
People releasing their code under MIT or BSD licenses might be able to give good answers to this.
Good answers like "it looks cool in my CV that big company XYZ uses my MIT licensed script"
It's extremely dishonest to compare someone voluntarily releasing their work under a permissive license with someone who is involuntarily having their content and effort stolen by an organization training an AI.
And I think it's erroneous to say having publicly disseminated content being read in an LLM training process is "stealing."
If I read a publicly distributed, copyrighted blog post of yours, learn something, then use that knowledge later on, did I steal your content?
If an author distributes something in public, the public is allowed to read it and learn from it, whether with their eyes, a screen reader, AI agent, or whatever. Any copyright violation occurs if they attempt to republish your content, not in the reading of it.
However, scraping illegally obtained non-public material - such as books an author is trying to sell or blog posts behind a paywall - could well be a violation unless access is obtained legally.
> And I think it's erroneous to say having publicly disseminated content being read in an LLM training process is "stealing."
It's clearly theft-adjacent. You're free to use "piracy" if you want, as long as it's clear that it's illegal and morally on the level of theft.
> If I read a publicly distributed, copyrighted blog post of yours, learn something, then use that knowledge later on, did I steal your content?
It's also extremely dishonest to compare AI to humans like this. AI are not people - morally, socially, biologically, or legally. What a human does with a piece of content is utterly irrelevant to the process of training an AI.
> If an author distributes something in public, the public is allowed to read it and learn from it, whether with their eyes, a screen reader, AI agent, or whatever.
Again - very dishonest to conflate a pre-trained AI agent (such as OpenAI's Operator) with the training process.
> Any copyright violation occurs if they attempt to republish your content, not in the reading of it.
OK, this is just factually incorrect. It is a violation of copyright law to make copies of copyrighted content, with very limited and case-by-case fair use exceptions - the claim that violation only happens in the republishing case is completely false.
This entire defense is a mix between deceptive and flat-out factually incorrect statements.
Your repeated use of the word 'dishonest' seems odd to me. I infer you think I'm making arguments disingenuously and without believing in them and/or manipulating the truth. I can reassure you this is not the case. I sincerely believe you are making your own arguments honestly also, and am engaging with them in that spirit.
as long as it's clear that it's illegal and morally on the level of theft.
It's not clear. I do not consider training an LLM on publicly disseminated text to be "morally on the level of theft." Stealing my car, or even a pen off my desk, is a much more reprehensible action than slurping everything I've shared in public into an LLM training process, purely IMHO.
It's also extremely dishonest to compare AI to humans like this. AI are not people - morally, socially, biologically, or legally. What a human does with a piece of content is utterly irrelevant to the process of training an AI.
People or corporations (which are usually treated as person-like) operate training processes and are morally and legally responsible for them. I believe training an LLM is "a human/corporation doing something" with a piece of content.
Again - very dishonest to conflate a pre-trained AI agent (such as OpenAI's Operator) with the training process.
Again, I am being honest. Whether I let an AI agent read your blog post or whether I write a program to read it into an LLM fine tuning process seems immaterial to me. I am open to being convinced otherwise, of course.
It is a violation of copyright law to make copies of copyrighted content, with very limited and case-by-case fair use exceptions
One of those exceptions (in many jurisdictions) is making temporary copies of data to use in a computational process. For example, browser caching, buffering, or transient storage during compression/decompression.
While many of the "pile" style of permanently stored and redistributed datasets are more than likely engaging in copyright violation, that's not inherent to the process of training an LLM, the topic of this thread. I believe that if copyright holders want to go after anyone and have success in doing so, they should go after those redistributing their content in such datasets, not those merely training LLMs which is not, in and of itself, violating any laws I can establish.
> It's not clear. I do not consider training an LLM on publicly disseminated text to be "morally on the level of theft." Stealing my car, or even a pen off my desk, is a much more reprehensible action than slurping everything I've shared in public into an LLM training process, purely IMHO.
The theft is that of effort, in the exact same (or a worse) sense as pirating media or stealing IP from a company.
It takes effort to write. That effort is being stolen by an LLM during the training process - the LLM cannot possibly exist without the work done by the authors who wrote content that it is being trained by, and the LLM can also be used to automate away those authors' ability to do work (and jobs) by replacing them. Which is worse - to have your car stolen (which is very bad, I'm not arguing that it isn't), or to lose your job, and not being able to afford anything?
Alternatively, if you believe that it's not bad to take someone's effort without their consent and without compensating them for it, then you shouldn't object to your employer withholding wages from you, or a client refusing to pay you, on the same principle.
> People or corporations (which are usually treated as person-like) operate training processes and are morally and legally responsible for them. I believe training an LLM is "a human/corporation doing something" with a piece of content.
That's not reasonable, and most people do not share your opinion (including the relevant group, which is the authors of the content being trained on). That's equivalent to saying that a human writing a program to perfectly reproduce a copyrighted work (e.g. print out the complete text of Harry Potter) is a human "doing something" with that copyrighted work (in the same class as reading Harry Potter).
> Whether I let an AI agent read your blog post or whether I write a program to read it into an LLM fine tuning process seems immaterial to me.
Those are categorically different. The vast majority of the population (again, including those writing the works that are being trained on without their consent) will agree that they are categorically different and incomparable, and they are logically, legally, and morally distinct.
> One of those exceptions (in many jurisdictions) is making temporary copies of data to use in a computational process. For example, browser caching, buffering, or transient storage during compression/decompression.
To use in specific computational processes for which you do not store the output because the output is subject to the same copyright laws. The implicit premise when you talk about training is that you're going to save the trained model, so this obviously doesn't apply, in the same sense that if you take a copyrighted work and transcode it, the transcoded output is subject to the exact same set of copyright laws as the original.
> not those merely training LLMs which is not, in and of itself, violating any laws I can establish.
That's the "law is morality" fallacy. Morally, this is clearly wrong, the point of the copyright system is to prevent exactly things like this from happening. The courts have not yet decided whether training an LLM is "copying" a copyrighted work, but if they do, then it's clearly illegal.
I appreciate your arguments and know they are in good faith. I think we would have an edifying debate in person!
I'm not going to reply to everything as I think our viewpoints are tricky to reconcile, since we find different things to be moral/immoral. That's fine, but it might not be productive. However, I acknowledge your position and know it reflects much popular sentiment; I cannot dispute that.
if you believe that it's not bad to take someone's effort without their consent and without compensating them for it, then you shouldn't object to your employer withholding wages from you
I think this gets to the crux of our difference. Employment is an explicit contract that binds two parties to honor their obligations. If someone posts a blog post openly, busks in the street, or does some graffiti art, I don't think observers have any obligations beyond an implicit idea of "experience this in any way you like as long as it's legal". Whether you prefer 'legal' or 'moral' there, it brings us back to the problem that we disagree on the morality/legality of the core issue. Given the constraints of this venue, not to mention our time, I'm happy to recognize this difference and leave it unsettled.
That's the "law is morality" fallacy. Morally, this is clearly wrong, the point of the copyright system is to prevent exactly things like this from happening. The courts have not yet decided whether training an LLM is "copying" a copyrighted work, but if they do, then it's clearly illegal.
If that should come to pass, I agree. However, your suggested fallacy then comes into play the other way around. Merely because a legal precedent may be set does not change my opinion that it is not immoral. That is a point on which we clearly differ and one I think would be fascinating to debate if only in a more appropriate venue as I may even be won over but have not been by any arguments so far.
*value to a private corporation that'll keep all of the profits, not pay for the environmental impact and then lobby lawmakers to stay untouchable.
You can still write, paint, compose and whatnot to create "value" – just don't put it on the Internet for scraping.
To me, the fact that a blog post would be used to train AI is a good thing. Hell yes I want my writing to inform the future zeitgeist! I guess it helps that the things I want to write about are novel things no-one has ever written about. I could see how AI would demoralize me if I were otherwise employed writing Generic Politics Blog #84773. But as someone who writes original unique content, I'm like, hell yes, the more readers the merrier, whether they be human or AI or some unholy combination!
The main reason I blog is because I work (or am starting out working in) academia and therefore I have to write papers. To write academic papers you have to write terribly, using passive voice whenever possible.
I blog so that I know I am still capable of writing coherently, rather than in horrible academic language.
You can avoid the passive voice and write coherently or even conversationally if you have something interesting enough to say.
There is also the gwern point of "You should write so that your voice is present in the future LLM": https://www.lesswrong.com/posts/PQaZiATafCh7n5Luf/gwern-s-sh...
The jokey last paragraph was probably based on a half-memory of having read that, now you mention it...
Anyone here seen an LLM actually produce a really novel thought that hasn't already been written about ad nauseam?
The well of new ideas, or re-formulation of existing ideas with perspective and prose that LLMs can't match is plenty deep to be worth hoisting the bucket still.
I don't know that I've seen a human blogger on HN produce a really novel thought that hasn't already been written about ad nauseam.
This blog post is itself a good example. It may have the author's voice, but it's just another rehashing of something that a million people have said before already. HN loves upvoting blog posts about why you should blog. https://hn.algolia.com/?q=why+blog
I read Stratechery. He’s had some novel thoughts. He just repeats them ad nauseam.
He’s made quite a living from blogging.
https://blockbuster.thoughtleader.school/p/how-ben-thompson-...
I can't find a less snarky way to say this, but why are you here then? I figure there's enough signal in the noise to make it worth it.
Do you mean on HN or in this particular thread?
Value is not measured in novelty alone.
I’ve seen novel thoughts, but typically they are coming from trolls who are taking impossibly contrarian opinions and presenting horrifying morally bankrupt perspectives. Maybe these ideas are common elsewhere though.
All the “good” ideas that can be said have probably been said. Maybe that’s why some people just enjoy trolling: for the novelty.
> Maybe these ideas are common elsewhere though.
I don't know which ideas you're talking about, but I bet they are. Humanity has thousands of years of recorded navelgazing (the blog post) and reactions to navelgazing (me) and reactions to reactions to navelgazing (you) and so on.
> Maybe that’s why some people just enjoy trolling: for the novelty.
Novelty is a morally bankrupt reason to enjoy trolling, so maybe you're onto something.
>Maybe that’s why some people just enjoy trolling: for the novelty.
Or a most general word 'fun' ? ( yes it's a catch all...)
> Anyone here seen an LLM actually produce a really novel thought that hasn't already been written about ad nauseam?
Personal anecdote. I stopped reading on most subjects that might be regarded as philosophical in nature a long time ago, perhaps in my early twenties.
Most writers have nothing new to offer in terms of our (philosophical) understanding of the world and I think themes are just repeated ad nauseam.
On the other hand when it come to the average Joe, to him it would appear that he is encountering novel/profound thoughts if he keep reading. The same goes for LLMs generated trash, he feels it's novel.
Prompt: Write a totally novel thought about how individuals can prosper in the age of AI. The thought must be truly new and unique. Write your answer like a semi-troll on an internet forum.
*Title: "Start a Personal AI Circus: The Next Frontier in Human-AI Collaboration!"*
Alright, buckle up, folks. Forget the usual "learn to code" or "adapt to the machines" spiel. If you really want to thrive in the AI age, here’s a spicy, totally unique idea: *start your own personal AI circus.*
Hear me out. We’re all aware that AI is taking over mundane tasks and repetitive jobs, but what if you could actually *curate* an ensemble of AIs that each specializes in one quirky "act"? Imagine this:
1. *Ensemble Cast of AI Performers* – You get a chatbot that does stand-up comedy, an AI that can whip up avant-garde digital art, and maybe an AI that can compose catchy jingles. Each AI has its own “act” and you market them together.
2. *Perform Live Shows* – Stream your AI circus on Twitch, TikTok, or wherever the cool kids hang out. Interact with your audience in real-time. Sure, AI can do things fast, but can it handle hecklers?
3. *Crowd involvement* – Use collaborative AI tools like GPT to let your audience write a portion of the comedy script live, or have an AI-driven talent competition where people can pitch their own AI's act. Let’s be real, everyone loves the thrill of unpredictability, and this gives a platform for some experimental creativity.
4. *Monetization Madness* – The revenue streams could come from subscriptions, sponsorships, merchandise (who wouldn't want a plushie of your AI stand-up comic?), and maybe even a VR experience where other AIs perform for people in their own homes.
5. *Building a Community* – This is more than just a circus; it's a community. Teach workshops on how to set up your own act, lead discussions on ethics in AI, and create a digital space where AI enthusiasts can share their journeys.
So, while everyone else is stuck in the grind trying to "adapt," you’re over here juggling your quirky AI acts, creating a spectacle, and engaging people in ways they never expected. You’re not just surviving; you’re thriving in the most entertaining way possible.
Let’s be real, wouldn’t you pay to watch an AI sing opera while your buddy’s AI attempts to juggle data? Welcome to the AI Circus, folks. Step right up!
This just made my day
I blog solely for my own memory. If people find it interesting, ok, but it's because I forget stuff and then will go somewhere thinking it's interesting only to recall I did this already. If no-one ever reads it, it's not really a worry.
The age of AI is the very reason it's even more important that we blog. AI in it current form ingests and repeats what we train it to. I don't think new ideas will "organically" come out of current generation models.
If you now make blog entries on new ideas, it will be consumed and regurgitated as new ideas with no attribution or benefit to the original author, giving the illusion of new ideas.
Feels dirty to have your creativity stolen easily by unforgiving machines.
I'm thinking the only reason to write may be to either have your ideas consumed by AI intentionally (so they are influence AI users) or to add negative value to the pool to make AI more useless when it regurgitates incorrect data.
> If you now make blog entries on new ideas, it will be consumed and regurgitated as new ideas with no attribution or benefit to the original author, giving the illusion of new ideas.
This was already happening though. I’m not saying it makes what’s happening now ok, I’m just continually surprised to see this raised as though prior to LLMs, average individuals were going around giving attribution all day long after consuming your content.
In my mind the only thing to have changed is the scale of it. I always accepted that the moment I published something online, I lost control of what was to be done with it or by who. LLMs have not really changed much in this regard for me.
That being said, I only ever really write for my own clarity. If I had to survive off it or something, I’d probably feel different.
I feel it is the honourable thing to reference where ideas come from.
Attribution exists so that we can understand the impact that we have on the world. I refuse to be gaslit that LLM's poor facimilie is a positive net outcome for the individual creator or humanity.
To use a stolen culture analogy, we can't put the genie back in the bottle, I don't think that anyone should be discoraged for not wanting to help Jafar become even more powerful and oppressive dictator.
> I feel it is the honourable thing to reference where ideas come from.
Sure, I’m not disagreeing.
> Attribution exists so that we can understand the impact that we have on the world.
Again, sure.
> I refuse to be gaslit that LLM's poor facsimile is a positive net outcome for the individual creator or humanity.
I’m not sure who’s trying to convince you of that, but I hear you.
I only intended to address a specific statement, which was:
> If you now make blog entries on new ideas, it will be consumed and regurgitated as new ideas with no attribution or benefit to the original author.
My point was that this isn’t new. This has been happening long before LLMs. People have always taken others’ work, repackaged it, and passed it off as their own. The only thing that’s changed is the scale. That doesn’t mean it’s ideal or that it shouldn’t be addressed, just that it’s not some new phenomenon LLMs have introduced.
Once you put something on the internet, you’ve always run the risk of it being reused without credit or in ways you may not have forseen. If avoiding that is the goal, then the safest move would have been to never publish online in the first place.
As an example, just a few months ago, I found out a vendor lifted code from multiple open-source projects, stripped the BSD licenses, and passed it off as their own for a “custom” software job. No LLMs involved, just plain old fashioned plagiarism.
> I don't think that anyone should be discouraged for not wanting to help Jafar become an even more powerful and oppressive dictator.
Fair enough. I completely get not wanting to contribute to Meta, OpenAI, Microsoft, etc. If you don’t want your content feeding into LLMs, that’s a valid stance, and I respect it.
While I'm not disputing this, chatbots citing my posts have sent a lot of traffic my way. Obviously though, this could all disappear the next time they tune the model.
What the hell are people blogging about that is so generic that an AI could write it?
The important part of any journalism is to introduce new facts and findings that aren't anywhere else to be found. An AI cannot do that.
> There were a lot of excellent comments, but one thing kept coming up: what's the point in blogging if people are using ChatGPT, Claude and DeepSeek to spoon-feed them answers? Who, apart from the AIs, will read what you write?
It's undeniable that AI has changed the utility of blogging as a means to spread knowledge to others. I agree 100% with the author and most of the comments below that blogging for yourself is great, and that showcasing what you are passionate about is also great. I don't think the position people are taking is that blogging isn't worth it because that is no longer important.
The point I gather, is that it used to be you could find a small nugget of something technical (for example, on stack-overflow, in a manual, etc) and explain it in a way that is approachable to a select audience. Then, over time that audience comes to appreciate learning new and interesting things. Take for example https://kyrylo.org/html/2024/10/25/why-does-target-blank-hav... . ChatGPT can practically write this same post by asking the question, and tailor the explanation to any level of expertise, and can break individual concepts down that is 100% tailored to the person asking the question. That is where I think blogging has been turned on its head.
I checked out his About page just because he mentioned that only 1% do it. For my writing, the observation that almost everyone moves on is certainly true. But those few who reached out sometimes become new friends. I've never met any of them in person though. It's like a group of friends for whom the connection is that they all read something that I wrote. Also, funny to now see that when doing research I've gotten AI's citing my own past posts as sources. Maybe this means I need to reread my own writing.
You writing something personal serves as inspiration to people who just want to write.
Many years ago I started writing as a goal to make it something big. But I was very far away from acheiving anything close to my goals, and writing started feeling like a chore.
Now I am working on my personal blog as a platform, not only for writing, but for self expression in design and programming. Because I just read your post - it makes me want to speed up the process of finishing my own blog (it contains many engineering quirks that need to be ironed out as it is not a simple website - hence the time taken to make it).
So what I am trying to say, when people read your posts - you are moving the needle. You are not changing anything spectacularly, but rather have some influence of people's perception over certain things. And now - you moved me closer to finishing what I started months ago - this should serve enough of a dopamine hit for you to continue doing what you are doing, however you choose to rationalize it :) (assuming you even get to read this comment)
Personally I don't have a blog, even though I really want to, but I newer seem to be able to get started. Still, I really enjoy reading what others write. One thing a blog post from a real live person can do, which AI can't, is to inspire you to dive into a subject and try out things for yourself.
There are so many things, books, technologies, tips, tricks, interesting solutions I've picked up over the years because someone wrote a blog post and it got posted on HN or some other site.
As more and more content is AI generated I've started to enjoy the writing and ramblings of people with weird little blogs so much more. So to those of you who are blogging: Please don't stop.
Same.
I do write stuff on Obsidian just for myself. I just feel too self-conscious to publish it, because it's just for me and not written in a way that (in my opinion) is helpful to others.
Never mind that I've fixed my own issues by finding an obscure post in a small blog multiple times.
> So, I don't think you can make a name for yourself by blogging alone
Or maybe you could if you post bangers like https://ciechanow.ski/
Nice blog either way!
Thanks! And that's a good counter-example... which made me realise, there's also https://waitbutwhy.com/
Related aforementioned thread:
Write the post you wish you'd found https://news.ycombinator.com/item?id=43154666
I will more likely believe a blog or website if I know it has been made by a real person, and not some hallucinating AI. The style of the prose is usually a giveaway of when AI is used to generate blog articles or websites.
I am starting a Masters in Computer Science, and going to learn Java for the first time. I wanted to know how to use default method parameter values. I searched for it and found this near the top of the results:
https://skillapp.co/blog/deep-dive-into-default-parameters-i...
Java developers will know that the language does not support default method parameter values. I only found that out when I tried to implement it in a test, was surprised it didn't work, thought maybe my version of Java was too old, etc.
But there is no syntax like "void coolFunction(int x=5){};" that other languages have.
One more reason that I will value a human-written article, blog or website. Please keep writing them!
Writing things down reveals the lapses in one's own knowledge, and those gaps are opportunities to learn something new.
It saddens me a lot that people are now relying on AI to fill the gaps in their knowledge.
Humans got where they are thanks to their innate curiosity. If you take that away from us, we become no different from animals.
Not just gaps. Blind spots too. Humans are driven by curiosity and limited by biases. You're only painting half of the picture.
> Humans got where they are thanks to their innate curiosity. If you take that away from us, we become no different from animals.
Your message is a bit late. We've had people zoned out in front of their TVs for decades.
The more that we have AI trawling our writing, the more important I think it is to realize that AI is part of the audience.
I have been thinking about writing more "cookbook" style posts so that users and AI have good modern examples to look at. CSS-tricks is a great example. It has good code and and is easy to search.
One thing that articles that include code need to tag good and bad examples. AI will get better at recognizing the bad, but we should have more comments that say "This is bad. Use the better pattern."
It is helping out the collective to give AI a tutorial on undocumented matter. Yes, you don't receive anything in exchange (except personal achievment and growth from writing, as many have commented). But making a world a better place is more important for some people than personal fame and recognition.
If AI systems are crawling everything we write publicly, we have a certain responsibility to teach it the right things. It's like our common child we're raising through what we leave open on the internet.
If we don't blog, what will the next big AI models be trained on?!
Good point. "Blog as much as you can, as otherwise the training set will be Twitter!"
I’m glad he goes out of his way to mention that, basically, effective blogging is downstream of accomplishments in a prior arena. It’s about channeling expertise and experience into a narrative, to summarize the core assertion he’s making, and he’s right. It also correlates to high-value Video content, such as the YouTube channel AmmoNYC (car detailing), where the high level of professional knowledge is yoked with high quality production. Very spot on follow-up to his initial post.
If you're not interested in letting your blog posts be used to train AI, remember to update your robots.txt regularly as AI crawlers come up. This is assuming they respect your robots.txt, and don't change their user-agent. There's a service for this called Dark Visitors[0], though I wish there was a FOSS project to track AI user-agents.
There was a good reminder in the Hanselminutes interview with Shaundai Person about blogging for yourself. It was a while since I listened so I'm fuzzy on the details - all I remember is that I enjoyed the interview and there was some relevance to this topic.
Also, blogs are no doubt a major source of AI training, so maybe more worth than before.
I've suggested this elsewhere too, but have you considered a mailing list instead?
It ticks the boxes about notoriety—which appears to be the main concern of the OP author—with the added benefit of being more difficult for AI systems to crawl for training data.
Perhaps we'll see more bloggers going the way of a bespoke mailing list.
I wonder if there is a practical test of this question? Are there any moderately successful blogs written purely by an LLM with a human just doing the prompts.
As far as I know there aren't any but I look forward to being corrected
Gwern said something along the lines of: If you don't leave marks in the online world, you're invisible to AI.
I have no idea what that's going to mean moving forward, but it makes sense to me.
Strangely enough, doing work with AI helped push me to blog and be more creative. Something about the artificiality of AI and if the machines can start doing what we do, what else is there?
My 2c: AI can give you the answers if you have the right questions. Having the right questions is a journey of human development through self reading, for which blogs are essential.
I think it's more worth than ever. The idea that some of my thoughts might be integrated into some AI model by being used as training data is exciting
One always writes for the potential readers. Even if the human readers are becoming rare, the opportunity to be read forever by all these AIs can only boost our ego :-)
Being read by one human: you deepened their understanding of the thing.
Being read by one OpenAI crawler: you deepened the understanding of ChatGPT, thus offering knowledge of the thing to everyone that interacts with it about the thing.
One of these scenarios has much more collective impact than the other.
What upsets me about AI is that the proof-of-value in basically articulate prose has gone to zero. I can still write better than AI, but I’m not sure the remaining margin matters, socially speaking.
That said, AI is less than 2 percent of what has enshittified the internet. It’s a factor, but corporatization and algorithmic optimization have already done plenty of damage even without LLMs in the mix.
While quite a few folks agree blogging is still worth it in the age of AI, please do your bit to support privacy-friendly services (ie. no tracking). Blogs don't /need/ modern web bloat, nor need to resort to adverts / paywalls.
I happen to build a privacy-friendly option https://LMNO.lol (my blog at https://LMNO.lol/alvaro or https://xenodium.com). I'm hoping it's fairly priced at $1.50/month. You pay for hosting and we provide that service (that's the extent of our transation).
There are others services to pick from like Bear, Ghost, Nekoweb, omg.lol...
I like https://mataroa.blog/
Human conscience has not and will not be duplicated by "AI" in the near future.
These are simply programs that regurgitate mostly stolen data.
I think blogging isn't just about getting views, it's more about organizing your thoughts and solidifying what you've learned.
I've been considering starting up a(n extremely) sporadic blog for funsies. What platforms do people use these days?
I built https://LMNO.lol and using it myself https://lmno.lol/alvaro (close to a year now).
Ah, the weekly post about if writing blog posts is worth it with the comments reaching the same conclusion.
...you can do it!
It's still worth doing wood work in the age of IKEA. Perhaps even more so.
"It's still worth selling steaks in the age of the hamburger."
Yea. This will never not be true.
I just can't wait for the age of the "silicon valley visionary" to come to an end.
yes. The research that goes into it helps be finally learn the concept better. Having a few subscribers also want to spend time to write something and put it out.
I believe its even more important now than ever before.
I dont blog for others, I blog for myself and my joy.
I deeply distrust anyone blames AI for "ruining" their creative hobby, or "creative" career. Mostly because the one absolute, will never ever change, always our trump card, is our individual perspective and creativity. AI is a sorting and statistics machine, it can only reproduce (my personal theory about AI hallucinations is pretty simple. AI has been a well known concept for longer than the Internet has been around, pop-scifi and post apocalyptic books, movies, and shows, and causing a societal immune response... And satire, sarcasm, bad photography, bad art, websites with gibberish or incorrect facts just to game SEO algos (many many many that still have generations old SEO text blocks), all being hoovered up by an AI and the techbros supervising it all are MBA dropouts, and/or marketing bros. All their AI datasets are so poisoned by the human condition, and as universal, AI hallucinations/errors are, it just proves there isn't a single model, or a single person, in the AI space who can identify irony. We may be the most destructive thing this earth has ever encountered, but like.. Causing mayhem and chaos so concentrated and complicated it's impossible (so far at least) to write code that can decypher it.
So yeah, anyways, I left tech almost a decade ago, I'm a photographer now, and if youve put thought into something. Whether it's writing, how a photo is edited/processed, a joke, its a gift you're giving to yourself, and hopefully some of whatever you felt like creating, a few pieces at the least haha, will be gifts to the world :)
most things are still worth doing in the age of ai
writing for thinking , share ideas with others
blogging is zen and zen is who cares?
am I so unusual in just skimming past and ignoring the AI search results? I didn't think a majority of people outside the silicon valley tech bubble were all that keen on getting their questions answered by a chatbot
Yes :)
[flagged]
Could you please stop posting unsubstantive comments and flamebait? You've unfortunately been doing it repeatedly. It's not what this site is for, and destroys what it is for.
If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.