Yes but this is incredibly competitive and undifferentiated.
It's a huge market but who will it be a profitable business for?
Likely a company or multiple who own some sort of platform that people are already on, so not OpenAI.
What they have right now is the strong ChatGPT brand and that does mean a lot. But how long will it last?
They're not the technology leader anymore, and that spells a lot of trouble.
They are at a stage where they need to dominate the market and then leverage the data that gives them, plus the brand, plus the tech advantage to establish a durable near monopoly, but it looks like it's not working.
It's a bit as if in 1999 3 equally strong Google competitors had popped up, with some pulling ahead.
I wrote this comment [0] very recently and when I wrote it had in mind that Cloudflare might very well end up being a key player in a more centralized Internet that has developed far away from its original architecture.
Defense against threats is a pretty strong centralization incentive in different kinds of networks - social, biological.
I could imagine that a lot of people are investing based on similar scenarios in their minds.
Joining the big three requires capital investments orders of magnitude beyond where they are. Nevertheless I'd like to see someone do it, and if not Equinix then it could be them. But sadly the #4 right now appears to be Oracle.
The Internet has really been an interesting case study for what happens between people when you remove a varying number of layers of social control.
All the way back to the early days of Usenet really.
I would hate to see it but at the same time I feel like the incentives created by the bad actors really push this towards a much more centralized model over time, e.g. one where all traffic provenance must be signed and identified and must flow through a few big networks that enforce laws around that.
"Socialists"* argue for more regulations; "liberals" claim that there should be financial incentives to not do that.
I'm neither. I believe that we should go back to being "tribes"/communities. At least it's a time-tested way to – maybe not prevent, but somewhat allieviate – the tragedy of the commons.
(I'm aware that this is a very poor and naive theory; I'll happily ditch it for a better idea.)
Little would prevent attacks by APTs and other powerful groups. (This, btw., is one of the few facets of this problem that technology could help solve.) But a trivial change: a hard requirement to sign up (=send a human-composed message to one of the moderators) to be able to participate (or, in extreme cases, to read the contents) "automagically" stops almost all spam, scrapers (in the extreme case), vandalism, etc. (from my personal experience based on a rather large sample).
I think it's one of the multi-faceted problems where technology (a "moat", "palisade", etc. for your "tribe") should accompany social changes.
We were very confident by ca. 2008 that Facebook would still be around in 2025. It's no mystery, it's the network effects. They had started with a prestige demographic (Harvard), and secured a demographic you could trust to not move on to the next big thing in a hurry, yet which most people want contact with (your parents).
Perceived quality is relative. It's roughly linearly related to rank position along some dimension, but moving up in rank requires exponential effort due to competition.
I would be surprised if anyone perceives quality like that. Like, are you saying that in a situation where there are only two examples of some type of work, it is impossible to judge whether one is much better than the other, it is only possible to say that it's better? What makes you think it works like this?
This insight, that perceived quality is relative, can be understood in a more literary way, in this fragment by Proust describing the Rivebelle restaurant:
> Soon the spectacle became arranged, in my eyes at least, in a more noble and calmer fashion. All this vertiginous activity settled into a calm harmony. I would watch the round tables, whose innumerable assemblage filled the restaurant, like so many planets, such as they are figured in old allegorical pictures. Moreover, an irresistible force of attraction was exerted between these various stars, and at each table the diners had eyes only for the tables at which they were not sitting,
The relative nature of perceived quality indicates that the order of representations (artworks) is judged only in relation to the order of castes (ranking); one can only judge what is equal by comparison to what is unequal, and equivalence is not understood through equality (even approximate) but through inequality (partial order). It is the absence of a ranking relationship between two entities that establishes their equivalence.
Maybe I should have just gone with "in this case, classification is more fundamental than measure", but I feel there is something interesting to be said about the structure of artworks and the structure of their reception by society, indeed Proust continues with:
> ...the diners had eyes only for the tables at which they were not sitting, with the exception of some wealthy host who, having succeeded in bringing a famous writer, strove to extract from him, thanks to the spiritual properties of the 'turning table', insignificant observations at which the ladies marvelled.
See ? Writers (and artists in general) take on the role of a medium. They are used to channel distant entities, like tables during spiritism sessions, and from what Proust tells us, that "the diners had eyes only for the tables at which they were not sitting", maybe we can infer that what writers channel doesn't just come from distant worlds, as incarnated in what their words represent, but as a delta in perceived quality, starting with our own.
This is why I'd like to elaborate on this idea of coupled SSR processes I developed in another comment.
A sample space reducing process is a process that seeks to combine atomic parts into a coherent whole by iteratively picking groups of parts that can be assembled into functional elements ready to be added to this whole.
In that sense, the act of writing a long work already has the shape of a SSR process in a very simple sense: each narrative, stylistic, and conceptual choice constrains what can follow without breaking coherence. As a novel unfolds, fewer continuations remain compatible with its voice, characters, rhythms, arguments. You are not wandering freely in idea-space; you are navigating a progressively narrowed funnel of possibilities induced by your own earlier decisions. A good book is one that survives this internal reduction without collapsing.
On top of that, there is a second, external funnel: the competitive ranking of works and authors. Here too the available space narrows as you move upward. The further you climb in terms of attention, recognition, or canonization, the smaller the set of works that can plausibly dislodge those already in place. Near the top, the acceptance region is tiny: most new works, even competent ones, will not significantly shift the existing order. From that perspective, perceived quality is largely tied to where a work ends up in this hierarchy, not to some independently measurable scalar.
The interesting part is how these two processes couple. To have any chance of entering the higher strata of the external ranking, a work first has to survive its own internal funnel: it has to maintain coherence, depth, and a recognizable voice under increasingly tight self-imposed constraints. At the same time, the shape of the external funnel, market expectations, critical fashions, existing canons, feeds back into the act of writing by making some narrative paths feel viable and others almost unthinkable. So the writer is never optimizing in a vacuum, but always under a joint pressure: what keeps the book internally alive, and what keeps it externally legible.
But what interests me more is that some works don't just suffer this coupling, they encode it. That's what you see in the Proust passage: he is not merely describing a restaurant; he is describing the optics of social distinction, the way people look at other tables, the way a famous writer is used as a medium to channel prestige, the way perception itself is structured by rankings. The text is aware of the hierarchy through which it will itself be read. It doesn't just represent a world; it stages the illusions and comparisons that make that world intelligible. That's a second-order move: the work includes within itself a model of the very mechanisms that will classify it.
If you like a more structural vocabulary: natural language is massively stratified by frequency. Highly frequent words ("I", "of", "after") act as primitive binders; extremely rare words tend to live out on the leaves of the tree; in between you get heavier operators that bind large-scale entities and narratives ("terrorism" being a classic example in the grammar of public opinion). Something similar happens socially. Highly visible figures – the wealthy host, the celebrated writer, the glamorous guest – play the role of grammatical linkers in the social syntax of recognition: they bind other people, distribute attention, create or close off relational triads. Proust's "the diners had eyes only for the tables at which they were not sitting" is exactly this: desire and judgment are mediated through a few high-frequency social operators.
A certain kind of writing operates precisely at that interface: it doesn't just tell a story inside the internal funnel, and it doesn't just try to climb the external ranking; it exposes and recombines the "function words" of social perception themselves: the roles, clichés, prestige tokens, feared or desired third parties (like the forever-imagined intruder in Swann's jealousy). The difficulty is not only to satisfy two nested constraints (a coherent work and a competitive position), but to produce a form that reflects on, and potentially perturbs, the very grammar that links the two. That's where the channeling comes in: literature not only represents something, it re-routes the connective tissue through which quality, status, and desire are perceived in the first place.
>Moreover, an irresistible force of attraction was exerted between these various stars, and at each table the diners had eyes only for the tables at which they were not sitting,
You're reading way too much into it. This is just a reprise of "the grass is greener on the other side". What it's saying is simply "a lot of diners were dissatisfied with their dishes and looked around to see what other people were eating".
>The relative nature of perceived quality indicates [...]
My whole point is that I don't buy quality is purely perceived relatively. If you start a sentence like this, whatever comes after is irrelevant.
You've brought up food, so let's go with that. If I'm a blank slate and I eat a certain food, am I unable to decide whether I like it or not until I eat a second, different food? Are the sensory signals my brain receives just a confounding mystery in the absence of further stimulation, to the extent that I can't even tell sweet from bitter?
One of the great understated reasons for using 'cool' programming languages is that it allows a business to hire selectively for people for whom programming is a passion.
Yes. I never had problems with Linux itself and compiled kernels constantly. What I did have incessant problems with was compiling GNOME 1.2 and 1.4. SO MANY problems, just non-stop... it was always something. I learned a bit though, although not as much as I could have if I paid attention more.
That they carry massive amounts of power so they'll help you to destruct themselves once the barrier is penetrated, are actually quite complex and can be sabotaged easily.
In a country as conservative as yours, if you're culturally in line with the 'coastal elites' (forgive my use of this term), you can't expect a stable majority by being uncompromising.
reply