Classic PG essay in a lot of ways, but I was expecting (a) the entire watch discussion to be 1/10th of its length, and (b) lead DIRECTLY into a social commentary around how every area -- politics, software dev, music creation & distribution etc -- are now firmly in their terminal brand ages.
A political admin lies directly to your face daily - well, the base doesn't care because they've bought into the brand. The Truth and principles are irrelevant.
A talented musician is making innovative, fresh music - but hasn't yet learned the ability to market herself on IG/TikTok - well, she won't reach an audience because she can't compete in a brand game dominated by marketers and AI-driven beat-makers. The music is irrelevant.
Creating software used to be a very human thing - empathetic understanding of the user's problems, a dash of inspiration and good taste, and you get some cool stuff and possibly a valuable company; but now, if all that's automated, what is it? The craft is irrelevant.
Kudos! This comment really gave simplicity to my feelings on this whole thing too! (Favourited your comment)
I just want to add a minor point that the marketing/advertising industry in general just sometimes feels very unethical to me and I am worried about that too.
Increasingly so, The only way to survive in brand age is to connect to two or three centralized platforms. Authenticity feels redundant.
> A talented musician is making innovative, fresh music - but hasn't yet learned the ability to market herself on IG/TikTok - well, she won't reach an audience because she can't compete in a brand game dominated by marketers and AI-driven beat-makers. The music is irrelevant.
Although I could've ended my comment previously, I do think that there are just very few cracks and ripples within the algorithm. At some point we have to step up to preserve authenticity. It isn't as if those people aren't trying but that you can't find them.
I found this person who makes music[0] on my youtube shorts on how his genz cousin told him to make videos of his everday life with his music in background (He sells pizzas and left college to make music) and there is a small community of viewers/music listeners who have joined him now. It made me genuinely think and stop from the doom scroll, Although small but it was one sure of an experience.
I do think that if there is a race to the bottom for code/infra/even marketing. At some point, the idea becomes of fulfilling/creating an experience but I do think that the passion of the people who create things isn't gone to waste because the passion that they applied into creating any form of software now, can be reflected back to the user in form of experience.
Atleast that's my take. Another product I found interesting like this was fluxer.gg[1](discord alternative)
Maybe there can be a platform/mass adoption of the thing where one can just share cool people doing cool things and have links of them which can say "hey check them out! What they are doing is cool"
(I actually tried out some bookmark app just right now but funny thing happened haha where it accidentally destroyed my whole collection as I tried to have a collection name be All, and they don't have support so that I can't even guide them to fix it :/ so its better if we use open source solutions in this maybe, Luckily I was only trying it out but this actually makes me feel like this sharing should be made easier and less error prone
it would be hard to see if average people realize this and if they do end up doing it at decent scale where it makes meaningful difference tho.)
Edit: I rediscovered Linkhut, it feels really good as well. I will be having some fun with it especially with the social elements of it, its front page is about what the official logo of html should be :D (https://ln.ht)
Nice points! You reminded me that, actually, contrary to my hand-wringing, the IG algorithm is excellent (for me) when it comes to recommending music - and it's often artists with very few monthly listeners on Spotify.
Why is IG so much better than Spotify itself's algo? Or YouTube's? Or... etc? No idea. And also - it's always paid ads from these artists that get me.
So on the one hand, <insert my original point>; but on the other, as much as technology is perhaps driving the brand hollowness, it's also allow for more long-tail discovery precisely because it makes powerful marketing/advertising channels accessible to that girl making beats in her bedroom - so long as she has a bit of $ to deploy. It still may not be profitable for her (because: Spotify) but I also try to support these artists with merch buys, physical buys, etc.
2) 'After processing' is wide enough to drive a truck through. What if processing takes a year? What if processing is defined as something involving recurring checks?
3) You have no contract with Persona or even LinkedIn beyond the fact that you agreed to LinkedIn's TOS (but didn't even read).
4) The company that acquires or takes-private Persona might have a very different of how it handles this.
5) What does verifying do for you, the user? I understand its value to LinkedIn and their ability to sell your attention to advertisers, but what do YOU gain?
"how incredibly good their google workspace suite of tools is" - is that a common sentiment on HN?
To me, Google Sheets is 10% of Excel on desktop (Mac), Slides are 5% of PowerPoint on desktop (Mac), and the integration between the two (copying and pasting linked charts from Excel to Powerpoint with formatting) makes it a completely non-starter to consider the Google alternatives as primary drivers.
I'm probably a power-user of both, granted, but I took for granted Sheets/Slides are still just toys compared to the real stuff, so curious if I'm missing something.
I've worked for years at companies that only use Google Sheets.
For 99% of people (sometimes we let Finance folks have an Excel license), it's more than enough. Google Apps Script is also reasonably useful, and the newer Smart Chips are a nice addition.
Competent is not the same as good. I can do the very basic things I need in Sheets, but the moment it needs more than =A1+B2 then it's uphill all the way. I also don't know how performant it is with larger datasets. I use Libre Office instead and despite the horrible UI it's been speedy and accurate. Desktop Excel is still king of the spreadsheets.
As for Slides, it's pure junk compared to the Keynote, but iCloud has it's own problems so I use this offline-only.
With the web version of Word 365 or whatever it's called, we've had so many problems syncing with OneDrive and sharing and whether it's showing the right version of the document that I'd be happy to never see it again, but their foothold in education means I'm forced to deal with it and provide technical support.
I am not a power user of either, and I absolutely detest when someone insists on using Excel. Sharing and collaboration is such a giant pain, and it's like going backwards in time to the 90's with e-mailed versions of files back and forth. Our org does not have a MS 365 license, so I'm unsure of Microsoft's web versions of Excel or how good they happen to be these days. I know users of it who complain though, and end up using it locally on their workstation like the olden days.
Most of my use is incredibly simple and used for project planning, inventory counting, lists of things that are split up into status/to-dos among multiple people, etc.
I've also never had a use for "Advanced" powerpoint, so the simplicity of google slides is a breath of fresh are as I only ever use the 10% most common feature set.
I actually get a bit of anxiety when someone sends me an excel sheet these days. It's usually going to be overly complex using clever methods, and that person is going to be a real pain to work with on iterating anything most of the time.
I've noted some very rare and specific times Excel is warranted though - such as our CFO creating complex financial modeling. For those uses I totally get that Google Sheets would be like working with handcuffs on.
An alternative perspective is that if someone sends me a Google sheet link then I know almost immediately that is probably not a “serious” document.
Similarly with Google Docs, as “serious” documents with proper tracking of changes and so on are in Word.
Of course the uses of serious spreadsheets are often in finance and serious documents are in law.
> “serious” documents with proper tracking of changes and so on are in Word
How does the tracking works in Word? I've never seen this proper setup so I'm just ignorant when it comes to this. If I hear "Word" and versioning in the same sentence, I'd just assume we're talking about the doc_v1_3_final_really_final_public_feb_2024.docx naming.
What a timeless banger, you fill in the blanks on this part:
'Over and over I have found myself asking: "What kind of people worship here? Who is their God? Where were their voices when the lips of __________ <American political leader> dripped with words of interposition and nullification? Where were they when __________ <American political leader> gave a clarion call for defiance and hatred?"'
A bit of nuance: yes, Carney said that but he didn't just offer up the opinion unprompted - it was in response to a direct press question about if China or the US is a more predictable partner right now.
And even then, he didn't lead with "China is!" but wandered his way into offering the assessment.
The context makes his comment on this seem less nakedly provocative (not that it'll matter either way - the headline will be the headline, and the Trump admin will use it however they see fit as usual).
Some of this analysis seems a bit lazy for the Economist.
Apple is in the "AI-related companies in the SP500" group? Microsoft too? Tesla too? Amazon too? But... if these companies' AI efforts fail, 95%+ of their revenues would be unaffected. So big stretch to paint them with that brush.
Nvidia, OK that one is obvious. Meta, Alphabet, OK.
But MOST of the companies listed in that chart are only "AI companies" in the sense that EVERY tech company building peripheral AI into their products is an AI company.
Case in point: if Apple stock goes 'on sale' as part of an AI-bubble sell-off, are you really deciding whether or not to buy based on their AI-ness?
Tesla for example: its stock price has fluctuated down 50%, then up 100% (relative to the dip), in this year alone. Clearly, that's market speculation, not capital + earnings. So how much of that speculation is AI-dependent? Depends on how much coffee investors drink before reading Musk's latest tweets, I guess.
Apple is HEAVILY invested in AI, but you're right: it's earnings are dependent more on iPads than AI right now.
> ...are you really deciding whether or not to buy based on their AI-ness?
Tales as old as time, especially in tech: rich monopolistic incumbents not seeing the writing on the wall of a new paradigm shift; seemingly invincible execs brazenly displaying their (incorrect) hot-takes; and the inevitable enshittification of the new paradigm as it turns from revolutionary movement to ruling-class incentives.
"Companies often have a demo environment that is open" - huh?
And... Margolis allowed this open demo environment to connect to their ENTIRE Box drive of millions of super sensitive documents?
HUH???!
Before you get to the terrible security practices of the vendor, you have to place a massive amount of blame on the IT team of Margolis for allowing the above.
No amount of AI hype excuses that kind of professional misjudgement.
I don't think we have enough information to conclude exactly what happened. But my read is the researcher was looking for demo.filevine.com and found margolis.filevine.com instead. The implication is that many other customers may have been vulnerable in the same way.
Ah, I see now that I read too quickly - the "open demo environment" was clearly referencing the idea that the vendor (Filevine) would have a live demo, NOT that each client wanted an open playground demo account that is linked to a subset of their data (which would be utterly insane).
"The best case scenario is that AI is just not as valuable as those who invest in it, make it, and sell it believe."
This is the crux of the OP's argument, adding in that (in the meantime) the incumbents and/or bad actors will use it as a path to intensify their political and economic power.
But to me the article fails to:
(1) actually make the case that AI's not going to be 'valuable enough' which is a sweeping and bold claim (especially in light of its speed), and;
(2) quantify AI's true value versus the crazy overhyped valuation, which is admittedly hard to do - but matters if we're talking 10% of 100x overvalued.
If all of my direct evidence (from my own work and life) is that AI is absolutely transformative and multiplies my output substantially, AND I see that that trend seems to be continuing - then it's going to be a hard argument for me to agree with #1 just because image generation isn't great (and OP really cares about that).
Higher Ed is in crisis; VC has bet their entire asset class on AI; non-trivial amounts of code are being written by AI at every startup; tech co's are paying crazy amounts for top AI talents... in other words, just because it can't one-shot some complex visual design workflow does not mean (a) it's limited in its potential, or (b) that we fully understand how valuable it will become given the rate of change.
As for #2 - well, that's the whole rub isn't it? Knowing how much something is overvalued or undervalued is the whole game. If you believe it's waaaay overvalued with only a limited time before the music stop, then go make your fortune! "The Big Short 2: The AI Boogaloo".
I was aware of that. Not everyone reads HN but enough people will surely know about this if they continue. If they are not fined then people should just stop being so naive. I'm tired of seeing well-designed lies on Facebook and people clearly believing the lies in the comment section.
edit: by well designed I mean that they put the lies in a chart or something. The lies themselves may be quite obvious to spot, like saying that 86% of marriages in Spain end in divorces (when the reality is hard to measure but more likely about 50%). Still, Facebook users don't seem to spot them (or maybe the ones spotting them don't comment?).
You've got your finger on the pulse of something that open source has always represented to me: freedom of the creator and others to just... do what they want with it (subject to the license of course).
Don't like what the main developer is doing with it? You're free to fork and continue on your way if they don't see it your way. If you lack the skills or time to do that, that's your problem - you're not entitled to the maintainers' labor.
The freedom cuts both ways, and by adding in elements of social contracts and other overlays onto the otherwise relatively pure freedom represented by OSS, you end up with the worst of both worlds.
THAT ALL SAID - there's an important distinction between a given piece of software that's open source versus a "true project", which is larger-scale, more contributors involved, might be part of mission-critical systems, etc, where the social dynamics DO need to careful thought and management.
But even that seems to be more a question of specific types of OSS business models which is related but not the same as the licenses and overall social dynamics around OSS projects.
Before it becomes anything else code is first and foremost art & personal expression.
Code is a very fun form of literature at heart.
Other attributes may be tacked on later, it may be integrated into and transform into an engine or company that has rules and regulations.
If the author treats it as only art, with license choices, etc. then they aren't entitled to treat it like anything at all, it's literally their personal expression.
And this is recognized in the physical world as well. More than people realize, some buildings that are incredibly dangerous are considered sculpture effectively. There is a rickety castle built by mostly one guy in CO that meets this criteria.
I think you are trying to make the point that there is the ethical argument to consider the impact of a decision if your project has grown large enough that there is major dependence on it.
I do agree there is an ethical obligation to make some effort to consider impacts like that, and make an attempt to inform users of your intent, but that's all it is. No one is obligated to be ethical, either, when it comes to a personal project with volunteered time and effort.
I think the argument of obligation becomes a stronger if your project has taken on a lot of contributions from other parties. And yet, those contributors must acknowledge that they are willingly doing so with no promises except the posted license terms.
A political admin lies directly to your face daily - well, the base doesn't care because they've bought into the brand. The Truth and principles are irrelevant.
A talented musician is making innovative, fresh music - but hasn't yet learned the ability to market herself on IG/TikTok - well, she won't reach an audience because she can't compete in a brand game dominated by marketers and AI-driven beat-makers. The music is irrelevant.
Creating software used to be a very human thing - empathetic understanding of the user's problems, a dash of inspiration and good taste, and you get some cool stuff and possibly a valuable company; but now, if all that's automated, what is it? The craft is irrelevant.
Hmmmm. Perhaps his essay wasn't dark enough.
reply