HNHacker News
TopNewBestAskShowJobs

mjr00

8,478 karma · joined May 25, 2019

submissionscomments
mjr00··on Americans don't know how to fight AI so they're fighting data centers
> Then there's the fact that many of those datacenter are being built over what would otherwise be usable farmland. I'm sure many will say "it's not that much land", but then tech billionaires would like to build datacenters the size of Manhattan. What for? To train a bigger LLM? Yay?

Sure but you can say this about everything. Where are the protests about the wine industry in California? 500,000 acres of land for vineyards, far more water used for growing grapes than cooling data centers, all so a handful of people can make fortunes selling empty calories to the rich?

If you want to focus purely on utilitarian "optimal land use for essentials only" arguments there's way worse offenders than datacenters, the anti-DC sentiment is purely part of the anti-AI wave.

mjr00··on It Will Never Be the Year of the Linux Desktop
I love Linux and think it's better than ever to have Linux as your daily driver, especially thanks to the work of Valve with Proton, but I'm gonna be real and say "the year of the Linux Desktop is a personal journey" is a retrofit. The "Year of the Linux Desktop" meme came out of Slashdot in the 90s/early 00s where people were insisting that Linux was due to overtake Windows as the way an average, non-technical user interacted with a computer.

Of course, this did turn out to be true... in the form of Android, which is maybe the most monkeys-paw-curling way YotLD could possibly happen.

mjr00··on I think Anthropic and OpenAI have found product-market fit
Right, the confusion is that the quote-unquote "subsidized" monthly pricing is often used by Anthropic/OpenAI skeptics as proof that inference is unprofitable, i.e. the API would have cost $2000 but you only paid $200 for a subscription, therefore OpenAI is selling dollars for 95 cents and the house of cards is about to collapse. As the GP says, this is faulty logic because we don't know what the actual cost of a token is; OpenAI might only pay $1 in inference costs, in which case they're merely "incredibly profitable" making $199 off you instead of "ludicrously profitable" making $1999 off you had you used the API.

But to your point, re-reading the article, this is not what Simon is saying at all; he's just pointing out that he got to use ~$2000 "worth" of tokens on his $200 plan. Which makes total sense! Subscriptions are sticky, that's why the entire software industry moved towards subscription models (as much as we hate it); the person paying $200/month is more likely to stick around than the person who paid $2000 using the API.

mjr00··on Last.fm is now independent
Yep, member since May 21 2005 here, still scrobbling with Spotify. Don't think I've ever used any of the radio features on the site, really; even back in the 00s all I used were the WinAmp/Foobar plug-ins.
mjr00··on Bill to block publishers from killing online games advances in California
> It seems like the fair solution to this problem is to open source server code if you are going to cease support for an online game. That way the community has the opportunity to run their own servers if they want to.

It's nice in theory, but in practice many (most?) games are using middleware they don't have the rights to redistribute as open source. IIRC when the source code for Doom, the first major commercial game that went open source, originally came out, it had all of the sound code removed because it was dependent on a third party library. Not that you're going to have sound code in a server, but you may be using third party libraries for networking, replays, anti-cheat, etc.

mjr00··on Amazon workers under pressure to up their AI usage are making up tasks
Yes, with the caveat that the 30% work allocation counts toward likability. You can be friendly, charming, well-spoken, fun, etc., but if you fail to deliver and make work for other people, cause your coworkers frustration, and make your manager look bad, you're not going to move up. You will be able to coast for a while though, as managers have a hard time firing people they personally like.

It's ultimately a combination. A pretty good software developer who is friendly and pleasant is, in most organizations, going to get promoted over the grumbling angry software developer who is brilliant but everyone hates talking to. A lot of this has to do with most work at more senior levels being communication.

mjr00··on Amazon workers under pressure to up their AI usage are making up tasks
I've heard similar stories from AWS and other non-AWS FAANG employees. All of the token leaderboards have a "this doesn't count toward your performance review" disclaimer, but there's an implied nudge nudge, wink wink after that statement.

One person I've talked to has someone in their org who is running GasTown and chews through tokens 24/7. They don't contribute very much, but they're comfortably in the #1 spot.

mjr00··on Rars: a Rust RAR implementation, mostly written by LLMs
This is Rust we're talking about. It doesn't even need to work; as long as it compiles, it's correct.
mjr00··on If AI writes your code, why use Python?
Python has had type hinting for quite a while, and adding validation with mypy/pyright/ty as a step in CLAUDE.md (as well as having it as part of your CI pipeline) can emulate static type checking pretty well.
mjr00··on Agents need control flow, not more prompts
> Some have expressed the opinion in this forum that the future of software lies in programs that are created and adapted at runtime, using genAI.

Good luck with that. Users will flood you with complaints if a button moves 5px to the left after a design update. A program that is generated at runtime, with not just a variable UI but also UX and workflows, would get you death threats.

mjr00··on Vibe coding and agentic engineering are getting closer than I'd like
> OpenClaw is a good example of a completely new project written using coding agents that made a significant impression on the world and would not have been built without them.

Nobody is denying that OpenClaw is popular, and nobody (in this thread, at least) is denying that AI rapidly speeds up the ability to make an initial release or prototype for greenfield projects. But the comment that spawned this discussion was:

> we still have no companies compressing 10 years into 1 year thus exploding past all the incumbents who don't "get it".

The issue is that you're extrapolating OpenClaw, which upon release was a month of pre-AI development work compressed into a few days, to cover the "10 years into 1 year" scenario. However, this isn't appropriate because software development is non-linear. As anyone who has worked on a greenfield project pre-AI should know, those first weeks and months have much faster development cycles. There's no tech debt to worry about; there's no urgent bug tickets or feature requests from customers; there's no thinking about whether it's okay to ship a breaking change.

mjr00··on Vibe coding and agentic engineering are getting closer than I'd like
> 41,964 commits is a lot more than "a month of greenfield work".

I meant a month for the initial release, not current state.

Regardless, much like lines of code, number of commits is not a good metric, not even as a proxy, for how much "work" was actually done. Quickly browsing there are plenty[0] of[1] really[2] small[3] commits[4]. Agentic coding naturally optimizes for small commits because that's what the process is meant to do, but it doesn't mean that more work is being done, or that the work is effective. If anything, looking at the changelog[5] OpenClaw feels like a directionless dumpster fire right now. I would expect a lot more from a project if it had multiple people working on it for 5 years, pre-AI.

[0] https://github.com/openclaw/openclaw/commit/e43ae8e8cd1ffc07...

[1] https://github.com/openclaw/openclaw/commit/377c69773f0a1b8e...

[2] https://github.com/openclaw/openclaw/commit/ffafa9008da249a0...

[3] https://github.com/openclaw/openclaw/commit/506b0bbaad312454...

[4] https://github.com/openclaw/openclaw/commit/512f777099eb19df...

[5] https://github.com/openclaw/openclaw/blob/main/CHANGELOG.md

mjr00··on Vibe coding and agentic engineering are getting closer than I'd like
OpenClaw is definitely not a "5 years" project pre-AI though. That was more like a month of greenfield work compressed into a weekend -- which is still really impressive, don't get me wrong! -- but I think the point is we're not seeing mature, legacy codebases get outcompeted by new, agile, AI-driven codebases; we're seeing greenfield projects get spun up faster. Which, again, is still impressive and valuable.

If agents could really compress 10 years of development into 1 year, you'd see people making e.g. HFT platforms and becoming obscenely rich, not making a fun open-source project and getting hired by OpenAI as an employee.

mjr00··on Roblox shares plummet 18% as child safety measures weigh on bookings
In this case, "child safety measures" includes not just "stopping child predators," but also "not letting kids use their parents' credit card to buy $500 of Robux" and "not letting underage users buy lootboxes, aka gambling".

It's completely understandable that the company, which profits off children, putting in measures making it harder to profit off children, would lower both its long and short-term valuations.

mjr00··on Spotify adds 'Verified' badges to distinguish human artists from AI
You'll notice in the article they mention that these AI artists got into the charts thanks to 1,000 downloads sold. 1,000 is a comically easy number to game; that's US$1,000, tops, and far less if you pay for false downloads from SEA and such where prices are lower.

As a concrete example of how gamed these are, look at one of the examples from the article, Enlly Blue[0]. The video for the song mentioned in the article, Through My Soul, has 10 million views. All four of her (its?) most recent videos over the past 1 month: 2.2k views, 3.3k views, 2.1k views, 2.1k views. The views stop coming when the creator stops spending.

[0] https://www.youtube.com/@EnllyBlueOfficial/videos

mjr00··on Spotify adds 'Verified' badges to distinguish human artists from AI
> Lie. You will not. You need to go through the distributor (1), and it has always been this way.

Er yes, which is why I mentioned LANDR and Amuse, both of which are on the page you linked. I mentioned those two specifically because I know they don't charge up-front and instead take a % of royalties, so they're ideal for flooding Spotify with AI slop. I'm not sure which part you think is a lie.

> You need to go through a distributor (1) that does due diligence first, and it has always been this way.

I see you edited your comment. Distributors do not do any sort of "due diligence". For the free distributors, you don't even need to give them personal information until you try to actually cash out your earnings. For DistroKid, when I first signed up I put in my credit card info, submitted my first song and it was up on Spotify 3 days later.

mjr00··on Spotify adds 'Verified' badges to distinguish human artists from AI
> The more appropriate question is why they published a AI artist at all.

Because they allow anyone to upload to Spotify. There's nothing stopping me, you, or anyone from generating AI tracks with Suno & friends, downloading them, and using a service like LANDR or Amuse to distribute them to Spotify, all for free.

> Like Spotify owns distribution, their largest investor Tencent Music Entertainment Group publishes AI-generated music = almost infinite profit.

This assumes that real people are listening to AI-generated music which does not seem to be the case. According to Deezer, 85% of streams on AI-generated music are fraudulent.[0] It's largely a vanity ouroboros where someone with more money than sense generates a song, pays bots to get fraudulent streams, and uses those streams to generate vanity metrics. Consumers are by and large not listening to AI generated music.

[0] https://techcrunch.com/2026/04/20/deezer-says-44-of-songs-up...

mjr00··on Where the goblins came from
> If you make a program which can solve many different classes of problems that's called AI.

What about Salesforce? That solves a ton of different problems!

And introduces a ton of new problems, too; which is strong evidence that Salesforce is intelligent!

mjr00··on "People who don't use AI will be left behind"
> Although I don’t think Google and Calculators are the main culprits, I do think it’s due to larger technology/internet landscape.

That's extremely speculative, especially given there was a major event in 2020 which massively disrupted education worldwide.

mjr00··on United Wizards of the Coast
> Nit picked, I suppose.

It's not a nitpick. There are some industries and jobs where employees truly are fungible. There's never going to be a security guard or Starbucks barista or janitor who does such an amazing job that they're able to negotiate significantly higher wages than their peers because of how much value they deliver. "Software developer" is not that kind of job, at least not right now.

> my overall point does not change that tech unions will depress wages for high earners.

And this goes back to the original point which is that unions only depress wages for high earners if the unions negotiate terms which depress wages for high earners. As has been pointed out in the thread there are many existing examples of unions for collections of high-earning employees.

mjr00··on United Wizards of the Coast
> Sure it does. I'm talking about the majority of tech employees who are fungible while you for some reason are talking about the few that aren't, I'm not sure why they're relevant to this debate.

Because I gave examples of unions which contain and are beneficial to high wage earners, and you dismissed them as irrelevant because tech employees are "fungible" except for the ones who aren't. Except those unions also contain far more "fungible" employees (background actors, backups, practice squad players) than they do star actors and quarterbacks, so it's a close comparison. There's currently 2,314 members of the NFL Players' Association but only 32 starting NFL quarterbacks.

Fungible means all employees are the same, that's the exact definition of fungibility. If certain employees have a measurably different impact on delivery, that is the definition of non-fungible. Just because most employees deliver near-average performance does not change this--it's tautologically expected!

mjr00··on United Wizards of the Coast
> Those 8 and 9 figure employees are the exception that proves the rule

??? This doesn't make any sense. There's a hell of a lot more background actors, which you can call fungible, than there are Tom Cruises, but they're all covered by SAG-AFTRA. "Tech workers are fungible except for the ones that aren't" is tautological and doesn't support your argument.

mjr00··on United Wizards of the Coast
> The difference is that the shitty companies there's never a route to do anything, however unrelated, without running it by HR. Because they think they own you.

"Shitty" companies have legal obligations to include these kind of clauses though, because investors and board members don't want to deal with a situation where there's exactly this ambiguity about what counts as "obviously not-related software". HR exists to protect the company. What counts as obviously non-related to you may not count as obviously non-related to a board member who hears about the new optimization algorithm you wrote and realizes the company can monetize it.

> On the contrary, using a nonzero amount of working hours to work on something pretty clearly makes it company property.

How recently have you seen a tech employment contract where you were obligated to work exactly 8 to 5 with a 1 hour lunch break? Flexible hours means any hour is potentially a "working hour".

mjr00··on United Wizards of the Coast
> Tech workers are fungible while actors or athletes are not though

Strongly disagree with this, and I'm in 'tech management' now so I'm not saying this out of self-preservation.

Tech workers are similar to athletes in skill disparity IMO. A handful of superstars makes a massive difference. Look at those superstar AI developers who are allegedly getting 8 or 9 figure compensation from Meta etc. They're still technically "just" tech workers.

mjr00··on United Wizards of the Coast
> High earners generally do not benefit from unions as they can make more negotiating their own labor value versus participating in collective negotiation which is more likely to drag their compensation down towards the average, even if it's still higher than average.

High earners benefit from unions if union rules are made which benefit high earners.

The most well-known example of a high-earner union is SAG-AFTRA and they provide benefits to established performers, the big one being name exclusivity. e.g. even if your real life birth certificate says your name is "Tom Cruise" you cannot advertise your name as "Tom Cruise". Obviously, actors are still negotiating their own pay.

Another good one is the NFL Players Association which capped the rookie pay scale in 2011 to ensure the salary cap was going more toward veterans, people already in the union. However they still independently negotiate their compensation.

I don't know what people would want out of a tech union specifically but the idea that "union = payscale based on seniority" just plain isn't true.

mjr00··on United Wizards of the Coast
> If you write logistics software, the company will say "we own all logistics software you write". You can't create a competitor. But if you work for a logistics software company and decide to go write a video editor on your own time, the company wouldn't own that.

The problem is there's no clear legal definition of what "logistics software" is. A video editor is a seemingly obvious example of what is not, but what if you came up with a novel optimization technique which is not necessarily only applicable to logistics? Could you spin off that software into a separate business? What about something more fundamental, like tooling? Think about something like Slack, which was just meant to be an internal messaging tool created as part of the development of a video game. Imagine after Slack took off, that an employee claimed that because they had written Slack at least partly outside of working hours and it was not "video game software", the company didn't have ownership of the software.

This is why the common approach is that the company owns everything except anything explicitly carved out, it avoids ambiguities like this.

mjr00··on Laws of Software Engineering
> If you add a new thing (constant, route, feature flag, property, DB table) and it immediately needs to be added in 4 different places (4 seems to be the standard in my current project) before you can use it, that's not DRY.

The tricky part is that sometimes "a new thing" is really "four new things" disguised as one. A database table is a great example because it's a failure mode I've seen many times. A developer has to do it once and they have to add what they perceive as the same thing four times: the database table itself, the internal DB->code translation e.g. ORM mapping, the API definition, and maybe a CRUD UI widget. The developer thinks, "oh, this isn't DRY" and looks to tools like Alembic and PostGREST or Postgraphile to handle this end-to-end; now you only need to write to one place when adding a database table, great!

It works great at first, then more complex requirements come down: the database gets some virtual generated columns which shouldn't be exposed in code, the API shouldn't return certain fields, the UI needs to work off denormalized views. Suddenly what appeared to be the same thing four times is now four different things, except there's a framework in place which treats these four things as one, and the challenge is now decoupling them.

Thankfully most good modern frameworks have escape valves for when your requirements get more complicated, but a lot of older ones[0] really locked you in and it became a nightmare to deal with.

[0] really old versions of Entity Framework being the best/worst example.

mjr00··on Deezer says 44% of songs uploaded to its platform daily are AI-generated
This is the very definition of unnaturally: the creators of those AI songs spent a ton of money promoting them, for whatever reason.
mjr00··on Deezer says 44% of songs uploaded to its platform daily are AI-generated
I'm not saying it's impossible, but at a minimum it's extremely hard to game the recommendation algorithm (primarily talking about Spotify, maybe Deezer's is less sophisticated). The best way to "game" the recommendation algorithm, to kickstart a new/less-established artist profile, is to get onto popular playlists. However these playlists either have actual quality barriers (so they won't put AI slop music on) or they take $$ (so this doesn't really work with the "mass generated AI slop" approach).
mjr00··on Deezer says 44% of songs uploaded to its platform daily are AI-generated
Yes, last I checked at least LANDR, Amuse and Tunecore all had plans where you release for free and they take a % of royalties.
← PreviousPage 4 of 34Next →