Something I’ve been thinking about, somewhat related but also tangential to this topic:
The more code gets generated by AI, won’t that mean taking source code from a company becomes legal? Isn’t it true that works created with generative AI can’t be copyrighted?
I wonder if large companies have throught of this risk. Once a company’s product source code reaches a certain percentage of AI generation it no longer has copyright. Any employee with access can just take it and sell it to someone else, legally, right?
In theory, companies are all going to have an increasingly difficult time suing competitors for copyright infringement. By extension, this is also why, IMO, its important to keep AI generated code out of open source/free software projects.
The recent rulings on copyright though also need to be further tested, different judges may have different ideas on what "significant human contribution" looks like. The only thing we know for certain is that the prompt doesn't count.
My guess is that instead of enforcing via copyright, companies will use contracts & trade secret laws. Source code and algorithms counts as a trade secret, so in your example copyright doesn't even matter, the employee would be liable for stealing trade secrets.
AI generated code slowly stripping the ability of a project to enforce copyright protections though is a much bigger risk for free software.
It becomes less interesting the more the “overweight” stocks correct.
The extreme concentration risk lessens as these 8 stocks fall in value compared to the rest.
I also don’t personally see the risk in the concentration. Risk of what? These companies are legitimately larger and doing more business than other firms.
Pick a median consumer. Which company are they sending more profit to than companies like Apple or Amazon?
10 years ago the average consumer maybe bought an iPhone from Apple every 3 years, so they gave Apple less than $100 of pure profit dollars per year.
Now that same consumer is giving Apple money for the iPhone, but also spending on services that they weren’t buying 10 years ago. If they’ve got an Apple One subscription they’re now sending Apple double or triple the profit they used to get.
These companies are big because they sell more things and are more diversified than they were in the past.
There’s no concentration risk. I’d actually argue that the concentration risk can be resolved overnight through antitrust regulation (e.g., force Apple and Amazon to split into multiple companies, as they already have obvious verticals that could stand alone).
The concentration risk relates to diversification in investing. Index funds are generally thought of as a way to diversify a portfolio. Cap weighted index funds are generally preferred because they are cheaper for the provider to maintain. Compare VOO with RSV for example. VOO is cap weighted. RSV is equal weighted - which means investors in RSV bear the cost of periodically readjusting all holdings so they are once again equally weighted - something no necessary with VOO.
I am not the only investor who has taken steps to offset the overly high concentration in the SP500 that raises the riskiness of an investment portfolio. I've done so by splitting my VOO holdings in half, split 50/50 VOO/VTV that strategically diminishes the impact of the high top 10 stocks in the SP500.
I certainly think it's a good thing to diversify investing, while recognizing that there is value in putting a lot of your bets into heavyweights that are very likely to do very well in the long term.
One of my main points here is that dumping a lot of money into one company isn't always something that represents lack of diversity in your investment dollars.
A company like Microsoft has its hands in so many business verticals that its stock by itself is a highly diverse asset.
I also think it's important to realize that massive companies like these have inherent advantages over smaller ones. A company like Framework literally cannot make a better laptop than Apple even if an angel investor dropped billions of dollars into their laps. Even if they pulled it off, it wouldn't come with a free trial for Apple's content subscriptions and other revenue-maximizing features, and the wholesale price they get from the factory can't match Apple's margins on the device until they convince a large enough mass of people to buy them.
That's the kind of stuff that big companies can do, and that's why they are worth more putting more bets into than smaller ones.
Obviously, companies like Tesla and Nvidia are far bigger risks in the S&P 500, but they represent a small minority of those giants.
There is nothing wrong with your desire to 'dump[ing] a lot of money into one company'. That is easy to do without an index fund. And it is not the investing theory behind the creation of index funds and their investing purpose. When 8 companies dominate an index fund, that means the index is not performing the intended function for which it was created.
I wonder what a solution could look like. Perhaps keep the market cap weighting, but cap the weighting at a max $500b (or some sliding scale to prevent the top X stocks from composing more than Y% of the portfolio)
That would certainly be a way to control escalating concentration but at the expense of keeping index fund costs low. The Vanguard Total Stock Index (VTI) has an expense ratio of 0.03 - almost zero. Low expenses is a critical factor behind why index funds outperform active investing. So, yes, your proposal would work, but the expense ratio would up to implement the cap.
But the index fund is doing what it was designed for, which is to index on the companies based on their relative importance in the marketplace.
And that’s really my whole point. Someone who is buying an S&P Index fund wants to own more Apple than GoDaddy, because Apple represents much more economic activity than GoDaddy.
I have read John Bogle extensively. I believe he would disagree with you about the purpose behind why Bogle invented the index fund. Index funds are cap based primarily because that saves on costs (there is no need to rebalance the index). But the philosophical framework is diversification. When 10 companies make the other 490 irrelevant in producing the annual return of the index, the index itself is no longer serving the diversification purpose.
Nobody is going to deny enjoying the monetary gains produced by the index becoming concentrated. But it comes at the cost of the portfolio risk that diversification (i.e. absence of concentration) is intended to eliminate.
I’ll make an analogy to maybe help explain what I mean further:
I own a somewhat diverse set of 50 company stocks, at least for the purposes of this exercise.
Let’s say a bunch of those companies merge, now there’s only 20 companies.
No product lines have been discontinued. The companies make all the same things with the same client lists.
Did my investments become less diverse when these companies merged? Perhaps in some ways yes, in many other ways no.
Is my investment portfolio more diverse if I own one stock, Apple, or if I own three stocks, Time Warner, Paramount, and Comcast? All these companies make media content, but Apple is in more industry verticals overall in addition to being a media company (or at least, we can say they are for the purposes of this analogy). If the content industry collapses, Apple is fine, the rest not so much.
Sounds like some lame ass tech founder bullshit if I’ll be honest.
If I had cancer the last think I’d be thinking would be to make a slide deck about it.
Can these robot people come back down to earth and have a genuine human experience for a chance? Not everything has to be framed in the view of a startup company or a data analysis exercise.
Maybe focus on spending time with your family and friends? If they still like you after years of being an insufferable tech bro.
He’s been public that he’s ten months clear now. Some prefer to accept undesirable circumstances. Others prefer to oppose them. He’s one of the latter. A little paraphrase of Dylan Thomas’ work here is something I’m fond of:
Do not go gentle into that good night
Rage, rage against the dying of the light
And if he’s successful, which hopefully now he has a much better chance of, there are all these new medical results out that are useful.
As an example, a close friend is using one of the personalized medicine companies that sytse’s “CEO of care” has invested in to diagnose a persistent debilitating condition with no specific cause.
Or to quote someone else: All progress depends on the unreasonable man.
> Serviceable, repairable, upgradable Macs are officially a thing of the past.
Well, not exactly. Apple’s desktop Macs actually all have modular SSD storage, and third parties sell upgrade kits. And it’s not like Thunderbolt is a slouch as far as expandability.
I can see why the Mac Pro is gone. Yeah, it has PCIe slots…that I don’t really think anyone is using. It’s not like you can drop an RTX 5090 in there.
The latest Mac Pro didn’t have upgradable memory so it wasn’t much different than a Mac Studio with a bunch of empty space inside.
The Mac Studio is very obviously a better buy for someone looking for a system like that. It’s just hard to imagine who the Mac Pro is for at its pricing and size.
I think what happened is that the Studio totally cannibalized Mac Pro sales.
And here I am with a gaming PC that has absolutely nothing in the PCIe card slots except for a graphics card. 1 out of 5 slots filled!
The truth of the matter is that there's basically no hardware on the market that actually depends on PCIe bandwidth besides graphics cards.
Furthermore, us computer nerds don't like to hear it, but making someone open up a computer is a barrier for many customers.
E.g., I'm selling a computer to a video professional and they want a processing appliance that connects to a computer to help with their video workflow.
Is that video professional also skilled in computer subjects like opening up computers? Probably not! A nice box that you plug in to Thunderbolt is way simpler.
Bonus points, you can plug that box into any other computer without taking your whole computer apart.
Apparently the Neo is surprisingly repairable - in that parts can be replaced, not that you can buy stuff at Microcenter or Fry's (RIP) and shove them in.
I didn’t phrase myself very well. What I’m saying is that the loss of the Mac Pro didn’t reduce the repairability or modularity at all in the product lineup.
It was exactly as modular as the Mac mini and Mac Studio.
The only difference is that it had some PCIe slots that basically had no use since you couldn’t throw a GPU in there, and because thunderbolt 5 exists.
Yeah, sure, there were some niche PCIe things that two people probably used. Hence the discontinuation.
I am an ex-Mac user, I own a Framework. Don’t worry, you’re preaching to the choir.
Those are all for Intel Macs, and not even the recent Intel Macs. You can't use a passive adapter to put a NVMe SSD into a current Mac like you could a decade ago, because back then the only thing non-standard about the SSD was the connector. Now most of the SSD controller itself has moved to the SoC and trying to put an off the shelf SSD into the current slot makes no more sense than trying to put an SSD into a DIMM slot.
Honestly I don't care, but Apples SSDs don't have a storage controller on them, and those adapters are designed to "bypass" the controller on m.2 drives.
You can argue that it's different for the sake of being different, but
A) I personally don't always hold that monopoly is a good thing, even if we agree m.2 is fairly decent it doesn't make it universally the best.
B) I'd make the argument that Apple is competing very well with performance and reliability..
C) IIRC there are some hardware guarantees that the new filesystem needs to be aware of (for wear levelling and error-correction) and those would be obfuscated by a controller that thinks its smarter than the CPU and OS.
if we're talking about Intel era Macs then that proprietary connector predates M.2 entirely and is actually even thinner and smaller (which is pretty important when the primary use-cases is thin-and-lights); though I suppose that the adapter fits is a sign that it would have been possible to use a larger connector...
That is an absolutely awful argument against what I just said. I can tell that you don't care.
Tens of thousands of mini PC and laptop boards ship with multiple M.2 slots. Apple can use both connectors, with the exact same caveats that normal M.2 SSDs have on ordinary filesystems. Apple does not have to enable swap, zram, or other high-wear settings on macOS if they are uncomfortable with the inconsistency of M.2 drives. Now, I'd make the argument that people don't complain about APFS wear on external SSDs, but maybe I'm wrong and macOS does have some fancy bypass saving thousands of TBW/year.
Whatever the case is, "the annoying thing is competitive" was not a justification for the Lightning cable when it reached the gallows. It did not compete, it specifically protected Apple from the competitive pressure of higher-capacity connectors. The same is true of Apple's SSD racket and the decade-old meme of $400 1tb NVMe drives.
I don't buy that argument, "a PC by any other name" is what made intel mac's somewhat uncompetitive when compared to the M-series laptops: which are currently dominating with total vertical integration of the OS and hardware.
Also: All things being equal, the lightening connector was technically superior to USB-C and arrived much earlier.. so it's somewhat on the same path.
USB-C succeeded due to a confluence of;
A) Being a standard people can get behind. (lightning was, of course, much more awkwardly licensed)
B) Lightning never got a sufficient uplift from USB-2.0 performance.
C) The EU eventually killed lightening through regulation.
It was, however, smaller, more durable and (as mentioned) earlier.
I'm totally not against our new USB-C everywhere situation w.r.t. phones, but if anything it reinforces the point: The technically superior thing being too proprietary caused its death (despite being early).
Well, this specific solution was only set up on specific hardware, and is Nvidia dependent, as the readme stares.
That doesn’t mean the 9070XT can’t do AI stuff, quite the opposite. ROCm gets better all the time. There are many AI workloads you can do on AMD cards.
Is it a card I would choose if I was primarily working on AI? Absolutely not. But it is the card I own and it’s been a great value for gaming.
Unfortunately AMD is much worse with supporting AI features like FSR4 on older hardware generations, despite the capability and leaked INT8 models being there. Totally unlike NVIDIA.
It’s absurd I have to use open source programs to get INT8 FSR4 support.
The issue isn’t the proof of citizenship. The issue is that poll taxes are unconstitutional and there is no state that I’m aware of that makes the acquisition of identification documents free of cost.
I’m honestly quite surprised that politicians don’t resolve this idiotic situation because it’s so damn simple, but I think it’s not solved because various state governments rely on small fees for revenue. And of course because there are many political situations in which making it difficult for specific opponent voters to vote is a campaign strategy.
Make fees for drivers licenses, birth certificates, and passports illegal, and ideally institute a system that makes these forms of identification automatic/stupidly easy to acquire and the whole issue is resolved. Now you can require voters to present them and you aren’t disenfranchising anyone.
Thanks for clarifying. Not being of USA I didn't even consider the angle about having to pay for government ID - it's a very alien concept for us in eastern europe at least.
Election fraud on the other hand.. this we are very familiar with. Reaction to the coverage of last three US presidential elections was mostly "oh, how cute, such naive first attempts". So from our PoV there most certainly were widespread attempts to rig them, mostly from Dem side, and so very unprofessional, that their existence cannot be denied in good faith.
> from our PoV there most certainly were widespread attempts to rig them, mostly from Dem side
Oof, I feel bad for whatever news network you are getting your American coverage from. You might want to look into who owns that news network. This is a very common political message specifically originating from the Republican Party’s media network (e.g., Murdoch-owned media, Turning Point USA, etc).
To "get coverage from a news network" is ignorance bordering on pure madness.
They are almost exclusively propaganda and manipulation and as such the only useful signal that can be extracted is something like "how those people chose to frame certain events they feel they can't ignore in hopes of them going unnoticed". Note I'm talking about our local ones, in my opinion yours do not differ materially in this aspect.
So no. I'm not parroting after a talking head on some network or other (the thought itself is mildly insulting). For an interesting incident (and election-related stuff was interesting enough) what one does is gather as much coverage as possible and then try to reconstruct what event could have lead to this set of framings.
What I wrote is somewhat of a consensus between us old hands of many years experience resisting election fraud, with hands-on knowledge of how it's done, how to fight it, how attempts at covering it up look like and how people that prefer to believe it never happens behave.
I just respectfully disagree with your view in a whole lot of ways.
I don't know how you can be aware of things like the January 6 insurrection and the fake electors scheme and believe that election fraud is "mostly dems."
We have a recorded phone call of Donald Trump asking the Georgia secretary of state to "find 11,780 votes" for him.
It seems like we are creatively bankrupt if we can’t think of any solution. I think many of us could think of a good solution in literally seconds.
And there’s a really good argument that a solution isn’t actually needed.
Does the NBA need a solution for Steph Curry being the best 3 point shooter of all time and dominating his competition? Did the NFL need a solution for Tom Brady winning the Super Bowl 30% of the seasons he played in his career? Did Ohio high school basketball need a solution for LeBron James only losing 6 games in his entire high school career?
Athletes dominating their league happens all the time without the issue of transgender and intersex players.
If there is some kind of mass influx of men playing women’s sports to win easy championships that’s when we can deal with the problem. But as of now there is no such problem on any kind of significant scale. E.g. there has never been a time when washed up NBA player that decided to try and join the WNBA. We don’t need to solve problems that do not yet exist.
But let’s say we have to solve this problem to make everyone shut up about it. Here’s one I just thought of off the top of my head:
Anyone who performs at a level of play at an abnormally high gap between themselves and their competition (a set statistical percentage better) can be forced to seek a higher league of play if it exists and they are eligible if and only if other competitors in the league request they do so with a strong consensus.
Is this a perfect solution? No, but I thought of it in literally ten seconds, it doesn’t even involve gender, and I didn’t resort to sitting on my hands and saying “aw shucks there’s no solution” or “I guess we’ll just have to ban trans people from sports.
I think not many people are arguing that we shouldn’t exclude people based on testosterone in elite events, but none of these were trans women, these were all women who lived their entire lives as women from the moment they were born
I'd argue about testosterone. High testosterone happens in some woman naturally, why exclude them? They still are woman, they should have a right to participate.
Height is also an advantage in sports, and women statistically are much shorter then man, should we ban tall woman from sports? Should we say "she exhibits a male amount of height, it isn't fair to let her participate with 'normal' woman"?
The more "fair" we make woman competition the narrower our definition of a woman gets.
If you want to make it fair, let's pick a random chemical in man exclude people from competition based on their readings. That surely would make sport career look more fun for everyone, training all your life only to find out that some committee doesn't consider you a man. And then we can celebrate equality by noticing that man-to-woman sport participation ratio got closer to 50-50
My view is that testosterone is a reasonable thing to discriminate on because:
1. It is causally connected to primary and secondary sex characteristics
2. It has a large impact on performance in many sports
3. It's easy to explain to most people and somewhat matches people's intuitions around fairness
But, yes, it is true that there are cis women with high T levels and it is somewhat unfair and arbitrary to include them when not excluding other random advantages that people have. I'm just not sure if I have a better solution
It's dumb because there are two types of hyper/hypo-gonadism. "Primary" hypergonadism is where you have way more of the hormone in your blood stream. You're advocating testing for only "primary hypergonadism" in women.
Secondary hypergonadism is where someone has a normal concentration of the hormone in their blood, but they have an unusual abundance of hormone receptors.
The effects are the same, but currently we can only measure secondary hypergonadism during an autopsy/dissection.
> But as of now there is no such problem on any kind of significant scale.
This is not the same as saying there's no problem.
A fraction of humans will ever compete in the Olympics. People train their whole lives for it. It's not about 'scale', it's about safety and fairness. It's not reasonable to expect them to 'shut up' about it.
I don't want to watch a man beat up a woman in a boxing ring.
I was going to say that I’d be happy to run a local Mac mini to be a runner but I noticed that Forgejo runners are only built for Linux.
It seems like to be a serious CI platform they really need to change Windows and Mac binaries for runners so you can build for those platforms.
And this is more of a Forgejo issue than a Codeberg issue specifically.
But also, I’d also throw out there the idea that CI doesn’t have to be at the same website as your source control. It’s nice that GitHub actions are conveniently part of the product but it’s not even really the top CI system out there.
Forgejo is committed to using exclusively Free Software for it's own project development. Windows and Mac versions of the Forgejo Runner are built in the project's CI system as a minimal check to ensure platform compatibility, but due to the project's commitment, the project doesn't do integration testing on these platform. And therefore doesn't distribute untested software.
Something about your phrasing is such hilarious techbrained spin.
Let’s be real: OpenAI is circling the drain.
The company with the fraudster serial liar CEO who said he was gonna spend a trillion dollars can’t keep a video service alive right after signing a $1 billion dollar with Disney?
What kind of a joke is that?
This is a company that has blown its opportunity twiddling around with zero product. They still just run a plain chatbot interface with zero moat and zero stickiness.
There’s no “pivot” for a company that is in this deep.
Sam was brought back because there was no one to replace him. The non-profit types on the board were living in a consensus bubble that didn't extend far beyond a small inner circle, and they discovered that they didn't have sufficient support from the engineers who had lots of other employment options and threatened to quit if Altman wasn't reinstated. Altman himself had no problem finding a replacement job in a matter of hours, and the board was looking at a business drained of talent in a cut-throat tech race.
I'm no fan of Altman or OpenAI, it's a pretty shady company and I am suspicious of their books, but this was a great demonstration of the uselessness of boards and how out of touch they are with the business they are supposed to be supervising. It's really rare to find an effective board, primarily they sit like a House of Lords enjoying ceremonial perks and a stipend in exchange for holding a few meetings a year.
The more code gets generated by AI, won’t that mean taking source code from a company becomes legal? Isn’t it true that works created with generative AI can’t be copyrighted?
I wonder if large companies have throught of this risk. Once a company’s product source code reaches a certain percentage of AI generation it no longer has copyright. Any employee with access can just take it and sell it to someone else, legally, right?
reply