You tell them to cancel their Astra subscriptions and get to work. Hundreds of mathematicians, including Peter Scholze, have already agreed to not use AI at all. There’s absolutely no reason to help train an AI that will just make your own life more miserable.
A friend of a friend had a grant project written up for the next 10 years, with habilitation on the horizon.
His end goal was one of the things released in the recent OpenAI publications. It took a model 3.5 hours to do.
Things are wild now on math and theoretical faculties now.
And for theoretical fields labs don’t need human data - synthetic one works well if not better.
The best thing you can do if you’re a scientist, imho - wrapup whatever grant you have now asap with ai, use ai to get more grants if possible, and spend the rest od the year learning what new science you can do with the new tools and participate in creating the new paradigm in your field.
That’s when you’re an established scientist. If you’re new in the field then this is the most exciting time you could wish for - an opportunity to make a name for yourself.
One natural step of AI usage, the way I see it, is that even mediocre researchers can use AI as a harness to become prolific researchers.
So if you're part of a pure "human only" researchers that publish 1/10th of what the rest are doing, how are you going to survive?
The more prolific researchers will eat your grants for lunch.
This is exactly the same thing a lot of software devs are going through now. AI models have lifted up EVERYONES ability to produce code, so there's simply no premium anymore for those that will only code by hand. And the people paying their salary are asking why they aren't more productive, when even business analyst Joe is pumping out new products left and right.
People will pay good money for artisanal, hand-crafted clothing, or watches, or cars, or whatever, but I doubt that the market for artisanal mathematics is now or will ever be very large.
It's just stubbornness, in the end, that is manifesting as a kind of gatekeeping Luddism. I see the same thing in my Mastodon feed from folks in the software engineering world, particularly people who have a great love for writing code and hold it as a core part of their identities. I understand being irritated at a machine having been trained on (potentially, inter alia) your work with no compensation to you, and now it is perhaps better at (some of) what you do than you are; or maybe companies are now making a lot of money off of something that benefited from your contributions; but I find it extremely improbable that we will be able to refuse our way out of further progress now that so much money has been invested and so much momentum has been gathered.
I don't think I feel any discomfort about a machine being better at anything that I can do than I am at it. Maybe it's because I'm a mediocre person. I personally think that if there's something that I'm better at than a machine, then we need to build a better machine. Imagine what I'd be able to do if I used that machine!
To be fair I see two separate ways CS students use LLMs, one is to just complete assignments in the same way you could by asking someone else to do it, the other way is by getting an LLM to explain certain concepts in another way, or create more examples of them to aid in understanding quicker/in more depth. I actually think in the 2nd case they can be incredibly useful and can assist in creating a deeper understanding, but in the 1st case they obviously create fundamental gaps in understanding.
In a way it's on the student to make sure whatever they're doing, they actually understand it, or come exam season the gaps in their knowledge sans Claude will become quite obvious.
You can't just handwave the fact that we're requiring what probably amounts to two orders of magnitude more discipline and self control out of students than we've ever done before. The educational system will collapse under this.
> To not bury your head in the sand hoping this will go away, because it won't.
Eh? As Greg K-H mentioned in that video from a few days back about how very disappointing Fable was when compared to its astronomical hype, LLMs that one can run locally are quite good enough for a great many tasks... including bug hunting in the Linux kernel. And -as LLM boosters keep saying- they're only keep getting better, right?
We absolutely should be telling people to stop using the LLMs from the major LLM manufacturers. Those manufacturers have spent so many billions of dollars on this project, made so many promises that they're going to have no way to keep, and now that they're running into resistance, [0] they're holding all of humanity hostage unless they're permitted to get intimately involved in the creation of new laws and regulations especially for them. [2]
Even if one only considers their recent threats to humanity, it's clear that these are not companies that deserve any of our hard-earned money.
[0] Some of that resistance comes from their ever-more-sharply-increasing cost to produce the next performance increment, some from folks asking the probing questions about their promises, claims, and business practices that should have been asked years ago, some from ongoing State AG's court cases in regards to their illegal conduct, and some from folks who -unsurprisingly- don't want enormous warehouses that suck up quite notable amounts of power [1] but provide dreadfully little revenue to the areas that house them in their communities, and still others who are starting to realize that benefits of cloud LLMs aren't worth the tradeoff of being unable to afford a new personal computer.
[1] Note carefully that I made no reference to water usage. The only counterargument you can make here is that 500MW -> 2GW is not a quite notable amount of power.
That's going to be a very short term solution, no?
You can't really swear off looking at problems solved by AI, or keep moving away from fields whenever AI starts contributing, and have a viable career.
Today's frontier models, are everybody's local models in the near future.
If this were a change that had expected stalls or reprieves it would be different.
Doing mathematics looks to be the next chess, er, I mean protein folding challenge. The latter's computation demands have dropped significantly.
There are an infinite number of problems that require a unique proof (you can prove this btw). Yes maybe you need to work harder to find a more obscure problem now but why give up the fight already? I believe that students should have the right to intellectually stimulating education. If you really need to use AI, go for something that isn’t trying to put you out of business or destroy your right to a decent education, like a local open weights LLM.
> There are an infinite number of problems that require a unique proof
Yes. The problem isn't that there won't be any problems. The problem will be the rate of general progress and solutions that at least some awareness is required of, to reliably identify a good new unsolved problem will just keep getting more challenging. And then an attempt needs to be make, to solve it in a very short time.
There is always another race to run too, but increasingly slow runners don't find that translates to winning.
This change is not going to slow down, it is going to speed up. Machines will be doing math systematically, checking off the meta-math theorems that ensure axiom combinations are covered, that the search does not stop where theorems are not exhausted, and does stop where it can be proven they are. Humans will never operate at that level.
You have many fields that will be protected from ai, or even enabled by it - if I could bet, sociology will be a hot new field that will by definition still need humans always, but will gain both speed and significance.
I think we should apply AI into human, so we could be better and then start doing more advanced stuff required for the Universe explorations. Current AI/LLMs can be considered as first step to that. Take and look about our history. We could never achieve so many if we stopped doing great things because of <any reason>.
Why do we ever use PCs, smartphones, Internet, electricity, medicine and so on? Why do we live? For what?
I don’t have a problem with using AI in the applied sciences once a researcher has proven their skills. I do have a problem with it damaging education, the pure math pipeline, and art.
I understand but I guess the future (far future or who knows) could be like that human just download SKILL and that's it - no more learning, study ing basic things. While basic things could be advanced math, computer science, biology and just everything (or nearly everything) what human do. I know it is science fiction right now but who knows. Smartphone + the Internet were step 0. AI/LLMs are next iteration. Still knowledge is valueable and so on but I guess eventually we get to the point which I said. Then galaxy exploration - if we still be here alive :)
I don't have a problem with AI to generate code..so long as the developer understands the output and would review his own mess rather than expecting those who know how to code will save the burning house after everyone's left in panic.
The best part is that when the academics fix OAI’s issues, the model gets better and OAI shareholders get richer and more powerful!
As someone who uses LLM tech occasionally, this is why I prefer using open local models. If I’m making myself obsolete, at least I’m not making some asshole richer and their closed model better.
You shouldn't be getting downvoted for something that a majority of pure math and art enthusiasts believe to be true. The truth is many of these entrepreneurs and VCs are obsessed with AI not for money or human progress, but because it makes them feel closer to being a "god" rather than a mere mortal. Much of it (especially AI art) is out of spite for human creativity, which is done by mortals with limitations.
Taking the stance of moral superiority is kind of funny. And pure math and art enthusiasts don't think they are closer to being a "god" from understanding/"discovering" math?
Stop coping and deluding yourself mate.
To begin with, whether AI is the one doing the discovering or not makes no difference. Any "pure math" person would aim to understand regardless - and would be quite glad that they have a longer paved path.
Any mathematician in academic or industry is more than likely not a "pure math" person (tainted by capitalism).
Lmao what a ridiculous response. Yes, some mathematicians and artists are in it to feel smart. But the vast majority also just enjoy the process. Having a computer do all the work for you and just typing prompts in ruins that completely. As Ronny Chieng said in his Harvard speech, the journey is the point.
>Any mathematician in academic or industry is more than likely not a "pure math" person (tainted by capitalism)
Ignoring that I meant pure as in non applied math, let's just make it clear: you agree that mathematicians who are against capitalism encroaching on this process should be allowed to dislike it without criticism of being pretentious?
I’m vehemently anti-genAI in creative fields but even I think this metric is stupid and unfair to younger generations and generations to come, even if genAI writing gets better.
If you can’t trust a contemporary writer to not use genAI, then find another interest with true zero trust creative verification (like improv) because copying and cheating on writing has existed for decades.
However, my unpopular opinion is extremist in another way: societies should probably not tolerate AI in creative fields like creative writing at all, if Sora can be shut down, so can other useless and toxic parts of LLMs.
As I explained in an earlier thread, MongoDB tried using AGPL. AGPL is not a barrier for Amazon, they still will resell your product without contributing. MongoDB ended up using a variant of AGPL that is even stricter (requiring the entire tech stack to be under the same license) but is no longer considered FOSS. Until the attitude changes around what FOSS is, this will keep happening.
Um. Mongodb changed its license before AWS offered a mongodb compatible service. And since I can't get the source code for documentdb, either it isn't actually using a fork of mongodb, or Amazon isn't complying with the AGPL. I think the latter is pretty unlikely.
AWS never offered MongoDB as a managed service, or used any of their server software when it was licensed under APLv3, or SSPLv1.
However, we have contributed patches to MongoDB even after their license change to improve its performance on Graviton processors. Because that's what's good for customers, and MongoDB is an important customer and partner.
AGPLv3 gives all the permissions needed to offer software as a manged service, just like every other FOSS license does. Unfortunately, in my personal opinion, the license has been co-opted by companies that do not care about Software Freedom, and rather hope that companies fear the license so they choose an alternative commercial agreement [1]. I don't think that's good for the community.
> Does Amazon have any policies against offering AGPv3 software as a service?
Each use of AGPLv3 licensed software has to be reviewed to ensure that the obligations of the license can be and will be met (and also screen for cases where it is known that the vendor of software does not prefer a company like Amazon import the software under a FOSS license). Today we use AGPLv3 licensed software internally, and include AGPLv3 licensed software in Amazon Linux (Ghostscript, in particular).
> Does Amazon contribute funding back to the software projects they offer as a service?
Yes, in varying ways. For example, it is not easy to provide "funding" for something like Apache Kafka. You need to have people working on the upstream.
> Does Amazon contribute code changes back to the software projects they modified when offering them as a service?
Yes, but not all changes are appropriate for upstream.
> But the (modified) source is available to consumers of the service either way, under the AGPL?
If Amazon made an AGPLv3 licensed program available to others over a network, it would have an obligation to provide to anyone that has access to that program the complete corresponding source.
Today, there aren't any services from Amazon that offer AGPLv3 licensed programs as a service. An example that may come to someone's mind is Grafana, but there is a partnership there, and AGPLv3 is not the binding license in that case.
In my personal opinion, AGPLv3 compliance is not difficult, so long as the licensor of the software is committed to community-oriented copyleft enforcement principles [1].
I think it would be awesome if Amazon could offer hosted AGPLv3 software, plus revenue share with the developers, code contributions to the original project and public forks containing non-upstreamable changes.
If AWS decides to copy your product, going closed-source or source-available just means they have to copy it from design docs or protocol specs. That's more friction than being able to reuse code outright, but it's not going to stop them.
If Amazon don't have a need to change anything in software, they'll just provide it as service without any problems. AGPL permits this.
If they have to change something, then they would likely want want to return hose changes in upstream to lower maintenance burden. Or just publish changes on github of upstream doesn't want to accept them. AGPL is fine with this too.
If Amazon would like create similar offering but with some secret sauce that they don't want to share, then they'll develop in-house solution from scratch and sell it as a service in AWS.
I heard there was a lawsuit about this, but can't find the outcome. Can someone please enlighten me how that story ended (if it had - but I think it should, it's been quite a while ago)?
I enjoyed the post and this is off-topic but when I was younger I was very pro-GPL and I still like copyleft, but to me the logic behind it makes less and less sense as time goes on. RMS compares software to a physical book you can modify, make copies of, and share with your friends, but in my opinion, software is much more like a power drill, a tool you can modify with difficulty and let people borrow. The analogy for software as media is also ironic because the FSF themselves claim that the GPL is not intended for media works and recommend other licenses (like Creative Commons which I support).
I do like the idea that GPL prevents big corps from taking advantage of programmers, but the idea that a GPL program is inherently better than a proprietary program written by an independent developer no longer makes sense to me, though independent proprietary software is often abandoned. But more importantly, it doesn't seem fair when a company makes an AGPL devops tool and then a bigger corp just sells that product as a service. When companies like MongoDB decide to change their licenses so that Amazon doesn't steal all their customers, I don't feel bad about it at all. I have no skin in this game but it blows my mind that there are FSF people who defend massive corporations essentially reselling a smaller company's work solely for the sake of ideological purity. Someone can correct me if I've been misinformed here.
This is not a barrier. Amazon is perfectly able to publish their changes if any. Because the moat is that its hosted inside AWS at a button click. They can also host it for cheaper as they don't have to pay anything to develop the product.
Then why did Mongo relicense? I assumed that AGPL was not enough to maintain their business which is why they are now using a more extreme form of AGPL that is no longer considered Free.
You can fully, legally use AGPL software as Amazon was doing as long as you follow the licensing terms.
Which doesn’t help the FOSS developers extract money out of the cloud users because they can just use the free (for whatever value of ‘free’ Amazon provides) bundled version without any support contacts or whatever.
If your entire business model is based on giving away the software you can see how this might be a problem, no?
Are you disagreeing with me or just giving more detail? Because I think that's what I was saying. My opinion is that this is not fair to a company like MongoDB.
For clarification: MongoDB is still Free as in beer, but not free as in Libre. IOW, Anyone can use MongoDB without paying money, but they cannot offer it to others as-a-service without providing the entire stack of hosting software back to the community. And of course AWS won't do that. So therefore the license is not 'free' to AWS, Azure, GCP, etc.
I think this is a reasonable take. To add a slightly more unreasonable one to it,
I would rather just never read GPL source fullstop. I don't want the copyleft to taint me and inadvertently reproduce copyleft code where I can't, so I will just avoid reading it entirely. I try to avoid using GPL libraries entirely if I can help it since inevitably I may need to circle back and read some of the source.
It's an interesting idea and I cheer those who manage to live in an ideologically pure bubble but I don't think it's fair to say GPL is strictly more free than BSD/MIT.
I saw Tom be the awards emcee at Yale's hackathon and I don't really get why he acts the way he does. It's some persona that makes people hate startup founders even more than they already do. Add the Yale tag to that and boom you've got your classic Ivy privilege douchebag. I know people from Ivies (including my own) and it's really rare for people to act like this. Is he giving these groups a bad name? I don't know him personally, and he's probably a fine guy, but it seemed like by the end of the event people were sort of sick of his antics.
>MRI executes Ruby via an interpreted stack-machine bytecode language known as YARV
On Wikipedia:
>Since YARV has become the official Ruby interpreter for Ruby 1.9, it is also named KRI (Koichi's Ruby Interpreter), in the same vein as the original Ruby MRI, named for Ruby's creator Yukihiro Matsumoto.
AFAIK, technically, ruby 1.9 and later are no longer MRI, but KRI. So, yeah, I would say that technically speaking, refererring to current versions of ruby as "MRI" is not terribly correct.
On the other hand, to tell the "default" ruby apart from jruby and rubinius, one might still refer to it as "MRI". Also, Robert Metcalfe once said something to the effect that "Ethernet cannot die - if something comes along to replace it, it will just be called Ethernet" or something along those lines. Maybe something similar is at work here? (Sloppiness is more likely, though.)
I am not 100% sure, but wasn't KRI a fork of MRI that at some point simply was merged? It would be a bit strange if the whole interpreter would be called KRI when most code was still written by Matz. Though Ko1 is the main developer currently Matz is still the visionary and decision maker.
A problem is that if you want to sync regular text files to apps on your phone we have formed somewhat of a Dropbox monoculture where DB is generally the main or sometimes the only syncing option available in mobile apps. We should really start to move away from these tools.
Syncthing is great, but the devs say they can't develop a mechanism for sync between operating systems on the same computer, which is something I really need.
You could do as I do: Syncthing on laptop <> Raspberry Pi <> phone.
Now all changes made on either the phone (OS #1, in your case) or laptop (OS #2) will be synced to the Pi (or VPS) ready to be picked up by the other device when switched on/synced. Placing a device between the two others ensures that all files are up to date.
Without it you can run into situations where you've updated a file on the laptop that is also open on the phone, accidentally saving the phone changes over the laptop changes, then saving the laptop changes back over the phone changes. Syncthing has versioning, but this setup negates the need for most rollbacks.
I use it to keep notes in simple text files with Vim and Notepad++. No need for markdown or images here, so I just use basic formatting, timestamps and hashtags to help with searching and browsing.
Including an encrypted filesystem, alongside Syncthing's encrypted syncing, can keep data that bit more secure.
All open source, all encrypted, all self-hosted. Give yourself 30 minutes to set it up, if that.
Notes:
The Pi is the Pi2, but until a recent upgrade it worked fine on the Pi1. The phone is an n900, still doing a great job 5 years on... but in need of a replacement, preferably another phone that can install arbitrary Linux software and is not walled into an 'apps' ecosystem.
The OP was interested in Android support. I don't think there's an app for Syncthing, but you could SSH to your VPS, with notes synced to your laptop? You're just missing that middle device in that case.
https://www.ahmath.org/members
reply