loose-cannon

1. Not everything is about efficiency.

2. People don't all have the same goals as you.

3. These kinds of posts are awful

I read quite a bit of this and I'm not getting the point or understanding what the author is saying. On the one hand babies are born with reflexes and learning mechanisms. On the other, we're not born with knowledge, rules, representations, algorithms, models?

Yea I agree. Thinking from first principles is a valuable skill in general.... but certain areas have received so much attention/work and are highly developed. It's not impossible to develop something new by thinking from first principles, but when I see it I often interpret that as either 1) lacking awareness towards the sophistication of the domain, or 2) arrogance.

I'd like to think I know a thing or two about math/cs... but I don't really understand what the author is trying to present. And it's not presented to be accessible...

I don't understand. This past week we saw them announcing a solution to the Navier-Stokes, which we're supposed to interpret as a success? But they're slowing down? They're slowing the development because they want to or because they have to? Because of ethical concerns? Or because of economics factors that prevent them from delivering results?

I don't understand the psychology of showing an extremely capable product one week and then announcing a slow down the next.

These kinds of responses aren't helpful. Your response essentially muddies the water because it sets an artificial standard that nobody can/will meet. Yes, you're correct. There likely is some grey area with how we decide to give credit. Are we talking about the same story? This isn't "a polynomial". This is a research program that has been ongoing for years or possibly decades with nontrivial contributions. If you're not sure what does or does not deserve credit, you should engage with the community instead of suggesting that the idea of giving credit is "vague".

1. Tristan's research direction was known to only a handful of other academics. It might help if you can be more specific regarding the origin of the contamination.

2. If someone did come out and claim that their ideas were used without proper attribution in Tristan's work, then of course, that deserves consideration.

3. What you're suggesting seems purely hypothetical. At present, there is nobody claiming that Tristan's work is "contaminated"

4. Tristan was very willing in his initial statement to give credit to the people who developed the ideas.

dude, proper attribution is part of scholarship and academics. Not giving credit to the people whose work led up to your solution is malpractice.

Yes, I also think it's alarming. And I don't think there is enough reporting/weight given to the circumstances in certain mainstream outlets. There are certain mathematical programs that have existed for decades (for example Langland's). These are endeavors that span lifetimes. A solution in any one of those areas has to acknowledge the people whose work was used. That's part of scholarship. In OAI's initial write-up, they didn't even bother to cite Cordoba and Zoroa.

It's very disturbing that we have no idea and can't know if OAI acted ethically. But also, like the author said, this is very damaging to the field because it devalues all of the labor which isn't directly visible in the solution.

In the PDF:

"The route to the Clay problem through a smooth force, options c and d in Fefferman’s statement of the problem, is the route Luis and Diego opened and the one Levent and I had quietly chosen to attack. Almost nobody else I know of was working on it. It is not the direction one arrives at in a few days by giving a model the problem statement. When I heard “forced,” it was a bright red flag."

Whether or not they were researching it before isn't the concern.

where does it pull its sources from? Should I expect some degree of exhaustiveness? Meaning: does it miss information?

Does it ever prioritize one source over another? Will somebody ever see a source from, say, a Joe Rogan podcast? What makes a source reliable?

I'm almost instinctively opposed to software that claims that it can "fact check".

>Memes are inherited. You brush your teeth because your parents do. Memes are subject to selection via biological fitness. The people who didn't brush their teeth died sooner of tooth disease, so those who brushed their teeth had more children. Memes mutate. You don't brush your teeth identically to your parents.

That's parenting and the passing of tradition. I wouldn't call it a meme. Also I can't tell if you're joking with people dying due to teeth disease.

I mostly agree. There are certain classes of tasks where the spreadsheet is more efficient. It's "more efficient" because it 1) requires less set up time (like you said, translating a computation in a more traditional system could take months), but also because 2) the spreadsheet is a highly interactive and accessible environment.

We have a programmer on our team who keeps trying to convert our excel calculations to more structured systems (database store). The issue is that the spreadsheet-work almost always requires navigating and making decisions based on unstructured data. The format isn't formally specified, so in the end our programmer makes more mistakes because he doesn't engage with the data - the processing gets treated as a routine algorithm, and it's not. The interactivity (and the fact that we're not spending energy on formal specification) is what allows us to more effectively engage with the data.

Another important point is that spreadsheets more easily allows you to consider data points individually without hand cuffing yourself to one kind of processing algorithm or software design. I can easily edit one cell to make a correction in a data set (which is a double-edged sword). The programmer has to create an additional handler to handle the exception. Depending on the design/quality of the code, this could just be adding undesirable complexity in the software. So we can either spend 30 seconds and edit a cell, or spend half an hour adding to code (which by the way may never get used again).

I work at a small business where probably 95% of the work gets done through spreadsheets. Spreadsheets are incredibly valuable, but yea, like you said it requires a certain mindset or discipline. Otherwise you end up with a disorganized mess which almost always has errors. The errors are bad, but the distributed arrangement of the formulas is what gets abused the most in my opinion. Modularity isn't a thing and there isn't formula re-use like in the software world. Something might be correct in one cell, but wrong in another. It could be due to bad assumptions or simply because there is so much formula repetition and the user just wasn't exercising care.

Show me instances where we've observed any of those mechanism occuring at the cosmic-level.

It's an interesting idea, but I feel like this is taking a lot of speculative (not observable) ideas for granted. What is a "variation in the basic parameters of matter"? Has anyone observed such a thing taking place? Also where would the universe birthing take place? I'm assuming you want me to believe the expansion of a brand new universe isn't occuring within the space of our own? I don't know how we would study new bubbles of space time, whatever those are.

Some computer science is mathematical. A lot of computer science is about abstraction: creating interfaces independent of implementation. For example, we have models of digital components and construct circuits that perform meaningful tasks (based on, or on top of, those models). How those digital elements are implemented with materials is often ignored. So what is a computer? I think a computer can be viewed as something abstract. The design of systems isn't traditionally a mathematical topic, so it is its own abstract area.

There are limits to how much you can ignore implementation below your abstraction, so yea obviously the field deals with physics and economics. The point is that I think it has a quality that makes it abstract.

I don't think "more rules" is the answer. There needs to be people in our government who have integrity, who can apply pressure and take risk. It's the people, not the constitution. The last 1.5 years is evidence that legality doesn't mean anything without people willing to take action.

I agree with your larger point. Though I think it's pretty natural to wonder how the poster ended up with a multi million line codebase.

The thing that you assume in a theorem is usually called a hypothesis.

I think "being concerned with knowledge" and "being invested in the process of creation/discovery" can occur at the same time. There isn't a singular "scientific mindset" or set of values that differentiates what is and isn't science. It's just silly.

there's a human element in discovery and in problem solving. Whether or not somebody chooses to sidestep work isn't necessarily an indication that the work isn't science. It's just an indication that some people are passionate and invested in performing the work.

Yea sorry. I didn't mean to criticize or diminish its value by saying it's just an educational tool. It's a powerful tool that lets you do some advanced things quickly. That's totally valuable outside of an educational setting. And I didn't mean to imply that there's any limitation to its use.

I just don't understand the criticism in the post I was replying to.

I thought Scheme was just an educational tool - a minimal language that allows the student to explore language construction & design relatively quickly. I'm not sure I understand your comment about homoiconicity leading people astray. Isn't that the characteristic feature of lisp? Doesn't that dramatically simplify generating code (in principle)?

I'm not sure what point you're trying to make. In science and engineering, being able to provide justification is a core skill. The comparison we should be making is against the human practitioners who are trained in their fields. There will always be a distribution of ability. Saying that there's evidence that people are capable of providing post-hoc rationalization doesn't say anything about the ability of experts to produce well thought out responses (in their respective fields) that don't immediately fall apart under scrutiny.

Dude, these two things are not at all analogous:

1. Asking a model why it did a certain thing, and

2. Expecting a human to say which neuron fired in their response.

So nature doesn't exhibit justice, but a society likely needs to implement it, yes? But you don't want to say who gets to receive those rights? Some might say that those millions of dying children is an injustice that we should try and prevent. In particular, the people whose lives are the most affected might want to have a word with you.

I'm not really sure how your reasoning here is in line with your previous post.

There will always be scarcity and inequality. The point is to minimize/mitigate their effects. Can't you make the same argument for justice? Why does everyone deserve justice? Isn't that just entitlement? What is the historical or evolutionary basis for justice?

I think it's slightly comical to see people from the tech community deal with the problems, as customers, that are brought about by the dysfunction of AI. No, I'm not making a sweeping statement on the usefulness of AI. It clearly is capable in some areas. But, surely, companies should have an incentive to have a functioning customer support? And doubly so if you're a company trying to automate labor?