The institution which is passing children through is not corrupt, it is serving the will and moral character of the people. It has not departed from its mission to do the best it can.
The people don't want the slightest fluctuation of whatever complex story surrounds the issue to means a chunk of children fail. In contrast parents would rather as much record fuzziness as possible if it means giving children a 2nd or 100th chance (putting aside the dooming issue that bad tests mean funding cuts).
So I think you'll find that it's not just that records are fuzzy in school, padded up by participation and homework and extra credit, it's that you'll likely be able to predict which regions are fuzzier than others in record keeping.
The idea seems like it should work but it is questionable what cases will it be better in.
B+tree and LSM-tree are very developed and are kind of optimal. They are also fairly easy to beat for a given specific use case.
I guess they have a concrete case that has benefitted from this design or this was an attempt at doing that. Would be interesting to read about that specific case they had. I just skimmed the paper, so I'm sorry if they explained it in the middle somewhere.
Also I tried some other databases that claim to be better than rocksdb but it just is miles better than other databases when I needed large scale (couple billions of 32byte keys mapped to 8byte values).
I tried MDBX(LMDB), sled (also claimed read AND write optimized).
Tried sharding and all configuration options with both.
Reading papers about database research unfortunately feels like reading LLM output because I have to sift through a lot of fluff, and I have to know exactly that the thing is about and the surrounding ideas. I am not super knowledgeable in this field so this might be just a skill issue, but I would recommend seeing it this way.
This paper also writes about variable sized pages so it might be relevant to understanding what the trade-offs might be.
Also another thing I highly recommend is to always judge by hardware limits vs db measurement instead of looking at graphs in paper.
If something is doing 1GB/s write on an ssd that can do 7GB/s than it is bad at writes. It doesn't matter if it looks cool on a graph. This is kind of a crude way of seeing it but it is at least reliable.
Another interesting tree filesystem data structure is the Bε-tree ("b epsilon tree"), which also tries to bridge the gap between small writes and the large pages of modern drives. The first paper/talk from 2015 has a fun name "BetrFS: A Right-Optimized Write-Optimized File System" and they published a few dozen times until 2022. https://www.betrfs.org/
I've read this paper and it's a neat idea. It hasn't been introduced into popular oss databases like postgres and mysql, and my understanding is it has some drawbacks for real prod use vs ths simplistic benchmarks presented in the paper.
Would love to know if anyones built something using it outside of academic testing.
People can think of ML on a government level, but it has an inescapably international dimension as a kind of gunpowder-like discovery. Relatedly, if war becomes increasingly automated and cheap, then civilian targets will be seen as obvious.
As we discuss policy ideas to pump the breaks on a domestic level, I hope we balance that against the arms race that's happening around the world.
You wish to lead with "dumb shit" in framing why people have a problem with Elon Musk? Why not lead with the Nazi salute at the presidential podium? That would more quickly get to the point.
You do not have to look beyond Elon’s own Twitter accounts posts, retweets, and likes, to see that he is a full fledged white supremacist. Calling him a Nazi is appropriate.
Nazi salutes are protected speech and not "beyond politics". Yes it's disgraceful, and it's reasonable to leave his platform. But it qualifies as "dumb shit".
I think the point is to distinguish ‘political opinions that I am comfortable disagreeing with people about, and can still be friendly with people who strongly disagree with me’ and ‘morally unacceptable opinions that I will neither listen to nor associate with anyone who hold them’.
There are many political opinions that I strongly believe in that I am comfortable disagreeing with people on. I believe everyone has a right to health care, and that society should guarantee basic necessities for everyone. I even feel that belief is a morality based belief. However, I can accept people disagreeing with me, and can accept that there are some strong arguments against my belief, and that good people can disagree with my position.
On the other hand, if someone believes that certain races should not have the same rights, or that women should be given less agency than men, I will not entertain that argument or accept that it is just a political dispute. That is a fundamental moral issue, and is beyond JUST politics.
Affirmative action and similar policies are examples of those sorts of political opinions that I can happily debate, and I definitely don't think I have the perfect answers for how best to obtain the goal of equality.
As far as your particular question goes, I don't agree that believing that all races should have the same rights is inherently in conflict with the idea of affirmative action. In most implementations, there are no rights that are denied to anyone when affirmative action policies are implemented. The entire point and purpose is to counteract existing norms, institutions, and system structures that are actively denying rights to citizens in particular groups/races.
For example, take the original affirmative action order (from which the phrase was coined) signed by JFK in 1961. The text stated, "take affirmative action to ensure that applicants are employed, and employees are treated [fairly] during employment, without regard to their race, creed, color, or national origin"
What rights are being denied if that is followed? The idea is that it is clear through observation that the criteria that was being used before was preferential to white Christian men, so they were instructed to proactively address that unfairness by changing their hiring process to attempt to eliminate those biases. How is that in any way denying rights to any group?
That JFK quote is not what it means. It means denying access to limited places in education based on race. Do you mean those aren't rights so denying them doesn't fall within you definition of intolerable ideas?
You don't need to explain what it's for because what what it's for doesn't change what it is. If I robbed somebody to use the money to cure cancer, it doesn't change the fact that I still robbed somebody.
That is literally where the term comes from. It isn't a quote, it was an executive order. That language is what it legally meant.
> It means denying access to limited places in education based on race.
Every person accepted is a denial to someone else. As you said, there are limited spaces. If you define it as a right to have a space at that school, then by definition you have to deny some people their rights, since you can't accept all of them.
Affirmative action means you are supposed to factor in the existing disadvantages that minorities face when deciding between two candidates. It doesn't mean accepting a less qualified candidate, it means acknowledging that our previous methods for choosing between candidates was inherently discriminatory already, and in order to counteract that, we need to take 'affirmative action' to make things more fair.
You can always argue about what criteria should be used to choose between two comparable candidates, there is no such thing as a perfectly 'objective' evaluation. Even if you chose to base everything on a test score, you still have to decide what goes on the test and how the questions are worded. There is no way to do that that is perfectly fair for everyone, even if we accepted the premise that test scores are an accurate and fair measure for choosing who to accept to a school.
Why shouldn't the pervasive, clear, and systemic racism and discrimination that many minorities face be used as a factor when determining school acceptance? How is ignoring that reality 'more fair', and how is acknowledging and compensating for that reality a 'denial of rights' to anyone? Wouldn't it be a worse denial of rights to ignore the discrimination and racism, and making decisions as if the world wasn't the way it is?
> That is literally where the term comes from. It isn't a quote, it was an executive order. That language is what it legally meant.
It doesn't matter what the history is - we both know that is not what it means in practice today. Continuing to lean on this incorrect definition is dishonest. If that's really what you mean, then I completely agree with you, but you've shown that it's not what you really mean.
> then by definition you have to deny some people their rights, since you can't accept all of them.
Yes but you can do that without using their race as a factor.
You are still literally justifying denying rights to people because of their race. You have some reason for it but as I said, the reason doesn't change the fact that it's still denying rights because of race.
To show why it's wrong, imagine you're a black immigrant from a black country and you've never suffered any of this discrimination you talk about. You now get preferential access to tops universities because some other people who aren't you did suffer discrimination. That really just entrenches the unfairness.
Do you also favor a Jew tax because Jews are rich? That's the logic you're using. Treating individuals according to their group's characteristics. It's also the core of modern leftism (wokism) which why I suggested leftists would hate you for your ideas.
> It doesn't matter what the history is - we both know that is not what it means in practice today.
That isn't true. The sort of affirmative action I am talking about is still used in the world today. People who are against any sort of action to counter systemic racism have chosen to pretend that all affirmative action is the "racial quota" type that you are talking about and that has been illegal for quite some time. Continuing to pretend that is what affirmative action proponents are talking about is how opponents are attempting to get rid of fair and reasonable affirmative action by treating it all as the illegal kind.
The 'Jew tax' example is completely disingenuous. This is not applying any rule or law to a specific race, ethnicity, or religion; it is simply taking into account the effect that discrimination and racism has had on people when evaluating candidates for limited positions. It is not the same at all.
Your black immigrant example also is quite the reach. For one thing, that immigrant is facing racism and discrimination the moment they step into the country.
You say you are worried about unfairness being entrenched, but this has already happened and is what we are trying to fix. Racial discrimination against minorities is CURRENTLY entrenched in our institutions, and affirmative action is the attempt to overcome some of it.
I find it very interesting that you are so concerned about any advantages that might become entrenched for minorities, but are completely fine allowing the entrenched advantages for the majority to persist. You are more worried about hypothetical future advantages rather than actual present advantages.
The whole point of affirmative action is action is to acknowledge that if two candidates are equal or close to equal in qualifications, the one that has had more disadvantages is probably the better candidate and should be chosen.
"if two candidates are equal or close to equal in qualifications, the one [whose race] has had more disadvantages is probably the better candidate and should be chosen."
Now that I've changed it to say what you actually believe and have stated before, you can see why it's controversial.
You are really struggling with the idea that race-based discrimination is something you actually favor even though society has told you you're supposed to be intolerant of it. It's leading you into all these contradictions and justifications. Modern leftists have resolved these contradictions by not making such bold simple claims as you did.
I'll just leave you with an example from my own country. It's a kind of quota (100% for a specific race) so you might not like it or maybe you will. I have no idea because your idea is so inconsistent.
Explanation and summarization without visual interactables is so much harder to do. A person can talk to an interface but I don't know how many people would like natural language back.
Those two tasks are just very different. In one world you have provided a complete specification, such as 1 + 1, for which the calculator responds with some answer and both you and the machine have a decidable procedure for judging answers. In another world you have engaged in a declaration for which the are many right and wrong answers, and thus even the boundaries of error are in question.
It's equivalent to asking your friend to pick you up, and they arrive in a big vs small car. Maybe you needed a big car because you were going to move furniture, or maybe you don't care, oops either way.
Furthermore, it is possible to build a precise mathematical formula to produce a desired solution
It is not possible to be nearly as precise when describing a desired solution to an LLM, because natural languages are simply not capable of that level of precision... Which is the entire reason coding languages exist in the first place
I have a feeling we'll care less about untyped languages going forward as LLMs prototype faster than we do, and fast prototyping was a big reason why we cared about untyped languages.
Pedantic but Lisp is not "untyped". (Neither are JS or Python.) All data has a type you can query with the type-of function. The typing is strong, you'll get a type-error if you try to add an integer to a string. Types can be declared, and some implementations (like SBCL) can and do use that information to generate better assembly and provide some compilation-time type checks. (Those checks don't go all the way like a statically typed language would, but Lisp being a programmable programming language, you can go all the way to Haskell-style types if you want: https://coalton-lang.github.io/)
The people don't want the slightest fluctuation of whatever complex story surrounds the issue to means a chunk of children fail. In contrast parents would rather as much record fuzziness as possible if it means giving children a 2nd or 100th chance (putting aside the dooming issue that bad tests mean funding cuts).
So I think you'll find that it's not just that records are fuzzy in school, padded up by participation and homework and extra credit, it's that you'll likely be able to predict which regions are fuzzier than others in record keeping.
reply