HNHacker News
TopNewBestAskShowJobs

the5avage

159 karma · joined November 1, 2020

submissionscomments
the5avage··on Trump wins presidency for second time
One very mundane reason a company I had worked for switched to Office365, was that emails from our own domain would often end up in the spam filter. It can cost a lot when that happens.
the5avage··on Trump wins presidency for second time
Is there some analysis why the polls didn't correctly predict the result?

A failure in representative polls like this should be avoided with statistical methods.

the5avage··on State of Python 3.13 performance: Free-threading
Thank you:)
the5avage··on Ask HN: I think I'm done with tech. Burnout constantly. Now what?
I am sure somewhere is a job you would like. Probably with less salary.
the5avage··on State of Python 3.13 performance: Free-threading
Is this implemented with lockless programming? Is it a reason for the performance drop in single thread code?

Does it eliminate the need for a GC pause completely?

the5avage··on More Oracle Layoffs Started Nov. 1, Cloud Unit Impacted
I once worked at a company, that bought a lot of unnecessary software, because they knew they need a database, but didn't knew what a database actually is.
the5avage··on U.S. chip revival plan chooses sites
Do you have some secret intel? They make the best chips in the USA.
the5avage··on U.S. chip revival plan chooses sites
That is not fair. They did not actively kill hundreds of people.

They just waste some energy on suboptimal chips and business decisions.

the5avage··on State of Python 3.13 performance: Free-threading
Can someone share insight into what was technically done to enable this? What replaced the global lock? Is the GC stopping all threads during collection or an other locking mechanism?
the5avage··on Tencent Hunyuan-Large
Where would you go when you would live there (as a programmer interested in ai)? Just asking for a friend.
the5avage··on Tencent Hunyuan-Large
I once triggered the ChatGPT censorship (by trying to manipulate an image of my face) and it also responded in english to a german query.
the5avage··on Former Bayesian captain on sinking, vessel ballast and stability
> As it pushed the boundaries of aluminum mast construction there were lessons that had to be learned over the years, but the result was a robust and well controlled rig in the end that performed well, within the limits of the design.

The mast of this ship was slightly higher than that of the Preußen, one of the biggest sail ships ever build.

While it clearly worked for a while, it shows how insane the design actually was

https://en.m.wikipedia.org/wiki/Preussen_(ship)

the5avage··on Interview gone wrong
"a == b" is the same as "not (a != b)"

Maybe I'm simple minded though.

the5avage··on An embarrassingly simple approach to recover unlearned knowledge for LLMs
Maybe they are not allowed to use uncensored LLMs, so they have to first develop this unlearning, before they can even use it.
the5avage··on Interview gone wrong
> I really don't want to have to pull out an operator precedence chart in order to read your code

Hard disagree here. I write code for people who can read the language. This includes operator precedence.

the5avage··on Interview gone wrong
I would say if you know you have to later work with the guy and he will be higher up than you this is actually smart.
the5avage··on Interview gone wrong
No, because when all cells are empty he returns "Winner".

The interviewer also lost his job, since he didn't spot it nor does he know basic python syntax.

the5avage··on Interview gone wrong
I think it makes sense in python for ==, <=, >=

I cannot see it for !=

"a != b != c" is not equal to "not (a == b == c)" which is a bit strange imo

the5avage··on C++, Complexity, and Compiler Bugs
Its generally a fun read, but

> I don’t know man, why are we writing C++?

Why do you write it in a way that you even blow up the most esoteric footgun in existence?

Using temporary lifetime extension - just because it exists or what?

const int& i_am_a_ref = return_a_value() // why do you need this?????

Compiling C code with a C++ compiler - it is no secret that it only works until C89

Unnecessary template disambiguation - why not remove it when it is not necessary

the5avage··on Did You Know That Apes Have Never Asked a Question?
So the goal of the education system is to produce monkeys.

The website where the article is hosted looks very dubious to put it mildly. Looks like SEO spam.

the5avage··on Please Publish and Share More
The problem I have is I don't use any social media.

I did write a very cool tutorial about how to apply machine learning to control problems, but not many people even looked at it I guess^^

https://github.com/the5avage/Q-Learning

the5avage··on Ask HN: What would you preserve if the internet were to go down tomorrow?
All the papers and books of library genesis
the5avage··on Eighty Years of the Finite Element Method (2022)
Have you heard of physics informed neural nets?

It seems like a hot candidate to potentially yield better results in the future

the5avage··on Implementing a Tiny CPU Rasterizer
I once wrote a tiny cpu rasterizer in dlang. It is more efficient than what is shown here, since it uses only integer arithmetic (based on bresenham's algorithm). E.g. for a triangle it outputs the points in the triangle without ever considering a single pixel outside the triangle.

It's based on ranges (similiar to the ones added to C++) and the concept works very well imo. Could be a good exercise to translate it to C++23 (using coroutines with the range "generator").

https://github.com/the5avage/shapes

the5avage··on Reinforcement Learning: An Introduction (2018)
I made a small project where I show how to train a Neural Net with Q-Learning for a simple temperature control problem. Still a hard read for someone with no background in math/control theory probably, but the code is actually much easier than the text^^

https://github.com/the5avage/Q-Learning/

the5avage··on Real World Haskell
Hey thanks for the nice reply. I know you can't generalize like that. There are probably a lot friendly Haskellers around ;)

I was just angry when I wrote my last comment, because I had to read this:

>> as a special case of most people are idiots

which is so low in multiple ways. It's also just wrong, because by definition most people have average intelligence. Even if you think this is true, a smart person would not use it as an argument^^

It's on my todo-list to dive a bit into haskell, when I find time.

the5avage··on Real World Haskell
> No, I didn't, and please refrain from imputing thought processes without evidence.

There where at least two other replies talking about accumulators. I call that pattern, like it or not.

Have you even heard of programming patterns?

> If we assume most imperative programmers are idiots as a special case of most people are idiots, sure.

The other reply from chongli seems to disagree with you ;)

Also suggesting that I'm stupid is no constructive style of argumentation.

> rather than that it wasn't a good/intuitive example to illustrate recursion

You are still misunderstanding me but a genius like you are should be able to figure that out.

I now understand what my teacher meant when he said:

"Haskell is a genius language but I don't like the community,"

the5avage··on Real World Haskell
Yeah it's ok ;) Going back to my original comment this code is exactly what I meant with "patterns to deal with other problems". You clearly used a pattern (accumulator) with wich you are familiar.

Do you agree that most imperative programmers would not come up with this solution on their own, if you didn't teach them the "accumulator pattern"?

For me personally a tutorial with a lot code like that would be interesting.

Showing me another version of a recursive descent parser doesn't teach me anything, because I know how to write it (with recursion!!!) in an imperative language...

the5avage··on Real World Haskell
I found something like splitting a string into words not very intuitive...

Also the classical (recursive) fibonacci function everyone shows has accidental complexity of O(2^n). If you want a sane version it looks definitely more complicated than the iterative version.

I never saw a recursive version of bresenham's line plotting algorithm so this might be a good candidate as well ;)

the5avage··on Real World Haskell
> Recursion is equivalent to iteration. You can solve any problem you like with it.

Maybe my comment was somehow badly written, because you (and the only other reply from chriswarbo) clearly misunderstood me.

I know that recursion is equivalent to iteration^^

My point was that some problems fit the one or the other better.

Look at fft for example. The algorithm is easy with recursion but hard with iteration. For other problems its the opposite.

← PreviousPage 2 of 3Next →