smokel

20 ms·

361.
by smokel 1y ago
> 500 years ago, what were people worried about? It wasn't so much different from our time. Read "Don Quixote" [1] and be amazed. Whether the updates you read are actually playing out live, or happening in a book doesn&
362.
by smokel 1y ago
There are still people with too much money who want to do good. In the art world, for example, it is pretty common to sponsor promising artists in one way or the other. Unfortunately, it turns out that the problem of picking the right pers
363.
by smokel 1y ago
The ugliness scares most people away, or at least it doesn't accidentally lure them in.
364.
by smokel 1y ago
Understanding is probably not much more than making abstractions into simpler terms until you are left with something one can relate to by intuition or social consensus.
365.
by smokel 1y ago
> Gemini's 1M token context window is really unbeatable. How does that work in practice? Swallowing a full 1M context window would take in the order of minutes, no? Is it possible to do this for, say, an entire codebase and then
366.
by smokel 1y ago
I have been in this industry for some time, and over the years I have only seen more people being glued to electronic devices, not less. It might have been a selling point, but the status quo is that we are inventing new jobs faster than ph
367.
by smokel 1y ago
Please do. It is unfortunate that those who shout hardest are not necessarily the smartest. There is a lot of nonsense being put out there, and it would be refreshing to read alternative perspectives. Edit: this was a generic comment, not
368.
by smokel 1y ago
Thanks for sharing. This probably means that search results that state "<language> programming sucks" or "<language> programming blows" both increase the relevant score in the TIOBE Index.
369.
by smokel 1y ago
I got by just fine without social media for most of my life. I look forward to society returning to that quieter and more fulfilling time.
370.
by smokel 1y ago
TIOBE is a famous benchmark, but it is also completely useless. There is a lot of legacy code out there that skews the benchmark, making the index an unreliable indicator of a programming language’s current popularity or the volume of activ
371.
by smokel 1y ago
Reserve it for your own friends, and just have fun? Seems the most genuine approach.
372.
by smokel 1y ago
Perhaps this is a way to channel money into someone's pocket, instead of keeping it in a non-profit?
373.
by smokel 1y ago
There are clearly some frustrated voices in this thread. I try not to take the downvotes personally. I hope to eventually learn how to engage more constructively with such responses. When someone claims to understand how a system works, yet
374.
by smokel 1y ago
> If your business does not have university affiliations with people that get existing grants, or connections to EU bureaucracy, then you can forget about getting a dime. So, then simply get in contact with your local university. It he
375.
by smokel 1y ago
Not an expert in the field, but apparently there has been some research into this. It's called inference-time intervention [1], [2]. [1] "Steering Language Models With Activation Engineering", 2023, https://arxiv.
376.
by smokel 1y ago
As a counterpoint, many systems that were originally implemented as native desktop applications have since been migrated to the web. The motivation for this shift is not particularly strong from a technical standpoint, but it is a practical
377.
by smokel 1y ago
Thank you for taking the time to respond to so many comments. Note that my remark is more about my own semi-philosophical thoughts on the value of entrepreneurship, and not so much about your particular effort. Perhaps I should've mad
378.
by smokel 1y ago
I thought "toi la mot con Lua" was French for "You, the word with Lua!", the well-known, but not to me, scripting bible. Turns out I misread the Unicode, and it is actually "Tôi Là Một Con Lừa", which is Vietna
379.
by smokel 1y ago
It is indeed the title that trips me up. When I was their age, we would call similar efforts "play", and we certainly didn't pretend to be the next Bill Gates. I'm not saying that one or the other approach is objectivel
380.
by smokel 1y ago
I think the reasonable explanation is that logic works great for simple systems, but once there are more than, say, seven variables involved, nobody can properly reason in real-time anymore. Personal relations, politics, raising a child, f
381.
by smokel 1y ago
Here you go: "Focus on value creation. Execution beats ideas. Leadership and culture matter."
382.
by smokel 1y ago
You might be interested in Reinforcement Learning [1]. By giving the system a negative reward, it may eventually start complying with safety rules. Still a good idea to keep the harness in place during production use, though. [1] https:&
383.
by smokel 1y ago
AG code?
384.
by smokel 1y ago
Indeed, this is not how language evolves. You can't dictate the meaning of words. Another unfortunate example is the increasingly negative connotation assigned to the word "algorithm".
385.
by smokel 1y ago
This explanation is only superficially correct, and there is more to it than simply predicting the next word. It is the way in which the prediction works, that leads to some form of intelligence.
386.
by smokel 2y ago
One could implement some pipelining to avoid that problem.
387.
by smokel 2y ago
This article is written by a recruiter. Recruiters make less money if everyone stays put. So repeating this trope about employees being transactions might be great for them, but it does not contribute to a more friendly society. A bit of
388.
by smokel 2y ago
Here you go: https://www.frontiersin.org/journals/psychology/articles/10....
389.
by smokel 2y ago
A paper on languages that begins with a grammatical error in the first sentence does not inspire confidence: > LLMs represents a disruptive technology
390.
by smokel 2y ago
This is a false dichotomy. You introduce two categories, state that LLMs are not part of one category, and then conclude that it must be in the other. In reality, the distinction between the two classes is not so clear. The transformer arc

More ›