onurcel

That’s basically why dictatorships are so hard to overthrow. In real life the game is slightly different, if you choose red (don’t protest) you get a negative outcome. But if you choose blue (protest) you risk being jailed (a very negative outcome) unless enough people also choose blue, in which case the outcome can turn highly positive.

That’s why dictators try to limit protests, not just because of the protests themselves but because they don’t want people to know how many others are willing to protest.

I agree with you. In B2B SaaS you don't sell the software, you sell your expertise in a specific domain and the responsability you take for owning that expertise. The fact that the development costs are nearly zero will make them more valuable and more protifable

I had a C64 and learned programming on it. I guess it's a question of destiny at this point :)

great idea! just ordered one, I will give it a try

First of all, great job! I think the inference will become more and more important.

That being said, I have a question regarding the ease of use. How difficult it is for someone with python/c++ background to get used to zig and (re)write a model to use with zml?

> It started by accident, with the original llama weights being leaked by two separate employees.

This is not true. Meta Fair has been built on openness from day 1. We published many papers and open-source d many repositories to reproduce the work

On a side note, if you are running something for days or weeks you should implement checkpointing anyways. I have no idea if matlab allows that for internal operations

These CRUD apps need complex business rules, requiring expertise in the domain and making them configurable on the application level for customer while trying to keep the app not bloated.

Scaling is not the only challenge engineers face, but somehow it's the one that is mostly praised.

same. I have these exact urls bookmarked for years

For popups I blame the website developer, I read these as "let me give you a small annoyance to push a personal opinion" => ctrl+w

To be accurate, IE was not hated by "end-users". It was hated by web developers (understandably) reluctant to support non-standard apis and by informed people.

Realistically, with the title you suggest nobody would have read the post.

A title is not "the most informative and complete sentence summarizing the article", it also has the goal to stimulate curiosity. I understand that we don't want misleading titles but this obsession on titles is not very helpful. I am participating in this useless conversation but I couldn't help myself. Now every single HN post has a comment on how the title is wrong..

I hope the magic of HN will bring one of the men in the photo to intervene here.

I don't get it. An open-source project don't need users? Sure someone could publish a work for the sole goal of publishing it, but most of the time you make something public for a reason (financial or not). In fact, like the other comment in this thread, I find the author's point contradictory to ask for sponsoring their project. Again someone could sponsor something that has no utility, but it seems to be far fetched to say the open-source don't need users.

However, while I disagree with the statement, I agree with the sentiment. Most open-source projects' authors don't care about their users, at least not as much as in the beginning, probably because of the said incentive mismatch.

argument by authority?

He also said pytorch was just a hype. It looks like he just doesn't like things he didn't do himself.

I am curious what makes the author think that expert reviews are free of any incentives for abusing their influence. Also, if you ever built anything useful in your life, you should know that it's not possible to count solely on word-of-mouth, even in a upotian marketing-free world.

Listen, I get you are disappointed by bad actors, but going to the other extreme is never a solution.

this is actually our recurring joke for our team meeting offsites!

This is one of the examples we keep in mind and that's also why we can't 100% trust public dataset labels. This motivated us to train a Language IDentification system for all the languages we wanted to handle in order to build the monolingual dataset. More details in the paper ;) Or here, if you have questions

hi @btheshoe, I work on this project in the data part. As others mentioned, the amount of data available for a language is not correlated to the number of speakers of that language, which explains the potential impact of focusing on these.

in this work we tried to rely not only on automated evaluation scores but also on human evaluation for exactly this reason: we wanted to have a better understanding of how our model actually performs and how it correlates to automated scores.

This is 100% Google's responsability. If you claim to have a feature but it is broken it's your fault, you should just not claim that you can actually do this. The restaurant guys provide exactly what they want: a pdf menu, if google can't parse it correctly it should show the raw information instead of trying to do something fancy

> Is it only me who’d like a 15 inch air?

yes

In my opinion Tesla has the largest part of the fault here. You can say "yeah we all know Elon Musk is lying and that is not a real auto-pilot", but I am having hard time to understand why the responsability of a blatantly lying company/CEO is not discussed.

that's really weird. It looks like the author wrote it with a more powerful system then partly disassembled it. In my opinion this hurts the narrative of this otherwise cool project/article.

I can't tell the difference with LinkedIn posts

This, I suppose, is implicit in the word "boring". I think the premise is: People could follow boring advice to become successful but they don't, precisely because it's boring.

Regardless the definition of success, we tend to consider it as being in the top <50%. So even if everyone was more succesful (regardless the metric) in today's standard, only the best ones would be said "successful".

why stop there? why wouldn't you make an OS that would turn the phone off if you use it more than 1 minute? Boum, you solved everything.

I thought he was working on AGI

On my TO-DO project list, there is "BASIC compiler/interpreter for raspberry(-like) computers".

I am nostalgic of Commodore64 too.

The thing that made many of us go down the rabbit hole was the immediacy of the interpreter, begging you to enter some commands 2 seconds after turning on your computer.

The perceived small gap between the commands you enter in a programming language and the games you could play with the machine contributed largely to the curiosity and willing to learn more.

On a modern computer, building the right interpreter is challenging. There is a trade-off between the simplicity of the language and what you can achieve with it. You won't impress today's kids if it allows you to only print text on a scrolling page.

But I think one can build something really great. There could be an integrated sprite editor (with a graphic UI) and the interpreter can use them right away with something like: Sprite1.visible Sprite1.x = 10

emphasising again the immediacy. Those "sprites" could be 3d objects as well.