HN user

tasogare

1,277 karma
Posts0
Comments1,260
View on HN
No posts found.

the model used and the approach is extremely simplistic

Seems like a constant in doomsday "science": making a model that bear ressemblance to reality as much as SimCity does, run it to get the expected conclusion, then extrapolate and conclude with newsworthy title that X or Y is true in reality.

Not everyone is interested in a humongous editor with a steep learning

Then why using a TUI editor at all? Each time nano was opened because of some default settings, I saw this whole bottom bar with shortcuts. It sure doesn't seems to be easy to learn either.

I don't know France

Then don't try to put discredit to the saying of an insider. GP is 100% right. I've been back after about ~5 years living abroad and the country is in an astonishing decline on every plan (educational, moral, economical, political, diplomatic, security, judicial, medical, etc. the list goes one). And most of it is stemming from the policies enforced in the last decades.

There're a bunch of corrupted inviduals with no counter power to keep them in check. No decisions are made for the benefit of the French people since a long time, every law passed is either to augment their power, rise taxes (already world top 1 or 2 depending on the year) or to the benefit their friends owning private companies. Of course technology freedom will suffer from this as well.

So quite similar to the US where oligarchs (oups, billionaires) are invited at the White House and companies provide algorithmic censorship favoring a particular political camp (obvious with Twitter before its recent purchase, YouTube actions against public free discourse on covid, Google removing pages related to a certain laptop, etc.)

In the face of current hype around LLMs and 'fear of AI', calling a Random Forest Classifier 'AI' is a bit... far

Just because the state of art evolved doesn't mean we have to erase history of the field. This is a ML algorithm so calling it AI is perfectly in line.

The fear you mention is built on the lack of understanding of what ML is. Showing that some AI has "dumb" yet useful implementations can help show the limits of this category of technology.

Wtf is this? As a White Christian (living as both a racial and religious minority in my country of residence btw) this is extremely offensive. I have the right to assemble with other people on the identity feature I want, especially on the internet. I bet wouldn't dare to say the same kind of things about Judaism because you know exactly how racist and discriminatory it is.

I tried writing a sentence (我想去買東西) with iOS pinyin keyboard traditional and while I had some luck with the first two letters, it breaks on "wxq" to work again with "wxqm" and then break again. So it's definitely not as good as you market it.

However it could be explained by the fact predictive typing on iOS is quite poor. Often a wrong letter in a word need to be corrected and the next predicted word (and it's conjugation; I'm speaking of French and English input) is often wrong. It may not be noticeable for someone who never used another keyboard, but as a former Windows Phone user the difference is very easy to see. The WP prediction was mind blowing and I could indeed write long sentences in French just by typing a few letter and using one of the prediction candidates.

Please keep fighting. The automatic association here on memory safe language with Rust annoys me as well, just as the one about mixing languages on a VM and WebAssembly. Both features are available since decades on C# and the CLR, and Java and the JVM to name a few.

Except this is why chemistry is less entrepreneurial - you can't just set up a lab in your shed (for better and for worse)

Given the number of wannabes PhD in chemistry having peaceful religious beliefs that are entering in EU every year, this is really for the better.

The only linked paper is from Journal of Experimental Social Psychology, so basically soft science whining about hard science with "muh men bad" argument.

So typical divisive BS gender study. I'll continue to work with coworkers from both sexes without taking offense in a disparity that isn't even proved to have influence on papers' quality and science output.

Yeah dropping Silverlight support for Mac was a shame, Microsoft finally had a good multiplatform story. They could have keep the out-of-browser mode and market it as desktop dev framework instead of browser plugin.

The biggest irony is that this tech would have allowed them to open a store on Windows 7, 10 and Mac which could have been way more popular due to market share than both the Windows and Mac app stores.

Product release. For instance there is for years issue with C#'s GUI frameworks mainly the lack of a multiplateform one. Microsoft disregarded developer feedback on UWP (it was stupid to propose a Windows 10 only framework at a time Win7 was still dominant and tie it to a mean of delivery - the Windows store). About a decade later Microsoft remembered with dotnet 5 and 6 that WPF and WinForms are the frameworks people are actually using and they got some improvements.

Thanks for the kind offer! I got it sorted out yesterday but it's really nice to give a hand to a stranger.

Ah yes, the specific form. Frankly given than in France test are valid only 24 hours for the "fuck your freedom pass" the 72 hours validity feels a long time and making an additional paper filled wasn't too much annoying. The highest ranking personal in the laboratory (not a doctor) signed and that was ok. I guess Japanese car more about the seal part.

Does Not Translate 5 years ago

Excellent.

One funny thing I've noticed with French EFL speakers of medium proficiency is that they too think all those loanwords from French aren't English words. It's always fun to troll them dropping a en route and look at their reaction.