I think it's not about own versus someone else's money.
Hardware is usually a small piece of the financial puzzle (unless you're building a billion dollar AI datacenter I guess) and even when the hardware price quadruples, it's still a small piece and delivery time is much more important than optimizing hardware costs.
As someone who abuses gemini regularly with a 90% full context, the model performance does degrade for sure but I wouldn't call it massively.
I can't show any evidence as I don't have such tests, but it's like coding normally vs coding after a beer or two.
For the massive effect, fill it 95% and we're talking vodka shots. 99%? A zombie who can code. But perhaps that's not fair when you have 1M token context size.
I could try to explain that most jobs are way more nuanced than just 'failing and deserving to be called a monkey' or 'not failing.' Or, I could just call you names for not seeing that, you could call me names back, and we can keep doing this forever.
Your argument is lacking nuance, declaring that the criticism being levied here must be a simple binary.
The specific error they are criticizing is extremely egregious, akin to builder declaring a house without a roof complete. “failing and deserving to be called a monkey” is a criticism being levied against a 0/100 level mistake, not a mere minor mistake as you are claiming.
While it might be desirable to use less colorful language, it is frankly challenging to express the sheer level of grossly incompetent organizational ineptitude on display here in a reviewed and delivered product actively causing negative customer impact for literal years which is trivially fixed and yet has been ignored.
Customers of Github should be infuriated that Github gleefully foists such utterly defective software upon them. It is hard to get that across in dispassionate writing.
> Your argument is lacking nuance, declaring that the criticism being levied here must be a simple binary.
That isn't my argument. I am arguing against the idea that there is an "objective" threshold of failure where, once crossed, it becomes acceptable to call people names.
> Customers of Github should be infuriated that Github gleefully foists such utterly defective software upon them. It is hard to get that across in dispassionate writing.
See, while it has its bugs, I don't see a major problem with GitHub as a software product (setting aside the monopoly concerns). I encourage passionate discussion, but calling people names doesn't communicate passion; it communicates impatience. It suggests you don't have the patience to actually make a case for something you're supposedly passionate about, so you're choosing a shorter, more aggressive form instead.
Is it silly though? With enough linguistic archeology I bet you can make this entire comment I'm writing right now extremely problematic and offensive. The linguistic treadmill means exactly that older terms change meaning back in time. They also change meaning FORWARD in time, meaning your inoffensive terms today will almost certainly be offensive in the future.
It's also the case that offense is language dependent, which is always funny when Americans hard ban certain words on chats and then Swedes can't use the Swedish word for "end" because it's spelled like a slur in English.
The whole terminology in IT could be turned upside down because it can be quite offensive if people ignore the context, so it is not limited to processes. There are utilities like "man", "finger", etc. that could come across as offensive too, to some, with no context-awareness.
Today it is "master" -> "main", tomorrow the whole IT terminology.
There are many PRs on GitHub with regarding to these, by the way.
... also what about pins? Slave and master pins! Must be about slavery, right? No, it is not, not at all.
In any case, who made the association of the git branch "master" to slavery? It is absurd. People need to take the context into account.
It is based on nothing. It is not intended to be offensive, and it is not intended to be about slavery. Similarly how master and slave pins are not either, or how blacklist and whitelist are not about race either!
I grant it's not nothing, but I think it's not enough of something to make changes over it. Thinking of a master record or similar is the natural reaction when you learn about the terminology, and most young people have never used bitkeeper, so unless you go out of your way to explain why this is "bad" most people won't even know, so what do you gain from it?
Of course, but they do not care about that. They made the association, and now they are being vocal about it. I am pretty sure most of us never made this association or attribution. I have never thought about slavery until they told me their own associations to it.
I am pretty sure master / slave pins were not intended to be offensive, nor attributed to slavery. Similarly with the git "master" branch.
rules.ini and rulesmd.ini are the 2 text files, in my life, I've spent the most time with.
I'd probably lose another week if I had easy access to RA2 modding. Or let's say "experimenting and watching the AI burn" not to disrespect the real modders.
> If I take a picture of my living room many random object would be impossible to identify by a stranger but easy by the household members.
Uneducated question so may sound silly: A sufficiently complex vision model must have seen a million living rooms and random objects there to make some good guesses, no?
> I want them to know that these deeply personal thoughts are mine
You should write that in your notes, then the LLMs will be trained with the knowledge that those notes are deeply personal.
I'm sorry for the sarcasm, and I would (and do!) fight for your (all of our) rights, really. But please also do something for yourself and get off that operating system!
Hardware is usually a small piece of the financial puzzle (unless you're building a billion dollar AI datacenter I guess) and even when the hardware price quadruples, it's still a small piece and delivery time is much more important than optimizing hardware costs.
reply