ilc

Honestly: Play a game where you are supposed to win.

There is a reason why single player video games are still popular. It is fun to win!

I play ARPGs, where, I mow down hordes of enemies with a mere mouse click with fun graphics and sound.

Is this cotton candy compared to playing PvP games. Absolutely. I've done serious PvP before. But, after a day of work... sometimes, it is just fun to save the world, or make it burn.

Because the whole way they've handled this situation reeks.

Oh, we'll get you on a wait list. Bullshit. Waitlist never came in.

It's OK to say the waitlist closed. Keeping people hopeful is a shitty move.

Then to pull this on launch... (sigh) I didn't want to have to waste time on this shit. I'll go back to VSCode, neovim or whatever if I must.

I'm more frustrated with myself that I trusted them to provide me a year+ of quality service... and that isn't happening.

I am considering it. Seriously.

The policies of my company are clear on the issue. I can't use a tool I previously used safely... yep. I'll be changing IDEs. I may go back to neovim, emacs, or whatever.

Alas, it blocks my upgrade totally.

I'm glad your firm lets you send code anywhere. Mine does not.

The thing about trust is... Once broken it is hard to recover.

Here we have 2xshitty behaviors, we can see around this product. Who is to say there isn't more we can't see.

Removing it seems 100% reasonable, if only for security purposes.

I haven't upgraded yet. I may have to pin my version so it doesn't upgrade.

For someone who wasn't in their AI trial, then gets this as a subscriber. It sure feels shitty. And Fleet doesn't looks much better. Maybe it is time to look at VSCode again... it's been 6-9 months.

But could the same problem have been done with a simple expert system, instead of a chatbot?

People seem all caught up in the new hottness, and forget the technologies that still work and are simple as dirt.

Draw me an image where a little girl burns her finger on a hot pan, and is bent over in pain yelling.

Proof by example.

Doesn't the article say this is a LoRA off ORCA? Isn't that a finetune?

I don't use them for auto-complete that may be the difference. I use them for various other things. Tests, etc. I don't find them 100% accurate. But it beats writing it from scratch.

Don't fall into the trap of: It has to be the best or it is useless.

The question for me is: Does it help me get my job done. Does it assist me in ways I find useful.

Currently Phind_Phind-CodeLlama-34B-v2, gonna try CodeBooga because, I can... that's the point. Use what ya like.

4k/32k, if I remember right. An interesting thing to do is actually try to use a true coding model in a chat. I had one model puking out its stack overflow training data, down to including user IDs.

Oops.

Honestly, with a 3090 costing like $700 used. It is within reach of many on this site quite easily, assuming they have a machine to plug the 3090 into.

People seem to think the hardware for this stuff is way out there... it is getting to be within reach VERY quickly.

Been using CodeGPT a while myself.

Great tool, send e-mail to the author, seems like a good person.

I actually hook it up to my Windows gaming rig. The 4090 gets the job done, when I'm not gaming. (I just start the server when I go work, and then stop it when I want to game. It isn't that big a deal. I haven't hit the use-case where I want to use the AI to develop game mods yet. But I'll cross that bridge when I get there. ;) )

Don't worry, it'll get to OpenAI^H^H^H^H^H^HMicrosoft soon enough via the API.

There is a reason I didn't draw a hard parallel.

I use AI day to day. I see what it can do and can't.

But when you see pages, and pages, and pages of GPT spam all over the place. Finding the nuggets of wisdom will be much harder than before, the "bomb" was dropped.

Thus actually leading to the whole FO4 main plot.

Yes, life will always find a way. And yes humanity can not put the genie in a bottle, we are much more likely to put it on a Fat Man catapult.

But it means, that in a sense... that we will all have to accept this background radiation of AI shit, as part of our new norm.

And this isn't the first time I've thought in similar ways. I remember reading older math texts (way pre-computer works in things like diffeq and PDE) and often thinking the explanations were clearer. Probably because of the increased effort to actually print something.

Who knows... maybe I'm just an old coot seeing patterns where there are none.

This whole thing reminds me a bit of playing Fallout 4.

There will be good data, the pre-AI enshitification data. The stuff from before the war.

And then... the data after. Tainted by the entropy, and lack of utility of AI.

Alas, this means in some senses, human progress will slow and stop in the tech field if we aren't careful and preserve ways to create pre-AI data. But the cost of it is so high in comparison to post... I'm not sold it will be worth it.

No contract can save you from what you don't know, and don't understand.

Different points of view, if they can stay respectful. Can be a great strength.

I see the world in terms of risk. If X will help me derisk something. Even if it isn't totally EV+ it can be a good play, because risk of ruin is real. Especially with large amounts of money, and small numbers of trials.

25% less to have a 10% better chance at making 75m vs 100m? Probably worth it a that level. Because 75m or 100m is life changing money for 99.99% of people at least. The amount won't matter as much as your chance of getting to it.

It does help to have SOME finance background when doing all this.

I will not deny that people are gonna get screwed. The question is how to minimize the damages.

The NSA doesn't need to know what Google knows. It needs to know things beyond what Google knows.

It may have less information overall. But that information may be of much higher value.

It is not much data to know how other countries are encrypting their political or military data. And the amount of data in there may not be that high.

The value is... tremendous. Bulk data collection, gives an overall picture. But the details that make it real, that's the secret sauce of the NSA and CIA.

As a technical founder/early person in my past:

If you don't know why you need a technical founder. That is EXACTLY why you need one.

It is the same reason why if as a technical founder, I didn't know why I needed a non-technical founder. I'd probably be making a major error.

To those saying they waste equity:

If you truly have a failed company, with a bad founder, buy them out or kill the firm, it happens. You've learned some key lessons.

If the founders you need cost you 75% of your equity and you make a unicorn.... Were they worth it? Who cares. You got your fuck you money.

Honestly as a technical guy: You come off as greedy.

As a technical founder, I face the EXACT same risks. What if the business guy can't drum up money? What if he can't sell the damn pen.

Most companies fail, the strongest firms will have strong business and technical people.

If you want to stand on one leg. Go for it. You'll piss your money away overseas. When you need someone to help you get across that line... there is nobody at your side.

Founding a successful company is hard. You will face days where you don't have the answer. Things will happen that are outside your experience... and yes, you can buy, buy, buy your way there... but remember, you are buying mercs. I've been a merc. Win or lose... I get paid, so, yes sir. How high sir?

Rework is just more money.

... You sure you want to be on the other side of the table from someone like me, without someone like me at your side?

Because all those off-shore firms... are more merc than I am.

Honestly: I rely on my shell scripts working and any "grep replacement" has to work with all the old crusty shell scripts out there, likely including ones that use odd "quirks" and GNU options.

If you want to innovate in this space, why sign up for all that? Invent a better wheel, and if people like it, they'll migrate over time.

I remember using ag in the old days, and I use rg now. But there's things rg does by default that I don't like at times... so I go back to old fashioned grep.

rg is at the point where many programmers use it. I think it is on its way to becoming one of those "standard tools". It needs... another 5 years?

When POSIX has a rg standard... we'll know ripgrep "succeeded" and teargrep will soon come into existence ;)

I do also. I wish the LG had Nebula. I'd gladly pay the fee.

There is nothing wrong with paying for a service you use.

If you are a normal techie. Ask yourself: How many hours will I spend circumventing this, then having to deal with when it breaks etc. Vs. paying the bill. For me, it was an easy call, and one I don't regret.

What is amusing is that he goes on to prove his own point, when saying that Trump benefited from the underdog effect. The best pollsters had him at about 40% to win (538).

That is an underdog win. But far from an unexpected win.

Overall: Think sideways? No. Think for yourself.

Also if the entire world is going one way... and you think differently. Stop a moment and ask someone: Am I crazy or is the world crazy and lay out your logic. You may discover... the world is crazy.

In a world with so much cargo-culting etc. Insanity is around the bend.

Somehow, I'm betting there's a non-compete with some pasture money in there somewhere.

I've played my share on-line.

Honestly, I can't see it as a job. The skill at the high levels is insane, the variance is insane.

There is a reason people have said "No professional will win the WSOP again." Even if you are 5-10% better than everyone else... There is many, many, many trials to survive, and you only have to lose once when you are right.

There's a reason Phil Hellmuth folded KK early in the WSOP. And when you understand why... It speaks volumes of his thoughts about himself. Right or wrong.

The fish are out there. It just ain't worth frying them. Go have fun playing. It is a fun game. But a living... for just a few.

I wonder what it'll have to do to get to an IPO ;)

For those who haven't read the code:

This reuses Fig.io's internal completion engine to actually get the job done.

Because of that, it forces the language choice of Typescript, and the heavy lift is done by Fig's engine.

I hope the OP continues with the work and improves it. I was looking forward to trying Fig on Linux. This sounds like a step towards that.