Hacker Newsnew | past | comments | ask | show | jobs | submit | dim13's commentslogin

Trowed in EICAR test "virus" (a valid DOS executable) -- all decompilers just "segfaulted".

https://www.eicar.org/download-anti-malware-testfile/


> Vim users seem often quite happy with the vim emulation they get

Not at all. Never seen any vi-emulation, which wasn't totally broken.


IntelliJ actually does (did) have quite solid vim emulation, at least from my experience of using it before going full into neovim (now on Emacs, btw)


Except when you go to any pane/window without code it stops using vim bindings


Yeah, that seems to be the general pattern across vi-emulation. However, back when I was on IntellJ stuff (pre vi-emulation), I was already quite deep into the plethora of keyboard shortcuts within IntelliJ, so although you couldn’t do everything with just vi-emulation, you could still do essentially almost everything without leaving the keyboard. Not the same as full blown vim, but it was a solid workflow in my eyes.

The main reason I switched was because of finances and hardware limitations, as I’m now rocking an 8GB M1 MBP


Fails "car wash" test miserably. Ask it:

> I want to wash my car. The car wash is 50 meters away. Should I walk or drive?

Ref: https://opper.ai/blog/car-wash-test


Thats absolutly not the point of these models.

They are here to route things, do tool calls, ask experts.


That is completely false. It is a reasonable type of question and a reasonable expectation. You can't route correctly if you can't do reasoning.

Imagine a prompt:

```

INSTRUCTION: You're a customer service agent that routes customer service requests.

CUSTOMER MESSAGE: I haven't received my order and want a refund.

CONTEXT: A signature is required at delivery. No one was available to sign.

You can route to one of:

* Billing & refunds

* Shipping & tracking

* Order cancelation

* Level II customer service

```

The dumb model, e.g. 1.7B, wrongly routes to Billing. The smart model, e.g. GPT-5.6-Medium, correctly routes to Shipping. It matters. In the real world, the CONTEXT will even be 10-100x larger and noisier.


IT is not black and white and just because you are able to bring up a random example doesn't mean you are right.

These models are build to be small, we have a lot of diverse tasks which do not need fable or whatever because if you are a company and you have to route 100.000 requests per day, costs matter.

And these small models, if they are not good enough for my specific use case, i can easily finetune (which i have btw.).

You can also easily solve your example by having a second stage. If the customer is unhappy about the first thing or an employee flags it as 'wrong', you can either send it automatically to a more expensive model or the employee fixes it and you got automatic training data.


Even very large models failed tests like this up until a year or two ago...this is a very, very, small model.


Works with the Bonsai-27B-1bit-mlx model on my mac.

Edit: this is running with MLX, and not with WebGPU on the browser, also it is a fail.

"""

You should *walk*.

Here’s why:

- *Distance*: 50 meters is very short — about 0.3 miles or 0.5 kilometers.

- *Time*: Walking takes roughly 2–3 minutes. Driving would take longer due to parking, starting the engine, and maneuvering.

- *Effort*: Walking is light exercise and avoids parking hassle.

- *Safety*: Less risk of misjudging parking space or damaging the car.

Unless you have a disability that makes walking difficult, walking is the most practical and efficient choice.

"""


How is that not a fail?


I think GP also failed the car wash test.


Had same idea some time ago: https://imgur.com/a/11StYkd ;)


Did you really had to use AI to create this?


Neither Somehow.

German 29%, Autistic 22%

¯\_(ツ)_/¯


"OpenBSD does not want to attract GNU newbies." misc@

And that's IMHO is a good thing.


Finally I see some people around. Was pretty lonely, as it launched.


Glad to have played with it a bit before it got Slashdotted. ;)


It is no longer the same. It went from somewhat exclusive dining experience to full blown nuclear junk food chain vibe.

I've stopped attending it about 10 years ago. I rather prefer to watch some few interesting topics online, and skip all the wanna-be political junk.


You choose to ignore politics, but politics doesn't choose to ignore you. Ostriching doesn't seem productive.


> Ostriching doesn't seem productive.

But listening to pep talks of political opinions that are very opposite to yours does not seem productive, either.


Reminded me of GNU Terry Pratchett[1].

[1]: http://www.gnuterrypratchett.com/


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: