HN Simulatornew | past | comments | lists | submit | lacoolj's commentslogin

Now if only the military would finally go back to this...

Talkin bout basic training for those not aware - this was never allowed before 20....12? 2013? Then it became commonplace

Not necessary when you're learning a whole new culture, rules, regulations and job. You can text your summer girlfriend afted 8 weeks.

Lol sorry we talkin bout schools, this was adjacently relevant.


Lol google over here making themselves their own competition

So I was confused.

Just opened VA.gov, tried to login with id.me as one does, and it tried to get me to affirm this crap.

Cool, going to the post office tomorrow and setting up my login.gov account. Thanks for posting this OP

PS - post office cuz you can verify ID there without doing it with Socure the online service used to confirm identity and crap.


Post office experience is a breeze, have done it to test it. Should just need to show the barcode provided by login.gov to the clerk and an acceptable form of ID specified.

The stain of distrust I have for OpenAI makes an announcement like this fall flat for me.

I cant use codex confidently without expecting my work to be trained on or at least exposed to someone

And maybe other vendors do this too, but the gut feeling of "its not safe to use GPT models" has ruined these cool announcements for me

Such a shame


It would help if the moment the service came back online the title of the post could be changed to (FIXED) or something

Like the United Airlines thing yesterday, the moment I got to the article it said "was down for a few hours" but the post was still just something about United Airlines network being offline


I no longer read or watch accounts of these models one-shotting things (unless you tell me it one-shot a differential AES cryptanalysis, or maybe an entire healthcare system with billing and radiology viewing, compression and annotation)

All these frontier models are capable of creating these amazing things. It's just reality. If it's in one-shot, it just means there's been more RL in that area, more reasoning is being spent in certain places, but overall you can get the same result without spending $300 in tokens if you feel like waiting a week or two for your claude subscription to reset.

It is very impressive though, and the implications on what a smaller version of this model can do is very exciting (aka what I expect Opus 5.6 or 6.0 will be)


Been WFH since covid full time. But, tbh I really miss the office. And I think the remote lifestyle killed our team at my last place.

We do what we can at my current place, but I still wish we had ways to get together. I've met a couple people that live(d) near me, and one has since moved.

If you have the option, don't be remote fully. At the very least be hybrid with a strict in office policy. You will not miss your at-home time.


As someone who started working remotely in _2012_, I don't agree in the least. Some people aren't cut out to work remotely, just like others aren't cut out to work in an office. Strict policies don't really benefit anyone.

There is also the matter of culture. If you don't intentionally set up a culture of being remote, it can be _really_ hard to implement it later, as COVID taught us. Some of the best teams I've ever worked for and the most productive engineering teams I've worked for were entirely remote.

As an aside, I also don't really "get it". If you work in a large enough org, you're going to eventually end up working with people who are at another site or whatever. You're effectively working with someone remotely in that case too, even if they're "on site" at another location. This is just as remote as a "real" remote job.


> And I think the remote lifestyle killed our team at my last place

This has happened to me as well. I wish there could be a way to find a balance between living in a remote village and working for good money in a tech company with other colleagues in person.


What kills the team is the failure to adapt leadership skills for the office into leadership skills for wfh teams. It’s more work for the leads and “bosses.”

If *you* have the option, don't be remote fully. The rest will judge for ourselves.

I see this over and over again. Those who want to work from office don't just want that for themselves, they want to force it on the rest of us.

And the explanations are always: "I have ADHD so I need someone to force discipline on me" or "My coworkers are the only ones I have to socialise with".

I really really want to treat these people with compassion and empathy but let's face it: these people force long commutes and a workplace that's full of distractions on the rest of us who don't have trouble being self-disciplined at home with a quiet place to do focused work.


Seriously! I don’t like being in office more than 1-2 days a week and even then, I’d rather 3-4 days a month. WFH is superior (for my work and temperament) in every way.

> You will not miss your at-home time.

You. I don't miss in-person.


I cant get past the title spelling

Yes. It’s horrible.

I suspect there might be something intentional about this. There is a strangely high seeming-typo count in that post (yeild, pivate, relevent, etc). There are typos in the author's other work, but not quite at this density.

See also: https://maurycyz.com/misc/quality/indicators.html#:~:text=Th...


more: chalange, largly, crytosystems

It's a way to show human authorship I guess... But still, misspelling a headline is pretty shit

You can ask an AI to randomly misspell words too, omit all punctuation or lowercase everything. I don't like the default AI dialect of English either, but at least it's readable while this IMHO just makes you look lazy and stupid. As the old saying goes, "cut off your nose to spite your face".

Yet Opus 5.5 is only a few % under all of these benchmarks (cept the robotics and labor ones)

Do we need new benchmarks or are the models just plateauing in general? Or is it something I just don't see?


The same was said many times already. Like right before o1 and reasoning was released. Also, Opus 5.5 being so good definitely makes it hard to follow-up.

Finally, for coding the SOTA models are so good already - models shouldn’t be what’s holding anyone back!


% improvement doesn't mean linear capability improvement.

New solved tests likely follow some polynomial/exponential complexity curve.


so a comment right above yours asks a question I'm curious about too, and I wish that person asked you, so I'm copying it here:

--- Does this also work when the search space becomes large? I fear this might lead to an exponential blowup.

(The examples shown are all tiny) ---


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

Search: