HN user

samstave

8,455 karma

sstave via proton mail

Posts194
Comments9,927
View on HN
news.ycombinator.com 1y ago

Ask HN: Serious] Risk of agentic local network access? (Cursor Agentic sysad)

samstave
1pts6
news.ycombinator.com 1y ago

Ask HN: How can we provide mass contextual knowledge rapidly to a brain/nervous?

samstave
2pts3
facelandmarks.com 1y ago

How to Setup ARKit Face Tracking in Xcode

samstave
1pts0
news.ycombinator.com 1y ago

Ask HN: Discernment Lattice [Prompting]

samstave
6pts17
news.ycombinator.com 2y ago

Ask HN: Subscribable LLMs for great justice and profit?

samstave
1pts1
news.ycombinator.com 2y ago

Show HN: HAL 500 gatekeeping: It "feels" my question is insensitive

samstave
1pts1
news.ycombinator.com 2y ago

Tell HN: We need a complete thread on CLI

samstave
1pts0
news.ycombinator.com 2y ago

Ask HN: Trained LLMs on Network Traffic?

samstave
2pts1
news.ycombinator.com 2y ago

Ask HN: Human) Parents' native – language as it pertains to AI

samstave
1pts1
news.ycombinator.com 2y ago

Ask HN: State of CD/CI Wrt AI? Anyone Using a GPT to Control Ops/SRE/CI Tasks?

samstave
1pts0
news.ycombinator.com 2y ago

Ask HN: At what point do the Cloud's datacenters become national security sites?

samstave
7pts0
news.ycombinator.com 2y ago

Ask HN: Serious: How to keep up WRT human education?

samstave
1pts0
news.ycombinator.com 2y ago

Ask HN: Should open data sets be behind 'token walls' (paid access AI analysis)

samstave
1pts0
news.ycombinator.com 2y ago

Ask HN: Crazy AI/"CoPilot" challenge: have a GPT build an app on Git posts to HN

samstave
1pts0
news.ycombinator.com 2y ago

Ask HN: Serious; Do you have self amnisia in comment discovery? (see below)

samstave
3pts1
news.ycombinator.com 2y ago

Ask HN: What will become of current population of AI's GPUs as they age?

samstave
2pts6
news.ycombinator.com 2y ago

Ask HN: Police use of drones and the 4th Amendment?

samstave
3pts7
news.ycombinator.com 2y ago

Ask HN: Anyone implemented an 'F8' keypress counter for Baldurs Gate 3?

samstave
1pts1
news.ycombinator.com 2y ago

Ask HN: Worst outcomes with self-driving being hacked?

samstave
4pts6
news.ycombinator.com 2y ago

Ask HN: [Serious] – I want a medical LLM of my own bionomics (genome, etc.)

samstave
1pts0
news.ycombinator.com 3y ago

Tell HN: Dont tie your life to your tech

samstave
10pts5
news.ycombinator.com 3y ago

Ask HN: I want every single thing. txt, VMs, emails. Period

samstave
3pts2
news.ycombinator.com 3y ago

Ask HN: Knowledge assesment – AI asking Person

samstave
1pts0
news.ycombinator.com 3y ago

Ask HN: Does Reddit pay Imgur for pummeling of its services/API?

samstave
4pts3
news.ycombinator.com 3y ago

Ask HN: Curious, do you see Holmes being either flight or suicide risk?

samstave
1pts2
news.ycombinator.com 3y ago

Ask HN: How is “AI Alignment” currently being(treated/discussed/eval'd) on HN?

samstave
3pts2
news.ycombinator.com 3y ago

Ask HN: Are there any willing HN mentors/tutors for various subjects available?

samstave
1pts0
news.ycombinator.com 3y ago

Ask HN: How close are we to Bard/ChatGPT/AGI crawling sites as HN for answers?

samstave
3pts7
news.ycombinator.com 3y ago

Ask HN: Do you have any input on the following, or a fact-checking of the claims

samstave
1pts4
news.ycombinator.com 3y ago

Tell HN: We need an HN EDU – an archive of ideas

samstave
1pts4

I read this in Sam Wattersons voice with a pipe abt maybey an inch from his beard,

(Fyi I was a designer at fb and while it was luxious I still hated what I saw in zucks eyes every morn when I passed him.

Super diff from Andy Grove at intel where for whateveer reason we were in the sam oee schekdule

(That was me typing with eues ckised as a test (to myself, typos abound

I want a tiny_phone_based LLM to do thought tracking and comms awareness..

I actually applied to YC in like ~2014 or such for thus;

-JotPlot - I wanted a timeline for basically giving a histo timeline of comms btwn me and others - such that I had a sankey-ish diagram for when and whom and via method I spoke with folks and then each node eas the message, call, text, meta links...

I think its still viable - but my thought process is too currently chaotic to pull it off.

Basically looking at a timeline of your comms and thoughts and expand into links of thought - now with LLMs you could have a Throw Tag od some sort whereby you have the bot do work on research expanding on certain things and plugging up a site for that Idea on LOCAL HOST (i.e. your phone so that you can pull up data relevant to the convo - and its all in a timeline of thought/stream of conscious

hopefully you can visualize it...

That is BEAUTIFUL.

So I applied a long time ago to YX with an idea of "Jot_PLOT"

Take this idea;

An app which maps a song to an interaction. The idea is that you have an interaction with a friend - and give it a soundtrack.

and then you can tie them together,

and you can hear a soundtrack of how you interacted with that friend or with group in a playlist etx

every time you rcv txt a song is assoc

--Better yet song emogid - a link to a song to express something, or feeling.

but dont make it monolithic - make it loving anf un with samples and refereences,....

(you understand, but I love your commment.. - lmk if you need help - nut this can be massive - imagine this thought on youth and dialougues on such...

Your ide has legs BUILD IT

+1

I found the mode.

Too early maybe. Your comment is profound, and loved and received.

You must watch the Telepathy series on JRE and such,

Your wording is well sorted. We are finding ourselves - ive been open on HN with resistance,

But the entiretity of tech build in Silicon Valley has been built on som dark shit.. (but its all around mind manip and Sex Magikx if you want to discuss that) ( Morehouse university and such)

regardless I appreciate you

(Ill freely give you all my creds if need... I know a lot)

Ive noticed in my life;

My brother has an exceptional sense of smell... and he had a wonderful memory.

I personally, have a compltely retarded_sense of smell, and a poor mem.

I was extremely involved in martial arts (pre-olympic, blah blah) and got hit in the nose a lot)

--

I had a photographic mem until I was about ~14 which I originally thought of as the first time I smoked pot -- but really it was when I started Martial arts...

((I used to take tests in Biology whereby rather than doing the test, I would re-draw the entirety of the chalkboard of the class for that test - and not answer a thing)

(I have memories from 6 months old... where a girl named Wednesday attempted to smother me in my crib with a pillow) ((hippy commune, 1970s, Sanf Francisco -- yes my mom was friends with Jim Jones -- I have incredible memories))

but I relegated my mem loss to various things -- but I have a REALLY poor sense of smell currently. and a poor mem as such.

(However, I can tell you every single passwd Ive ever typed into a machine. The first ISP passwd I was given in the early 90s by netcom auto gen was "FblQ00Ho" in ~96-ish... ((part of my frustration in IT when someone says they cant recall they passwd))

We should have a device that provides a scent when someone selects a secret - and then have them select the passwd induced by the scent.

(I dont give a heck about how nuts the above sounds... its real. thats why its called NOSE-To-LOGIA nostalgia. (I love dissecting etymology)

--

I wrote the following joke eons ago about photographic mems as such:

"I have a photogenic memory... Whenever I think back on myself - I look fantastic"

An interesting thought to think about is how sense of Smell is related to memory (from an evolutionary perspective the sense of smell-mem link was to ensure we Humans knew when something was not OK to ingest)

But in modern times, now you can have smells link or lock memories in...

So for you as a line cook - you have a bunch of experiences with these songs linked to whatever you were smelling at the time which then locked the 'nostalgia' into memory - so now in the Reverse RAG situation, you hear a song and your memories of those times are really strong.

And to quote the magnanimous Steven Wright:

"Memories... they're really the only thing you have to think back on"

Another thought: Can it do this:

"give me a list of the latest podcasts about/from [subject/channel] {{from the already subscribed channels}}"

--

Or a crontab of schedule "play the latest X at Y time" (so you can tell it to put on your bedtime playlist starting at 9pm)

sort of thing?

Wonderful!

(Also, to all the other posters who have done the same for themselves)

--

I have been mentally building a UX I want out of YT over the last few weeks. What I want to do is have it go through all my history and categorize it and give me a local page and sqlite3 of my browsing hist with various meta-data..

My YT experience has gotten so poor, that even browsing which channels I am sub'd to and finding newer vids in them is a nightmare of a dark pattern...

I thought I wouldnt be able to pull off my vision - but this gives me new hope - and I had told myself that this week I would make an attempt.

One thing I want to do is include VoidTools 'Everything' Search into some MCP tools for Cursor -- and this inspiration ties it all into a more formulated vision for what I want out of a YT ux.

I look forward to trying this out and seeing if it fills the void - or still build my own thing.

(There was an HN SHOW: that was "what if YT channels were like a TV some time ago and that always pops into my head)

--

EDIT: With the postings of GH repos and such, and my comment on categorizing and searching hist -- I also want to be able to have a dashboard of GH repos that I click on, and then have that click in hist be sent to my history categorizer automatically and give me a summary of the thing and category. maybe even from which site I found the repo -- so much like broawsing a YT hist of vids - being able to see all the repos I have been interested in.

Anyone build anything like that for themselves?

This is great!

I posted the following the other day -- and if you like this post and the maths involved - you will really enjoy Randall Carlson's Sacred Geometry lecture whereby he talks about all the Ancient Sacred Geometry and how it was used to measure the earth, moon, sun, stars and Earth's precession -- and how it all relates down to the measurements from the Human Body - whereby the 12-inch foot comes from, Cubits, etc.

---

Speaking of "feet" as a measurement - Randall Carlson has an AMAZING video[0] on the source of the 12 inch foot.

And how its all related to the measurement of the precession of the earth. And yes - its specifically how to measure things in space. And its all from Sacred Geometry.

[0] https://www.youtube.com/watch?v=R7oyZGW99os

I do this as well.

There was an article about Sam Altman that stated that ex/other OAI employees called him some bad_names and that he was a psychopath...

So I had GPT take on the role of an NSA cybersecurity and crypto profiler and read the thread and the article and do a profile dossier of Altman and have it cite sources...

And it posted a great list of the deep psychology and other books it used to make its claims

Which basically was that Altman is a deep opportunist and showed certain psychopathological tendencies.

Frankly - the statement wasn't as interesting of how it cited the expert sources and the books it used in the analysis.

however, after this OAIs newer models were less capable of doing this type of report, which was interesting.

Ingestion of context.

Period.

If you dont understand the benefit of an AI augmenting the speed and depth of ingestion of Domain Context into a human mind.. then... go play with chalk.||I as a smart Human operate on lots of data... and AI and such has allowed me to consume such.

The most important medicines in the world are MEMORY retention...

It s youd like a conspiracy, eat too much aluminum to give you alzheimers asap so your generation forgets... (based though. hope you undestand what I am saying)

Your definition of "truth" is limited;

Truth: The real meaning behind the words which may or may not be interpreted by the receiver in A/B/N meaning

Truth: The actual structure of the nature of whats being presented.

When you can manipulate an individual's PERCEIVED reception of TRUTH between such, you can control reality... now do that at scale..

Dont look at it as such, mayhaps;

Look it at literally who will have GPU dominance in future. (obv who will hit Qbit at scale... but we are at this scale now - and control currently is controlled by policy, then bits, then Qbits.)

Remember, we are witnessing the "Wouldnt it be cool if..?" CyberPunk manifestations of our Cyberpunk Readings of youth?

((I buildt a bunch of shit that spied on you because I read NeuroMancer, and thought wouldnt it be cool if..."

And then I helped build Ono Sendai throughout my career...

OMG Thank you - hilarious. TimeCube is a legend...

---

I am saying that AI can be used very beneficially to do a calculated dissection of the Truth of our Political structure as a Nation and how it truly impacts an Individual/Unit (person, family) -- and do so where we can get discernible metrics and utilize AIs understanding of the vast matrices of such inputs to provide meaningful outputs. Simple.

EDIT @MegaButts;

>Why is this better than AI

People tend to think of AI in two disjointed categories; [AI THAT KNOWS EVERYTHING] v [AI THAT CAN EASILY SIFT THROUGH VAST EVERYTHING DATA GIVEN TO IT AND BE COMMANDED TO OUTPUT FINDINGS THAT A HUMAN COULDN'T DO ALONE]

---

Which do you think I refer to?

AI is transformative (pun intended) -- in that it allows for very complex questions to be asked of our very complex civilization in a simple and EveryMan hands...

>favor the other side, to obfuscate and protect falsehoods.

Unless they delete the Internet Archive of such info?

>*and they don't do math.*

Have you ever considered the "Math of Politics"

Yeah - they do that just fine

((FYI -- Politics == Propaganda.. the first iteration of models is CHAT == Politics...

They do plenty of "Maths"))

Thats awesome.

See my post below -- I have been tracking Eschelon since the early 90s...

Guess what NSA router backdoors have become (mobile phones with socialifelog media apps on them)

---

And like @tptacek said

>If it was the NSA, their van wouldn't look so goofy that people took one look at the photo and assumed it had to be an NSA van, which is what happened here. This is a bad movie plot trope: the bad guys can't simultaneously be omniscient and so dumb they're trivially outed like this, just like the real supervillain isn't going to monologue while you free yourself from the chains lowering you into the shark tank.

---

Love that but, I do think that both is true...

Look incompetent so they don't think you're competent (Stuxnet/DUQU)

CDroms were a luxury addon when W95 was released - but every machine had a 3.5

I want to say it was in the ~20 disk range...

There were a lot of really fun things that happened with W95 - a lot of "mischevious" cyberwar...

Like taking image of desktop as background came out with that - so nothing was clickable as a prank.

There were several backdoor utils

There were several prank links to something that seemed serious/work -- but then switched to a really loud voice yelling "IM WATCHING P*RN"

(The backdoor utils were really powerful though, and they remind me of a thing I am doing with Cursor/Claude -- Agent mode access to a fresh windows laptop as admin and having the bot fully config my new windows machine to my specs.

I knew I would be in that 2013 thread...

I mistook the building, but I do remember details that Twitter had a direct fiber connection to that room...

Also, we have a LOT of evidence of prior NSA backdoors and interceptions...

in 1998 I had to hire a CSIE (cisco expert) (like a 3 digit uuid) to help me recover a router password from infra I inherited... and during the password reset procedure on a 3640 - he was telling me how "the NSA requires Cisco to put in back doors into all the routers)

((The passwd BTW was Feet4Monkey))

--

Then recall Carnivore? (and its predecessor eschelon - and a whole bunch of reveals) -- what was interesting was that the only company to refuse to install Carnivor was Earthlink.net (ISP) -- and the reason they stated they wouldnt put in Carnivore, was because they stated they already had their user tracking system (They were owned by the Mormon Church) ((and for some reason Whoopi Goldberg was one of their large notable investors))

And recall how they stated that the NSA specifically likes to hire Mormons?

And recall North First Street DC that was purchased by Cerberus Group which was the ~Bush-Cabal hedgie, and the reason they bought it because it housed MAE WEST and they wanted to inject NBAR/Surveilling into it -- once they completed that, they sold it off again (To one of their subs, IIRC)

I hate that I am getting old and I start to forget a lot of the malfeasance I have witnessed in my ~30+ years in SV.

One of my "hero memories" was a time when I was a master of Win95 - and a friend had accidentally changed ALL of her display options to black - so all the UI was black, but I knew Win95 so well I could navigate the entire OS via keyboard - and was able to from memory navigate through the start menu, to settings, knowing how many tabs to hit to get to display and change that back to default.

The people watching thought I was a magician.

(I also had several sealed original W95 boxes on floppies...(we shutdown an office, and as IT mgr - I had to go liquidate - and we had ~50 boxes of original release W95s there - so I took several home) and I held them for ~10+ years then sold them on eBay, I only got $25 for each - but I sold them as pieces of "computing history")

The weak point in the OAI armor is SLAs.

So - are people forming relationships with OAI which include an SLA, and if so - what do those look like?

Having gone through quite a number of compliance audits... the one thing that is good in that advice, is that many items in an audit are just a checklist of questions, such as

do you have a policy for XYZ?

or confirm you have a process for "thing"

So what ends up happeneing is if you feel stressed about an audit, just getting a list of the audit, you will realize how much you can just say "yes" to and feel less daunted by the audit.

So, its a good self-check even if youre just crossing out the things you should have already have a framework for.

This would be fun: Put an AI bot with an input and a Deep Draw output with Protein Folding logic to make a cellular wall...

("make a deep draw wall based on protein folding

This is fun:

https://i.imgur.com/JRCE4Xr.png

https://i.imgur.com/aEgNRih.png

https://i.imgur.com/4w8H9Tg.png

https://i.imgur.com/FE6GkJH.png

I need to work on it though.

---

https://i.imgur.com/ORc6ueK.png

https://i.imgur.com/ClH4Coy.png

----

https://i.imgur.com/CTslAxE.png

------

https://i.imgur.com/pUhSGho.png

I need to attempt in Cursor or Windsurf Agent mode and have it build a scene

=====

https://i.imgur.com/vQm92yC.png