HNHacker News
TopNewBestAskShowJobs

yourapostasy

3,287 karma · joined October 3, 2011

submissionscomments
yourapostasy··on How well does the money laundering control system work?
> ...an AML form that needs filling in if the sales go through...If that’s not overreach, I don’t know what is.

Absent a complete dismantling of AML oversight (and I do have empathy for those free market purists who want to wholesale toss out KYC and AML, but for now in practice we have to deal with what are on the books), these are difficult use cases to address. These kinds of edicts don’t usually say, “any retirees buying even economy cars must fill this out”, it is usually broadly applied like, “all car dealerships must do this for all transactions no exceptions”. And these laws are often a lagging reaction to various sham transactions uncovered as a result of crime busts.

Once you start weaving “reasonable” exceptions to address the overreach, the scammers start to come up with sham transactions that fit the exceptions filter. It’s a pretty fascinating problem.

There are many who would still object to a system where your mother doesn’t fill out a form. Instead her banking app pings her to confirm that she is purchasing a Yaris from the dealership (looking at the other comments here, it seems anonymous large transactions scattered through many people with otherwise clean records are a common laundering pattern, so metadata on the nature of the transactions might be one way to counter that kind of structuring, but alas that’s overreach for many), and uses her financial history in the background as the AML controls rails.

I’d love to see AML professionals participate in this thread to help us learn what they’re really facing. Assuming we have to put up with it for the time being, might as well design and make its implementation as low friction for lawful people as possible.

yourapostasy··on Cognition (Devin AI) to Acquire Windsurf
> Current situation doesn't sound too good for "scaling hypothesis" itself.

But the “scaling hypothesis” is the easiest, fastest story to raise money. So it will be leveraged until conclusively broken by the next advancement.

yourapostasy··on Private sector lost 33k jobs, badly missing expectations of 100k increase
This happens everywhere with lots of people in many different contexts. I call it the “‘why don’t we just’ disease”, or WDWJ Disease. When you’re in any leadership position, you have to stay especially careful to catch yourself from falling prey to this pernicious effect and behavior, and its equally debilitating sibling yak shaving when you over index on preventing WDWJ.
yourapostasy··on I convinced HP's board to buy Palm and watched them kill it
This form versus substance issue is a really deeply embedded problem in our industry, and it is getting worse.

Time and again, I run into professionals who claim X, only to find out that the assertion was based only upon the flimsiest interpretation of what it took to accomplish the assertion. If I had to be less charitable, then I’d say fraudulent interpretations.

Promo Packet Princesses are especially prone to getting caught out doing this. And as the above story illustrates, you better catch and tear down these “interpretations” as the risks to the enterprise they are, well before they obtain visible executive sponsorship, or the political waters gets choppy.

IMHE, if you catch these in time, then estimate the risk along with a solution, it usually defuses them and “prices” their proposals more at a “market clearing rate” of the actual risk. They’re usually hoping to pass the hot potato to the poor suckers forced to handle sustaining work streams on their “brilliant vision” before anyone notices the emperor has no clothes.

I’d love to hear others’ experiences around this and how they defused the risk time bombs.

yourapostasy··on How Big Is VMS?
Would you happen to know where developers who want to integrate against MCP and OS 2200 running on a variety of hardware and OS versions (compile targets for validation and distribution, really) can lease time from?
yourapostasy··on [dead]
Due to Betteridge's law of headlines: "No". But I loves me some Sabine sometimes, and watched anyways.

TL;DR: "No, unless this Doppler stuff later on turns out to have legs and this time around they were just incredibly sloppy."

I swear, half this conspiracy pot-stirring has better storylines than the Hollywood pablum out there these days. I can't wait for these folks to get their hands on cheap LLM video generators and make full multi-hour video serials from these storylines. Now, there's some AI-proof work, as their hallucinations don't go nearly as far as the best of these fabulist universes.

yourapostasy··on MCP vs. API Explained
Under this thesis, then SLOP would win, except I don’t yet see how it can be composed by the user, which MCP is supposed to have moved the composability into?

https://i-love-slop.com/

yourapostasy··on Svelte 5 is not JavaScript
So how up to date and idiomatic all the React code shoveled into production servers by LLM's can be, if their training data is mostly riddled with deprecated styles?
yourapostasy··on Using Linear Programming to find optimal builds in League of Legend
Googled around and yep...there are EVE Online gamers using LP as well. What I find interesting is from skimming the content, I don't see anyone doing statistical sampling of opposing force specs, then using that to LP-build that has special applicability towards those specs. I am fascinated by whether it was tried and rejected as not a good strategy by gamers.
yourapostasy··on Blue Origin reaches orbit on first flight of its titanic New Glenn rocket
20 stops dynamic range is about the human eye's range. Achieving that for digital capture and display would be mind blowing.
yourapostasy··on Ghostty 1.0
Neat, this [1] looks like the project uses vttest [2] to test DEC VT terminal compatibility?

[1] https://github.com/ghostty-org/ghostty/blob/main/TODO.md

[2] https://invisible-island.net/vttest/

yourapostasy··on Can hunters' donations help deliver high-quality meat to Colorado food pantries?
If someone can figure out how to scale up deboning Asian Carp and turning them into fish sticks [1], then there are many infested waterways that would feed many, and the ground up remains make great plant fertilizer and animal feed. They're easy to catch at volume in the infested areas.

Hopefully, AI advances will help make it economically easy to kill in the wild invasive Burmese pythons at scale with robots, though unfortunately, they contain excessive mercury and lead [2] and are more unsuitable for eating than tuna. I have no idea what could be done with their carcasses. Still boggles my mind to this day that we polluted our planet with coal smoke to the extent that it is dangerous to eat some fish too frequently, a science fiction/fantasy notion to our ancestors a mere 1-2 generations ago, and we still outstrip by 2X volcanic eruptions and undersea vents pumping out mercury into the atmosphere.

Colorado has controlled its feral pig population, but the infestation is out of control in the southern US and could do with someone figuring out an economical way to scale up killing them in the wild. Frozen to -10 degrees F at the innermost parts for five minutes, kills the parasites that infest them, and many people are accustomed to eating pork can handle the gamier feral version.

Lionfish in the Caribbean are invasive, but also difficult to economically kill at scale in the wild. Tastes like buttery grouper.

There are lots of invasive species in the wild that need to be harvested, but the cost per kg is way higher than current conventional sources of protein, and hunter/fisher-sourced channels, while appreciated for the gesture, still lack the scale we seek to foundationally address the income precarity that pervades our civilization.

[1] https://www.youtube.com/watch?v=GaZHPToxb3Y

[2] https://www.naplesnews.com/story/news/2024/08/16/florida-pyt...

yourapostasy··on Bpftune uses BPF to auto-tune Linux systems
Record changes in git and then git bisect issues, maybe?

Without change capture, solid regression testing, or observability, it seems difficult to manage these changes. I’d like to how others are managing these kinds of changes to readily troubleshoot them, without lots of regression testing or observability, if anyone has successes to share.

yourapostasy··on Netflix buffering issues: Boxing fans complain about Jake Paul vs. Mike Tyson
What I’m seeing in large organizations is tracking dependencies within a team’s scope is better than dependencies between teams, because so many developers between teams are punting on tracking dependencies upon other teams’ artifacts if there isn’t a formal system already in place at the organization to establish contracts between teams along these dependency routes, that automatically handle the state and notifications when changes announcing the intended state change are put through the system. Usually some haphazard dependency representation is embedded into some software and developers call it a day, expecting the software to auto-magically solve a social-technical logistical problem instead of realizing state transitions of the dependencies are not represented and the software could never deliver what they assume.
yourapostasy··on Hertz's Tesla Fire Sale Continues as Depreciation Costs Mount
My US friends share with me that car repair costs have followed inflation and not general population wage growth (unfortunately, mechanics other than independent mechanics likely aren't seeing a commensurate increase in invoice bottom lines). Unless you wrench it yourself, on some vehicle like an old 80-90's diesel pickup, 4 figures invoices are common now.
yourapostasy··on Show HN: OnAir – create link, receive calls
For customer/client communications, absolutely.

Internally however, once I figure out someone is haphazard with meeting time and meanders around in meetings getting precious little if any action done, I take proactive measures.

When they request a meeting, I ask for the agenda and the outcomes they want to take out of the meeting. Most of the time, everything is hashed out in advance in async channels and the meeting is either highly abbreviated from the original invitation or cancelled altogether because it turns out, the rest of the invited attendees also weren't crisp on what the meeting organizer wanted.

The number of people who call for a meeting simply because they haven't organized their thoughts, asked the right people the right questions, and are simply waiting around for someone to tell them what the correct next step is...is thankfully not an alarming number, but nonetheless simply a personally idiosyncratic annoyance to me.

yourapostasy··on OpenObserve: Observability platform for logs, metrics, traces, analytics
> ...conversations with multi-billion dollar companies...

Slight clarification here. Might not apply in OpenObservability's case but might help others on their journey to enterprise sales with their projects.

Those are typically conversations with managers holding $X purchasing authority, typically like $500K for a US director'ish level, within multi-billion dollar companies. These managers usually aren't averse to spending on open source projects. They're averse to cutting a check not tied to a support contract with responsive, polite, helpful support with published support policies at 0300h local time on a break-fix line with 75 other people from other support teams in the company watching. A surprising number of open source projects won't offer that guarantee, and instead only offer the option to "donate" with vague promises of priority support. More projects are getting better at this more recently, but it takes a surprising amount of red tape to onboard as a vendor into these organizations, and a lot of open source teams don't have the appetite for putting up with that.

Until kind of recently, the conversation switching to the SSO Tax is really about accessing that level of guaranteed support delivery.

yourapostasy··on Understanding how bureaucracy develops
A key feature I see that rampantly grows bureaucracy: the ability to impose costs upon other teams without accountability. This takes many forms, but common ones include the following. People don't say the following outright, you spot these a mile away when actions speak louder than words.

"I don't have to know how what I tell you to do fits into the overall flow of what you want to accomplish between multiple different teams. That's your problem, go talk to all the different teams you have to work with to get approvals and figure it out yourself, that's not my problem."

"I acknowledge absolutely nothing has changed, and I will never use what you create, but format A has changed to format B because everything new is in format B, and thus you must spend time to re-create everything in format B so everything in my filing system is consistent."

"Yes, this is recording the information twice in two different ways. By hand. And with no enforcement to keep it up to date. But I'm only measured on recording the information, and as long as Number Go Up on the recordings, it is not my problem."

"Yes, the new form has not been published. Yes, the new requirement for these approvals have not been put into existing processes that take months to grind out a decision, much less the processes debugged. But in six months, everyone must follow what we publish in several months leaving only two months for you to implement in 3-4 months a new, untried process. No, we will not talk about this possibly creating a bottleneck and exacerbating the flow, that won't happen."

If there was a chargeback of effort to this kind of mindset and decisions, then a different political battle would emerge, but at least it would make these kinds of unilateral, unaccountable decisions in their aggregate form visible to the organization. A tremendous amount of this friction goes away if organizations had the capacity to automate many processes, but the cost hurdle is currently too high for most organizations. I'm hopeful LLM's in the hands of non-technical staff can mitigate that, but I'm likely dangerously naive there.

yourapostasy··on Unit tests as documentation
> ...increasingly moving toward Donald Knuth's literate style of programming.

I've been wishing for a long time that the industry would move towards this, but it is tough to get developers to write more than performative documentation that checks an agile sprint box, much less get product owners to allocate time test the documentation (throw someone unfamiliar with the code to do something small with it armed with only its documentation, like code another few necessary tests and document them, and correct the bumps in the consumption of the documentation). Even tougher to move towards the kind of Knuth'ian TeX'ish-quality and -sophistication documentation, which I consider necessary (though perhaps not sufficient) for taming increasing software complexity.

I hoped the kind of deep technical writing at large scales supported by Adobe Framemaker would make its way into open source alternatives like Scribus, but instead we're stuck with Markdown and Mermaid, which have their place but are painful when maintaining content over a long time, sprawling audience roles, and broad scopes. Unfortunate, since LLM's could support a quite rich technical writing and editing delivery sitting on top of a Framemaker-feature'ish document processing system oriented towards supporting literal programming.

yourapostasy··on Ask HN: How big of a problem is unstructured data for companies?
> What's the workflow like when these issues arise?

The workflow is by hand, though not quite all the way down, it damn near is.

> At what point in that workflow would a possible generative ai be of help?

Wherever the non-deterministic output of Generative AI still delivers benefits that exceed the transactional value of getting it wrong some X% of the time. Highly data and context dependent. That's a way of saying the "It Depends" cop-out, but I have a hunch we have to solve each industry segment individually, if not in finer gradations.

yourapostasy··on How big of a problem is unstructured data for companies?
> ...and the information in the document is old and out of date?

Not just old and out of date. Plain wrong, like referring to internal systems, processes or intranet URL's that are (often poorly done) duplications of enterprise standard services, processes or pages, and should be deprecated instead of incorporated into training data.

Or the unstructured data sits within structured data. Some development teams stick giant strings with their own internal, proprietary formatting into a database blob column because...I dunno, they thought it was expedient or whatever for Very Good [Undocumented] Reasons. Invariably, these fields hold all sorts of nastiness you either want to tease out, or do not want to exist at all because they expose what you wish they wouldn't in cleartext. And even though your company pays the vendors of these monstrosities (oh yes, my PTSD suppressed memories of multiple vendors committing these same terrible, no good, very bad software sins) three commas a year of "support", they adamantly refuse to share their precious proprietary format with you, much less give you the courtesy to update you when they change the byzantine format that gives the three body problem a run for the money for a deterministic solution.

If you make a Generative AI sort sense out of that, then you can practically segment the market by "how much before the customer starts to put away the blank check" and you'll make bank. Everyone I know of in this Data Loss Prevention (DLP) or adjacent spaces are deer in the headlights when the conversation turns towards these operational realities and what we might do with their solutions to put a dent in these multiverses of pain. No one has roadmaps talking of iterative pipelines, search outcome behavior analysis and automated training thereof, or really much beyond various degrees of pattern matching, whether that be regex's or LLM's.

yourapostasy··on Convert AWS console actions to reusable code with AWS Console-to-Code, now
Shades of AIX SMIT/smitty [1] command generation [2]. I wish more products would not just implement parity as much as possible between GUI, CLI and API, but also document them side-by-side ("To do this same REST API operation in the GUI, see <foo>, to do it in the CLI, see <bar>".

[1] https://www.ibm.com/docs/en/aix/7.3?topic=concepts-system-ma...

[2] https://www.ibm.com/docs/en/aix/7.3?topic=smit-command-gener...

yourapostasy··on COBOL has been “dead” for so long, my grandpa wrote about it
Sounds like they’re talking about running IBM Wazi on Red Hat OpenShift Virtualization. As far as I know, there isn’t a System z-on-a-container offering, like you install from a Helm Chart or comes to you from an OCI registry. If it is the IBM I know, it’s completely out of reach of most homelab’ers and hobbyists.

IBM Wazi As A Service is supposed to be more affordable than the self hosted version and the Z Development and Test Environment (ZD&T) offering. ZD&T is around $5000 USD for the cheapest personal edition, so maybe around $2500-3500 USD per year?

yourapostasy··on AGI is far from inevitable
Peter Watts in Blindsight [1] puts forth a strong argument that self-aware cognition as we understand it is not necessarily required for what we ascribe to "intelligent" behavior. Thomas Metzinger contributed a lot to Watt's musings in Blindsight.

Even today, large proportions of unsophisticated and uninformed members of our planet's human population (like various aboriginal tribal members still living a pre-technological lifestyle) when confronted with ChatGPT's Advanced Voice Option will likely readily say it passes the Turing Test. With the range of embedded data, they may well say ChatGPT is "more intelligent" than they are. However, a modern era person armed with ChapGPT on a robust device with unlimited power but nothing else likely will perish in short order trying to live off the land of those same aborigines, who possess far more intelligence for their contextual landscape.

If Metzinger and Watts are correct in their observations, then even if LLM's do not lead directly or indirectly to AGI, we can still get ferociously useful "intelligent" behaviors out of them, and be glad of it, even if it cannot (yet?) materially help us survive if we're dropped in the middle of the Amazon.

Personally in my loosely-held opinion, the authors' assertion that "the ability to observe, learn and gain new insight, is incredibly hard to replicate through AI on the scale that it occurs in the human brain" relies upon the foundational assumption that the process of "observe, learn and gain new insight" is based upon some mechanism other than the kind of encoding of data LLM's use, and I'm not familiar with any extant cognitive science research literature that conclusively shows that (citations welcome). For all we know, what we have with LLM's today is a necessary but not sufficient component supplying the "raw data" to a future system that produces the same kinds of insight, where variant timescales, emotions, experiences and so on bend the pure statistical token generation today. I'm baffled by the absolutism.

[1] https://rifters.com/real/Blindsight.htm#Notes

yourapostasy··on Qocker is a user-friendly Qt GUI application for managing Docker containers
I imagine they'll run into some of the same issues lazydocker [1] did. It seems non-trivial for seamless, comprehensive compatibility.

[1] https://github.com/jesseduffield/lazydocker/issues/4

yourapostasy··on Amazon tells employees to return to office five days a week
> ...a lot of the good Amazonians in mid-level management got poached...

I hear a lot of complaints of Amazon management going to other companies bringing the Amazon culture with them, and turning off the collaboration, communication and innovation spigot between and even within teams with their imported leadership style. Have others seen this first-hand and seen effective counter-measures they can report upon that deflect that energy towards more positive ends?

yourapostasy··on Small asteroid to hit Earth's atmosphere today
Thank you for your useful comment. It sent me down a brief rabbit hole of looking for a grid of amateur telescope owners cooperating for NEO object detection. I found eSTAR for the big boy telescopes [1] [2], but nothing for enthusiasts. I ran across NEO Surveyor that launches in September 2027, will run for 5 years, and is aiming for 90% coverage [3], so I suppose a grid of amateurs contributing telescope time into a software-driven grid and detection system won't be helpful?

[1] https://indico.cern.ch/event/423169/contributions/1890157/at...

[2] https://ui.adsabs.harvard.edu/abs/2002SPIE.4845...13S/abstra...

[3] https://www.jpl.nasa.gov/missions/near-earth-object-surveyor

yourapostasy··on Some thoughts on OpenSSH 9.8's PerSourcePenalties feature
> Many experts don't trust OpenBSD's security implementation...

Does this distrust go to the extent where confronted with a public Internet-facing, stock OpenBSD with a TLS-secured SSH connection, and a GSA hardened RHEL 8 [1] with a similar SSH configuration, they’ll pick the RHEL instance?

[1] https://github.com/GSA/ansible-os-rhel8

yourapostasy··on All of Earth's water in a single sphere (2019)
When I see these infographics I think of non-technical audiences like policy makers and politicians consuming the same information, and I avoid making these fine distinctions. In this case, I would not mention the water in the mantle at all.

In a separate, private graphic, I’d show the available water next to the number of 1 GW reactors, the pile of annual uranium mining output to feed those reactors, and annual calendars it would take to assemble all that to extract the water and dispose of the waste in a way that won’t harm our ecosystem further to express, “if you want this water in a form you colloquially understand, the species possibly can’t afford it”. In case some wise ass decides to bring up that mantle water. But that additional detail would even help technically inclined audiences reading the infographic.

yourapostasy··on Building data infrastructure that will last
> It's a little unfortunate I didn't hear more about it as I was writing it...

This is usually because most people find it faster and easier to ping the random access memory that is you while you are there instead of serially reading the documentation. I have found a non-linear correlation going in the wrong direction between the comprehensiveness of the documentation written, and the amount of requests for live assistance on the very matters covered in the documentation, when the requestee is the author of the documentation.

This is where an effectively-segmented Table Of Contents comes in handy. As the author, instead of rewarding unwanted organizational behavior by giving the requestor the spoon-fed answer they seek even if the requestee knows it off the top of their head, encourage the organization a "teach them to fish" posture by pointing them in the right direction by the chapter section.

"It will be somewhere in Chapter Foo" gets the requestors to familiarize themselves with the documentation structure at least in piecemeal fashion.

Junior engineers who make a habit of asking questions already addressed in documentation and don't take the "LMGTFY"-type responses as a hint they should search first, ask later, I start helping by asking, "what keywords did you try searching upon in the documentation", and start a conversation about why they used only such keywords. I'm excited by the RAG-family generative AI I am able to start loading documentation into that will let me ask, "what prompts did you try upon the documentation fine-tuned AI", as these AI's are powerfully effective with functionally fuzzy keyword searches. I'll fine-tune the documentation when it looks like various terminology is used over and over by searchers, but in general this approach works great to teach juniors how to search for information.

Engineers who don't quickly learn and keep asking for spoon-fed answers, enter an endless loop where they might occasionally help fine-tune the documentation, but otherwise are only pointed to the general area to search, and delegated to their colleagues who have learned to search for peer-led instruction. In over 20 years of applying this strategy, I've yet to run into an engineer who would not eventually learn, though if I did, they probably self select out of this field.

In a dog fooding fashion, when I ask someone for help, after my specific ask of them, I tell them what documentation I tried availing myself of first, what terms I searched upon, how I reasoned through to the point where I'm stuck, and what my theories/models are of what I think is happening. This helps rapidly prune the tree of responses, and 80% of the time the response is along the lines of, "oh, yeah the official documentation is out of date, use this technical note instead".

← PreviousPage 3 of 34Next →