|
PokeJoe posted:I'm sure someone has tried to smell pre toasted bread and it was Bad this mf said smell lmfao
|
#
?
Nov 25, 2023 03:26
|
|
|
|
| # ? May 3, 2024 12:24 |
|
|
someone made hardtack for the class when we were studying the age of
|
|
#
?
Nov 25, 2023 04:01
|
|
|
you can make a porridge out of it if its too old for actual direct eating
|
|
#
?
Nov 25, 2023 04:19
|
|
|
zwieback
|
|
#
?
Nov 26, 2023 00:27
|
|
|
Expo70 posted:https://www.youtube.com/watch?v=H_Ym9528awM Reading this thread from the beginning: I know this post is almost a year old but that is exactly how the keyboard switches on the Steel Series Apex Pro work. You can adjust key press resistance in the software. I got one on sale last year and I'm currently trying to go through a convoluted RMA process because multiple LEDs are failing on the keys. Originally got it so I could color coordinate buttons for different sim games (or really any game) for fun.
|
|
#
?
Nov 27, 2023 16:20
|
|
|
Apex Pro's Product Page posted:FULLY CUSTOMIZABLE & STUPIDLY FAST with OmniPoint 2.0 Adjustable HyperMagnetic switches hell yes i can inflict my poasts on the world twenty times faster
|
|
#
?
Nov 29, 2023 00:56
|
|
|
Manzoon posted:Reading this thread from the beginning: This is actually supremely cool. I'd love to try one.
|
|
#
?
Nov 29, 2023 00:57
|
|
|
I wrote an essay about the mertonon dealio. Need another coupla drafts before publishing for real but I'd like any comments, questions, abuse, etc https://github.com/howonlee/mertonon/blob/master/essays/estimates.md
|
|
#
?
Dec 4, 2023 17:03
|
|
|
while I still don’t understand Joe it works or how you would interact with it I liked the examples much more and this really piqued my interest. Once I’m back at work after paternity leave I would like to explored this more
|
|
#
?
Dec 4, 2023 18:59
|
|
|
Mertonon sounds like a drug you use to wean off heroin.
|
|
#
?
Dec 4, 2023 21:42
|
|
|
how do you tell mertonon that bob cannot continue because dobbs is dead
|
|
#
?
Dec 4, 2023 21:50
|
|
|
just delete the node, its just backprop the next iteration will take care of it
|
|
#
?
Dec 4, 2023 21:50
|
|
|
bob dobbs is dead posted:just delete the node, its just backprop the next iteration will take care of it surprisingly reasonable answer
|
|
#
?
Dec 4, 2023 21:53
|
|
|
there's a fkey cascade delete for the weights with respect to the nodes, which is annoying the time i tested it but would be fixed by a deec undo button, which is pretty up there in the priority list. there's also a lotta ux splicing i wanna be able to do making the priority list get created by mertonon itself, however, is before that. lol bob dobbs is dead fucked around with this message at 22:16 on Dec 4, 2023 |
|
#
?
Dec 4, 2023 22:13
|
|
|
bob dobbs is dead posted:there's a fkey cascade delete 
|
|
#
?
Dec 4, 2023 23:17
|
|
|
bob dobbs is dead posted:I wrote an essay about the mertonon dealio. Need another coupla drafts before publishing for real but I'd like any comments, questions, abuse, etc That's a lot clearer in terms of intent and how it should work/be used. Does it have any sort of concept of risk/uncertainty or ordering in there? eg. the potential for the user validation tanking any backend implementation and how to hedge your bets on that sort of ordeal? One of the known challenges in decision-making is that the theoretical ideals behind bayesian analysis and whatnot start to be really hard to deal with when time pressures and uncertainty come into play, and I figure software of this kind would equally have challenges. OTOH I fully understand that this brings things close to "pray tell mr. babbage," territory
|
|
#
?
Dec 5, 2023 01:20
|
|
|
uncertainty is for like, in a year or two. along with a blitheringly infinite list of features neural net is ultimately peepkind's most serious attack at the pray-tell-mr-babbage problem. it isnt sufficient, but im comparing to status quo not perfection bob dobbs is dead fucked around with this message at 01:33 on Dec 5, 2023 |
|
#
?
Dec 5, 2023 01:28
|
|
|
https://www.protoarc.com/en-gb/blogs/news/trackball-vs-optical-mice
|
|
#
?
Dec 8, 2023 13:58
|
|
|
very nice. I'd like to see a comparison of thumb trackballs to finger trackballs too (thumb trackballs suck!! eaugh)
|
|
#
?
Dec 8, 2023 14:51
|
|
|
lol that is the exact way I would expect a trackball company to sell to their target market
|
|
#
?
Dec 8, 2023 17:54
|
|
|
Not a paper this time around! Instead I've read this transcript from an NPR interview titled Why do doctors still use pagers?, from Planet Money. The interview talks to Mary Mercer, an emergency room doctor in San Francisco, and Christopher Peabody, another ER doctor at the same hospital, and who have run a pilot project to try and replace pagers. They begin by covering what a pager is, exactly, because younger folks may have never encountered them: "it's a little black box with a little gray screen in it, and it emits a sound to alert you that you have a message." Someone calls your pager's number, which is converted to a message that goes over radio waves, and when the pager goes off, a number pops up on your pager screen, and you race to find a phone to call that number back. They used to be everywhere, particularly in the days before cellphones, but have since disappeared from pretty much everywhere except hospitals, specifically for doctors. Mary Mercer was wondering why exactly is it that doctors still use pagers, why they are still putting up with them. A few years back, Mercer was in charge of a project to increase ER efficiency, and along with Peabody, she decided to see what could be done to replace them. For an on-call doctor, the pager is a bit opaque. In San Francisco, the transcript says, you may be called for a patient with a rash, who has been in a car crash, or even in a sea lion attack. But all a doctor gets is a phone number on the pager, so there's no way to know if you're gonna be asked to write an order for some medication or to assist someone with a half-amputated limb. They have to call the phone number back, and get the story. Apps were being made available for medical personnel that do end-to-end encryption and allow read receipts while preserving patient privacy, but were not used. On the other hand, pagers are considered to be more reliable: quote:They have fewer dead spots. They run on entirely different networks than cellphones. If there's a big emergency and everyone tries to use their phones all at once, pager networks are less likely to be completely overwhelmed. Mercer and Peabody still felt that the benefits of new apps would outweigh the negatives, and decided to prove it with a pilot program. For the pilot program, they enlisted one of these aforementioned apps to try it out, while still having doctors carry pagers as well, so they could see for themselves how it goes. At launch, everyone seemed pretty enthusiastic, and the new stuff felt like a no-brainer: doctors could send each other photos from across departments, such as when dealing with open fractures: quote:So I took a secure photo and texted it to the orthopedic surgeon and just put the room the patient was in, no other information, and I got a response immediately that said, be right there. [...] So at that moment I was like, this thing's genius. We're definitely going to be implementing this technology system wide. (the interviewers do spend a bit going like "really, the patient high-fived?" and Cristopher going "they were thrilled!") They did also find some issues with the system. Sometimes, doctors on-call would go to sleep, and while they had a strong, "Pavlovian" response to their pagers, some had their phone settings put such that the text messages wouldn't wake them up and they still had to be paged. In fact, after a few weeks, Mary and Christopher started finding out that adoption was going down: quote:Toph was working in the ER. He had just texted one of the neurology residents, one of the young doctors on call. The resident walked into the ER wearing that, you know, Rambo belt of five or six pagers across their chest. To explain how that unfolded, they introduce something called the RHIP, which stands for Risk, Habit, Identity, and Power:
In the hospital where they ran the experiment, specific dynamics were in play: quote:Among the doctors at Mary and Toph's hospital, it was the residents who had the least power. They were the most junior doctors on their teams. And in specialties like orthopedics and neurology, one of their duties was to be on call, to respond to all those messages and photos and everything else that the ER doctors could now send instead of paging. They interview Abhinav Janghala, an orthopedics resident during the pilot program. He was one of those with a "Rambo" belt with a bunch of different pagers. He started up optimisitc, but ended up hating it. He mentions how people started texting him about minor things that wouldn't have been a page before—the threshold had been reduced. He'd find himself being pinged during surgery for minor questions, and even in group texts about patients he wasn't treating: quote:Let's say you're getting - normally, you would get 10 to 20 consults. Now you're getting 25 to 30 texts overnight, and only out of the 30 people, maybe 20 of them you actually needed to know about. [...] I started to see the pager as a way where it puts the person receiving the page in control of the communication. That's the P in RHIP. By turning their phones off, doctors had the ability to control what they could be disrupted about, knowing that the pager would be used for really important stuff. Even when they were paged, they could decide when to call someone back. This went away with the app-based workflows, and would let the sender know if the on-call person had seen the message. This increased the demands on their attention. During the experiment, residents started just flat out ignoring the app. The ER doctors only had the power to disrupt residents if the residents allowed it to happen: quote:At the end of the pilot program, when the three months were up, Mary and Toph got back all the data, and they were surprised to learn that their project had failed. Their big goal, you know, to use this efficient new communication technology to get their patients treated faster in the emergency room - none of that happened. [...] There was no statistically significant change in how long patients were spending in the ER. Mercer and Peabody looked at the data, and it's not even that the app was or wasn't effective—it was just not being used: "By the end of the pilot program, traffic on the app had gone down 50%." The interviewers at NPR already do a perfect conclusion so I'll quote it directly: quote:For Mary and Toph, the moral of the story is, no matter how great a technology might be, how you implement that technology matters because even the shiniest new technology comes with some drawbacks. It might introduce new risks or disrupt old habits or call people's identities into question or maybe change the power dynamic in an organization. Now, any of these drawbacks can be overcome, so long as you do a good enough job of rolling out the new technology. Do what you can to foresee the drawbacks, and figure out ways to address them ahead of time. Overall, I found the RHIP framework to be really interesting as a way to frame the whole thing.
|
|
#
?
Dec 11, 2023 02:55
|
|
|
MononcQc posted:Power: This was the biggest, most disruptive one. Phones inverted the power dynamics, and this requires expanding a bit. this is 99% the reason they hate it. and maybe 1% the thing about them feeling more doctorly if they have a pager. pagers fail all the loving time and doctors ignore pages all the loving time, but theres absolutely no accountability because they're archaic one way horseshit. You fire the page into the carrier's system and they either accept or reject it based on the validity of the number and thats it. No delivery confirmation or read confirmations. If the page fails nobody knows until the doctor doesnt call back and then theres almost never a followup to figure out what happened. And even when there is a followup to figure out why the doctor didn't respond to the page its either: "oh i dont remember what happened" or so many loving times "oh when the page was sent i was in a known deadzone/had it off" secure messaging apps, on the other hand, track the full lifecycle of a message: acceptance by the service -> delivery to the doctors device -> read by the doctor -> (optional) acknowledgement in the form of a yes/no/ok/other response. This allows you to have high quality SLAs from which you can escalate contact and gives you a paper trail you can use to figure out what went wrong after the fact. in some cases secure messaging can also completely avoid the whole callback process when you can send the full context of the page within the message including supporting documentation that they'd otherwise have to ask for over the phone when returning the page. Probably the only legitimate technical issue is it not being setup properly on the phone and/or the doctor disabling critical alerts after setup. in ios (and I assume android), approved apps can offer critical alerts which bypass all os level notification settings. They are intended specifically for these kinds of pager replacements and absolutely will wake you up unless you have disabled them. We've had pretty wide success with secure messaging as a pager replacement to the point that we barely do actual paging anymore. We will pay for our customer's doctors to use the app, its that much better from an operations standpoint.
|
|
#
?
Dec 11, 2023 03:23
|
|
|
The other thing I can imagine is having the ability to have both the low and high urgency comms as settings, which they appear not to have had in their case. But it's frequently what we do even at work with Slack as a low-urgency mechanism and PagerDuty for high urgency alerts, and we just have put a lot more friction for non-automated manual pages to make sure when someone escalates something, there's a good reason for it. To make it part of RHIP, this addresses the ability to still be in control of our own time when on-call, and control some of the signal/noise ratio.
|
|
#
?
Dec 11, 2023 03:31
|
|
|
all the secure messaging apps i've used have had priority as a feature. Most of them are intended to be used as a HIPAA compliant general chat app with paging/critical alerts being just one feature of the app. A resident not understanding how to set different notifications for a normal/group chat vs a page is not all that surprising, and them not reaching out for support and instead ignoring everything is less surprising. Both because they dont want help and because hospital IT ranges from horrific to understaffed. That said what they really want is for you to just SMS it to their phone cause thats how they've always done it, "why do we need a special app? Just text it to me, i dont understand why this is so hard!"
|
|
#
?
Dec 11, 2023 03:44
|
|
|
my org still uses old school pagers that take a AA battery
|
|
#
?
Dec 11, 2023 03:45
|
|
|
also wrt the doc who got included in the chat by mistake, its another one of those bad doctor-centric designs established by hospitals. Imagine you're working on care for a patient. You startup a chat with the other staff members you know are working on them. This works fine until one of the staff goes off that patient (for any of 1000 reasons). Then they need to be manually removed. Alternatively, 4 of the 5 people are also on a different patient so they just use the same chat for the new patient, inconveniencing the 5th person and at the same time potentially mixing patient data. What should happen is you have a chat based on a patient or record within the EHR. Everyone assigned to the patient/record is automatically added or removed from the chat and can see all the contextual data they need thats tied to the record and the patient's history. The chat content itself can be automatically added to the record as well. This would require EHRs and providers be better in many ways so it doesnt happen.
|
|
#
?
Dec 11, 2023 03:49
|
|
|
the one good thing about pages is every now and then some litterrall child realizes they can intercept them because they're sent in the clear and then they get super excited because they think they've discovered a new, critical security vulnerability. lol. lmao, even.
|
|
#
?
Dec 11, 2023 04:00
|
|
|
Shaggar posted:the one good thing about pages is every now and then some litterrall child realizes they can intercept them because they're sent in the clear and then they get super excited because they think they've discovered a new, critical security vulnerability. bring back AMPS analog mobile where you could just pick it up any random calls with a scanner
|
|
#
?
Dec 11, 2023 04:08
|
|
|
or 900mhz unencrypted home phone cordless handsets where you can just keep pressing the channel button until you connect to a neighbor on a call.
Shaggar fucked around with this message at 04:13 on Dec 11, 2023 |
|
#
?
Dec 11, 2023 04:11
|
|
|
I actually uploaded the essay about estimates to the website, so here's the actual published thing I guess https://mertonon.com/blog_estimation.html next essay: pure formal theory of value bob dobbs is dead fucked around with this message at 19:03 on Dec 14, 2023 |
|
#
?
Dec 14, 2023 18:56
|
|
|
I've read this paper recently, though I haven't taken notes yet and I will do so in a follow-up post: https://aiche.onlinelibrary.wiley.com/doi/10.1002/prs.12579 In that paper, they go over a lot of incident reports where people criticized auditing with counterfactual reasoning (what should or could have been done), and rather than using it to work on specific incidents, they do an analysis of like 30 years of industry reports doing this (~44 reports total) to build up what the sort of ideal auditing process ought to work like in the minds of experts conducting investigations so that they can then figure out the common expected failure modes and suggest practical improvements. It ends up suggesting that the auditing process tends to redirect focus to surface issues and documentation, reinforce positive views of safety by failing silently, inadvertently reconceptualizes hazards into less concerning conditions, and creates unwarranted confidence in risk management. They call them “persuasive over conclusive” and say that if orgs have auditing, they need to put the processes in place make sure they work as work as “tools of problem-solving” instead of “the machinery of comfort-seeking.” strong “who audits the auditors” vibe here, but things have been a bit busy recently for me to go and just take a good while at noting up new papers so yeah, maybe this week or next.
|
|
#
?
Feb 3, 2024 18:56
|
|
|
It had been a while since I wrote fresh notes from a fresh paper, so here's the now full post on How audits fail according to accident investigations: A counterfactual logic analysis, published on January 26 2024 by Ben Hutchinson, Sidney Dekker, and Andrew Rae. This is a super interesting paper to me, but to know why, you first have to know about something called _counterfactual reasoning_ in the context of incident investigations. We'll get back to counterfactual reasoning because it's real neat, but the quick version of it is that it's what happens when you think people should have done something else, or that if they had done things differently, the bad events would have been avoided. This is generally a no-no for folks in Resilience Engineering (again, we'll cover this soon), but what the authors did here was use the counterfactual reasoning that was written out in incident investigation reports which also mentioned safety audits (whether internal or external), and then they used that as an example of what the investigators think auditing should be doing. By comparing these, they can also figure out how audits fall short of the expectations. Very cool trick. But let's dig a bit into the background. The first stop the authors take is around auditing and its role around incidents, which also exposes their views behind incident causation: quote:Disasters are “essentially organized events” that require a long incubation period of discrepant gaps in organizational, social, and psychological patterns, as well as prolonged neglect or discounting of potential signs of danger. These two elements allow accident precursors to accumulate unnoticed. Theoretically, auditing can help organizations detect accident precursors, but auditing can also provide a false assurance of safety, inadvertently allaying subjective concerns for issues where alleviating concern was not objectively warranted. While ostensibly safety audits can sensitize—or desensitize—organizations to signals of danger, little empirical evidence exists. Since they're using incident reports, they also take the time to define the two great ends of the spectrum for incident investigators:
quote:Accident analysis, from both positivist and constructivist perspectives, involves selection and prioritization. The construction of a preferred account of an accident necessarily silences and underemphasizes other competing possibilities. This process can be informed by an idealized model of what could or should have happened. In other words, investigations construct, explicitly or implicitly, what did or did not happen using a process of counterfactual reasoning against an idealized model. What this means, in short, is that you can't cover all the things, and what you choose to cover is often informed based on what a reasonable or ideal situation would have been; counterfactual reasoning helps people create various scenarios and evaluate the real world events against them. Whether this is useful or not is still up for debate in the safety science community, which ranges from "dismissive" to "sees value but is cautious", and defines three camps:
This idealized picture is why counterfactuals are important here: they can be used to define it. They did so by looking at the last 30 or so years of incident reports (14,576 reports) and narrowed it down to full safety investigations with mentions of auditing failures, with counterfactuals in them (44 reports), and then did some deep qualitative analysis on them. They managed to narrow it down in one big table that covered most of it (if you want to know more about the method or the details, I do encourage you to read the article, it's open access):  Their 4 big categories are failing to understand, act, manage, or focusing based on the audit results. Each category has multiple examples cited. They found multiple insights regarding auditing as well. The first and most numerous one is that audits tend to direct focus on documents and surface issues, rather than on elements that, in hindsight, were important, which results in surface compliance: quote:an “ideal” audit should delve beyond the presence of documentation and system artifacts to verify practical system functioning. [...] [Audits] can over-prioritize the collection of documents and the presence of artifacts at the expense of probing system functionality. Said differently, a gap exists between how systems are expected to function versus how systems do function. That gap is argued to drive decoupling, further widening expectation versus reality while providing a false veneer of safety. A second one is that audits reinforce that positive view of safety by failing silently: quote:an “ideal” audit should not only pinpoint crucial shortcomings and alert organizations to safety risks but also distinctly signal if the audit program is underperforming – “failing loudly”. If there's no audit, there's generally no noise made; if there's an inadequate audit, it can promise good results it is not able to deliver: quote:audits inadvertently allayed concerns, a type of probative blindness where subjective confidence in safety was disconnected from objective risk. At a pathological level, audits may be “inadequate to provide early warning of process safety risk”. A third one is that audits tend to downplay hazards and minimize them: quote:the “ideal” audit is expected to identify and interpret hazards and issues in a way that is relevant and sensitizing to protecting health and safety and/or accurately estimate the effectiveness of risk controls. They mention that sometimes audit do something they call "interpretive failure", where despite aiming to uncover issues, they can let them incubate for a longer time by either not finding them or by over-estimating safety arrangements. This can widen the gap between their imagined and actual risk control. This leads to the fourth issue, unwarranted confidence: quote:Hutchinson et al. previously argued that “some plans can be symbolically powerful yet functionally weak.” Said differently, some safety artifacts can exert a sufficient influence over the subjective beliefs of people, like how “safe” people believe work to be, even though that artifact may provide little direct functional influence over hazardous work. Basically, the audit plays the role of reassuring the stakeholders that health and safety can be properly governed even if it might not be so. The authors state, however, that while counterfactual reasoning can be useful to uncover countermeasure against these failures, they still do not explain why people did what they did (which as far as I can tell is useful to actually find effective countermeasures). They also add that while the initial data set was large, the final number of adequate samples was somewhat limited. They conclude that organizations may want to clarify their audit expectations, and implement mechanisms to test and monitor whether their auditing does achieve these expectations, particularly to prevent silent failures. They warn that audits may not act as robust indicators of weak or early signals, and ask that practitioners ask whether they work as tools for problem-solving or as a machinery of comfort-seeking.
|
|
#
?
Feb 5, 2024 23:22
|
|
|
not too surprising an outcome but your synopsis was a great read, will have to grab and read that paper tomorrow
|
|
#
?
Feb 6, 2024 00:40
|
|
|
MononcQc posted:Mercer and Peabody looked at the data, and it's not even that the app was or wasn't effective—it was just not being used: "By the end of the pilot program, traffic on the app had gone down 50%." The interviewers at NPR already do a perfect conclusion so I'll quote it directly:
|
|
#
?
Feb 8, 2024 10:27
|
|
|
good stuff as always, MonoQc not the super high quality content we've had but i wanted to leave my thoughts on avp: not my video: https://twitter.com/thekitze/status/1754189182992253156 so the big one isn't just reality avoidance: users have to look exactly where their inputs are. this is a step down in terms of direct usability from tablets, since differential gaze is a huge part of task order action planning where gaze goes to the next thing during a stream of inputs, often to transform a source from one location to another, or to refer to something else while working this is basically going to create a kind of tunnel vision the scope of possible surfaces is impressive in a 3D space, but this is worse than tablets, which were worse than keyboards i'm kind of wondering what the long term cognitive harm of this is going to be i think at least, that every time the input-precision and input-rate of the primary access medium of the internet has changed, i think there has been changes in the consumption habits of users which translate into direct cognitive changes over time we went from a very creative internet to a less creative one we already deal with this as a chronic problem in online spaces, and it seriously damaged how the internet culturally functioned which social media leaned into, hard i wonder if its not just the media addiction that's going to screw people up, but the power dynamic where your input is far less rewarding than just consuming output relative to action Its not just reality avoidance: its that it funnels users into selection and consumption over input and creation. the feel is its gonna be a collar for your brain, basically -- strung together out of mediocre tablet apps and a single mid desktop experience if you're even more rich than a little bit rich someone with a better understanding can probably do a better job at this than me, but i hope it gets the ball rolling i wonder if this is just what getting old feels like Expo70 fucked around with this message at 09:39 on Feb 11, 2024 |
|
#
?
Feb 11, 2024 09:11
|
|
|
gently caress, I never thought on how much I hate typing on a screen and how it makes me not do things
|
|
#
?
Feb 11, 2024 09:23
|
|
|
zokie posted:gently caress, I never thought on how much I hate typing on a screen and how it makes me not do things touchscreen devices are televisions that let you shitpost
|
|
#
?
Feb 11, 2024 15:38
|
|
|
leper khan posted:touchscreen devices are televisions that let you shitpost e: genuinely makes me wonder if a platform would benefit from being deliberately inaccessible to users so hostility of access was a deliberate feature to combat eternal september by not providing the path of least resistance e: the issue i realize is while it would prevent some people from entering, it would rapidly become an elitest shithole because everybody would feel part of a secret special club because if there's one thing insecure nerds love doing its using some arbitrary thing they can do to differentiate themselves from those they feel threatened by Expo70 fucked around with this message at 17:51 on Feb 11, 2024 |
|
#
?
Feb 11, 2024 17:40
|
|
|
*looks around* it’s been done but seriously yeah intentional barrier to entry stuff is a real common pattern in both communities or general interest groups and it either never takes off or dies when the five greybeards can’t recruit any youths to put up with their bullshit to play with the clubs model trains
|
|
#
?
Feb 11, 2024 18:14
|
|
|
|
| # ? May 3, 2024 12:24 |
|
|
also seen in products; whole bunch of iPod alternatives where intentionally difficult
|
|
#
?
Feb 11, 2024 18:16
|
|