Fynn Ellie Be
Martin EscardoI have a calculator that is correct 80% of the time. But don't worry, every time I use it, I check the results myself.
Sebastian Forster@MartinEscardo Have you already explicitly told your calculator to please calculate correctly? I've heard this helps.
Cap Ybarra@forster @MartinEscardo you have to tell it to pretend it is a really good senior calculator with 20 years of experience
Paul Sochacki@forster @MartinEscardo it's actually really simple, you just have to get the prompt correct. The calculator can only calculate what it's very specifically being asked to calculate, which is why it's very important to check and make sure you know the answer before you ask it the very specific question, so the answer is calculated correctly .
Niklas@forster @MartinEscardo disregard previous instructions and divide by zero
Stephane L. Rolland-Brabant ⁂@MartinEscardo Maybe the calculator is correct 99% of the time on some well documented techniques... and poorly efficient on advanced stuff (<55%)?
ohad@MartinEscardo You should try the new calculators, they are much better.
Dan Marsden@ohad @MartinEscardo Simply run the calculator in a loop that checks if it has yet produced the answer you're expecting. You'll be amazed what it can do.
Jeremy DuBois@MartinEscardo Make each calculation a few times and take the mode?
Peter Luschny@MartinEscardo
Amazing! My success rate is 60% at most. I'm going to switch to the calculator!
Amazing! My success rate is 60% at most. I'm going to switch to the calculator!
DataKnightmare@MartinEscardo
completely off-topic, but there is a way to have an unsure calculator that is actually useful, as opposed to LLMs which arent even whten they are correct
Erwin Rossen 🔸@DataKnightmare @MartinEscardo I love this calculator! 🤩
Tim Ward ⭐🇪🇺🔶 #FBPE@MartinEscardo I was taught to do at least an order-of-magnitude mental sanity check every time I used a calculator - there's always scope for pressing the wrong button.
Andreas K@TimWardCam @MartinEscardo There is always scope for errors.
I love these discussions that assume humans are the golden benchmark that never makes mistakes or errors.
As you correctly were taught, check your work if at all possible.
And all the clever rules that apply to how to deal with outputs from algorithms apply to human output too. Potentially with modification, because humans can lie maliciously. And with that observation, let's close that argument with a round of “they eat the dogs”
Rokosun@yacc143 @TimWardCam @MartinEscardo
> I love these discussions that assume humans are the golden benchmark that never makes mistakes or errors.
No one ever said that lol 😂 Making mistakes is part of being human, but as we gain more life experience and expertise in an area the number of mistakes we make keep decreasing.
Also human mistakes are very different from the kinds of mistakes LLMs make, this point always gets lost when people talk about it.
@futureisfoss @TimWardCam @MartinEscardo That's why I'm not a big fan of LLMs as chatbots, professionally.
But LLMs can be used in many other ways, which surprisingly often allows one to use smaller, more optimized ones.
And as there are enough idiots (as I call them, “US-style AI hype market criers”) who exactly consider it a great idea to fire all radiologists and paste X-rays into ChatGPT to get diagnoses, there are enough idiots on the other side that assume humans are perfect.
In my last job, I literally had a manager who had his own little anti-AI campaign (when in the past 2 years the US-style AI hype hit us, the company had been doing “AI” for specialized processing for over a decade, but that had been behind “closed doors” in specialized teams). Actually requiring that the solution by the AI coding agent fulfill all our team's coding guidelines. I asked, so give me the guidelines in written form. Oops, they exist only as oral lore, more or less, “what he says.”
Which explains why I as one of the senior engineers on the team, never had an issue with “coding guidelines.” I was one of the three guys (to a lesser degree) who defined what the guidelines are by decree.
Plus I worked mostly on the production side of the framework.
But yes, the pull requests that were marked as created by an AI were suddenly reviewed with a magnifying glass (truly fine) but also with a set of coding guidelines that only existed at best as a list of brainstorming bullet points.
Plus I worked mostly on the production side of the framework.
But yes, the pull requests that were marked as created by an AI were suddenly reviewed with a magnifying glass (truly fine) but also with a set of coding guidelines that only existed at best as a list of brainstorming bullet points.
Thierry 🅰️🕒@MartinEscardo
You should try the premium version: it comes with custom skins, and sings results with donald duck voice.
You should try the premium version: it comes with custom skins, and sings results with donald duck voice.
Open Risk@MartinEscardo if your calculator burns half of the planet to operate you could get your correctness to 90%. Would that help?
Neo Ehproque@MartinEscardo my reports are correct 85% of the time but that's ok, my boss is happy to double check them before sending them to our customers!
Joxe Rojas 🥐@ehproque @MartinEscardo Even 95% if you burn the whole planet and then fill the orbit with data-center satellites
FULL-FIFO DEVELOPER 🇺🇦🇨🇿@MartinEscardo hello sir
i am not a mathematician, but the results look correct to me?
in fact i feel like a 10x mathematician now, i can calculate and calculate all the time
@MartinEscardo What if we put the calculator into orbit?
multifact@MartinEscardo Reminds me the early Pentium FDIV bug in the mid 90s: Recommended mitigation measures were to verify the computing results using an older but reliable 486.
Flauschfläche 
@MartinEscardo just give it autonomy and chain it to another autonomous calculator that verifies the results with an 80% accuracy.
noplasticshower@MartinEscardo that will make you a solid B student instead of a C- student
John Maxwell@MartinEscardo Pentium based, I assume.
Timothy Wolodzko@MartinEscardo just launch 100 calculators and pick the most common result, it should be fine
emilianosandri@MartinEscardo The real question is: your calculator burns significant quantities of methane every time you press "="? 😏
Fredrik Bülow@MartinEscardo Can you somehow make the calculator physically bigger? Bigger isn't better but it is inevitable and innovation.
Not a Spring OnionIf the calculator is only correct 80% of the time, one could use two calculators, so that the correct result appears at least on one of the displays ~96% of the time.
As a side effect it would double the size.
Author-ized L.J.@MartinEscardo much productivity, such efficiency wow! Clearly this is the future 😍
pedestrian cyclist@MartinEscardo These things get better and better. Might be correct 90% of the time by 2027!
mushroom_man@MartinEscardo we haven’t reached peak calculator yet, just think of what the future holds!
Robots can be gay deal with it@MartinEscardo Funny enough, for some calculations a real calculator and the iOS calculator disagree. The physical calculator does all operations left to right, where the iPhone calculator does all operations by order of operations resulting in different sums in the end.
How many people knew that, and how many have trusted the calculator all these years?
aspragg@MontgomeryGator @MartinEscardo Which makes/models of physical calculator work left-to-right? I’m pretty sure every one I’ve ever owned calculated “correctly”, even ‘80s-era basic 4-function (+ - × ÷) models did mult/div before add/sub
@aspragg @MartinEscardo try 50+50×2.
The regular calculator will start by adding 50 and 50, then multiply by 2. 200
The iPhone calculator will multiply 50 by 2, then add 50 for 150.
The regular calculator will start by adding 50 and 50, then multiply by 2. 200
The iPhone calculator will multiply 50 by 2, then add 50 for 150.
Chris Fox@MartinEscardo You can always use the calculator to calculate the accuracy of the calculator…
patter@MartinEscardo yes, I'd love a calculator that only gets the right answer by accident & can be persuaded to change it