I was watching Ben Krasnow making iron nitride permanent magnets and was struck by the fact that about half of the video was about making a magnetometer – a device for measuring and characterizing the magnet that he’d just made. This is really the difference between doing science and just messing around: if you want to test or improve on a procedure, you have to be able to measure how well it works.
When he puts his home-made magnet into the device, Ben finds out that he’s made a basically mediocre magnet, compared with samples out of his amply stocked magnet drawer. But that’s a great first data point, and more importantly, the magnetometer build gives him a way of gauging future improvements.
Of course there’s a time and a place for “good enough is good enough”, and you can easily spend more time building the measurement apparatus for a particular project than simply running the experiment, but that’s not science. Have you ever gone down the measurement rabbit hole, spending more time validating or characterizing the effect than you do on producing it in the first place?
And…?
They really should start a parallel site.
HackaDay for middle schoolers.
Their thesis is wrong.
Measurement is not ‘science’.
Science is a process, that often includes measurement, but can just be ‘cipherin’.
The key step is ‘publishing’ IMHO.
But it’s a feedback process, the whole thing has to work.
Open loop ‘science’ gets you sociology.
Yeah, well. Publishing is not science, either. Although publishing is necessary. For me the key is making verifiable predictions. With math. About measurements. An publishing the method.
If there are no predictions, you’re doing history.
If there’s no math, you are doing philosophy.
If there are no measurements, you are doing pure math (or metaphysics).
And if you are not publishing your methods, then you’re an snake oil seller.
My 2c.
I agree that the heart of the scientific method is making a prediction (“hypotheisis”) and testing to see of that hypothesis is true or not. Pass of fail, the objective is learning (generating “knowhow”).
I was surprised to see “publishing” listed as “ket part” of the scientific process. I happen to be in the private sector, where successful commercialization is the desired outcome, and knowhow from the scientific process is often treated as Confidential Information. I do not consider myself a snake oil salesman.
If you call a failed hyphothesis a “mistake”, and discovering a verifiable hypothesis a “success”, then our job as researchers is to make mistakes as fast as we can.
Publication is not a necessary part of science, but sharing knowledge is essential to make advancement possible.
If every scientist would keep all discoveries to himself others would need to reinvent the wheel over and over again.
So, publication may not be a necessary part of doing science, but it is absolutely necessary as an enabler for progress.
No. The key is being able to make falsifiable predictions.
If you make a hypothesis (a ‘prediction’) that only produces answers that verify your point, you’re not strengthening the case. You’re not proving anything. You need to ask a question that allows an answer to contradict you: You need to make a falsifiable prediction. Only then can you tell if the result actually supports (or refutes) your theory.
A long time a go, Hackaday used to have a specific filter which didn’t show any content with Arduino, as some people believed anything build with an Arduino was either ‘not a hack’ or it ‘could be done with a 555’. I guess you can filter these types of posts as well. However, you can also ask yourself, why do you keep reading Hackaday? (And more specifically, replying to them.)
The paragraph from the newsletter is quite interesting, and it doubles as a way to bring the newsletter to the attention of more readers. I sometimes listen to the podcast, and rarely read newsletters, but I’m glad they exist.
Back in the day when Arduino was new, all Arduino hacks were basically just second hand AVR hacks, where someone else had done the hard work and the author was just copy-pasting it in the form of a library, adding some trivial code on top to make it their project.
Come to think of it, that is still true to this day. It’s just that the standards have dropped a bunch to the point that actually reading the datasheet instead of relying on the Arduino programming reference and libraries is now considered a hack.
to be fair the atmega 168 datasheet is 288 pages long, kind of a bit of a read if you want to do most simple shit.
yea I was a kid in the 80’s and reading an encyclopedia was normal while your parents fell asleep watching “”the LawrenceWelk show” after the news then dinner, but still that’s still a bit of a read in order to rig a do dad to flip a output state every half a second
and especially in todays world, its just about impossible. Even the most impressive projects using the absolute cutting edge technology only skim the data sheet in a fact finding mission to accomplish a goal.
And not much else
That’s a short datasheet as far as MCUs go, and you don’t need to read all of it every time. There’s a table of contents and the list of registers for a reason.
Reading data sheets has just become a lost art, or have people become that bad at reading that 300 pages is too much? In there it’s like 30 pages worth of actual text to read and the rest is diagrams and tables. It’s perfectly possible to read through such a datasheet and understand most of it in a couple hours if you just start reading it.
That seems to be the problem. People try to treat the datasheet like a google search, only paying attention to specific answers to specific queries to find fast information. They’re reading the paper like they were committing a random burglary. Smash a window, grab some stuff, run. Trick is, if you’d been to the house before, you’d know where they keep the valuables.
Even a 1000 page datasheet isn’t actually that hard to read through, and once you do you’ll have a much better overview of how the thing works and what features are available.
With all due respect to the writes responsible for these newsletter article, this comment is what pops in my head every single time I read one of these.
Was really hoping this would be a run down of strategies to get better at arriving at accurate measurements, using different measurement tools etc
I have a lowly certificate in metrology. Measuring is absolutley a science. Why is there a thimble on the back of a micrometer? It gaurantees that us lowly humans are applying the same amount of torque to the measurement with repeatability. There are many things that can’t be measured directly but only by “comparison” measurements. Once you’ve entered the realm of metrology it is hard to look back. How accurate is your tape measure? how acccurate is your calipers? If you measure something with one tape measure and compare it to a second tape of the same brand from the same store at 3 meters you could easily be out by 3-5mm. I was working on a construction site and the foreman measured a piece os siding at 112 7/8 inches. He would yell it out to the siding cutter who would repeat the measurment back before cutting the part. After four incorrectly cut pieces I finally told them I wanted to the tape measures. Both were new and good quality. and one of them(the foremans) had a really wierd long inch at the 87 inch mark. making the tape wrong by almost a 1/2 inch after that mark. One of the parts of metrology is “Are you using the correct measurement method?”. Are you trying to be accurate to 5mm or 1mm or .100mm? This will help determine what you are going to use. Sometimes a scale or a ruler is good enough. Past 64ths of an inch the human eye has trouble differentiating the lines on the scale.
It does not. That is why there is a ratchet clutch in the micrometer that clicks and releases when you’ve done enough.
Thats called a thimble.
I apologize. Its called a ratchet stop.
Yes. The thimble is just the knurled cylinder at the end. Some may get the wrong idea that the size or shape of the knob is somehow supposed to limit the amount of torque you apply.
nice wall, you should build it on a border, its dense enough that no one could possibly pernitrate it let alone comprehend it
Oh go away and pernitrate something. It’s standard English and contains a lot of useful information.
u mad bro?
getting trolled must hurt so much XDDDDDD