Why Tap a Wheel of Cheese?
cheeseprofessor.com
cheeseprofessor.com
It is interesting how important it seems to us that jobs like this remain done in the traditional way. For all their expertise, I am sure a technical solution would also easily be able to detect what they are looking for: voids within the cheese, or lack of uniform density. This does not seem to be a case where the human expertise and artistry is actually important to the final product, besides the feeling of tradition.
Perhaps the best argument for keeping traditional jobs like this is that, even if that exact job could be done by machine, replacing these humans with machines would be the start of a short process that would end up with indistrially-produced bad cheese.
It sucks.
I used to get paid to do OCR on tax forms. I used n gram models, BERT,etc and it took weeks to get the forms right. Now you can do it in seconds with an api.
I wouldn’t say the profession got destroyed. I just work with different tools now. Instead of running ngrams, I’m testing different apis or embedding models.
The old job doesn’t exist anymore, in many ways I am the classic example of losing your job to AI-but I wouldn’t say it destroyed the profession. People just use different tools today to get intelligence from PDFs.
I assume the people doing that work might have been lighting lamps in the past.
In the same vein, you doing OCR and now doing other work is not a profession change. You're still a computer scientist doing computer engineering work just on a different problem set.
Lamp lighter to cheese tapper is a profession change. A lamp lighter with bad hearing or attention to detail is not a problem per se. Tapping cheese to see if its ripe (and all the other skills that come with cheese making) will potentially be different enough.
Now the problem comes in, when lamp lighter is supposed to start doing OCR engineering work. Unless the lamp lighter was horribly over qualified, most lamp lighters will not be able to do OCR engineering or any computer scientist style engineering work at all.
Professions do get destroyed all the time. Hopefully while a new profession is born that has similar enough skill level requirements as the previous ones, just in different form. Like a lamp lighter might do well in an Amazon warehouse but not as an AWS software engineer.
It's somewhat still held by the same people that somewhat-traditionally did the job. In some places, it used to be job of the night watch constable to light the kerosene lamps. Police now days just report a broken traffic light.
some professions spawn from viable hobbies and some hobbies spawn from unviable professions.
“The particularity of this profession to me is that it’s like it was 2 years ago, and it’s a skill that’s handed down from dev to dev. You go around with the most expert, most experienced BERTitori, and you watch and listen to them, and slowly they start to give you the keyboard. You try with them next to you, piano piano, and gradually, you begin to do more on your own,” explained Stocchi. “It’s a big responsibility, you have to be really capable of doing it, you can’t damage the forms.”
Our tendency to 'progress' doesn't have to be the case, we all could collectively decide to hold ourselves here.
Also, I don't think that all our technology has always been good for us either. But we are blind to the downsides mostly.
Engineering, the cause of, and solution to, all of life’s problems.
It does't have to be, of course, but the people with enough capital to set up automation tend to care more about money than cheese.
The cheap bad cheese wouldn't last a second in the market if we didn't all rush to buy it, accepting it's failings, but rejoicing in how close it is to the "real" thing and, of course, deriding the original as being "so expensive"
On the other side you have those wealthy enough to enjoy as much as the want out of it at its current price. Those would be the losers if quality were to deteriorate.
Edit: I can point to cases of that happening
Ferrari, who couldn't turn a profit so sold to another car manufacturer that created a "profitable" version that, quite frankly, is a shadow on its former glory
It has been fun watching Starbucks' various attempts at cracking into the Italian market.
The US government operates several special tasting departments.
It probably also helps that they are downing a demitasse in a minute or two instead of a giant venti size beast that lasts for an hour.
I have a Flair Espresso machine and get beans from a nearby roaster. I dunno. It seems decent. I don’t consider myself a snob, but I put some effort in.
Maybe Italians would be able to tase the defects that I add. But, they are the people who popularized Moka pots so they can’t really be so perfect.
I pretty much drink solely espresso, but will sometimes have cappuccino for breakfast (which is the only time of day that Italy accepts!)
Ultimately, it’s about what you like. It doesn’t really matter what Italy thinks. But as with anything, getting good ingredients and preparing them well makes a difference.
I wouldn’t consider myself a ‘coffee snob’, more a coffee enthusiast :D. Good coffee from ethical sources is just better. People talk of coffee jitters. I don’t get any. And the variety of flavours is awesome.
btw, anyone in the UK: I really like Coffee Bean Shop [1]. It’s a small family owned coffee roaster, with a good variety of flavours. And on every delivery there’s a hand written note from the packer. Usually talking about the weather, but can be a variety of things. It’s very cute in our internet age and genuinely brings a smile to my face every time I open the package. The coffees are excellent too.
I'm unfamiliar with the regs on this, but can the duties of the battitore be written into the standard for the cheese?
So what's written in the _disciplinare_ is the expectation that the wheel is flawless and that the experts will validate it at certain times. The details of how the validation is performed is vague, it mentions using the hammer but doesn't specify stuff like "it sounds X".
The watch on my arm is (mostly) handmade and worth a fortune, there are plenty of other cheaper watches that are arguably better due to them being made by robots.
Same principle, I think.
In the world of watches, this logic extends far beyond "handmade vs made by robots". Quartz, for example, is better than automatic & mechanical in every single quantitative metric, yet perceived as lower value due to the lack of craftsmanship.
I embrace life’s imperfections :D
Out of pure curiosity, how would an industrial process engineer approach this problem, de novo?
> A study on structural quality control of Swiss‐type cheese with ultrasound is presented. We used a longitudinal mode pulse‐echo setup using 1–2MHz ultrasonic frequencies to detect cheese‐eyes and ripening induced cracks. Results show that the ultrasonic method posses good potential to monitor the cheese structure during the ripening process. Preliminary results indicate that maturation stage could be monitored with ultrasonic velocity measurements.
https://pubs.aip.org/aip/acp/article-abstract/894/1/1328/953...
https://acs-international.com/instruments/ultrasonic-pulse-e...
Now, I don't think ultrasound would work (it's a harder cheese than Swiss cheese), also the mechanical interface would be a complication (you can't gel the cheese). CT would work but be expensive
But tiny percussions and analysis of echo/transmission delays would work in principle
Ultrasound would also be a solid bet, but it depends how many points you need to sample if it would be time efficient.
> Thus operating frequency is always a trade-off between resolution and penetration. Optimal depth of subsurface penetration is achieved in ice where the depth of penetration can achieve several thousand metres (to bedrock in Greenland) at low GPR frequencies. Dry sandy soils or massive dry materials such as granite, limestone, and concrete tend to be resistive rather than conductive, and the depth of penetration could be up to 15 metres (49 ft). However, in moist or clay-laden soils and materials with high electrical conductivity, penetration may be as little as a few centimetres.
It's not only the sound, it's the sound, the bounce, the response to different strengths, the smell, the color. Humans are multimodal, machines are not, yet.
The moment we have a Michelin star level robot cook, then we can start thinking about automating this kind of stuff. For now, we have better results with humans!
Italians have absolutely zero problems replacing manual processes with technology. Creating each wheel is more science than art, everything is done in highly sterilized environments with exact temperature control, as an example.
I agree. As soon as you replace humans with machines, the next step is so-called "value engineering," where squeezing pennies out of a process becomes more important than the product.
Let the tech people do tech. Let the artists do art. Food is an art.
While I respect the point you're making, food is not an art. It is usually a perishable commodity and a requirement for survival.
Also, the crust can be chopped up and added to risotto (as you're cooking it) and they turn into wonderful little chewy chunks.
I’m French and something that surprise me is that both Italy and France have very good cheeses, but only in France we eat cheese on its own as part of the meal: the traditional French meal is: starter / main plate / cheese (sometimes with salad) / dessert. In Italian restaurants you sometimes find them as antipasti, but not always, and (at least in my experience) at home people don’t really eat cheese on its own.
In America eating cheese by itself is usually seen as a very fancy activity.
I can’t think of the last time I had a spread of just cheese in the US, and it was probably a fancy place.
It was weirdly fancy (eating good cheese), and weirdly not (I don't think of March madness brackets as a particularly refined thing). It was all delicious though!
[1]: https://ericbirlouez.fr/index.php/activites/articles/42-une-... (fr)
We usually lean over the sink, with the bag of shredded in one hand…
that's sad. i love at home charcuterie boards and a nice bottle of wine. while much more fancy that what my dad did. I did grew up with blocks of cheese pretty much always available as he loved cheese. and no, we're not from Wisconsin
You can see remnants in pop culture, e.g. in the '80s we had a tv quiz show called "il pranzo è servito" ("lunch is served") where participants had to win rounds for each of: first course, second course, cheese, dessert, fruit.
There was also a saying in central Italy "la bocca non è stracca se non sa di vacca" ("the mouth is not tired unless it tastes of cow", which in hindsight is odd cause there's mostly sheep cheese there) but by the '90s the cheese plate had switched to antipasti only, IME.
Appenzeller medium, young Asiago, sake trappist cheese, blue brie, extra sharp English cheddar, and of course mozzarella and halloumi.
I think both the Grana and the Parmiggiano are great in some places but not everywhere, so I couldn't pick them as sole winners.
Thanks for the risotto tip; gonna have to try that. I've never tasted the rind after cooking it in soup: it's not an appealing look to my eye.
What size should the rind be chopped up for risotto?
From italian its pecorino pepato (specifc, I know) and then black truffles variants, usually also of pecorino.
Maybe cheese you mention are more of an acquired taste rather than love at first sight, can't tell but will keep trying :)
Also I always add the rind to my risotto too!
Parmigiano Reggiano is more tightly controlled, and has larger total cheese produced (by a fair bit I think).
Very good Grana Padano can be just as good, but there are lots of bulk producers who turn out younger, cheaper cheese, that is yummy, but not as good as aged Parmigiano Reggiano.
FWIW, in the USA in particular, there is quite an industry in fake Parmigiano Reggiano.
https://english.elpais.com/economy-and-business/2023-08-25/c...
I mean, you do you, but you'd find me reaching for a lot of other cheeses first to go with my bread.
Obviously, there's some balance in technology that is about right, but where do you draw the line? Because this could absolutely be done fully by hand or fully autonomously.
The episode after that showed a machine to milk cows which was fully automated with no human involvement at all!
A person with a spoon, tediously and laboriously stirring ingredients in a pot, is a poor way to make use of that person's intelligence, creativity, and flexibility. An electric motor just does the job better.
On the other hand, by the Anna Karenina principle, cheese inspection is one of those tasks where there are a thousand unique and unexpected ways for a cheese to be wrong but only one way for it to be right. It's very hard to design an inspection that would catch everything and miss nothing that a human would trivially see, smell, or feel, while also minimizing false positives.
The robotic wheel flipper is somewhere in the middle: humans are great at navigating complex environments, and while you can design a uniform, controlled environment that a complex AMR can navigate, and space the wheels out regularly, it seems like half the task (rotating the cheese) is something ideally suited for a robot arm and half the task (getting to the cheese) is something better suited for a human. Humans can maintain the environment and debug the process, the robots can flip the cheese.
With respect to the cheese testing, I think a good middle ground is tool-assisted human inspection. Instead of/in addition to a hammer, give them an ultrasonic transducer and audio analysis toolset. Let them manipulate the cheese, but also give them objective numerical data on the frequency response and calculated porosity. It's easy for a person to recognize when a cheese is more hollow-sounding than the previous, but the first cheese of the day might be hard to for a human to recognize, and better tools than a primitive hammer can help with that.
It did not go over well with the former lawn mower operators, who found themselves no more employed than they would have been if the robotic lawn mower took their job instead.
many farmers don't use automated milking machines because you still need a human to inspect each cow and thus the machine doesn't save much labor.
Try to generate something new based on that pattern, and they tend to have massive limitations. But just a pass/fail reply? They're amazing at that.
"Ford, whose electrical engineers couldn’t solve some problems they were having with a gigantic generator, called Steinmetz in to the plant. Upon arriving, Steinmetz rejected all assistance and asked only for a notebook, pencil and cot. According to Scott, Steinmetz listened to the generator and scribbled computations on the notepad for two straight days and nights. On the second night, he asked for a ladder, climbed up the generator and made a chalk mark on its side. Then he told Ford’s skeptical engineers to remove a plate at the mark and replace sixteen windings from the field coil. They did, and the generator performed to perfection.
Henry Ford was thrilled until he got an invoice from General Electric in the amount of $10,000. Ford acknowledged Steinmetz’s success but balked at the figure. He asked for an itemized bill.
Steinmetz ... responded personally to Ford’s request with the following:
Making chalk mark on generator $1.
Knowing where to make mark $9,999.
Ford paid the bill."
https://www.smithsonianmag.com/history/charles-proteus-stein...
I’d argue the most correct answer is: up to the customers ability to tolerate, such that they’d be inclined to want to become repeat customers.
(heck, even with a written deal, the fine print often matters less that people might think in court, though a badly written contract will sure draw out much more protracted arguments about such details)
They’re usually a pleasure to work with, and I’ve found they’re typically the sort of people who are more than happy to share their knowledge.
Changing a line: 1 minute
Knowing which line to change (which includes making sure that the change doesn't break something else): 23 hours, 59 minutes
One thing I've also wondered about this process. When I buy Parmigiano Reggiano, it's just sold as "Parmigiano Reggiano". There's no discernable branding beyond that (compared to the Pecorino Romano I buy, which has "Locatelli" plastered all over it). Is this true all over? Do any HNers seek out Parmigiano Reggiano from their favorite dairy?
https://www.parmigianoreggiano.com/product-guide-seals-and-m...
Not a "favorite dairy", but if it's convenient to do so, I usually look for the "vacche rosse" type, which is more expensive but also noticeably different (and better). Of course it would be a waste to grate it over pasta; better to eat it on its own.
The 36 months is too strong in my opinion, still delicious but I like the 24 more.
I suppose wheels of cheese can last a lot longer than normal table cheese, so that's why it makes sense to make this distinction.
There are some youtube videos about people making artisanal cheeses:
https://www.youtube.com/watch?v=vM102CO8JL0 and https://www.youtube.com/watch?v=ImpROVueIcE
Kids! Do this at home!!! I've been making cheese at home for the last year. It is hard - but not as hard as programming and tastes much, much, better.
From the headline, I immediately thought of an answer I swear I learned about in some machine learning class years ago. People were struggling with a food inspection device that tapped (cheese? fruit?) like this, trying to emulate what a human expert did.
The punchline was that the human expert didn't really know how to articulate their decision either, and it wasn't listening to the drumming sounds at all, but merely dispersing some odors to do a better sniff test.
https://en.wikipedia.org/wiki/Rennet
Contaminants are a common problem in industrial food manufacturing: citric acid (fungal contaminants), vitamin C (heavy metals), and "enzymes" (?).
I'm glad Italians insist their cheeses be made following the traditional methods.
How can you be so certain? I did not find any credible source correlating microbial rennet to rash. Thus I would not rule out that this was simply a coincidence or at least not applicable for most people.
"Vibe diagnosing"
In 2018 i was renovating my house in Little Italy Toronto (Canada). There was this 91 year old Italian woman, Assunta, living alone in the house next to mine. She was always curious (or nosey?), but only spoke Italian, so we struggled to communicate. She would always say in broken English encouraging statements like "You make it nice", "lot of work, you do so good" to which I would say "thanks" and often talk about the amount of work ahead of me. She would always follow up with "eh, piano piano...".
I had no idea what she meant until one day I Googled this term and i learnt it essentially means "slowly slowly" or "take it slowly".
Assunta is gone now, but she was a lovable character. I think my dog misses her treats, and I miss the snacks she would bring me when I was working on the house.
It is probably just a joke for show and to fool naive competition.
Behind closed doors, they must do some actual quality tests.
Hey! I don't think it's a good idea. But if it's cheap and effective, guess how long it will take?
But there is most likely also large amount general QA of environment going on. Are things done right, is temperature right, is humidity right, is there weird touch. And so on.
All of this is much harder to understand and gather under scope of AI... Lot of it is likely not even conscious...