Scientists of Lemmy, how would you standardize or improve cooking recipes?
I'm sick of having to look up what country an author is from to know which variant of teaspoon they're using or how big their lemons are compared to mine. It's amateur hour out there, I want those homely family recipes up to standard!
What are some good lessons from scientific documentation which should be encouraged in cooking recipes? What are some issues with recipes you've seen which have tripped you up?
I'm an American biochemist, I also never learned the english system because my school transitioned to metric too fast. The mental burden of trying to cook using english units after working all day in the lab using that same part of my brain leads me to just not want to cook 95% of the time. But when I do cook I have optimized processes for my few simple recipes. When I bake I usually use a metric recipe or convert a English one, and optimize it before making a large batch of something.
I was a professional chemist for around ~7 and love to cook. My suggestion is to stop expecting precision with an imprecise and natural product like cooking. Are your lemons larger? They also might be sweeter, tarter, juicer etc. than others. Same thing with teaspoons. The spices you are using may be more or less concentrated than who wrote it.
Lean into the uncertainty and be free. Double or even triple spices to see if you like it. Measure with your heart
That’s just people who know how to cook, beginners want to follow recipes to a T and almost always come up with sub par results to someone who knows how to cook because they already incorporate what you’ve mentioned. This is just “make sure people cooking know how to cook” lol
Cooking is not a standardized or reproducible process at home, because the variables outside of anybody's control. Modern mass recipes give only the illusion of being reproducible algorithms, but they will never achieve that.
Grappling with the complexity of different tooling, supply chains, seasonality and so on, all within a recipe, is a futile effort. That complexity must be handled outside the recipe.
Why would you want anything by volume? Mass is so much easier. 50 ml of honey is way more annoying to get into a recipe than dumping it right into whatever container the rest of the ingredients are in while it's sat on a scale.
This would only make sense, if all people were baking with the exact same ingredients, in the exact same environment, with the exact same equipment. You know, like in a factory.
For households and the like, it makes sense to have a bit of variation, until you find the way that makes it perfect for you.
I've been cooking at home, and occasionally in restaurants, since I was about ten or so. So, 40ish years.
No single standard is better than the others. It does suck that there isn't a single one that is used as a base, and then gets converted by the cook into their preferred units and structure, but even that has issues.
The good news is that most cooking, and even most baking, is very forgiving of the kind of discrepancies between sizes of lemons, onions, etc. You don't really run into trouble until you're dealing with things that react chemically based on the ratio of ingredients, which is still most common in baking, and not even all baking.
Even in those types of recipes, it's usually flour that's the problem, not leaveners, since flour compacts readily and to a high degree. But, then again, most modern recipes like that are going to be in weight measures, or in baker's ratios. You'd be using a scale for the fiddly recipes.
So, generally, just guesstimate your produce size the first time you make something. It's not going to be so far off that the results will suck if the dish itself doesn't. Then you tweak things until it fits what you prefer, which is what happens anyway as you build your recipe book/collection.
My old recipe book had scribbled notes in the margins from years of refinements. When I copied that into a digital recipe manager, I added them in directly. Now, I'm able to just enter the original recipe, then add my notes as parentheticals or whatever as I refine.
Even with those detailed notes, a given recipe won't always be reproducible as exactly the same. That's because you just can't standardize everything. You use good produce, there's going to be varying water content, slight differences in flavor compounds, more or less sugars, so to get the same results over time, the cook has to know how to adjust for those things on the fly.
Of equal import is that no matter how scientific your process of recipe development is, the table is never the same as the cook. My taste buds and brain aren't the same as my wife's, my kid's, my cousin's, etc. So there's limits to the benefits of standardized recipes on the plate.
Now, formatting? That's a huge help.
You want your ingredient list to include instructions about when an ingredient is used in multiple places. You want lists broken down in sections when a recipe calls for multiple procedures (like making the main dish, a sauce, and a crust).
In the instructions, make sure the ingredient quantities are included for redundancy.
If there's an instruction about duration that's variable explain what the variables change. As in: bake for 10 to 15 minutes. Okay, great. What's the difference? If my stove runs hot and I go for the short time, will I see golden brown, and will 15 be burnt or just really dark? Yeah, you can't expect identical results from one circumstance to the next, but at least drop an "until golden brown" at the very minimum.
That applies to any variable, imo, but it can get to be too much detail in complicated recipe.
Cooking and baking are chemistry, physics. But they're also an art. The more you try to strip a recipe of flexibility, the less successful it's going to be for the next cook.
Not any kind of scientist, but an adventurous home cook
I'd really like the USDA/FDA/etc. (maybe not under the current administration) to publish sort of a food safety handbook full of tables and charts for stuff like canning, curing meats, cooking temps, etc. targeted to people like me.
I've recently been experimenting with curing meats, I've done bacon, Montreal style smoked meat, corned beef, Canadian bacon, and kielbasa.
And holy fuck, is it hard to find good, solid, well-sourced information about how to do that safely.
And I know that information is out there somewhere, because people aren't dropping dead left and right of listeria, botulism, nitrate poisoning, etc. because they ate some grocery store bacon.
I just want some official reference I can look at to tell me that for a given weight of meat, a dry cure should be between X and Y percent salt, and between A and B percent of Prague powder #1, and that it needs to cure for Z days per inch of thickness, and if it's a wet brine then it should be C gallons of water and...
When I go looking for that information either I find a bunch of people on BBQ forums who seem to be pulling numbers out of their ass, random recipe sites and cooking blogs that for all I know may be AI slop, or I find some USDA document written in legalese that will say something like 7lbs of sodium nitrite in a 100 gallon pickle solution for 100lbs of meat, which is far bigger than anything I'll ever work with, and also doesn't scale directly to the ingredients I have readily available because I'm not starting with pure sodium nitrite but Prague powder which is only 6.25% sodium nitrite.
Seconding the national center for home food preservation document.
One thing that I like experimenting with that i have to search for every time is the time/temperature curves for pasteurization of different foods. Every "knows" you are supposed to cook chicken (and most "prepared foods") to 165 °F according to the FDA/USDA. What most people don't know is that that temperature is what your food needs to hit for 1 second to have the proper reduction of bacteria (e.g., 7-log for chicken, which is a really high bar). You get the same reduction with 15 seconds at 160 °F or an hour at a little over 135 °F. You can easily do that with a sous vide bath.
It's really cool for people who are immunocomprimised or pregnant because you can cook a steak to medium rare, but hold temp for a couple hours, and it's just as safe as if you cooked it to way hotter and ruined the meat. You can also do runny egg yolks.
Here's the first link that came up when I looked for it, but I'm sure you could find the actual government publication.
I basically want that kind of guide for curing meats and other such things
Also there are some blind spots, something I was just looking for recently is canning some of my home-cured meats to save some space in my freezer. I know it's a theoretically possible undertaking, I can go to the store and buy a can of corned beef after all
But reputable sources like the USDA and NCHFP are kind of silent on it and pretty much leave it at "we can't recommend doing that, curing can change the density and water content and such and we haven't gotten the funding to test it."
I can find people who have canned their own bacon and such, and apparently not died of botulism, but I don't exactly trust the processes cooked up by some off-grid homestead tradwife mommy-blogger.
And holy fuck, is it hard to find good, solid, well-sourced information about how to do that safely.
I have a similar experience with some basic fermenting (e.g. kombucha, pickling). I'm growing cultures of microbes like yeast and bacteria and while I've been able to spot some obvious unwanted cultures on failed batches, there's a surprising absence of reputable info and unfortunately I've had to get by on the brewing equivalent of gym broscience, mostly on reddit, some of which I've spotted is misinformation. The SEO AI-generated articles plaguing search results don't help either.
They do publish pretty good information about home canning, though in batch sizes more and more of us aren't going to do because we're not putting up 10 acres worth of vegetables.
Too many cooking sites are let's exchange your recipe and end up with either stuff missing or absurdly high amount of sugar (as a rule of thumb divide by 2 the amount of sugar) or a lack of salt/spice even when they're notsimply forgotten.
Published books tends to be a bit better as in principle they're revised.
Peer review is how scientists correct that. Often it's as simple as on figure 2, the labels are too small and sometimes it's I don't get how you've built your experimental setup can you clarify this section? It's rarely catching biq mystake but really improves overall quality
At the end of the second paragraph, you're missing a space between "not" & "simply".
In you third paragraph, you used the singular "tends" instead of the plural "tend". In addition, though I believe the sentence to be grammatically correct even without them, adding commas before & after "as in principle" would make the sentence a bit clearer.
Finally, your last paragraph. The second sentence is quite long, it would be more readable if you added commas before the "and" & after the second "it's". A comma could be placed just after "Often", but the sentence remains legible even without it. The sentence could use quotation marks to improve readability further, which would end the sentence on a question mark followed by an ending quote. This would be grammatically correct in American English, but as the sentence is not a question, a period should be added to the end. While it may have been intentional, for comedic effect, "biq" should be "big" & "mystake", "mistake". If I've understood the sentence correctly, the newly-corrected "mistake" should be in its plural form, "mistakes", and be followed by a comma. The sentence should also end with a period.
If you're asking scientists about writing protocols, you clearly don't know how scientific protocols work. If anything, scientists need to take lessons from recipe writers on how to write protocols. Scientific protocols are notoriously difficult to replicate.
Here's a burger recipe written like a scientific methodology:
Raw beef patties (Carshire Butcher) were prepared on a grill (Grillman) according to manufacturer's instructions. The burger was assembled with the prepared patties, burger bun (Lee Bakery), lettuce (Jordan Farms), American cheese (Cairn Dairy), and various toppings as necessary. Condiments were used where appropriate. Assembled burgers were served within 15 minutes of completion.
Methods sections are limited in word count, and if a lab is hoping to get a few more papers out of a paradigm, they may be intentionally terse. There's a big difference between how we write protocols in-house and how we write limited-length methods sections.
I don’t share this idea. Especially not in industry. SOPs are extremely detailed to the point of including lot numbers, etc. If done right it leaves no room for interpretation.
Chemistry might not be much better. It's because scientists generally assume that readers already know how to do the techniques, and so the only information they would care to provide are the ones that wouldn't be considered obvious. Such as equipment brand, the name of the technique if there's multiple techniques that do the same thing, or experiment-specific modifications to the technique.
My understanding is that it's a holdover from older times, when scientists were charged per word, and so methodology would be cut down to remove anything considered "general enough" knowledge
So you're basically telling chefs to research and write out for you all the variables?
Baking is a science, cooking is an art.
Every recipe handed down through generations has notes, changes, etc....that's what makes it beautiful.
I am lucky to have my grandmother's cook book with 3x5 index cards hand written, with the date and whom the recipe is from....but I don't use lard in her Ginger Bread recipe from 1932.
There is no exact science you're looking for, the garlic grown here won't be the same as the garlic grown there, your experience won't be the same as someone who has cooked for years saying 'fuck it, throw that in there and let's see what happens'.
(to be clear, I was saying 'amateur hour' tongue-in-cheek ;)
I am lucky to have my grandmother’s cook book with 3x5 index cards hand written, with the date and whom the recipe is from…but I don’t use lard in her Ginger Bread recipe from 1932.
That's wonderful! All I got was a disintegrating notebook of delights. I do like deciphering it but not when I'm hungry!
Now to really boil your noodle I used to work with a lot of (French) chefs who when they wrote out recipes for magazines and such (pre internet) they DGAF if it was accurate or not... "if zey screw eet up, zey sink it is zere fault"
Are your oven's thermometer and kitchen scale within their calibration due dates? Is that timer NIST traceable? The measuring cups ARE Class A glassware, aren't they? Please, at least tell me you're getting your ingredients from certified suppliers... No, the spices from the dollar store down the road are most definitely NOT on the approved list, no matter how cheap they were! Dear Lord, how are you going to blame the recipies when your kitchen is still operating in the Dark Ages?
examples from professional recipes – measurements are given as weights (in grams) – no worrying about how much brown sugar in a “packed cup” or if your cup of flour has been sifted enough or what exactly is meant by a “cup of spinach”
examples from baking recipes – measurements are given as percentages – allows easy scaling up and down
Parametric recipes are great. The central ingredient is quantity 1 and everything else is a ratio by weight. You then scale it to your needs. So an equilibrium brine would be.
1 meat
1 water
0.03 salt
Brine for 1 day per 2 inch of thickest section.
They don't work for everything. So when baking a loaf of bread time and temp are spefic to loaf size. It still works for a batch of bread dough however.
This also helps you think in ratios which help general recipe construction. Once you know what flour to egg radio you like for your bread you can alter recipes to your preference.
On thing that drives me nuts is weight measures for dry ingredients vs numerical egg measurements. Just give me ingredient ratios normalized to the egg mass.