"Ultraprocessed" is at least a tangible definition though (even if it's a proxy) where you can empirically show that a certain product is ultraprocessed or not based on the way it is manufactured.
It also has enough overlap with addictive food to be a useful criterion.
In contrast "hyperpalatable" is more precisely describing the problem, but seems much more difficult to proof / easy for manufacturers to wiggle out of.
How would you prove that a given food item is "hyperpalatable"?
> "Ultraprocessed" is at least a tangible definition
The Nova system's classification for UPFs seems to be what the majority of people who refer to them use as a definition.
In the Nova system, there are four main groups of food:
- Group one has 'unprocessed or minimally processed' foods, e.g. grains and fresh fruits.
- Group two has 'processed culinary ingredients'. These include foods that use naturally-derived ingredients like salt and flour.
- Group three has foods that combine the first two, like salted nuts, and can also include things with some added preservatives or flavourings.
- Group four is ultraprocessed foods. These are defined as industrially-manufactured foods made with multiple ingredients (typically multiple oils, sugars, fats, and salt) and ingredients with minimal culinary use.
The issue with group four is that it's far broader than it should be. For instance, under the Nova system sparkling water is a UPF because it's carbonated, and carbonation is considered a chemical additive. It also classifies anything with, say, Stevia as a UPF even though it's a perfectly safe artificial sweetener. It's broad enough that it covers tofu, various cheeses, and various breads, to name a few.
It also ignores the actual nutritional content of the foods (which the original Nova paper touches on, I think, specifically saying it's not meant to be used for nutrient profiling).
> How would you prove that a given food item is "hyperpalatable"?
I was recently looking at a study about this [0]. The three criterion that have been found to best define hyperpalatability are as follows:
(1) Foods with over 25% of calories from fat and more than 0.3% sodium by weight
(2) Foods with over 20% of calories from fat and more than 20% of calories from simple sugars
(3) Foods with over 40% of calories from carbs (not counting dietary fibre and simple sugars) and more than 0.2% sodium by weight
Even if you do eat them, there's no evidence (or I suppose I should say no evidence yet) of microplastics being harmful when ingested. Nanoplastics, on the other hand, have been found to impact animal embryos and cells grown in labs.
A problem with microplastics is that they come bundled with numerous xenoestrogens and other harmful contaminants. This too is what makes them very harmful, especially when conceiving a child.
Very cool project! When I was regularly using a multiplexer on my personal machines, I did something similar with `abduco` [0] for session management and `dvtm` [1] for the actual multiplexing.
This is awesome! I was thinking it would be neat to have something like abduco but on a more reliable foundation, like libghostty-vt.
For my agent management scripts I use zellij since it is more ergonomic than tmux. Abduco sounded good in principle, but implementation is too limited. However, zellij is quite huge in the order of tens of thousands LOC and I am using only small part of it. It looks like zmx might implement just the right amount of features for this use case, I am going to try it. It is always nice to achieve same functionality with leaner tools.
Do you also think about dvtm part alternative? I wonder if once libghostty proper gets finished it would open possibility to level up textual multiplexing and unlock some cool features with graphical UIs.
I have thought about writing a separate tool that resembles dvtm but I’m not exactly sure how I would build it.
I don’t want to maintain a monster project like terminal multiplexing. Zmx is basically a single file with 1500 LoC and is “production grade” with just a few quirks I haven’t figure out yet.
I would want something of similar scope.
With zmx I created two commands you might be interested in: zmx run and zmx history. Run lets your execute commands inside the PTY and history lets you read from the session history.
What do you mean by snapshots? There’s a “zmx history” command which will print whatever is stored in libghostty as plain text, or with ansi escape codes, or even html
I'm rendering a few dozen terminals in a website, and for all of the inactive ones i render and serve a jpg of the "current screen" of ansi escape codes from kitty.
I've found this to be a difficult thing to get. abduco doesn't have current state, and I dont want all of the complexity of tmux. I also don't want the entire scrollback history (until i click into a given terminal and connect with xterm).
This article equates ultraprocessed foods and hyperpalatable foods (foods designed to make people want to eat them more). While many hyperpalatable foods are classified as ultraprocessed, simply being hyperpalatable does not mean it's ultraprocessed.
Worth noting that the Nova food classificationvsysten (which this article references) completely disregards the actual nutritional content of foods.
For a good primer on a lot of the misconceptions around UPFs, check out [0].
I agree that many hyperpalatable foods are ultraprocessed so that they can be made more cheaply, but I don't think that's reason enough to say that the, uh, process of processing foods is entirely aligned with the concept of hyperpalatability.
[0] https://discuss.grapheneos.org/d/27068-grapheneos-security-p...