Curious question, in this case for verification they ported the software into a formal mathematical model. Why can't software development start from the other end instead ? Start with a formal model with the properties you want, then translate it into software.
I'm glad you said this because it's exactly the problem I've been obsessed with
I think it's absolutely viable and arguably should become the standard approach in time
I think it's overwhelmingly due to very very few people developing new greenfield software who are comfortable working with formal models
The talent pool for ATP devs is very small and specialized and they've only been financially justifiable when working on an established mission-critical system, like Lean's creator working at MSFT to catch Windows 7 bugs with SMT solving
Ironically I think these cases where formal models are clumsiest: using FP langs to retroactively model to C/C++ legacy programs
The benefits of formal verification are 10x easier to achieve if they are the foundation vs shoehorned in after IMO
You can. It ends up really slow normally which is why we don't do it. For example for a proven Deflate, it took an hour to process a 1MB file https://ar5iv.labs.arxiv.org/html/1609.01220
That’s not a great source to support your claim. From the paper
> The resolution (decompression) of such backreferences in an imperative context is trivial, but uses lots of invariants that make it hard to prove correct. In a purely functional context, it is non-trivial to find data structures that are fast enough. We decided to stick with purely functional algorithms, as they can be verified directly using Coq, and optimization of purely functional programs is interesting for its own sake. In our current verified implementation, this is the slowest part. The benchmarks in Appendix 0.A support this claim. We already have figured out an algorithm with better performance, but we are not yet done proving it formally correct; we will not get deeper into this algorithm in this paper.
There are many ways verify programs; the authors of your linked paper just chose a method that resulted in a slow binary. Alternatively, one can build a model of an imperative language (C) and prove that an imperative programs meets some specification.
They chose a method that's easier to prove. Imperative implementations are much harder to verify than functional. It's not that we can't do it, it's just really hard to do in practice so even people who verified the first compression bailed on the idea.
Refinement logics and rewriting logics take this approach directly, and it can go straight down to mixed-signal circuits or all the way up to distributed epistemic logic.
I've been increasingly interested in language-oriented programming and denotational design. So i get claude to create a domain specific language in Racket modelling the software i'm trying to build, we create nouns and verbs whose meaning fits the behaviour i'm going for, then when we find bugs or notice we need more exceptional (less-elegant) one-off behaviour to tackle edge-cases, we iterate on the language in such a manner that it these are tackled emergently with preferably less concepts, more cohersion and composition. It's somewhat similar to how memory safety is handled at compile time in Rust but for business logic.
There's no universe this doesn't negatively affect the UX of the OS for non-touch users. You can't simply support (or rather you can half-arse) new affordances without contorting the rest of the already well worn paths.
If you take an Apple computer from the early OSX releases (an iMac G4 from 2002, for example), you will be able to use it exactly as you use macos today. This indicates a success for UX.
You cannot say the same about GNOME3 or Windows 11 versus XP - Windows 11 has some baffling Settings app and also a Control Panel, being a mashup of different ideas and UI elements (and toolkits! see MMC snap-ins versus the UI of Settings).
The only difference is recent trends to dumb-down the UI to be more iPad-like, and this trend of Etch-A-Sketchification of the macos UI is not something to be welcomed.
I can actually see a possibility where this improves UI/UX a bit. One being the removal of some many hover-over UI elements, e.g., the (x) button on notifications do not display unless a mouse is hovering over the notification badge.
So, it may makes things a little easier for some GUI scripting. I can also see the UX improving for some disabilities, but I know that is merely coincidental rather than intentional.
> e.g., the (x) button on notifications do not display unless a mouse is hovering over the notification badge
Same applies to spaces where there's also delay applied before it appears for whatever reason. It would be nice if this fixed (on non-touch Macs as well).
It already has, see the massive increase in UI padding in Tahoe and later. Windows and GNOME both sacrificed their UX to support touchscreen laptops years ago, Apple was one of the holdouts before that point.
Windows did this seemingly by not caring that it's also a touch screen, so not everything in the system is touch-friendly. It's still fine for many situations.
Does that mean you also believe the same to be true for mouse affordances ruining it for keyboard users and vice-versa? Or "touch" is special in this regard?
Mouse affordances don't usually ruin it for keyboard preferrers. Keyboard shortcuts and control are often an afterthought or downright written off because "they can just click" but thats down to laziness or lack of care than some hard and fast rule. However, a keyboard affordance is usually invisible, and a mouse affordance isn't, so they can very happily co-exist without affecting one another. For instance, the Alt modifiers in menus in classic Windows, those have almost no effect at all on anyones ability to click the menu item with a mouse.
Touch affordances and UI does have to be visible, and there is only one screen. So the touch affordances and mouse affordances are mutually exclusive: targets get larger, padding expands, UI gets dumbed down, etc etc.
I can't think of cases where mouse affordances necessarily, intrinsically make things worse for keyboard users.
Of course, there's the extremely common case of keyboard support being half-assed or missing from GUI apps. But that's not mouse affordances making things worse, that's developers just not having the time or inclination to have great keyboard support.
Whereas, touch affordances seem like they'll be a lot more disruptive.
I'm still on Sequoia for this reason. I want a UI optimized for displaying text content and mouse and keyboard control. Mostly keyboard. I definitely don't want the larger border radius on the windows that would seem to be designed for touch.
Isn't that his point though ? Their plan seems to half-ass touch support (i.e. the menu bar, minimize/close buttons, etc. all stay in their current size), so it seems like a waste of hardware to include a touch screen just for very little functionality. If instead they make the UI friendly to fat fingers, they'd very much affect the mouse/trackpad experience.
Density is way down on UI elements for one. OS X used to feel like a Workstation. With every release macOS feels closer to a toy. Touch is great for convenience, but not getting into the weeds with “real” work.
Er… Yes they have. Liquid Glass is awful. It’s not only ugly (that’s preference, though), it has definitely negatively affected the UX. That’s why they’re adding sliders and options and reversal left and right.
Excellent article. This incoherence is a side-effect of a system built far faster than it is consolidated and thought about from an external behavioral standpoint.
I've created a skill that the agent runs after significant codebase work has been done: https://gist.github.com/scroobius-pip/adb87dffca64b11bad3dd1...
Not so much perplexing if they considered the AI rendition an accurate representation. It's somewhat quite internally consistent in its own illogical way
You mean: the family members personally wanted the maximum sentence, but they wanted the (dead) victim to be able to express that the victim wants leniency?
For me credit assignment without a global clock is extremely important, i believe it's a necessary step towards end to end training of models In-materio.
I've been thinking about this issue. One thing i realized early on is that the presentation layer is transcient it doesn't matter as much as data representation and storage.
Along the lines of presentation layer being transient,I’ve been experimenting with LLM and accessibility and screen reader to generate cleaner UIs and APIs. Haven’t spent too long on it, but parts of it seem to work. Speeds the biggest issue.
Hi, I'm currently planning on launching a product like this "Grammarly for Design" in a few weeks. Would you be interested in being part of the alpha group ?
I probably wouldn't be able to for data privacy reasons unfortunately, unless it's using an open-weight model (if it is, let me know). But I wish you the best of luck!
reply