The issue with color to grayscale conversion for human consumption is in most cases there is no well-defined ground truth. People don’t see in grayscale, so the appearance preservation approach doesn’t work. And the source image was most likely heavily color corrected to match certain aesthetic. So the problem becomes “to preserve as much information, both content and aesthetic, within constraints of the target grayscale medium”.
The bottom line is, use some standardized conversion (like described here — just to avoid surprising users) if images don’t actually matter, some contrast-preserving method if content matters, and edit creatively otherwise.
Also it has a proper builtin flashlight which is surprisingly useful. Amazing watch, especially if you get a comfortable aftermarket strap e.g. from Hemsut.
And there are 2 kinds of those: the other is refreshable materialized views, which run on schedule, can have dependencies between them, thus can implement quite complex data transformation pipelines.
For me the problem is less the UI but weird bugs / unimplemented cases in its modeling kernel. Some lofts and tubes just refuse to work for no apparent (to me, a non-expert in CAD) reason. Even worse, simple fillets sometimes don't work as expected on moderately complicated bodies. Maybe if you're an expert you can avoid these case, but it's extremely frustrating to spend several hours successfully modeling the general shape only to be unable to apply simple dress-up features.
I'd really like to use it and master it further, but CAD is not something I need often, and in F360 any geometry I've needed just works.
Maybe not the best name — ARIS is a relatively well-known business process management software suite, so it's definitely international, web and AI, so trademark conflicts may arise.
For $450 you can have Instinct 2X Solar, which is a very useful tool watch, surprisingly functional without phone connection and with a month of battery life. Can probably be used without ever connecting to a smartphone, I'm not 100% sure but everything seems to be configurable from the watch itself. Almost useless for reading text messages though. Basically a polar opposite to Apple Watch for a similar price.
Pebble is nice as a concept, with E-Ink and easily programmable watchfaces, but its hardware is arguably quite ugly, and has way less sensors.
UX does not seem to be a strong side of Metabase. I’ve recently started creating dashboards in Metabase, having used Grafana extensively, and it’s not very pleasant. Too bad Grafana has very limited options for not time-based visualizations.
100% this. At that time if one needed a simple custom database client with GUI and analytical reports, the go-to solution was MS Access and it was working well. Could even be placed on a shared drive. Now, as accepted solution, you either pay for different SaaSes, each with a fraction of functionality, or have to code a bespoke website using overcomplicated frameworks.
It's USB-C, "you'll never know" is the answer, often it's "kinda working, but extra slow". I consider having a cheap USB power meter a necessity at this point.
What's your FPS/LPS in this setup? I've experimented with similar imaging with an ordinary camera, but LPS was limiting, and I know line-scan machine vision cameras can output some amazing numbers, like 50k+ LPS.
With some luck, yes. I’ve had o3 in cursor successfully diagnose a couple of quite obscure bugs in multithreaded component interactions, on which I’d probably spend a couple of days otherwise.
From what I see in Russian-language news, only relatively minor damage. I've lived in Petropavlovsk, it's an ugly city in various states of disrepair, but they do take seismic reinforcements seriously, like mag 7 should cause zero damage according to plan.
It's basically immune to tsunamis as it's protected by a bay with narrow entrance that extinguishes the waves, also most of the city is raised at least 10m above the sea.
R/S is also heavily influenced by Lisp. Haven’t written it in 10 years, but AFAIR it even has proper macros where argument expressions are passed without evaluation.
I've tried to use various OpenAI models for OpenSCAD code generation, and while the code was valid, it absolutely couldn't get spatial relationships right. Even in simple cases, like a U-tube assembled from 3 straight and 2 curved segments. So this is definitely an area for improvement.
The importance of repairability is underestimated here. All new infrastructure must be built under assumption that there will be multiple attempts at sabotaging it by actors of various level, and multi-megavolt unrepairable cables that can be fully disabled by one smallish unmanned sub don’t win here at all.
What's the primary use case you had in mind here? In the example I see it generating a palette of 256 colors and then using them, but it doesn't seem to correspond to any modern use case. AFAIU one currently needs dithering either as part of print/display process (but then you have a fixed palette), or for compression, but I think this makes sense nowadays only with very low color count, like 16 max?
I have a lot of lumber from previous construction projects, and all the basic woodworking tools. I also have some relevant skills. Would I be able to build it right now? No, need a wide and sturdy board, widest I have is 150x30mm pinewood. Would I make one if I had a proper piece of wood? Also no, I just don't want to accept the structural weakness here. And this joint actually requires some precision, which is not trivial (at least for me) with handheld circular saw!
Haven't built a chair yet, but my go-to design would be something like Adirondack.
Given the compose sequences are mnemonic, I’d prefer it over Mac every time. Compare Compose+<< and >> for «» to Opt+[ and Opt+Sh+[ on Mac. Which may or may not work depending on locale.
This is a very good analogy. And this interest rate can probably be significantly reduced by applying TDD and reducing the size of isolated subsystems. That may start to look like microservices. I generally don’t like both for traditional development, but current LLMs both make them easier and more useful.
And the “rule of three” basically ceases to be applicable between components — either the code has localized impact, or is a part of rock-solid foundational library. Intermediate cases just explode the refactoring complexity.
While this is a good idea in theory, one needs quite a lot of patience to deal with its bugs and kernel limitations. It has definitely become much better since 1.0, but the inability to put chamfers and fillets wherever is extremely annoying — whether the features compute is order-dependent and they routinely conflict with each other for unclear reasons.
So, maybe it’s not a bad idea to start with a free version of something more ergonomic, just to avoid getting too discouraged.
Your Point example is indeed good, it shows one of the drawbacks of small 'proper' objects. How do you handle the case of offsetting thousands of points with a single operation, which in most cases will make sense and is readily vectorizable? It's better to expose the internals and use external functions here.
There's probably a deeper question, how to make objects 'transposable' in general case (like going from row-based to column-based representation) without duplicating code or exposing internals?
The biggest time saving I've had in woodworking was switching to modern self-drilling screws with Torx head [1]. With a highish-rpm impact driver, these screws require no pre-drilling even with plywood and are a pleasure to work with in general.
I didn't follow it close, but had an impression that C# and .Net still have a significant Windows ecosystem bias. But this article, and comments from happy C# users, may just convince me to re-learn modern C# and start using it in everyday work. I'm kinda tired of C++/Python ecosystem, and have fond memories of working with C# around 2.0-3.0.