Nope, it's too late. It sounded stupid when they called it AI in the start, but now AI means something like "tools that can perform tasks that typically requires human faculties to complete". Just like a computer does more than compute, words change meaning, sometimes influenced by marketing.
Assuming you used neovim, did you try neogit? It promises to be a magit clone for vim, but I'm curious how faithful it is. I've never used magit, so I don't understand what's so nice about it. Would be interesting to hear a long time emacs user's opinion on neogit.
I never used neogit so I can't comment. There are a couple things of magit that I really like; it's super easy to deal with individual hunks, and you have access to all the command line args of all the git commands. The defaults are sane, but you can get to the non-default stuff easy. (eg: force push with lease; diff ignore whitespace; commit amend, extend, --no-verify, etc.) It's all there, all the time. A lot of git UI's don't let you access those, or make you type the entire command to do it.
I used a debloater on my windows 11 laptop which also removed edge. No big deal since Firefox is my browser if choice, but I'm a Web developer and need to test my apps in different browsers. I've tried to install edge multiple times, but I can never get it to work.
But that's (+ AV1) about it. I wouldn't install edge just to test whether there are any files encoded with these codecs. (unless there are a LOT of media encoded in different ways e.g. youtube)
Just test on chrome and keep in mind to avoid AVIF
…Which you should be using the picture element or image() to fall all the way back to PNG or JPEG in most scenarios anyhow. New & improved formats like JPEG-XL should be considered ‘enhancements’ rather than relied on.
Don't remember what it was exactly, but I did run into some obscure scenario (something with tab focus I believe) where Edge had a bug that prevented my code from working.
Just spin up a VM for testing Edge, the differences are small.
Honestly, as much as I avoid it as a primary browser, I think all OSes should have a default browser that one could use as a preinstalled webview. Some way to avoid having to ship a 100 MB electron with simple web apps.
I don't think web apps are ideal, but having edge installed and available as a native webview is relevant.
I don't think this dream works yet, I don't think one can actually make THESE kinds of apps without shipping electron yet, but that should definitely be an option. I have no problem with edge used as a random weview in random apps. I only hate it when edge is opened as a dedicated browser.
I don't think an OS API to use Firefox as a webview is a reasonable ask :(.
How about you get VMware and install in there a vanilla Win10/11 just for the testing of Edge? Yes, it is heavy. Yes you will only fire it up when you need it. And this won't contaminate your machine.
I am not a color scientist, I do work in VFX though. Pixels and their generally-vector-valued contents are what we've mostly delivered to our clients for many years.
I'm not code-shy, but I haven't worked on our color pipeline very deeply myself - I did definitely get the impression from those around-me that did, that some of the hoop-jumping we've had to do to deliver the expected result color-wise have been quite complex indeed (especially if you rewind 10-15 years!)
For example our desire to do-color-well more easily led our sister-company 'Rising Sun Research' to develop a sophisticated color-management, calibration and viewing-environment software product called: 'CineSpace'! (which was ultimately acquired by Cine-Tal and is now owned by THX.)
I guess the above is all to try to help illustrate that color (and its representation/manipulation as values on a computer) can be a very complicated topic indeed!
I do think though, that making-things-look-good might not be something that color theory is really going to help directly with?! Of course, knowing your tools better is always a good thing! With greater capabilities can come increased flexibility and perhaps speed/efficiency advancements too. Definitely worth exploring if it piques your interest!
Not particularly good links I apologize! But hopefully somewhat-of-a-springboard for investigation?:
If you're going to delve in, you might like to look up some of the following topics too:
"Linear-to-Light Color" (this is probably the most important one for VFX/CG in my opinion, it affects everything from CG rendering/lighting to comp or even mip-map generation.. warping, image resizing (anything that 'filters' values), the list goes on! It's very important (and helpful!) to use values that are linear-to-real-world-light-intensities in cases where it is called-for!)
"Display Gamma" (This should explain why colors on most computers are all stored pre-raised to a particular power!)
"HSV/HSL Color Spaces" (transformation of color between spaces, sometimes the operation you want to do is trivial in another space and if you have good to/from transforms you can use, that can unlock some really powerful things!)
"High Dynamic Range Color" (This seems obvious in a way, but how bright is looking-at-the-actual-Sun compared to a full-white image (or an image of the sun) on a computer screen?!.. Using a less-limited-range (ie float values instead of integers, simplifying somewhat) to represent color is hugely important in some cases. It's mostly an issue on the very-bright-side but is also sometimes important on the dark-side too.)
Maybe for fun could you look into 'Spectral' color models?.. I find the idea fascinating but have not played with it much yet myself.
None of that is gonna help heaps with the creative side much though I wouldn't think.. but hopefully interesting if you're into this kinda stuff!
What books I would recommend would depend upon what application you would put them to. User Danwills mentions color spaces, which are the means by which colour is expressed and also the means by which color is 'transcribed' from one device to another. For this I would highly recommend Kuehni, R.: Color Space and its Divisions: Color Order from Antiquity to the Present. Wiley-Interscience, 2003. A great read.
It sounds as if you have an interest in the aesthetic application of colour. The standard tomes are Itten's The Art of Colour and Goethe's Colour Theory. In my opinion, both of them are semi-incoherent messes that have done more harm than good to painters. Bruce McEvoy's website 'Handprint' evidences a man who knows more about colour that Itten ever did.
https://www.handprint.com/HP/WCL/water.html Its good to see that he is finally getting round to writing a book.
The way I teach color application to my students is broadly as follows.
Firstly, the 'thinking' color space for an artist (digital or traditional) should be Hue, Saturation and Lightness (HSL). Using these values the colors of an aesthetic image can be 'conceptually disassembled'. To demonstrate this, I ask my students to describe the colors they are each wearing in terms of pairwise comparisons of HSL (e.g. 'A' is the same saturation as 'B' but of a different hue and slightly lighter).
For lightness, the simple rule of thumb is to ensure that the tones are approximately separate into three bands: light, middle dark (LMD). To this one might add black and white as the top and tail. Of courses, there is not an absolute rule, and without even blinking I could show you painters who have successfully employed only two-tone bands. Novices tend to overload the lower end of the tonal range. To demonstrate tone organization, I draw a posterise curve on Photoshop's Curves. It is important to know that the M of LMD is not absolute (with M as 50% with 0 as black), but instead tends to lay around 40%. It can also be observed that most paintings extend from black to white. However, again without even trying I could name a dozen exceptions (for example, check out the low tonal range of some of Gwen John's painting).
Saturation is the intensity half of the chromatic component of colour. In common with lightness, it exists as a 'ramp' value bound by a min and max. For this reason, it can mostly be treated in much the same way as lightness. For most of art history, the saturation map of a painting followed its lightness map. It was the development of the chemical and dying industry in France that introduced artists to a wider range of intense pigments. The saturation map of Gericault's Still Life with Lobsters is remarkably different to its lightness map. If you don't have access of MATLAB or Nuke or suchlike, you can make a saturation map in Photoshop using the Selective Colour adjustment. Just cycle through the colours in the drop-down menu and set the black value of each one to -100%. For the white, neutral and black, set the black value to +100%. Seeing the saturation of a painting expressed in this way is something that the impressionists would have murdered for.
Hue is the difficult beast in the HSL triumvirate. To get a grip on it you have to side-step into the RYB colour space. It is in the RYB colour wheel that the hues are arranged in their perceptually antagonistic pairings: red/green, yellow/purple, blue/orange. These pairings have been know for 100s of years, even before Newton's colour wheel was a thing. It is uniquely difficult to get a handle on hue... it defies easy conceptualisation. There is also a huge amount of bullshit in the wild on the subject of hue. Check out adobe's colour wheel: color.adobe.com/create/color-wheel See all that stuff about triadic complementary etc? Most of it is mostly hot air. How do I know? I have personally reviewed the hue histogram of hundreds of paintings. From this I can say that the organisational strategies that artists employ are, at best, approximate. Certainly, they are not as clear as the art books tell us. However, the following is generally true... painters avoid: hues that span no more than one half of the RYB hue wheel, hue that are evenly distributed on two halves of the RYB hue wheel, hue that occupies 360 degrees of the RYB hue wheel.
The final key wisdom that I expect of my students is to understand contrast as a structural phenomenon. This can be understood as manifesting in two (main) ways. The first is ratio contrast, which is simply the contrast between two values. In painting, these values are usually averages. Separately considered can be the contrast between neighbouring regions and those which are not. The former are perceptually prime and can overwhelm our ability to perceive those regions which are distant to each other. The second contrast is global contrast. An image which extends from black to white has higher global contrast that one which extends from dark grey to light grey. These two ways of understanding colour contrast can be extended in many dimentions. For example, the way that they manifest between (ratio) and within (global) the depth planes of a landscape.
As for what constitutes effective colour contrast: one observation holds true of all painters, writers, filmmakers etc.: that their task is to exaggerate life. This they do in two dimensions: exaggeration of difference and exaggeration of similarity (though Ruskin expressed this as a difference between affinity and contrast).
Which they absolutely are doing, but it's difficult to provide public transport anywhere else than the big cities when only about 2% of Norway is inhabited. The price of public transport is not what's keeping people from using it, but the fact that many people live outside the big cities and have to travel to them for work. Depending on where you live, commuting to the city may be very unwieldy. Oslo has been building out more train tracks recently which make travel time much quicker for a lot of people living outside the city.
I have never given emacs a fair shot, so I can't compare the two. Neovim makes it easier to get modern editor features like treesitter and lsp support integrated into the editor. The Lua api seems to have been very well received, so the plugin ecosystem for neovim is thriving. It has also attracted a lot of people outside of the original vim niche (including me!) recently, which means there are more plugin developers who care about how things look and feel to use.
There are certainly some constrains on what plugins can achieve due to nvim being exclusively a TUI, but in my opinion the pros of the editor outweigh the cons.
As to what makes me choose neovim over vscode with a vim plugin, it's a combination of things. One thing I love is that neovim is extremely easy to configure. I also enjoy that everything follows vim rules, rather than some things like file trees or consoles having their own unique rules. I have also become dependent on some vim plugins that don't exist in vscode.
I don’t know. I’m assuming I should install jq (or see if I already have it?)
Here’s the problem I have: I don’t often have to format unformatted JSON, but when I do, it’s almost always an error message I’m trying to parse in order to track down a bug. So I tell myself “I should really figure out how to automate this…I know it’s obviously possible, but I just really wanna fix this bug right now, so I’ll do that later.”
And then the next time it happens, I curse myself for not having researched that, and promise to do it before the next time.
So I’m just happy that I stumbled across this page. Now I don’t have to remember to research it :)
reply