Computer screens and paper are not comparable, they're physically different and they're not really even used in the same fashion. Yes, you can read off of both but you get the point.
You sound like a person who uses a terminal with a black bg and neon green text and claim that as superior. Maybe to you but surely not for the majority. Most likely that's not even optimal for you but you've just convinced yourself of that.
I get the frustration you seem to carry towards designers, but if you're blaming _all_ designers because some might utilize anti-patterns or you just have personal preferences that differ, you're kinda missing the woods for the trees.
You do realize that the chair you're sitting on, device you're writing on and building you're (likely) in are all designed by designers.
What's actually different between a photon of a given frequency that enters my eye after reflecting off a surface and a photon of the same frequency that enters my eye after being emitted from a display?
> You sound like a person who uses a terminal with a black bg and neon green text and claim that as superior. Maybe to you but surely not for the majority
BTW, in the CRT days green on black was actually objectively superior if you were using small fonts on most displays. That was because most displays would have slight misalignment of the red, green, and blue, especially near the edges, leading to color fringing around white things. For small text, the fringing could make text blurry and hard to read.
Pure green on black had no red or blue, so no fringing. Red on black or blue on black would have worked too, but most people find those much harder to read than green on black.
Nothing.
OTOH, the spectrum of photons emitted from maximally white and maximally black pixels on an LCD screen is probably nothing like those reflected from black ink or bright white paper under any realistic lighting conditions.
About the photons, you could look into a regular paper sheet's or book page paper's albedo values and compare that to the light emitted from a led display, it's not even in the same ballpark. Why do you think people enjoy reading on an ebook reader rather than a normal tablet with a lcd/led display?
Artificial light strains our eyes way more than normal sunlight (or even indoor lighting) reflecting off of surfaces indirectly lighting everything for us.
Well, none, of course. But the point is that white paper and a display's #FFFFFF probably won't send the same wavelength or number of photons into your eye.
One of them gives me a migraine after 5 minutes. I assume because light being reflected off a sheet of paper gets weaker, though I'm not sure how.
> You do realize that the chair you're sitting on, device you're writing on and building you're (likely) in are all designed by designers.
ANY designer having gone through design school can tell you that low contrast text is a no-non.
So maybe we should just say that the people doing designer jobs without learning about the basics of color, contrast and accessibility are unqualified to do their job in the same way a architect is unqualified if they design streets to small for cars or similar.
Btw. if you have problems with to bright whites (which I increasingly feel is related to OLEDS but that's just a wield guss). Make the background non-white but keep the text black or very close to black.
Also just to be clear I think no on is blaming all designers but just such which do thinks like gray on gray. And then believe it's all okay because their expensive OLED display displays it still kinda readable (again I'm just guessing about OLEDS, I think I have to make sure to get my hands on one to check that).
In the example (yellow on black) I mentioned has too much contrast and actually hinders readability for most people with a normal visual acuity as well as certain types of impairments.
Designers are also the ones responsible _for_ all the accessibility features and standards set to help developers implement accessible solutions.
If you're a software engineer who cares about contrast and studies different vision problems to figure out solutions for example, you're a designer in my book.
This is why I also specifically separated aesthetic and usable design. The former is more akin to "art" as it aims to simply look nice and I don't personally consider it design as design is problem solving and bad design just creates more.
I've always said that if a designer or web developer doesn't actively hinder accessibility, it's already a win in my book. It requires someone to make the decision to slap a #fafafa shaded text o top of a #ffffff background, semantic html and OS default colors go a long way if you don't know better.
That's part of what this article gives you, why it's talking about art school and not technology. Miscommunication - incommensurability of how to see the world really - is the pitfall when engineers are tasked with creating a subjective thing and asked to try to do something simultaneously functional, familiar and original.
Just put yourself in this guys shoes. Just like you are excited to share Linux display color control commands, this guy is excited to share a new way of thinking about black and dark shades that instantly expands your vocabulary - both for creating new things and understanding the decisions of others.
Anyway, the negativity of others in this thread, it's ironic because that's exactly the point of the article, really, that the pedantry gets in the way of being creative - or at the very least, seeing things originally.
I work with an artist daily. I appreciate and value his opinions that there should be no pure black in the creative product we make together. It elevates what we make above all our competitors. It's an extremely insightful perspective.
Design is always a trade off. Do you want more focus on content, or more information with less focus? Is your default setting more friendly to smaller or larger screens? Every person has different preferences, and every trade off has cost benefits you can run. Yet your statement here is very “black and white” (ahem), assuming everyone should think like you. That betrays a lack of nuance in understanding how design works.
In the programming world, you also have these trade offs, yet it’s easier to avoid thinking about them at times. Static vs dynamic languages are full of trade offs like this.
To avoid coming across as yelling at clouds, attempt to frame why it’s a trade off you prefer, rather than a platonic ideal others don’t see (because it isn’t).
It would not be the right choice for a default config, or in a context where sometimes things are scrollable and sometimes they aren’t. (This is often the case on mobile phones; if there’s no scroll bar, the only way to tell if something is scrollable is to try it, if you think to.)
Well, ok, sometimes we do, but...yeah.
and some designs come out of relentless targeting the impact of the bottom line at the expense of aesthetic or "pleasurable to use", some designs come out of snake-oil sales-people, etc etc.
Different populations, different needs/wants/desires/cultural expectations. This is what focused advertising is all about. You can create one interface and say "everyone must use and learn to love this one," and that can work well. You can also gain immediate appeal by creating "one interface for each population of potential users" and relentlessly target your designs for each of those populations. I don't think we've fully recognized it in software engineering circles, but I think a lot of what are producing today is really about creating different experiences for different populations when the core capabilities have been there, available for free or on extremely cheap computing platforms, for anyone to use, for decades.
I imagine that's fundamentally what the job of most software engineers is today: find the design that best matches the needs of the customer, and build a workflow for their needs.
I agree with your chagrin for vanishing scrollbars and hamburger menus, by the way. There ought to be a good term for designers who do the opposite of what good design does (which is make things which are usable).
Is this why the most fashionable websites are also the most difficult to read? I ask purely for information.
WCAG2.1 has some great guidelines when it comes to color contrast in digital text and as long as those are adhered to, most people will have a near optimal experience consuming that text.
And yes, too big of a contrast difference is not great either. Just try to throw bright yellow on top of a black background, it's gonna hurt your eyes pretty quick.
So the end result is that Windows users can’t adjust the brightness of their non-laptop monitors without third-party software, or reaching around the side of the screen and fiddling with the infuriating controls of the frustrating OSD interface. So almost no one ever adjusts their screen’s brightness at all, and almost no one of the remainder adjust their screen’s brightness to match conditions. And manufacturers have (understandably) decided that too dim is worse than too bright. So most people’s screens are genuinely painfully bright. So then people say “don’t use actual white” and “don’t use actual black”. And those that have calibrated displays end up sad.
It allows you to change brightness, contrast (and volume, never tried it), from the taskbar.
It works for a Dell and iiyama monitor I own, but not my Agon one, so YMMV
The brightness keys could be managed by BIOS? If you can modify brightness during early boot, or while BIOS settings is showing, then there is a chance Windows doesn’t see those keypresses.
I do wish I could intercept it, not just for external monitors’ sake, but also so I could make it logarithmic or similar, like I did in Arch Linux + i3 with XF86Brightness{Up,Down} mapping and an xbacklight wrapper script. On my Surface Book, the smallest effective brightness change is 2%, and each 2% step between 0% and about 20% has at least as much effect as 40–50% or 80–100%, yet the seemingly-hardcoded behaviour is that the brightness keys just adjust the level by a flat 10%, which is what every platform does (OK, I can actually only directly attest to Android, Ubuntu from some years ago, xbacklight by convention and Windows, but I think the rest do it to), even though such a linear scale is obviously stupid and wrong. On my phone, two millimetres of adjustment on the left has about as much effect as the whole right hand side of the scale (several centimetres).
Not sure what are you talking about. I can adjust brightness (and lots of other things) of my monitors just fine using their own buttons.What on Earth does that have to do with the Windows OS?
But if you’re in a normal house in a room that gets plenty of lighting from outside, absolutely. I tend to run my external monitors at 0% in the earlier morning, 20% in the mid-morning, 40% by the afternoon, sometimes up to 70% on a particularly bright summer day, and down to 0% in the evening (and I’d take it towards −20% at times if there was such a thing, their minimum brightness is nowhere near as dim as my laptop’s screen’s minimum brightness, which is even acceptable for use in a completely dark room; their 0% is its 25%, while their 100%s are similar). Leaving them at the 100% they came at would be awful, it’s a little too bright for my situation even in the middle of the day, let alone the evening.
Yup. That would be me. My office is in a basement of my house and has been for the last 20 years.
Is this your opinion or is this widely accepted as fact? I've read countless design suggestions to reduce contrast and avoid using pure black or white on an LED display. I'm curious whether there's research that supports either side.
I have seen those suggestions too, but they seem to be just parroted off of each other. I have never seen a single one link to any reasoning or studies that support the claims of better readability. They all seem to say it because someone else said it in a game of design telephone.
In my personal experience, I find gray text on an off-white background to be very difficult and straining to read, and additionally, I don't think it looks good.
I can see some merit in reducing the whiteness and thus brightness of the background, but I don't see why reducing black to some lighter black or gray helps at all. Having text in a nice solid black or near-black with an off-white background seems reasonable to me. That matches more closely with reading things on actual paper, where dark black inks are possible but the background is not pure white. That leaves legibility in place and reduces background brightness.
It's possible this whole "reduce the contrast" thing has gotten off track, when to me at least, the goal seems to be to reduce the blinding background blasting you with light and not to simply reduce contrast.
I so wish I could find that research that once and for all debunked the "UX expert" urban myth that #fff + #000 = evil. It made me consider past mistakes and I go for all black on all white whenever I can. Doesn't hurt from a WCAG perspective.
Rello, Kanvinde, & Baeza-Yates, 2011. Layout Guidelines for Web Text and a Web Service to Improve Accessibility for Dyslexics.
If you suffer from astigmatism you might experience halations or if you have contrast sensitivity, visual distortions.
Well, in either case, perhaps nobody should hard-code the darkness of text in their CSS but instead allow the user to set it in their browser?
I agree that the user should have browser defaults that they prefer, but motherfuckingwebsite.com still looks awful in most browsers.
>"I've read countless design suggestions to reduce contrast" I would say that modern design suggestion often the opposite of usability/ergonomics. On many sites you now often see undecipherable light grey text on white background. It sure is soothing. It also sure is useless as many people who do not have perfect vision are unable to actually read it without blowing text size out of proportion.
https://uxmovement.com/content/why-you-should-never-use-pure...
The user can easily change brightness of their screen just as the user can easily change volume of their music. But producers want their creation to stand out. Make their music a little bit louder, or their website a little bit softer than the rest.
In both cases the outcome is a race to the bottom and reduced dynamic range.
And finally, in both cases it's fairly inconsequential. Things get worse but only a little bit.
Please do some research instead of doubling down on opinions you're presenting as absolutes.
This is just not true.
100% (minus rounding error) of devices can easily adjust volume.
Phones and laptops can adjust brightness, but dodgy automatic brightness adjustment is the rule of the day. And basically no desktop monitors can easily adjust brightness (not because they can’t, but because OSes have frustratingly ignored functionality that has been standard for a decade or two; on-screen brightness adjustment is absolutely consistently painful enough to use that I disqualify it for the word “easily”—oh for the brightness wheels of the CRTs of yesteryear!).
That this is the top comment and yet no one has refuted it boggles me. I cannot think of anything I've read recently that uses bright white paper. Sometimes I see reams of bright white paper for sale for a few extra $. However if I'm printing a resume, it's on x% linen slightly off white. If I'm reading a book, it's on warm-colored paper. If I'm reading music, it's on even warmer colored paper- and dare I say it as a musician for decades, the stuff on bright white paper (a photocopy, or self-published octavo for example) is the HARDEST to read!
For books, paper isn't the brightest white; it's only copy/printer paper that got in to a ‘horsepower race’ for brightness.
This, absolutely! Monitors usually come with an eye-searingly-bright default that's meant to look good in a showroom and wear out the backlight more quickly, not at all a good brightness/contrast for normal use. I set my monitors to 0 when they were new, and that was already very bright; now, many years later, they've gone up to 20-40 to compensate for backlight degradation, but are otherwise still more than bright enough.
For Mac users, look at https://lunar.fyi/. There are a bunch of other options as well.
For Windows, look at https://github.com/emoacht/Monitorian
No need to yell.
Pure black is often not the correct colour for text, unless it’s monokrome (“B&W”) printing, rather than CMYK (but more on that later, because B&W monokrome black is often not “true black” either).
If you are printing CMYK you rarely use key (“K” - black) on its own, but rather use a CMYK mix to build a black shade, to give a rich black (more M) or cooler black (more C) or balanced black - neutral but not stark. The different mix can really change the way a page feels - and the right mix on the right paper (“stock”) can make the black pop or muddy, and have a huge impact on readability. This is really important when you’re designing for print. It really does make the difference between someone continuing to read and chucking your print aside after a paragraph. Good print designers know how to build different blacks for different stock.
When you are printing in monokrome black you are really printing g a shade of grey, to a greater or lesser extent, determined by the absorbency of the paper you are printing on.
And, indeed, different inks have different qualities so some “black” ink for monokrome print maybe be built from various pigments giving it a slight blue sheen or a brownish tint under certain lights. Plus inks - and the paper substrate - are more or less reflective meaning they will behave and appear differently depending on the light conditions.
I loved working in print.
Dark/black background with lighter text doesn’t bother me. White background with black text does.
Are you really going to tell me that I’m wrong for doing what is right for me because of your opinion?
The monitor only goes so low. And this works. Paper is very different to a screen. There’s a reason e-readers exist.
Maybe people should just do whatever they prefer/works better for them instead of insisting that light themes or dark themes are objectively better.
Because of this, for me the biggest benefit of upgrading to a phone with an OLED display was that it enabled me to adjust my phone's brightness without such dramatic loss of contrast and vibrance. Unfortunately I don't think OLED panels will ever become practical for desktop monitors… they're just now becoming somewhat available in laptops and even then sometimes come with issues similar to those seen in early OLED phone screens. We're probably going to have to wait for microLED tech to mature to get this benefit in full sized monitors.
There's far fewer desktop monitors and laptops being shipped than phones or TVs, and the majority of desktop and laptop panel sales are tied up in basic office-grade econoscreens and gaming monitors focused on high refresh rates and low input latency. Midrange/affordable laptop/desktop OLED panels fit neither of these categories and as such haven't received nearly as much investment and R&D.
To only apply the tweak to the dark theme, you can do:
@-moz-document url-prefix("about:reader") {
body.dark {
color: #FFFFFF !important;
background-color: #000000 !important;
}
}I which they would not, it's annoying when reading outside.
But I'm all in for the darkest blacks.
And yes I agree that if the contrast is to bright it most times you monitor which is generally set to bright.
Hm, maybe one exception might be OLED screens, but reducing the general screen brightness should still work.
If you are still convinced that it's to bright DO NOT make the black not black but make the withe less bright, preferable some subtle bright non gray color like for example hacker news does it.
That makes the contrast less strong without being a accessibility nightmare which requires people to open their web-console and modify your CSS.
Some people complain about excessive brightness being hard on their eyes. They likely have their monitor brightness set too high. Most monitors can have their brightness set using DDC/CI (this sometimes needs to be enabled from the monitor's OSD). Most Linux distributions include "ddcutil". Install it and give the user who'll run it permissions to access I2C. In Debian, this can be accomplished by adding them to the i2c group.
Find out which I2C buses your monitors are attached too:
Now you can change brightness from the command line: The first number after setvcp is the feature value to set. "10" is the standard feature value for brightness. The second number is the percentage brightness. I suggest picking values for low/medium/high brightness for each monitor, and binding a script that sets all them at once to the otherwise rarely used PrtScn/ScrLk/Pause keys. Presumably there is some way to do the same thing on other operating systems.