fadeyev

FontSquirrel's generator will strip out a lot of unneeded characters. By default this means including just the basic Latin subset with the main punctuation glyphs. If you look at the character set for Clear Sans (http://www.fontsquirrel.com/fonts/clear-sans) you will see that it includes Cyrillic and Greek letters, among other things like musical notes and smilie faces, so those can be removed if you don't want them (or included if you do).

I think his point is that low DPI devices don't render detailed fonts well so we're forced to use fonts that are optimized for the screen (Verdana, Georgia, etc) if we want to see decent results (i.e. not a blurry mess). Because these fonts are optimized for low pixel density they tend to be pretty simple and boring, or at least not as rich and expressive as they can be if there was no resolution constraint. As a result of using simpler fonts designers compensate by adding additional details to the interface in the form of skeuomorphic effects. Yes, the effects also often communicate function, but they also act as decorations that make the interface more interesting to look at.

Now, on HiDPI displays we can go back to using more detailed fonts that we use in print without loss of quality, so the text part of the interface gains a lot in aesthetics and character, so much so that we no longer need all the stuff around the text to make it look good.

Additionally, because HiDPI displays are able to show that much more detail, it is a lot more difficult for designers to create skeuomorphic interfaces that look believable. The crudeness of the old displays helped mask the simplicity and lack of realism of the skeuomorphic design. Our eyes and mind knew that what was being represented is not real, just an impression, and so we evaluate it as an impression, not as something real. On contrast, HiDPI resolution is so high that what's being represented actually looks real, but if the smaller detail is off the effect of realism is lost, so the overall thing ends up looking like a poor fake rather than a good impression.

Hey Eric, I'm the author of the linked post. I actually do mention that light fonts on dark backgrounds look terrible on OS X and suggest using the antialiasing mode as a fix. Dark text on a light background is another story however.

The margin under the subheading should be the same (or at the very least not smaller) than the bottom margin of each paragraph, otherwise you end up with the subheadings looking like they're attached to the paragraphs that directly follow them, creating a false hierarchy with a single paragraph rather than a selection of paragraphs. If you want to give each section extra separation increase the margin above the heading instead of decreasing the one below it.

Not an expert in typography so I may be wrong, but I'd say it's a modern take on Humanist (also known as Venetian), one of the earliest Roman typeface styles. Here's a good article about this style: http://ilovetypography.com/2007/11/06/type-terminology-human.... I'm not sure if it can be classified as such though because it is built to work well on the screen and so has a lot of straight lines, closer to Transitional typefaces like Times New Roman, but even though the overall design is very modern, the feel of it seems to me closer to the Humanist typefaces. For example, the bar on the 'e' is sloping a tiny bit upwards, which is a distinctive characteristic of Humanist typefaces.

Thank you for your comment saraid216.

I realize full well that I am prescribing my own moral framework, but that is the second part of the main idea, the first being that there be a moral framework in the first place. I emphasize this split because this way we can discuss: 1) whether or not morals play a role in design decisions, should they, and if they should then to what extent, and 2) what moral framework to use (i.e. how to live). Obviously the latter invites a very long and difficult discourse better suited for a moral treatise than a blog post or a few comments, so by taking it aside I focus on the former point which is more relevant to the discussion at hand.

Hey AshleysBrain, author of the post here. Thanks for your comment. To address your point:

The users' enjoyment isn't really the issue. It is not immoral to play the game, nor to pay for it, nor to have fun doing it. The issue is whether the motivation of the creator are to build a product that gives their customers value, or a product that is effective at making money (there is a large distinction here). If your product provides entertainment, it is pretty much essential that your customers enjoy it, but the key is whether you abuse their enjoyment by turning it into addiction (which you can then exploit for your own monetary gain, e.g. drug dealers), or whether you aim to give them the best entertainment you can while keeping their consumption of it in moderation -- in other words: your primary motivation for making the product is the product itself, not the monetary reward.

I posit that Zynga falls into the former category, that is: their primary motivation is to make money (and I may be wrong, but this is not a wild assumption), and they achieve it by developing an addictive product. I also posit that this leads to overconsumption on the part of the users (again, not improbable). Making money isn't immoral, but exploiting another's loss for your own gain is. I am prepared to argue further about the specific example, and perhaps in doing so we may discover that Zynga do actually care about making a really great product, in which case I will be happy to retract it, but the general principle for which the specific example is given still stands.

At this point some will make an argument along the lines of: the responsibility for moderate consumption falls on individual users, not on the creator of the product. Now this is a very important point, but it is by no means a resolved question. I side with the camp that says that some of the blame does fall on the creator because it is only through the combination of their intentions to make the product addictive and the user's weakness that overconsumption can occur. My argument is that by choosing not to make addictive products -- that is, to make moral design decisions -- we can make products that benefit society rather than cause it ill.

This is not some moral condemnation of the creator, far from it, for I care about the creators just as much as I do about the consumers (I am one, after all), and as much as the question is about what products are good for society, it is also this: What products should we be building? What products will truly matter? What design will live on and be cherished after we die? Resolving those questions will help us develop a society we all want to live in and be proud to be a part of, both as creators and consumers.

Author of the post here. Thank you for your comments.

I'd like to address the main criticism that I see in the comments, which is that I did not specify what it is that I mean by morality, and what moral system I'm using to make my judgement.

The post is not meant to be a complete essay, rather it is a short outline of an idea, much like the rest of the stuff I publish on my blog, most of those posts building on each other like bricks rather than standing alone as a finished structure. That's absolutely my fault and it doesn't help with public consumption, and this is something I need to work on. I am, and will be, writing more thoughts on the subject as the post in this submission hardly covers much ground, but nevertheless, I do stand by the idea presented there.

To address the actual point: I do not prescribe a specific moral framework, only that there be one. Morals can come from religion or philosophy (or both), and they should be used to guide design decisions rather than those design decisions being made in isolation. Design decisions made without a moral foundation focus on effectiveness alone, which is not a useful indicator for the good's true worth (again, assuming your life is not amoral, in which case none of that would matter), and what I mean by "true worth" is how that good benefits society and man in light of your moral framework.

Thanks for pointing this out, you're right that the example is now outdated, though it's probably not too difficult to find more browser specific code to pick on :) I'll change that.

Hey aditya42,

I'm the creator of LESS Elements. Nothing is wrong with Bootstrap, it's just that when I wrote these mixins a while back Bootstrap wasn't yet available.

Thanks for the feedback! Glad you're finding it useful.

Name should be pronounced like "Use Aura" but without a gap. It's a play on the word Aura, combined with User, or Usability, as the heatmap is a bit like an aura.

There are a couple of angles I see for a paid version:

1) Beef up the features by adding more tests types (e.g. a preference test like: "Which do you like more: A or B?" -- like a poll but visual) and more customisation (e.g. ability to chain tests together, easy way to compare test performance for iterations, priority in the public test pool etc.).

2) Hook it up to Amazon's Mechanical Turk and basically get instant outsourced testers. Perhaps you can even do some demographic filtering here, although I don't have any experience with Amazon Turk yet. This option can probably tie in with 1) since the single page tests right now are probably too simple to pay for--you'll want to get more value out of each one, so getting a chain of tests with comments will make more sense.

New project I'm working on. Wanted to make a site that's as simple as an image uploader or a code paster, but for click testing. You upload a screenshot of your interface and ask people to click somewhere. The app will generate a heatmap of the clicks, as well as show you how long they took.

Would love any feedback or suggestions. Also: would you find a tool like this useful? Would you pay for it given more functionality? Cheers.

It hasn't stopped working, it just behaves differently. It stopped working on shadow blur amount of 0, but it works at 1px or more.

Yes, but the whole point of subpixel rendering is aesthetics. Its job is to make fonts look smoother and crisper, and in majority of cases it does just that. I'll accept that OS X's rendering of bold fonts on dark backgrounds is terrible, which is where the text-shadow trick comes in handy (or antialiasing, which I think is the same thing in that case), but I'm not sure that for body fonts this is the right way to go. I see this technique popping up here and there and in many cases the fonts end up looking worse. Sure, it's up to the designer to decide what's best--it always is--which is why it's good to have this discussion so that people can better understand which choices to make.

Conventional CSS

LESS or Sass are great in specific use cases. They're especially useful when you're using a lot of CSS visuals (where you have to write out about 3-5 lines just for one property because each browser has it's own version, i.e. -moz-, -webkit-, etc). With LESS you can create a mixin that does this thing (e.g. rounded corners) that also takes an argument and just call it in all the places where you need rounded corners. Sometimes you want to re-use the same color over in many different places. If you change your mind, you have to go back and change it everywhere, making sure you don't miss a spot. Nesting makes the code much more organized (you have to try it to see) since you can see which bit of code relates to which area of the page. It's just neater and easier to view. Each of these benefits on its own isn't very big, but when you start saving time here and there they do add up, especially on large stylesheets.

Basically, it's not really about reducing syntax or size of the file (which does get smaller for large stylesheets), it's all about making the stylesheet easier to build and maintain.

Less.js will cache the CSS on modern browsers after the first load using HTML5 local storage, so subsequent loads aren't an issue. First load is also very fast and won't make a huge difference. Like you, I wouldn't normally recommend running the compiler live but the performance here is really impressive. Please don't assume how fast Less.js is before you try it.