It seems like ALL CAPS shouldn’t be any harder to read than all lowercase letters, right? What’s the difference? Is it just because we’re used to reading a larger ratio of lowercase letters?
Because people read by matching the shapes of words. There is less variation in the profile of words when they are typed in ALL CAPS vs. lower case.
SEE FOR YOURSELF.
See for yourself.
I think there’s a few reasons. The first is that yeah, we’ve gotten used to it. But capitals have an actual function, in making the beginning of sentences clearer, making names stand out etc. If you write in all capitals, you lose that function, and it no longer becomes a useful device.
I would like to know though, if our mixture of upper and lower case was specifically designed because it is more coherent and visually easier. It seems like it’s quite an old convention. It might have something to do with the shape of a word. Upper case is all at one level, whilst lower case gives words a more distinct form. It makes individual words easier to spot.
Then of course, there’s internet etiquette, where the use of capital letters is usually seen as a cry for attention or shouting, and is often used as emphasis instead of italics.
I’m unsure if it appeared elsewhere before hand, but I’d be interested in examples, if it did.
Our court’s computer-generated civil judgment forms all come up in FULL CAPS and it drives me crazy. The longer the form, the more difficult it is to skim with my eyes and to see what it’s about. I have to read it word by word rather than have key words leap out at me. YMMV, of course.
That’s a good question. I do know that a study was done which–if I recall correctly–found that serifs were typically harder for the eye to process than sans serifs. Yet the Times font seems to be more popular–or it somehow “holds more authority” than other fonts. This is probably connected with the tradition in typography which sees fonts as an art form valued for its own sake, over readability.
Times New Roman is not more popular because it is “an art form”, but because it is readable in a wide range of sizes, especially for reading a large amount of continuous text. And I’m not sure how significant it is that “serifs were typically harder for the eye to process than sans serifs”: people reading large chunks of text rapidly are not processing serif by serif, but are scanning whole words and even phrases. For that kind of reading, serifs are just a small part of a big picture.
I’m not an expert, but the impression I get is that upper case and/or sans serif is good for short pieces of text (e.g., signs and headlines), while lower case serifed text is better for body text, just from the point of view of readability, and not for artistic appearance.
I submit that upper case fills more visual space than lower case, therefore causing more strain on the eye. The same reason that a big block of text is easier to read if you break it up into paragraphs and add a little white space.
Yep. Originally people wrote using just lower-case letters, but reading them can be a PITA–for example, imagine the word “minimum” written in small font by hand, potentially with no dots over the "i"s. Adding capital letters breaks things up a bit.
Other way round, actually; in general, blocks of print are easier to read if they’re printed in a serif typeface.
There was a lot of research done on the legibility of typeface back in the '80s in particular.
I’d be curious to see a site either way. I’ve always been taught serifed typefaces are easier to read (as you say) and my personal experience seems to confirm that long stretches of Times New Roman feels like less labored reading than the same in Arial.
Here’s a cite to a site that addesses the serif v. sans serif situation.
ALL CAPS however is undoubtably harder for reasons already put forth. Mainly that it removes “top-down” pattern clues of whole word shape and meaning based on the presence or absence of the cap. We do not sound out most words. Instead we read whole words or even phrases all at once without processing much in the middle. That’s why editing is sometimes so difficult (especially self-editing) - our brains know what is supposed to be there and fills it in despite its lack. All CAPS prevents that process even more than all lower case would.
It’s been ages since I’ve looked at that research, so you’re probably just as well off simply going to Google Scholar and poking around.
The comment re serif typefaces being superior is a generalization (it mostly applies to continuous reading of large blocks of text in a moderate point size–your basic book/newspaper-story scenario); there are areas where there may be no differences or where sans serif is superior, such as small pieces of text, large or very small font sizes, or poor-vision/low-light conditions.
Additionally, there have been some other sans serif fonts developed in the intervening years that are pretty good .
On the serif vs. sans serif question:
The medium in which the typeface is used is important. In PRINT, serif is easier to read. The leading and trailing “tails” and variable stroke weights produce a visual flow, guiding the eye from left to right. Sans serif does not have the tails, and thus the eye slows down. You can see that sans serif thus is more useful for things that should grab attention or for small bits of text, where serif is more useful when there are larger blocks of text.
These principles are reversed when you are talking about text on screen. Because computer screens render text in pixels, traditional serif fonts developed for print use can appear blurry or run together. Sans serif fonts, with their cleaner lines and even stroke weights, thus are more useful for onscreen display.
Now, however, computer screens have better resolution than they did in Ye Olden Days, AND new serif and hybrid fonts have been designed specifically for use onscreen. So, this is less a factor than it was even 5 years ago. Still, something to keep in mind when designing a web page or application that will be viewed onscreen.
As for Times New Roman specifically, it is a variation of Times Roman, which was developed for the London Times IIRC. It’s very old – I don’t remember how old, but maybe even a couple of hundred years. It was designed specifically for squeezing a lot of text in a very small space while still being readable. It is so common these days because, for reasons that are unclear to me, Microsoft made it the default font choice for Word in Windows a million years ago. TNR is a TERRIBLE font for most everyday uses, AND it’s a bad copy of TR at that, just as Arial is a bad copy of Helvetica. There are plenty of other serif fonts that would be fine choices for print uses in common office and personal documents, I wish to heaven people would move to them and away from TNR, but that genie is long, long out of the bottle.
Which ones do you prefer?
I’d be interested in some suggestions here. I suspect that part of the reason why TNR is popular is just that it is a pretty bland type face, which doesn’t stand out in any way, and in many contexts that’s good: you don’t want people asking about the type face, you just want them reading your document. But often I like things to stand out a bit, without losing too much readability.
I thought the opposite was true - the minims came into play with medieval manuscripts, which were themselves a reaction to Roman writing, which was all upper case AND didn’t have breaks between words.
Tehre is eevn a faomus eaxpmle of tihs, wihch I am donug in tihs setnecne, taht deomntsrtaes how our mnid raeds sahpes of wrods mroe tahn the atcual lteters tehmsevles. Msot pepole dno’t hvae mcuh dfificutly porcessnig snetecnes in wihch the mdilde lteters hvae been jmulbed, as lnog as the frist and lsat lteters are peresevred.
It’s donig and presevred! 
This said, I’m not sure I’m convinced. In your sentence what you did was usually just transpose two consecutive letters a few times. As an example, “tehmsevles” is just “themselves” with letters 2 and 3, as well as 7 and 8, inverted. It still looks similar. It could have been possible to jumble the letters much more. For example, “tesvehlems”.
Which I have no problem reading, either. I just didn’t have the time to get so carried away.
DSeid and Giles:
You’re right, if people take notice of the typeface, you’ve probably chosen the wrong one. 
The deal with TNR, though, is that it’s not “bland” – ubiquitous, yes, but not a good choice in a lot of cases because it’s so … constipated, for lack of a better word.
(Note: The below references Windows systems … Macs have different choices and different defaults. One advantage is that Apple pays to license time-honored typefaces, Windows has been fond of creating knockoffs. It’s sort of a new ballgame today, though.)
One thing I forgot to mention is that, with Vista and with Office 2007, Microsoft released a set of new typefaces as well as a new screen rendering technology called ClearType. (See links below at the end of this post.) A lot of the new typefaces in common use start with C – Calibri, Cambria, etc. Calibri – essentially a sans serif face – is the new default face in Office. They did finally get rid of TNR as the default. The new faces are all hybrid-y designed to work fairly well in both mediums, print as well as onscreen. If I were developing a doc where I wanted a fairly formal, traditional look, I’d probably pick Cambria or Constantia. For daily stuff I’m churning out, I’ve found Calibri to be OK.
If you don’t have Vista/Office 2007 and need to stick with old-school with the standard choices, serif fonts I typically choose would be Book Antiqua or Palatino. For sans serif, sigh usually Arial since I wouldn’t have Helvetica. Before Office 2007, I usually used Trebuchet. It was something not a lot of people used, is clean and worked onscreen, and has enough of a suggestion of serifs to work OK in print, especially for the types of documents I tend to produce.
This is from a handout I developed for a class I taught last summer:
Learn More about Typography and ClearType
FOR THE LAY READER
A Comprehensive Look at the New Microsoft (Vista) Fonts
http://neosmart.net/blog/2006/a‐comprehensive‐look‐at‐the‐new‐microsoft‐fonts/
About Microsoft ClearType
http://www.microsoft.com/typography/cleartypeinfo.mspx
Examining the Legibility of Two New ClearType Fonts
http://psychology.wichita.edu/surl/usabilitynews/81/legibility.htm
The Next Big Thing in Online Type
http://www.poynter.org/column.asp?id=47&aid=78683
Wikipedia Article
FOR THE SLIGHTLY GEEKY
Microsoft’s ClearType Font Collection: A Fair and Balanced Review
http://typographica.org/001021.php
Wikipedia Article
Typography in Windows Vista (video)
Interestingly enough, it seems you are right to remain unconvinced. This does appear to be an urban legend, at least in the way it’s normally presented.
edit: There’s also this interesting article about how we process words linked to in the link above. Fascinating stuff. Apparently, the word shape model leaves a lot to be desired and is not the currently accepted model (although it is the oldest):
Fascinating stuff.