Your ideal font, coming right up.
Wouldn’t be nice if this was the only typeface people used?
1 Like
Wow.
Well, ok, it’s true. Well done. You’ve exposed our plan and/or plot. I do feel clever! Enjoy your pain—I will. Mine is an evil laugh!
I really should go join the Evil-AIGA or something. Or Designers who live to grief engineers or IT people (or Whichever) With Our Bad Fonts, or DWLGEITPOWWOBF. Is that an I or a lowercase-L though? Does it even matter?
Hahaha! Nope.
Not a bad one for a fixed size font. Could derive a decent sans-serif proportional font, too.
Some plots work even when exposed. It’s a good idea to favor those in planning.
We have our own people for that. We also handwrite all our passwords in cyrillic.
Edit: cursive cyrillic.
2 Likes
As an East-European I learned that in third grade. Neener neener! 
2 Likes
Now I’m jealous. I don’t know a lick of Cyrillic except that it makes Moscow look like Mockba.
1 Like
Well that’s just great. Somebody posted my password.
6 Likes
How much usability testing has cursive cyrillic been subjected to, I wonder?
Maybe I just like cursive because it’s so pretty (and I can leave off the diacrits). But this does go back to my point a bit: an experienced reader uses context to see the differences in the letterforms. Passwords in your sample? Have none. (Also, if xkcd is correct, they aren’t really much better than “password” for security.)
We used to make wider spacing between the letters to make the inter-letter boundaries obvious. If the papers wouldn’t be too deep in the geological layers, I would be able to unearth my old school writing books.
Do you understand the concept of unnecessary cognitive overhead? Font is a servant of conveying information. If you convey every character properly, you won’t need poor crutches like context.
And then there are the things like machine vision and OCR. Stupid design of fonts can cause unexpected problems. But the “designers” rarely think ahead.
And the problems of transcribing text without understanding it. The computer text recognition, again, but also attempts to retype the word into a translator manually.
Fonts are a servant of conveying language. Your password examples are not language. If the passwords represent a problem, blaming type designers does not solve that problem.
No indeed. They, instead, have created numerous font-based solutions for the issue. The problem is that engineers have failed to implement them. In such cases, designing a process around this particular limitation is your best option. Or not. All good by me.
Language is a subset of information. Claiming that fonts are for language, and not for information, is imposing an unnecessary and onerous restriction.
Passwords and acronyms are information. The latter may also count as language. There are mixed-case acronyms out there.
It’s the same kind of blindness as if car designers lived in a flat country, and did not add brakes because nobody will ever go downhill. And then complained when called about the vehicle crashes in hilly areas.
There are usable fonts. Some. True.
The implementation of the font rendering engines is just peachy. (Okay, it sucks but manageably sucks. Still peachy in comparison with the font designs.) But then the Designers rolled in and specified the default font sets, and in their ignorance and cluelessness chose the wrong ones.
Hanging the designers and rebuilding the world sounds like a good option.
Clearly, we are not communicating here. Let me try again.
Blaming the designers of screwdrivers for the screwdriver’s unsuitability for driving nails is not reasonable.
Screwdrivers drive screws, and fonts display language. Things that are not language may have to make use of different fonts. Fonts are generally functioning exactly as intended, exactly as designed. If your exaptation is not supported, you must find alternatives.
This distinction is the stumbling block. Most things can pass for “language” in one way or another. When you start narrowing it, you end up in a corner.
2 Likes
We are. You are attempting to push an unnecessarily restrictive use of a method for character displaying in order to justify bad decisions in its design.
That’s true. However we’re talking about the inability of the screwdrivers to drive screws, and your attempt to define screws as only a subset of screws with low driving force for which the lousy screwdrivers you’re peddling are good enough.
Here, FTFY.
Fonts are specified for characters, not for languages. (Granted, characters are specified by languages, but this is indirect. We also have a lot of special characters that are not present in languages.)
It is stupid to expect fonts to be used to display only what you call language. Default fonts should be able to display any character in any context in an unambiguous way. Artsy fartsy fonts can be a special case but they should not be anywhere close to the commonly used ones.
If this claim is true, then the intentions and designs are stupid, made from dumb and limited perspective. No wonder, when “designers” are in charge.
Sure, passwords can “pass” for language, but when they’re amongst only their own kind it’s all “did you see the plaintext on Ms. Thang” and “you did NOT just 6 character me in lowercase no symbols?!!”
2 Likes
Putting words in my missive doesn’t actually advance your argument. Yes. Fonts can do that. Perhaps not to your personal satisfaction, but they can. (You can also use a font for making pictures composed of glyphs, like ASCII art. Is that really the primary purpose of Futura or Courier? I don’t think so.)
[quote=“shaddack, post:95, topic:64896”]coherent_light:
Fonts are generally functioning exactly as intended, exactly as designed.
If this claim is true, then the intentions and designs are stupid, made from dumb and limited perspective. No wonder, when “designers” are in charge.
[/quote]
Well now. Since the analogy has failed, here’s an example: spaces.
Spaces are crucial to comprehension and reading (let’s ignore tabs, leading, or indents); spaces are obviously as much a part of the font as any visible glyph.
I’m sure you’re aware that spaces are also valid characters for passwords.
If you’re given a PDF as a record of a password which includes spaces—set in a proportional, serifed font like Times: “pAss[n-spaces]wORD” you have a problem. (N-spaces can be many or few, as you like, or as far as you know.)
We could have required the notification to be hand written, enforced clarification, or obliged the use of a font explicitly designed to accommodate and eliminate ambiguity, but for this oversight with implications for password transcription we’re going to zero in and…blame font designers. Yes, this actually makes sense.
So we need to establish usability testing for the spaces in fonts, too. Perhaps all spaces in all fonts should be a visible glyph that guarantees unambiguous passwords—by default. In spite of centuries of prior art of fonts as elements of media in language, font designers are to blame for this unforeseen complication of security engineering. Anything less is bad design.
Yeah, obviously, I don’t think so. “Fixing” spaces in fonts in order to satisfy a novel need that passwords be universally legible regardless of font, usage, or context—I think that—perhaps—would be bad design, “made from dumb and limited perspective.”
There are ancient examples of typewriter ASCII art. It seems to be a natural outcome of existence of a fixed-width font.
Now you’re grasping for straws, and clutching just empty space.
Yes, and this can get fairly complicated.
Whitespace character - Wikipedia
Indeed. Rarely used, I didn’t see it in any password any service sent to me, I don’t use them in my passwords generated for my users because they are confusing, but indeed they can be used.
All sorts of numerical-code-entered ASCII and unicode douchebaggery can also be used in passwords. That does not make it a good idea. Or maybe does… if it is for your own use and doesn’t involve the plebes and techsupport calls, and you won’t run into character-entering method compatibility or input filtering issues.
And that, dear children, is what the ␣ symbol is used for.
When you need to explicitly write a space as a space, use the ␣ entity, aka unicode U+2423. (There is apparently also the U+2420, aka ␠, but I never saw it in actual use. In editing, · is used to visualize spaces when needed.)
But you usually just don’t use space in passwords communicated to the general public.
That’s not really true. Many typefaces incorporate ligatures, which combine one or more graphemes. The rules which govern which ligatures should be displayed are often language dependent-- Swiss German, for instance, does not use ß.
See this article for more details
2 Likes
Is there a Falesehoods Programmers Believe About Fonts comparable to Falsehoods Programmers Believe About Names and Falsehoods Programmers Believe about Time?
(For Programmer you can substitute the superset Engineers)
2 Likes
