Slashdot Log In
Edward Tufte Weighs In on Apple's iPhone
Posted by
Zonk
on Thu Jan 24, 2008 04:00 PM
from the word-from-on-high dept.
from the word-from-on-high dept.
An anonymous reader writes "Via Daring Fireball, a post from design guru Edward Tufte's site discusses his views on the interface used by the Apple iPhone. The post includes a video presentation by Tufte on the subject of video resolution on the phone. His argument is primarily that while the iPhone does a lot of things very well, Apple hasn't quite realized the platform's full potential by making screen real estate all it could be. "
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Design decisions vs. 20/20 hindsight (Score:4, Insightful)
Chinese saying -- step too far, fall on face. A little more familiar is the phrase "perfect is the enemy of good". Attempting to release a 1.0 product that has everything absolutely perfect and without compromise is the surest way to never ship.
Perhaps iPhone2 will address some of these issues?
Re:Design decisions vs. 20/20 hindsight (Score:4, Interesting)
Parent
Re:Design decisions vs. 20/20 hindsight (Score:4, Funny)
I really hope this is not true.
Parent
Re: (Score:3, Funny)
Re:Design decisions vs. 20/20 hindsight (Score:4, Insightful)
You ever hear a baseball fan recite numbers? that same guy who has a hard time understanding what a directory tree is can rattle off 30 years of numbers, statistics and tell a surprising amount of information about their team.
Parent
Re: (Score:3, Interesting)
My mother shies away from the DVD player we got her... but she can look at nearly any 3d object, suc
Tufte didn't seem to be bagging on the iPhone... (Score:2)
Re:Design decisions vs. 20/20 hindsight (Score:5, Interesting)
I'm sure when you first saw an OS GUI thoughts went through your head on how to improve it. Perhaps YOU should have thought of something more novel than critiquing previous works.
Parent
Re: (Score:3, Insightful)
Re:Design decisions vs. 20/20 hindsight (Score:4, Informative)
Did you read on below the video?
The mock-ups are included, too.Parent
Re:Design decisions vs. 20/20 hindsight (Score:4, Insightful)
That's a stretch. My understanding is that to record video with an iPhone, you have to hack the thing. I don't know what other oversights were made but a new high end phone with a video camera that won't record a video clip in 2008 is a joke.
Parent
Re:Design decisions vs. 20/20 hindsight (Score:5, Interesting)
Besides, it's always easier to critique someone else's work than create something novel yourself.
I'd call both sparklines [wikipedia.org] and the data-ink [infovis-wiki.net] ratio [tbray.org] pretty good and novel innovations.
You can't credit the man with "creating" information design as a discipline, but he's done a great deal to evangelize it, and you certainly have to give him plenty of credit for its currently elevated profile.
Tufte is not just some crank. Intelligent, useful, compelling information display is what he's all about. You don't have to agree with him, but his thoughts are usually worth weighing.
Parent
A different take, ``good enough, isn't'' (Score:5, Insightful)
For my part, ``good enough, isn't'', and I far prefer the Zen parable of the archers --- three archers compete for a prize, all strike the mark, a fish, and are then asked ``At what were you aiming?''
The first answers, ``The fish.'' as does the second, but the third?
``The center of the fish's eye.''
You can't be any better than you try to be and I'd much rather wait for the efforts of a person striving for perfection than accept those of someone willing to be mediocre.
William
Parent
good & detailed constructive criticism (Score:5, Interesting)
Personally as a product developer myself, I would welcome such good detailed constructive criticism for free from a UI guru such as Tufte. Remember that there are all innovation is based on prior innovation, so it is good to have analysis done on existing products in order to improve on future versions.
BTW, on a side note, I hope that someone at Slashdot deletes the offences racist postings above.
Re: (Score:3, Interesting)
Re: (Score:2)
Tufte is from the old school of paper displays... (Score:4, Insightful)
But with dynamic displays, the game is all about minimizing the amount of retrieval time, not space. Users can tell the computer to pull up a graph for a new stock, or scroll the page downward with their finger to view the info under the buttons, or completely off-screen, with minimal effort. The biggest limiting factor is interaction. Let's keep the buttons visible, because they enable far more information than they hide.
If we sacrificed usability for screen real-estate, we'd end up with marginless documents and 4-pixel icons, which incidentally would look like windows mobile.
Re:Tufte is from the old school of paper displays. (Score:5, Insightful)
When you are trying to browse a web page on a screen that is an order of magnitude smaller than what the author expected, it is absurd for a full 10% of that precious space to be permanently devoted to a mere 4 buttons, only one of which sees frequent use. In the case of the stocks, once the user has selected what they want to know about (be it a single stock or a set of stocks) it makes sense to display as much information as possible about them. After all, the user has already asked for the information. The only reason to leave relevant information out is if it won't fit without sacrificing the readability of the report. Tufte has never failed to understand that point, and he certainly didn't leave it out of TFA.
You are right that paper's primary limitation is space, and that this is not the case with digital displays. This is not because the digital displays are less limited in space (they never are, and in this case, the computer display is downright tiny). The reason is that the resolution of digital displays is so much lower than that of paper that the overall size doesn't really matter anymore.
Parent
Re:Tufte is from the old school of paper displays. (Score:4, Insightful)
Parent
Tufte is cool, BUT... (Score:5, Insightful)
I liked the video as well, with a couple of exceptions:
1) In the video, Tufte has to bust out his Sparklines (the infographics that look like lightning bolts that he mentions in the section on stocks.) He claims these have thousands of pieces of information in them but the reality is that they're merely zig-zags. As the inventor of the sparkline, Tufte thinks they're the be-all and end-all of I.D.
2) I found it hilarious when Tufte showed how he would redesign the Weather program to show more information. He said something on the order of, the only bad information design is that which leaves out important information. Sorry, holmes, I don't need to see a time lapse of cloud patterns. The Apple weather design is elegant and succinct, yours is crowded, ugly and excessive.
Re:Tufte is cool, BUT... (Score:5, Insightful)
Parent
Re: (Score:3, Insightful)
Tufte's still working with sheets of paper. (Score:5, Insightful)
For example on the stock market page, drag stocks over each other to compare them, dragging a stock all the way to the top of the page would give you more information on that stock and let you drag the screen left or right to get other stocks, flip it sideways to get the graphs, and drag left and right to compare with other graphs.
On the weather page, use the same approach, and flip sideways to get the weather map, drag up and down to see different maps.
A video screen isn't a static device, and you don't need to cram data into a single static view. Data clutter is as bad as administrative clutter.
Parent
Re: (Score:2, Interesting)
I have to disagree (Score:5, Interesting)
Part of the reason that people BUY the iPhone is that it's simple and stylish, rather than the existing information heavy devices like Pocket PC phones. In particular, look at his example about the Weather- Apple's widget is small and sleek. It shows you the vital information, and it does it in strong fonts and bold styling. It's clear, and it's easy.
He squishes all of that information into a tiny corner, so that he can add a large repeating satellite view- Sure it's useful in some cases, and it's certainly a neat demonstration of the iPhone's abilities, but it fails when it comes to the task of quickly giving me the important information.
It makes me squint to see the tiny version of the temperature, and shows off, rather than helping.
Sometimes developers fall into the problem of working so often because they can, not wondering if they should.
Note- He dismisses this argument, saying that information density isn't the problem, it's laying it out clearly. I agree with him in general,in that complex information can often be presented simply, but in most of his cases, increasing the density would diminish it's usefulness.
Oh, the irony! (Score:5, Insightful)
How is it exactly that, in the same page where he tells us "Better to have users looking over material adjacent in space rather than stacked in time.", he puts most of his information in a fscking video?
Re:Oh, the irony! (Score:5, Insightful)
And above all, I have no interest in taking MINUTES to have information spoon-fed to me in real-time. I can read orders of magnitude faster than I can listen. And if I'm reading something that doesn't interest me I can easily skip ahead, because any decent author will include titles, section headings, paragraph breaks... and other cues to allow skipping ahead and finding the interesting parts.
Parent
Re: (Score:2, Insightful)
Tufte's books concern interactive interface design. This primary purpose of this video is not to be an interactive interface. Instead it is a pre-recorded ('stacked in time') presentation designed to demontrate specific features of the iPhone's interface.
For example, the ability to touch screen and drag the iPhone's screen 'real estate' and the subsequent transition between interfaces, can only be demonstrated effectively in a video like this.
Note this v
Re: (Score:3)
That's the point: Its a terrible interactive interface for the purpose of demonstrating specific features of the iPhone interface. It crams the entire set into a linear interface stacked in time without even so much as a chapter selection function, leaving
Mirror (Score:3, Informative)
armchair UI ideas (Score:4, Interesting)
Tufte makes a good point about the hidden potential of iPhone's brilliant display. But I feel the answer lies less in resolution, and more in depth. We have been exposed to much web content that is layered (for example, pop-up windows that appear on top of existing screens that fade into the dark) that we can now discern depth on a 2D picture provided it is clear, sharp and bright. There is this 3D real estate that is not exploited in iPhone (and something that it is quite capable of).
As an example, I sometimes find it a tad annoying to keep going to the Home screen on the iPhone when switching between applications (typically when I am viewing a website and quickly need to look at maps). A dock with all Home icons down the side that appears overlaid (and magnifies each icon on fingerscroll, just like on a mac) would eliminate the intermediate step of going to the Home screen. To take it a bit further, the Maps can open in a 75% window on top of the Safari, so we can get back to Safari by one fingerstroke (Tufte's idea would be to open two windows each 50%, because there's resolution). This is, as you can see, nothing new - just something that iPhone doesn't currently have but can quite possibly do.
MIRRORS! (Score:3, Informative)
Also, I've mirrored the video, as that was the slowest loading element of the page, here:
http://g.appleguru.org/iPhone_Resolution-desktop.m4v [appleguru.org] (58MB)
WiMax? (Score:2)
3G? (Score:2)
Re: (Score:3, Informative)
Re: (Score:2)
Re: (Score:2)
better use of the screen? (Score:3, Insightful)
now people want to clutter it up.
text is too small (Score:4, Insightful)
Tufte... (Score:5, Interesting)
"to clarify, add detail" (Score:5, Insightful)
"Drill-down" is automatic to the eye (Score:3, Informative)
Tufte's view, consistent for decades, is that the information display should be designed around the human visual system's abilities and preferences, not the designer's prejudices or what's easy for the display system.
The human eye automatically "drills down" in an information-rich visual field by focusing the fovea on anything that is noticed as being of interest. Further information on the subject of interest is gained in a dozen milliseconds by the act of focus. No jumping to new pages over a second later.
A couple of posters offered the absurd assertions that
a) Tufte is stuck in the paper era - when he's been commenting on computer displays for 20 years. His criticisms of the screen real-estate forgone to 'computer administrative debris' in Mac and, later, Windows, go back to their inception.
b) That space is limited on those paper pages when they are far more information-rich than screens. Multiply 8.5x11x300x300 and get over 8 Mpixels, guys. (And an open magazine is twice as big; an open newspaper, 10x that!) Why do you think most people prefer to read on paper even now? Richer colours, too; compare TIME print edition photos to the web pages printed out.
People who think information-thin combined with drill-down is the way to go are responsible for those frustrating answering-machine menus.
And definitely have never taken a look at Craigslist, where there are a maximum of index words per page, using smaller print, and every piece of information in the index is also a 'control'. a link to another information-dense page. You rarely have to go more than three clicks in until you are looking at a list of the things you want, out of all the country and all the products and services there are.
Bottom-line: provide the user with as much information as possible, use visual cues (size, colour, position) to prioritize, and have trust that they will pick out what they want. Providing them with less information so as to lead them by the nose down your little trail insults their intelligence and human abilities.
He's a bit full of himself (Score:4, Interesting)
I get the distinct feeling that Tufte understands data visualisation, but not interface design. These are different things, and he's letting his expertise in one area make him think he can make pronouncements from on high in other areas and comes out with some real bullshit as a result.
His "to clarify, add detail" rule could be applied to his comment on the photo browser. He says they should be grey not white, and only one pixel wide, but gives no reason why. I'd like some detail to clarify why he says that! It would not fit more images onto the screen, it would add no information content, it's barely even an aesthetic change to the design. It's news to me that arranging images against a plain white background is a bad approach. I've met a lot of smart people that like to "show off" by making detailed comments like this, without any actual substance or empirical evidence to back up what is simply their own preference. Tufte seems to be doing so here.
He criticises the stock app for being "cartoony" and "PowerPoint" like, which seems again a mere preference rather than an objective comment, uses words designed to provoke an emotional reaction rather than an intellectual one. He claims his app has more detail - which of course it should when it only has three stocks, not six. But I don't see how x thousand points of data points in a tiny little graph is of use. First of all, if you fit thousands of data points into a single graph, it's going to need a damn big piece of paper before I'm capable of distinguishing them, combined with a ruler and a set square if I want to get the value for a specific data point. Second, why would I want this level of detail on a phone app? Personally, I find the iPhone's red light / green light view combined with percentage points useful - it jumps out at you when e.g. the market crashes as it did recently. In Tufte's example, it's impossible to tell what recent market changes have taken place, and there is no obvious way to quickly see data for e.g. the last week. The "modest data graphic cartoon" conveys just as much information to the viewer as his "image resolution" with thousands of data points, and is the kind of thing a portable stocker checker would be used for. Tufte is letting his expertise get in the way of understanding the use case - all his catchphrases are there for the converted, but his use of them here just annoys me.
Here's a nice little piece - take a look at his site at http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=00036T [edwardtufte.com]. He criticises the iPhone browswer for having 10% of the screen used for buttons, but in his own designs he comments "about 90% of the image is substance". Clearly he's happy with that 10% sacrifice when it's his own work. And if you look at the designs, you'll note that in each case there is a navigation bar of some form at the top or bottom of the page. What a hypocrite.
Finally, he's very keen on getting rid of computer admin debris. The problem is, he treats looking at a web page the same as looking at a picture. But when I'm looking at a picture, I don't want to bookmark it (it's already in my collection), and I don't want to make a webclip of it. I don't need the back button with photos, because I can navigate via the photo collection. But I do need those functions in the browser, and I need them large enough to easily hit with my finger. We're all used to scrolling down webpages, so having a mere 90% of the screen available, and an intuitive flick of a finger to scroll down, is perfect. Commenting that the button bar should at least be transparent strikes me as just one of those condescending little compromises some people like to make when they know they won't convince the other side of "the right answer". It would be bad interface design to have application buttons hovering over hyperlinks, making it distinctly ambiguous what would happen when you touched that bottom 10% of the screen.
In particula
Re: (Score:3, Funny)
Re: (Score:3, Informative)
From the Fine Article:
"The iPhone platform elegantly solves the design problem of small screens by greatly
intensifying the information resolution of each displayed page. Small screens, as on
traditional cell phones, show very little information per screen, which in turn leads
to deep hierarchies of stacked-up thin information--too often leaving users with
"Where am I?" puzzles. Better to have users looking over material adjacent in space
rather than stacked in time
Re: (Score:2)
if you've made your mind up to buy now, fine, but don't bitch when the new ones comes out "zomg Apple hates customers I had teh no ideas things didn't not change!!!1" like so many others do.
Re: (Score:2, Interesting)
Re:Screen resolution should be increased for sure (Score:5, Interesting)
Just because an average human eye can tell apart a 600 dpi print from a 1200 dpi print doesn't mean it is more 'usable'. 1200 allows fancier or more elegant fonts (like a subtly waisted Optima) and nice printouts, but I have never met anyone who would have printed all his texts at 5 pt size just because his printer could.
UI design regarding resolution is mainly about legibility. Using sub-pixel anti-aliasing and optimized fonts you can get excellent, and I mean very excellent results below 200 dpi. Anything higher can show more detail in theory, but not more significant information. Nobody would chose a waisted font over a optimized one on a portable device, so why waste money (which could be used for other features) on displays which -could- display them.
Parent
Re: (Score:2)
Re: (Score:3, Interesting)
Because it is a selling point even though most people can only barely see the difference over 300 dpi. That was the resolution of the first laser printers, and most people thought they looked as good as typeset. And that was for black and white printers with no way of controlling dot intensity. With a screen that can produce halftones for antialiasing, it is hard to see much i