The Stanford Poynter Project Study 96
sredding writes: "The Stanford Poynter Project has some interesting conclusions after a study of Internet news readers. 'Two years ago Stanford University and The Poynter Institute researchers began collaborating to learn how frequent Internet news readers went about perusing news online.' It's an interesting read for Web designers." Cool info and interesting statistics, especially the one about how people jump for text first, not pictures. Take that, Mosaic! Lynx forever! ;)
Re:Eyetracking Study in Alertbox (Score:1)
Man... I need to pay more attention to AlertBox. Nielsen's abridged version is much easier to read than mucking through the Poynter page.
Wired also has a good article [wired.com] on the study.
In the Wired article it states, "The number of surfers who click on ad banners has already dwindled to an average of about two-tenths of 1 percent. If readers don't even look at graphics on a Web site, that number might not go up anytime soon."
If this is true, maybe we'll see a decrease in the use of banners.
A word about Lynx (Score:2)
OT: Monitor size (Score:1)
#ifdef RANT /* RANT */
Noooo. Why do people always do this? A 21" monitor will give you around 211 square inches of screen real estate. A dual headed setup with two 17" monitors gives you around 277 square inches. That's 30% more usable screen space. Two 17" Iiyama VM410 Pro monitors are also 25% cheaper than the Iiyama 22" VM510 Pro. That's more than enough to cover the cost of a dual headed video card (like the Matrox G400 MAX) or a second regular video card. So tell me again why you want a 21" monitor? Of course, if you can afford it, a dual-headed setup using 21" monitors would be rather nice...
#endif
Re:Lynx (Score:3)
Lynx (Score:2)
---
Re:news = usenet? (OT?) (Score:2)
No, you're not. I read the word "news" and immediately thought usenet.
The internet was so much nicer when the general population wasn't on it.
I couldn't agree more. I used to long for the day when everyone would be online. It was a very civilised place to be, and a nice way to communicate with others. Well, now they're all here, and I wish they weren't. The unwashed masses have changed the character of the net. It's not the same place it used to be :-(
Re:Width of text (Score:1)
--
:wq
Re:Banner ads (Score:1)
Mind you with adzap [zip.com.au] running on squid I am rarely confronted by this issue.
Re:I told you so!!! (Score:2)
Donny
Re:news = usenet? (OT?) (Score:1)
BTW, be careful what you wish for. I still read news and I'm really pretty happy that most clueless newbies don't. And when they do show up they usually end up needing a flame retardent garment after they've blundered their way into a group without reading the FAQ.
Re:Whatever happened to beer? (Score:2)
--
I blame literacy myself (Score:1)
The (western) world was indeed much nicer when all the books were in Latin, Arabic, and Greek. ;-)
Kept out the unwashed masses insisting on membership of a proper guild or showing your lineage to get access to a library, things really went downhill when they learnt that printing lark from the Chinese, all sorts of handbills and flyers and other such spam all over the place...
Study and GUIs ignore the visually impaired (Score:1)
Like most GUIs, the study ignores an obvious group of news-users who cannot easily physically read the online news. Most web sites make life only more difficult for the visually impaired, who rely on audio interpretations, Braille readers, or extreme magnification. (Not to mention anyone who doesn't have the bandwidth or hardware to handle your typical flash-bang site.) Not enough people know about the NIST standards for multiply-accessible web sites, or even about the Linux adaptive technologies efforts. How many installathons have someone ready to answer questions about making Linux work for everybody? Even geeks can have physical differences.
And while I'm ranting, sign your organ donor card. Your parts have a much longer shelf life than your computers' parts.
Reading on the web (Score:1)
The early eye-tracking research is mentioned in this article [useit.com]. I used to have an url for the study, but I can no longer find it. The article just referenced berates user registration forms, or at least it says you should delay registration, says that most web advertising is a waste of money, and the best (most cost-effective) way of attracting users is through affiliate programs.
Probably most interesting and useful is this 1997 article [useit.com] on how users read on the web. It gives great advice on laying out content in a way that users will be comfortable with.
Enjoy!
J
Websites "moods" (Score:1)
On the other hand, my company's website promotes "calmness." Lavenders, softer colors, very little black, and integrated, soft pictures. People comment that when browing our e-commerce store (3000 lines of Perl written by myself over 2 weekends, yikes!) they feel very calm and "safe" and ready to make purchases.
Very interesting stuff.
Hey, anyone wanna bring back ing text?
;)
Re:Multiple windows (Score:1)
"We will run this with the same kind of openness we have run Windows,"
that headgear warps the data (Score:3)
OK, here's the Poynter Institute bolting a metal frame onto a browser-user's head and monitoring his every eye motion. Don't you suppose that all that intensive surveillance might just possibly have some small effect on the user's behavior? Like, all alone, without the head gear, what could be more natural than if he might have headed on over to playboy.com [playboy.com] to spend a few minutes or even hours grazing among the bitmaps [playboy.com] and multimedia [playboy.com].
Now imagine that this same browsing subject has a birdcage studded with electronic doodads bolted around his skull, and a variety of all-too-serious sociologists peering over his shoulder. I'd guess that under those circumstances our lab rat will spend a good deal more time perusing something serious and scholarly like that lighthouse for the investment class the New York Times [nytimes.com], or even our own grave and stately slashdot [slashdot.org].
Yours WDK - WKiernan@concentric.net
Usability (Score:1)
So they might be right.
__
Re:Lynx (Score:1)
Banner ads (Score:1)
Utter rubbish. I never look at them. I do feel an urge to rush out and buy SuSE 6.4 for some reason though.
Multiple windows (Score:3)
Re:Multiple windows (Score:1)
I always find, oh, an interesting link, -> new window, I want to read this reply and its subcomments, -> new window.
Maybe it's because of my latency but I am always opening new windows. When I end (or get bored)with the current one, I pass to the next one, which, by that time, should be fully loaded.
__
Re:The Stanford... (Score:1)
My guess is that (at least for news sites) readers (viewers?) tend to assume that graphics are ads, or don't help much. Honestly, most news photos, while interesting, don't actually add to the story (then again, my mind is more verbal than visual). This is reversed if I know I'm going to a site like UserFriendly or Fluble, where the primary interest is a graphics file.
Re:Multiple windows (Score:1)
Re:Multiple windows (Score:2)
But I use two or three windows all the time. It allows me to read an article and the link in it at the same time, or to go back to the original article in a way that guarantees that it's still in the same state.
Re:An experiment (Score:1)
Yes, but sites that are dependent on ad revenue take advantage of the way that browsers render a page. Load order is very important to keeping some advertisers and media buyers satisfied. If you can find a way to load two things immediately, marketers and ad sales people will cheer (well not really, but they will not say as many nasty things about you). Those two page elements are: the "branding" (your logo, and major partners' logos); and the revenue-generating ads.
So, if you go to a news site and spend a second or two staring at a banner because it's the only thing on the screen while the rest of the page loads, it is probably by design. Someone tweaked and played with the page design in order to get the browser to load things in the order they wanted to. Of caourse, this can happen by accident too, but all of the times I've been involved with designs that render this way, it was a matter of choice, not accident.
Re:sig translation (Score:1)
Re:Pictures load last... (Score:2)
Re:Whoo hoo! (Score:1)
Re:The Stanford... (Score:1)
None of these have any useful function to a person interested in reading the news that is on the page.
Re:Self dilusion (Score:1)
Re:Multiple windows (Score:1)
If I'm at home, as soon as I see the modem activity drop to 0 for more than 10 seconds, I'll assume everything's loaded & cut the link.
So, first thing in the morning, I'll load up
Then I'll have around 20 browser windows open. I'll read one or two things to start with, and then work some, and read the rest when I'm waiting for builds to finish, or having a snack, or 'percolating' to problem solve, etc...
This is on top of MS Dev Studio, Netscape messenger inbox, MSDN, MS VSS, Delphi, Explorer, Bash (under cygwin) plus whatever else I happen to be doing.
Alt+Tab is my friend. As are tooltips.
Re:Multiple windows (Score:1)
Re:news = usenet? (OT?) (That's what I thought) (Score:1)
Re:Self dilusion (Score:1)
PostalNews.com, not Slashdot.org (Score:2)
Re:Multiple windows (Score:1)
Now that I think about it, I almost never left click a link... If a page is mostly content, I read, then close it. If it's full of links (front pages of sites mostly), I go through it, opening new windows...
Re:Self dilusion (Score:1)
Self dilusion (Score:4)
picture with an awfully small sidebar of text.
Lynx (Score:1)
Whoo hoo! (Score:1)
Not images??? (Score:1)
-- The Hollow Man
connection speed..... (Score:1)
I'm still stuck on a 56k modem... so on just about any page I look at... the text renders before any images....
Therefore I look at the text first.. since it is there first... But I still look at the pictures..
Link to full screen attention grabbing data:
http://www.poynter.org/eyetrack 2000/toc/frontpg.htm [poynter.org]
I told you so!!! (Score:5)
Ha!! I told you so. I go to read the articles at playboy.com, not look at the pictures :)
Re:Self dilusion (Score:2)
The article broken down into 3 or 4 pages with the graphs and data tables inline would have been much better...
No reason I need to scroll 20 times to read a single article.. (and guess what.... I didn't)
Eyetracking Study in Alertbox (Score:3)
The Poynter Eyetrack study was discussed, with interesting commentary, in May 14 edition of Jakob Nielsen's Alertbox [useit.com].
Direct article link (Score:3)
Eye trackers.... Hm...... (Score:2)
A way to get more specific answers than one can from videotaping is to track eye movements. So we began another study. This one used eyetracking equipment that recorded where the eyes stopped to absorb information. That tells us what our subjects read. We also could track movement from site to site.
News watching log digest for user B1FF:
TEXT: In this latest issue of Sports Illustrated, we have Natalie Portman modeling the latest bikinis, and
Follows link, eyes hover over pictures for 100 minutes.
TEXT: In today's 1M A K001 D00D cooking section, a new recipe for hot grits!
Follows link, eyes follow text at rate of 2 words a minute.
TEXT: H0W T0 B3 AN 3133T HAX0R revals the pros secrets on how to turn your computer on!
Follows link, computer powered off 30 seconds after eyes are on main article page.
Suppose you were an idiot. And suppose that you were a member of Congress. But I repeat myself.
Pictures load last... (Score:3)
Oh well, why can't they waste money in my direction? =-p
Re:Width of text (Score:1)
graphics vs text (Score:3)
This said, the study doesn't really make a point against embedding text in a graphical environment. No matter where people look first, the overall impression of a page has a huge influence on wether people come back or not. And the "commercial web" is all about making people come back. That for sure doesn't excuse these glossy macromedia flash company presentation websites that take longer to load than anyone would be ready to wait - but I support a sensible use of graphic elements on websites. Btw...anyone working on a text-only version of "userfriendly"? With ASCII graphics?
Re:Banner ads (Score:1)
I'm sure I'm not the only person who's learned to tune out any image that's set off from the page and wider than it is high.
news = usenet? (OT?) (Score:2)
Am I the only one that thought of usenet when reading this? Have people totally given up on usenet? I remember the first time that someone sent SPAM to every newsgroup. His ISP was DoSed with all of the replies complaining. The internet was so much nicer when the general population wasn't on it.
How to 'peruse news online' has become an art of dodging spam and looking for uuencoded binaries. -sigh-
Re:An experiment (Score:1)
Hmmmm
Re:"OSM" deserves what he gets! (Score:1)
I remember some fantastic trolls from those days ... you could always get people going by starting a flamewar between Texas Instruments and Hewlett Packard fans. There was this guy called opensourceman, who used to be obsessed with Gloria Gaynor ... wonder what the hell happened to him
Re:Whatever happened to beer? (Score:1)
Am I the only one who notices ... (Score:1)
Re:Lynx (Score:1)
My www reading behaviour... (Score:1)
2. quickly reading the 'Stanford Poynter Project Study' headline while remembering having read about it this morning in the newspaper. Otherwise this headline would have been bad for me since I wouldn't have known what 'Stanford Poynter' was all about.
3. Clicking on the 'read more' link using the middle mousebutton (=open in new window). I use the middle button far more than I use the left one.
4. Returning to the main slahdot window and middle click on the link to the project since I know that on our slow compagny line the slashdot comments take a while to load.
5. Skimming through the 'introducery highlights' table on the right of the screen (why do they place that on the right of the screen and not on the left???).
6. click on the slashdot comments window. Nice, the comments have loaded. Skimming through the comments, noticing the comment about Jacob Nielssens alertbox.
7. Middle click on the alertbox link.
8. Skim through the alertbox reading the part dealing with the main findings (inverted pyramid scheme at work!!!)
9. Send the alertbox text to the printer so I can read the whole alertbox on my way home.
10. Add the Poynter page to my bookmarks.
11. Close the stanford window and the alert window and start writing this comment.
Re:Multiple windows (Score:1)
Re:Lynx (Score:1)
No quite (Score:1)
Everyone has experienced net lag and knows intuitively that pictures take a long time to load. So we have gotten used to looking for the text first, then looking at the pictures that may actually be loaded by then. If everyone would just use the ALT tags like they're supposed to, I wouldn't ever look at pictures.
Jakob Nielsen (useit.com) mentions this too. (Score:1)
Re: Small Study Size (Score:1)
This could explain the higher-than-I-would-have-expected attention to banner ads. Also explains what baffled one reporter: why the Sun-Times site beat the online Chicago Tribune two-to-one.
Re:An experiment (Score:1)
Re:connection speed..... (Score:1)
Re: The Stanford... (Score:1)
Sure thing.
Way back before the earth's crust had cooled, the two first popular web browsers were Lynx [trill-home.com] and Mosaic [delphi.com]. Other early web browsers included the CERN line-mode browser and Cello. Lynx is character-mode, a.k.a. a text browser. Mosaic was the first popular graphical browser. Both Lynx and Mosaic came out about the same time, in 1993. The developers of Mosaic went on to Netscape, and Micros~1 Internet Explorer is based on a licensed copy of Spyglass Mosaic.
HTH.
Researchers to speak in Silicon Valley July 12th (Score:1)
The meeting is free and anyone is welcome to attend. Check the www.webguild.org [webguild.org] website for more information.
Re:Lynx (Score:1)
Yeah, BACK IN THE GOOD OLE DAYS (Score:1)
w3m load/render (Score:2)
The problem with w3m is that it has to load the page fully in order to know how to render it. Somewhat like Netscape. You buy some time by skipping graphics, but lose otherwise. I think you're stuck.
Mozilla (and IE 5.x) both render and redraw on the fly. This makes for much faster percieved page loads and redraws (say if you resize a window). If the console-mode browsers could emulate this behavior, things might be cool, but I'm not sure how 'zactly you'd go about doing that.
What part of "Gestalt" don't you understand?
Scope out Kuro5hin [kuro5hin.org]
Windows and /. display modes (Score:2)
The other value of multiple windows is that you can choose different display settings for, say, full page vs. a thread.
For full page, I tend to set threshold at 2 (or 3, if things are really bad), mode to threaded, highest moderated and most recent first. For a thread, I prefer threshold=0 (filter nothing but trolls), and mode to nested. I do this by hand-editing the URL (eg: http://slashdot.org/comments.pl?sid=00/07/04/04302 49&mode=nested&threshold=0&cid=46), and changing the cid to view different threads.
It would be kinda nice to be able to set seperate front page and thread view modes ;-)
What part of "Gestalt" don't you understand?
Scope out Kuro5hin [kuro5hin.org]
Re:My www reading behaviour... (Score:1)
Get it posted on slashdot! (Score:1)
News flash! (Score:2)
Where do eyes go initially after firing up the first screenful of online news? To text, most likely. Not to photos or graphics, as you might expect. Instead, briefs or captions get eye fixations first, by and large. The eyes of online news readers then come back to the photos and graphics, sometimes not until they have returned to the first page after clicking away to a full article.
News flash! When people read news, they actualy do read!
Wow. Who'da thunk, huh?
Suppose you were an idiot. And suppose that you were a member of Congress. But I repeat myself.
Small Study Size... (Score:2)
The Stanford... (Score:1)
Re:Small Study Size... (Score:1)
This current work is neither a survey such as an exit poll, nor an experiment with controlled conditions, but rather a study. We went out and looked at the way things were for a certain subset of people, a sample of those who read online news at least three times a week. The main advantage of a study is that it allows the researcher to look around in an open-minded way, without a predetermined agenda.
However, studies are not predictive in the way that an experiment or a survey is. They require more judgment about what constitutes "a lot" or "a little." And they require more judgment about what aspects of the data are stable and noteworthy, and which are ephemeral.
This study is not statistically valid as would be a survey, but we are able to draw general conclusions in some areas and make inferences in others. As researchers, we try to present carefully what we say 'flat out' and what we hedge or qualify.
Suppose you were an idiot. And suppose that you were a member of Congress. But I repeat myself.
Width of text (Score:1)
From personal experience being a news junkie, I love reading from sites which presents text in nice 4" (approx.) columns e.g. is /. main page and byte.com.
Loathe those left margin to right margin full of text! Hard to read on screen. Have to resort to printing e.g. HOW TOs on Linux.
start rant I would love to reward myself with a big 21" Sony monitor! end rant
Re:Lynx (Score:1)
I use "lite mode" (it's in the user preferences somewhere). Sometimes it's a bit hard to see the layout of the threads, but it looks fine with Lynx for me.
An experiment (Score:3)
The content loads first, and its unusual positioning grabs your eye. I read a paragraph or two, until I headed to the innocuous grey caption -- before I noticed the picture with the cyborg style eye tracker. The TOC doesn't load until clicked, presumably to lead you into the introduction without distractions. Oh, there was a logo up in the corner.
While their results were about news sites, the principles are pretty well demonstrated on there site. Like the site or not, it seems to re-enforce their findings (coincidence? I think not.)
Of course, given a particular browser choice, bandwith, and individual makeup, YMMV.
zaugg
Links 0.92 does (some) frames now as well... (Score:1)
--
-Rich (OS/2, Linux, BeOS, Mac, NT, Win95, Solaris, FreeBSD, and OS2200 user in Bloomington MN)
The study MUST have been flawed. (Score:1)
.
.
.
Showbizwire
Skinema
Sonic Net
South Bend Tribune
.
.
.
No Slashdot? Aww, come on people. We've got to infiltrate these studies if we want to count.
Re:"OSM" deserves what he gets! (Score:1)
Re:Lynx (Score:1)
Re:Multiple windows (Score:1)
Re:Eyetracking Study in Alertbox (Score:2)
success always occurs in private, and failu
Please post the rest of your
---
Re:Direct article link (Score:1)
Re:sig translation (Score:1)
Damn 120 character limit...
Re:Self dilusion (Score:2)
Right. Even I know how to make the text wrap around an image on a web page.
Congratulations on one of the least readable sites I've encountered in recent weeks.
Also... Are 67 individuals enough to draw much of a conclusion? What was the age distribution? I have a feeling that if they managed to find a bunch of people who regularly read newspapers that don't have a ton of photos, say the Wall Street Journal, that could skew their results enough to make the study's result quite questionable. Testing a population that prefers People magazine would, I'm betting, result in a different conclusion.
--
Re:Multiple windows (Score:2)
What's interesting is that nobody in the studies pointed out that this is most likely due to sluggish load times. You start something loading, and then flee to something else while it's in process. I do this all the time, more (of course) on slow sites or those with huge amounts of content that load slowly (Slashdot being a good example of this).
My behaviour, incidentally, is almost exactly like the other user on this thread (opening multiple articles in different windows) except that being in the US, I don't disconnect from my net connection, even when using a modem. This is most likely because I'm in the US with unlimited local calling, while he's in the UK with per-minute charges for use.
D
----
Re:(OT) Question about online news (Score:1)
Each baby'll get you enough for like two years of web hosting.
Re:Congratulations! (Score:1)
Re:Whoo hoo! (Score:1)
Keep in mind, the study focused on internet NEWS readers. I would assume that you can take some of this and apply it to the behavior of the average web-surfer. My guess is that there will be more studies with other types of webpages.
Re:graphics vs text (Score:1)
Yes, content is king. Content in the sense of *usable information*. But also keep in mind that people love to get things presented in an attractive way. And that also includes graphics.
Even if the pictures don't have much (or anything) to do with presenting revelant information, the "commercial web" would probably keep them there, because they're expected. Sorta like background noise; if it's too quite, you get distracted.
Suppose you were an idiot. And suppose that you were a member of Congress. But I repeat myself.
Re:connection speed..... (Score:1)
There are still a good number of people out there that have slow connections. They will read the text on the page while the graphics load...
Something to think about, especially with "alt" tags....