Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Usability and the Internet

By cheekymonkey_68 in Internet
Mon Oct 23, 2000 at 10:37:03 AM EST
Tags: Help! (Ask Kuro5hin) (all tags)
Help! (Ask Kuro5hin)

In these days of "flash" interfaces and serious eye candy is their much hope for the web developer who wants to put usability first ?

I know traditional HCI paradigms say design around your intended audience, but with the web you have no control over who views your web sites.

All the web developer sites I see give you advise on how to develop sites, but seem to forget that knowing html is only a beginning not an end in itself

So my question is I know W3C can vet your site against the html standards, but are their organisations who will vet your site in terms of its usability ?


Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o Also by cheekymonkey_68


Display: Sort:
Usability and the Internet | 38 comments (38 topical, editorial, 0 hidden)
Target audiences... (3.63 / 11) (#1)
by Miniluv on Mon Oct 23, 2000 at 06:03:13 AM EST

There is no medium of which I'm aware where you can control your audience access. I've never come across a book written in a language I do not understand which physically prevented me from opening it. Nor a research paper targetted towards quantum physicists that didn't let ME, mr. uneducated in physics, read it.

Despite the fact that MORE than your target audience will read it, you SHOULD, imho, target your interface, writing, etc towards said audience. If they are in fact "targetted" you owe it to them to do so. Target audience implies you've got people in mind for viewing this website, and if that's true why do you care if other people view it? If they do, you ought to have a useable interface, and I'm sure they'll figure it out. You do not HAVE to buy into the lowest common denominator theory with websites...unless they're e-commerce for John Q Public, in which case you better let it do everything for them.
"Its like someone opened my mouth and stuck a fistful of herbs in it." - Tamio Kageyama, Iron Chef 'Battle Eggplant'

Usability and Jakob Nielsen (3.38 / 13) (#2)
by Aquarius on Mon Oct 23, 2000 at 06:06:04 AM EST

Given that you're talking about usability, you're quite probably already aware of Jakob Nielsen's columns at the Alertbox. Nielsen makes a very, very big point of usability, and some of the columns there discuss how to set up a usability study and how to assess your site for ease of use.

Aq.


"The grand plan that is Aquarius proceeds apace" -- Ronin, Frank Miller
Nielson is a good start (3.50 / 6) (#8)
by cheekymonkey_68 on Mon Oct 23, 2000 at 09:43:07 AM EST

Yes I've read his book on web usability, I visit the alertbox where time permits, read his famous articles such as the 'anti-mac interface'.

His works are of a good standard, its just a shame that most web designers misunderstand usability does not just = graphic design.

My slant is that I am trying to implement web based learning materials at an English University. Being a bit of an HTML newbie I was surprised that nearly all the content we used was just traditional materials transposed
to web materials.

There had been no attempt to validate the web sites for usability (for the main site it had just been left to a team of graphic designers...fine it it was a magazine but as Nielson would say this paradigm doesn't suit web pages)






[ Parent ]
Strangely... (2.00 / 6) (#11)
by farmgeek on Mon Oct 23, 2000 at 11:09:31 AM EST

I've always found that site to be very difficult to use. I do read the articles there from time to time, but I find the site itself tends to overwhelm my eyeballs. The whole page being split down the middle without enough white space between content just annoys the piss out of me.

Just had to rant.

[ Parent ]
Webpages for people with disabilities (4.15 / 13) (#3)
by Lionfire on Mon Oct 23, 2000 at 06:21:04 AM EST

Bobby is an analyser that checks pages for usability by people with disabilities. It checks for alt tags on images (to ensure that your page will be processed correctly by a text-to-speech processor) and other text equivalents for multimedia, that information can be seen with and without colour (and by colour-blind people), that graphs and charts have summaries... okay, it checks a lot of stuff  :)

You can also use the advanced version to check for compatability with various browsers and versions, as well as download a copy to run over batches of webpages (uses Java, so it's portable).

[ blog | cute ]
standards, usability and clients (4.15 / 13) (#4)
by jesterzog on Mon Oct 23, 2000 at 06:41:57 AM EST

I don't know if this comment would be particularly useful, but I've always considered the w3c standards to be one of the biggest measures of usability. (At least in theory.) Personally I don't think usability vetting organisations are really needed any more than consultansy firms.

The thing is that nobody's really forcing anyone to comply with w3c standards. The w3c standards aren't a dictatorship - they're simply a well accepted definition of what sort of pages are "usable". I had a big argument the other day with some newish designers about why html was supposed to structure documents, and not format them. (Unfortunately they had a louder voice. :)

If a page will validate properly to the w3c standards, it means it's more likely that the user's client will be more capable of presenting the page to the user in a useful way for that user. To me at least, this is the same thing as usability because it gives the user more control over how they want to use your information.

There are arguments for things like designing the presentation so that people can use it easily, and they have their place too. As long as 99% of the world is using crappy web browsers, it'll be up to the page designers to make sure their pages are compatible. Ideally though, web pages should just be information, optionally accompanied by external style-sheet/formatting information, and the usability aspects should be in the browsing client's ability to convert your structured document into a brilliant and useful presentation for the user.


jesterzog Fight the light


...and arguments with other web designers (3.50 / 4) (#14)
by janra on Mon Oct 23, 2000 at 02:40:16 PM EST

I also hassle people on one web designers board (I think they're getting sick of me ;-) but a few of them don't seem to take me too seriously because I don't *work* in web design. "oh, it's just a hobby, she doesn't know anything" and then wonder why their pages don't work in netscape...

And the wierd thing is, I've got this tutorial that I wrote up that I've been told was 'better than the "for dummies" book' (exaggeration, I think - some people are *too* free with praise for something half-decent)

I always run my pages through the validator, and I look forward to the day when putting the little 'valid HTML' gif and link isn't a mark of an unusual web site, and I can leave it off.
--
Discuss the art and craft of writing
That's the problem with world domination... Nobody is willing to wait for it anymore, work slowly towards it, drink more and enjoy the ride more.
[ Parent ]
Validated sites (3.00 / 1) (#25)
by AndrewH on Tue Oct 24, 2000 at 09:46:24 AM EST

I always run my pages through the validator, and I look forward to the day when putting the little 'valid HTML' gif and link isn't a mark of an unusual web site, and I can leave it off.

A couple of months ago I started auditing the site my programming team run on our corporate network as XML Strict (new pages) and HTML 4.01 Strict (older stuff with tags in capitals), not that we were ever that far off standards. When I audited the release notes I added the valid XHTML PNG. A little while later, my boss expressed surprise at seeing it.


John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Improving browser standards (3.00 / 1) (#31)
by AndrewH on Tue Oct 24, 2000 at 11:49:10 AM EST

As long as 99% of the world is using crappy web browsers, it'll be up to the page designers to make sure their pages are compatible.

Compatible, yes. Optimised, no. Anything else is unfair to users of browsers that do work.


John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]

Online architecture. (4.28 / 14) (#5)
by Chakotay on Mon Oct 23, 2000 at 06:43:20 AM EST

Basically, you can see websites as architecture, in a way. A good website is designed in such a way that at least its entire target audience can access it, preferably even everybody. When I made my site, I put a lot of care in making sure it didn't break in 640x480 or in text browsers like Lynx and W3M, even though it's designed to be viewed at resolutions between 800x600 and 1024x768. Sure, it's not very flashy (pun intended ofcourse), but I do use some javascript to spice it up a little with some mouse-over effects. Thing is, javascript is simply ignored by browsers that don't understand it, and I tested mine on almost every browser that does support javascript.

Flash websites are generally just that - flashy. A nice front, sure, but they don't have much information, and waiting for a flash animation to finish every time you click a link becomes tedious after a while. On my site I have a lot of information to convey, so I held back on the eyecandy. Generally, the amount of eye candy is inverse proportional to the quality of the information that the site gives. What I hate most though is people using broken browser detection scripts that tell me my Netscape 4.75 isn't Netscape 4 or higher.

Ahhhh, I love a good web design rant once in a while :)

--
Linux like wigwam. No windows, no gates, Apache inside.

The web is unusable (3.54 / 11) (#6)
by evvk on Mon Oct 23, 2000 at 07:46:45 AM EST

There hardly are pages anymore that can be viewed with any browser without it looking like a mess (table kludges). People seem to have forgotten to specify structure and instead use all the features for layout tricks. Can you find hip-looking pages that would use the <Hx> tags for headers? No? They use <font> tags and such instead.

This common brain damage must originate from the WYSIWYG paradigm: look at Word and such programs - truly they know nothing about structure.

And what about real programs with fast _keybindings_ vs. "web page programs"? Truly the web is unusable compared to keyboard-based programs. (But most modern GUI programs are as bad as the web ).


Lamescript (3.87 / 8) (#7)
by evvk on Mon Oct 23, 2000 at 07:54:50 AM EST

Oh, forgot to mention javascript. There are far too many pages that don't work without it. And what is it most commonly used for? To check that you have the latest browser, although the later content may be barely viewable with older ones too! Or for links. Really, there's a lot better way to implement links. Another common use is to leave out the submit button of forms by using javascript from a pulldown. Why is it so damn hard to put the submit button there? Oh, and for opening new windows. *If I want I new window, I can open it myself with the middle button, new media idiots!*



[ Parent ]
Forms with JavaScript (slightly ranty :-) (3.50 / 2) (#19)
by Aquarius on Tue Oct 24, 2000 at 07:10:31 AM EST

Now, me, I like JavaScript. The key thing is that you should have it degrade properly. I don't see what the problem is with having extra stuff done with JS, as long as you don't set up pages so that they can *only* be used with it.

To pick an example that you gave above, that of skipping the submit button on a form with one pulldown item in it; I can think of two important reasons for this. Put them in order of importance according to personal preference.

The first is that leaving submit buttons in means that your form has to submit to something, by which I mean a CGI or other server-side process. If you're using the dropdown for site navigation, which is what pretty much all single-dropdown-forms-with-JS are used for, then it's a whole lot easier to stick an onChange() handler on the <select> and do it that way than it is to write a CGI. Some people don't have the ability to run CGIs at all; what they should be doing is, perhaps, not displaying the dropdown to non-JS-enabled browsers.

The second reason is that menus that do their thing when you choose an item from them is what the user expects. Look at all the drop-down menus on the menu bar in applications. Do you have to choose an item and then click "Go" to make it happen? Hell, no. A user-interface standard already exists for menus, and it is this: when you choose something off a menu, that thing happens. It doesn't store it up and wait for you to press a button to kick it off. Having the "Go" button is a step back, not forward; it breaks a UI that users already use and understand, and that's bad.

I've got no complaints about your views as regards badly-degrading pages, but I really cannot see why people denigrate JavaScript-enabled pages just because they have JavaScript in them. If a page doesn't work without it, fine, complain. But don't complain just because someone decided to use some scripting to make a page easier to use. If you've got some kind of moral objection to ease of use, then plug in that 3270 terminal and browse with text. :-)

Aq.

"The grand plan that is Aquarius proceeds apace" -- Ronin, Frank Miller
[ Parent ]
If the pages just worked without JS... (3.66 / 3) (#20)
by evvk on Tue Oct 24, 2000 at 07:34:59 AM EST

Well that is exactly the problem: I the pages _cannot_ be browsed without the fsckin javascript. As much as I would want to I cannot browse them in a text terminal with lynx without much effort. The browser check for example: sometimes all that the main page contains is javascript, and one has to look at the source to find the page where the redirect would go.
I keep javascript off for mainly three reasons 1) javascript is too often used for annoying things like opening new windows without my permission 2) stability 3) navigator 3.x (I use it because it is more stable - over month "uptimes", faster and especially lighter than the newer versions) doesn't know all the latest things and I hate error dialogs.

I'd have nothing against JS, if the pages would work without it, but too often they don't. As for the pulldown thing, HTML should then have an action=something for it, that would not require JS, if it was supposed to be used so. Very often pulldowns are useless anyway - links are a lot easier to use (Modern GUIs with complex widgets suck etc.).



[ Parent ]
Badly coded JS is bad (3.50 / 2) (#22)
by Aquarius on Tue Oct 24, 2000 at 08:26:32 AM EST

I agree with your point, then; writing pages that require JS to be used is Bad. That's a "moan at developers" issue. Having said this, a lot of Windows programs require the VB runtimes, for example, or MSVCRT.DLL; a lot of Linux programs require Perl, or glibc6 rather than libc5. Do you have the same objections to these programs as you do to pages that can't be viewed without JavaScript?

HTML might well have a <navmenu> tag coming up, but these things appear precisely because lots of people use them; if something becomes a UI standard, it can't hurt to make it directly available from HTML.

The reason that I'd disagree with your statement that "Very often pulldowns are useless" is that a dropdown menu takes up the same amount of room whether there's five or fifty elements in it. If you've got lots of different site areas, then putting in text links to them all from every page wastes a lot of screen real-estate, and breaking up your site into areas and sub-areas just adds an extra annoying navigation layer, if the fifty areas are all equivalent.

Aq.

"The grand plan that is Aquarius proceeds apace" -- Ronin, Frank Miller
[ Parent ]
Links, layers and browsability (4.00 / 3) (#23)
by evvk on Tue Oct 24, 2000 at 09:37:37 AM EST

Libraries and other requirements do not constantly make my life harder -- just for the time installing them, which can sometimes be a painfull experience and sometimes not -- enabling javascript does.

A long list of links would not "waste screen space", if the pages were normal long documents, not some flashy kludges that try force everything in a window of certain size. And often that window is wider than tall - exactly the opposite of my ideal A4-sized window. (And I want other visible windows on the screen at a time than just the browser, and I don't want overlapping windows). HTML is for documents, not graphical "things" with practically no text content. Flash or something can be used for such.

As for another layer of navigation, I don't care if it is logical. (Heh, having a sitemap often seems to be an excuse to make rest of the navigation totally illogical). You just can't put everything on the front page if one is supposed to actually find something. SGI's page, for example, is quite easy to browse for a new media page - they have not tried to put everything on the front page. M$, Sun or Compaq, on the other hand, for example, are not easy to browse.



[ Parent ]
Back to usability again (4.66 / 3) (#29)
by Aquarius on Tue Oct 24, 2000 at 10:14:50 AM EST

Here, I disagree with you, on two counts, and both of them related to the usability of sites.

1. Document length
You say that "[a] long list of links would not "waste screen space", if the pages were normal long documents". Documents shouldn't be long; a couple of reinforcements of that fact (taken from Jakob Nielsen's columns, since that's where we started ;) :
Instead of cramming everything about a product or topic into a single, infinite page, use hypertext to structure the content space into a starting page that provides an overview and several secondary pages that each focus on a specific topic
Ten Good Deeds in Web Design
There's a whole section called "Scrolling now allowed" in Changes in Web Usability Since 1994; it states that documents can now overrun a single screenful, as users have become more proficient at using the Web, but that "it should be a rare exception to go beyond three screenfulls on an average monitor", "[s]crolling still reduces usability", and "I still recommend trying to design navigation pages to make all major choices visible without scrolling on the monitors used by the average visitor to a site".
There's a big difference between documents that are short and documents that are flashy. Not all pages that are short are unnecessarily flashy, but all pages that are short are easier to see at a glance than a fifteen-screen scrollable document. My point was that, if you break up documents into small, easily-digestible sections (as the above blockquote describes), then you haven't got the screen space for fifty links down the left hand side.

2. Navigation layers
"As for another layer of navigation, I don't care if it is logical"
Your users care, though :-) The navigation on a site is vital. Without it, it doesn't matter how good your content is, because no-one ever gets to see it. My point here is that you seem to be classifying any navigation system that uses things other than plain links as bad navigation, and I do not believe that that's the case. Obviously, you can design atrocious and pointlessly overfeatureful navigation systems in a combination of Flash and Java, but that doesn't mean that all systems designed with those tools are bad, nor does it imply the corollary; that all "plain, simple" text-link systems are great and wonderful and easy to use. Usability is a thing almost entirely apart from the methods by which it is implemented. Whether you can fit everything on the front page is a measure of how good your usability assessment is; if you can do so and remain usable, that's great! It's certainly not a priori impossible.

Aq.

"The grand plan that is Aquarius proceeds apace" -- Ronin, Frank Miller
[ Parent ]
The problem with header tags (2.83 / 6) (#12)
by Karmakaze on Mon Oct 23, 2000 at 01:06:51 PM EST

The big problem with the H tags is that they tend to make the header text default too big. I mean ugly-big. I mean offending the delicate sensibilites of the graphic designer big. So you have to override it with the font tags. Or put up with ugly.

That's why you see so many documents starting with level H3 or not bothering with H tags at all.

Personally I tend to use the H tags and then tame them with css, but it's not an optimal solution.


--
Karmakaze
[ Parent ]
Why isn't it optimal? (3.60 / 5) (#16)
by Miniluv on Mon Oct 23, 2000 at 07:43:59 PM EST

Speaking as someone just beginning to learn CSS and what it can do...I see using the various H tages and "taming" them with CSS as a quite optimal solution. It's the usage of CSS that W3C is intending, from what I can see.

Please correct me if I'm wrong, but we're supposed to be abstracting layout from content, and CSS is the perfect way to do that, then we just use minimal layout formatting with H and P and other tags.
"Its like someone opened my mouth and stuck a fistful of herbs in it." - Tamio Kageyama, Iron Chef 'Battle Eggplant'
[ Parent ]

RE: Why isn't it optional (3.00 / 3) (#28)
by Karmakaze on Tue Oct 24, 2000 at 10:09:19 AM EST

CSS is the best solution for my needs. However, it's also not completely implemented yet, and implements differently on different browsers.

It's not that CSS isn't a great thing, it's just not done yet.


--
Karmakaze
[ Parent ]
CSS (4.00 / 3) (#35)
by Sunir on Wed Oct 25, 2000 at 02:19:43 AM EST

Yes, add CSS to impose a style on the semantic header tags is exactly the right thing to do if and only if you need to force the style of your page.

For instance, recently I wrote:

<STRONG STYLE="font-weight:normal">
which is totally bizarre on first blush, but it definitely served my purpose.

I question the objection that CSS is not ready. The major failure in terms of CSS is Netscape (but we already agree that Netscape must die). The other important sector are the growing number of text-based browsers on PDAs. However, that is why you use semantic tags like STRONG and the Hn tags instead of style tags like B and FONT: you gracefully degrade to the correct behaviour.

Once again, Lynx friendliness is good.

"Look! You're free! Go, and be free!" and everyone hated it for that. --r
[ Parent ]

Blame it on Nutscrape (none / 0) (#38)
by J'raxis on Thu Nov 02, 2000 at 11:07:46 AM EST

CSS is the way to control document styling and form. Unfortunately, browsers like Netscape 4.x still exist, which can't even handle a:hover right.

WebStandards project had an interesting letter about Netscape's rediculous 4.x browser and their failure to release their supposedly standards-compliant 6.0 browser in a timely manner here. Good reading.

-- The standards-compliant Raxis

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

Tags don't *do* anything (3.80 / 5) (#17)
by driptray on Mon Oct 23, 2000 at 09:39:34 PM EST

The big problem with the H tags is that they tend to make the header text default too big.

OK, I'm being pedantic here, but...

"Tags" don't do anything. Its the browser that sizes the text. If your browser makes Hx text too big then you should reconfigure your browser, or use another browser.

A Hx tag is silent as to the size of the text the heading should be rendered in. In fact its silent as to whether the heading should be rendered visually at all.

I'm making this point to emphasise that web designers don't control the appearance of web pages - the user, by their choice of browser and its configuration, does. Attempts by the designer to control the appearance usually end in failure due to the (wrong) assumptions about how users will access their content. By this I mean assumptions about the user's choice of browser, and the way that browser is configured (fonts, font sizes, colours, user style sheets etc).


--
We brought the disasters. The alcohol. We committed the murders. - Paul Keating
[ Parent ]
Yes, you are being pedantic (3.00 / 2) (#27)
by Karmakaze on Tue Oct 24, 2000 at 10:06:20 AM EST

"Tags" don't do anything. Its the browser that sizes the text. If your browser makes Hx text too big then you should reconfigure your browser, or use another browser.

Fine.

The big problem with the H tags is that the most common browsers render them larger than designers like.

And, like it or not, unless you're doing plain text, designers are going to want to have some control over presentation.


--
Karmakaze
[ Parent ]
But the distinction is important (5.00 / 2) (#34)
by driptray on Tue Oct 24, 2000 at 09:45:08 PM EST

The big problem with the H tags is that the most common browsers render them larger than designers like.

You've got this the wrong way around. The H tags do not have a "problem". Its the browsers that have the "problem".

And, like it or not, unless you're doing plain text, designers are going to want to have some control over presentation.

Then they're bound for frustration as their designs will fail to render as they wish in other peoples browsers. If they really want "control" they're using the wrong tool. HTML can't give it to them and they might be better off with PDF or flash or something.


--
We brought the disasters. The alcohol. We committed the murders. - Paul Keating
[ Parent ]
And yes, you're still being pedantic (3.00 / 1) (#36)
by Karmakaze on Wed Oct 25, 2000 at 02:10:22 PM EST

For one thing, this may have been lost in the chain of replies; I was not saying that I have a problem with using H tags (in my original post I mentioned I generally use them and then apply formatting via CSS), but was theorizing as to why so many developers use FONT tags to separate content rather than logical structure. You probably missed this because you were playing a game of pet peeve.

My theory - that they had what they perceived as bad results from using the header tags and retreated to the old typesetting mindset.

Then they're bound for frustration as their designs will fail to render as they wish in other peoples browsers. If they really want "control" they're using the wrong tool. HTML can't give it to them and they might be better off with PDF or flash or something.

Now, here, I appreciate where the negativity is coming from, but, come on... There is a world of difference between insisting that every pixel be in place and just wanting the design of a page to be not-ugly for most viewers.


--
Karmakaze
[ Parent ]
Structure in word processing (3.00 / 1) (#21)
by AndrewH on Tue Oct 24, 2000 at 07:53:59 AM EST

look at Word and such programs - truly they know nothing about structure

Not entirely true, actually.

This may be bias on my part, as a lot of my exposure to MS-Word has been through WexTech Doc-to-Help, which forces you to think that way, but it has in fact got significant style facilities.


John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Re: Structure in word processing (4.00 / 1) (#26)
by evvk on Tue Oct 24, 2000 at 09:48:01 AM EST

Yes, word and other word processors do have some structure kludges, but 1) word is buggy 2) structure doesn't really work in a WYSIWYG environment, where you can't control it. Word for example often loses structure information when I try to change some styles. Infact, it seems to me that all the "structure" is just the named styles. In LaTeX or even in HTML-as-it-used-to-be, on the other hand, I can explicitly specify the structure and don't have to deal with the layout at all.


[ Parent ]
Best if viewed with.... (4.20 / 10) (#9)
by unstable on Mon Oct 23, 2000 at 10:09:08 AM EST

The 4 words that piss me off to no end... what ever happened to testing on all browsers (or at least Exploder and netscrape).

Fancy graphics/java/flash/etc is nice once in a while but make sure that those that use alternatives OS/browsers can still get info they need. or there is one more GUARANTEED lost customer.

People get so caught up with making it look good that they forgot how to make it work.




Reverend Unstable
all praise the almighty Bob
and be filled with slack

Best if viewed by... (4.11 / 9) (#13)
by bgarcia on Mon Oct 23, 2000 at 01:25:18 PM EST

My hero is the guy who wrote the following in a little perl cgi script:
# Make the user feel inferior
my $browser = $ENV{"HTTP_USER_AGENT"};
if ($browser =~ /mozilla/i) {
    print "Best viewed with Microsoft Internet Explorer!\n";
} elsif ($browser =~ /msie/i) {
    print "Best viewed with Netscape Navigator!\n";
}


[ Parent ]
Mozilla/4.0 (compatible; MSIE 5.0; ...) (none / 0) (#37)
by J'raxis on Thu Nov 02, 2000 at 10:57:35 AM EST

Will this work? Example MSIE user-agent:

Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt)

Both MSIE and Netscape will match the first condition. The if ($browser =~ /msie/i) should come first, no?

-- The PERL-writin' Raxis

[ J’raxis·Com | Liberty in your lifetime ]
[ Parent ]

An example of a site with a good interface... (3.44 / 9) (#10)
by Sunir on Mon Oct 23, 2000 at 10:38:15 AM EST

An example of a (complex) site with a good interface would be kuro5hin.org itself. There aren't many (any?) parts that are strictly superfluous or out of place. It is intuitive. It's not dying from bad feature karma. It is even Lynx friendly.

Once, when asked for a good example of web interface design, I pointed someone here. They responded that k5 was just another boring news site design, so naturally it was easy to use. Well, that's the point. Don't be ever so clever. Content over form after all.

P.S. Flash is evil.

"Look! You're free! Go, and be free!" and everyone hated it for that. --r

Flattery (3.75 / 4) (#15)
by rusty on Mon Oct 23, 2000 at 07:27:32 PM EST

...still won't get you better Mojo, but thank you anyway. ;-)

They responded that k5 was just another boring news site design, so naturally it was easy to use.

That's pretty amusing. IMO, the largest single component of "Easy to use" is *familiar*. If you don't have to think much to know what stuff does, then it will seem easy to use, because you've already learned the interface. People always want to do something new and exciting, but frankly, site design never adds users, it only repels them (in the case of bad design). Keep it simple, keep it familiar, and you hopefully won't scare too many potential readers off. Leave the artsy bullshit to Word and that ilk. Does anyone read Word by the way? I've always found it impossible to get past the front page. Are they any good? Do they even have, like, content? And if so, what on earth is it about?

____
Not the real rusty
[ Parent ]

Nope (2.00 / 1) (#33)
by skim123 on Tue Oct 24, 2000 at 09:42:22 PM EST

Leave the artsy bullshit to Word and that ilk. Does anyone read Word by the way

Never had heard of it until just now... damn, it took too long to load the frontpage (on dialip), so I still have no idea what it is. Anyway, that image that was half-loaded looked way too busy...

Money is in some respects like fire; it is a very excellent servant but a terrible master.
PT Barnum


[ Parent ]
Could do better (3.25 / 4) (#18)
by AndrewH on Tue Oct 24, 2000 at 06:08:20 AM EST

Disclaimer: This is a personal opinion.

An example of a (complex) site with a good interface would be kuro5hin.org itself.

Flattery, flattery, flattery. It may be better that a lot of other sites out there, but:

The background colour is forced to white. If I set the colour override on my browser, I lose the highlight colours at the same time.

It misuses tables for text positioning, so the top is untreadable until the whole page is downloaded.

Where are the <Hn> tags?


John Wilkes Booth, Lee Harvey Oswald, John Hinckley Jr — where are you now that we need you?
[ Parent ]
Light version (3.00 / 1) (#24)
by evvk on Tue Oct 24, 2000 at 09:41:23 AM EST

K5 needs a "light version" with no layout tricks, like the other site has. I find it much more usable and nicer to my eyes with the netscape default gray background and all.

[ Parent ]
FONT tag abuse (3.66 / 3) (#30)
by Sunir on Tue Oct 24, 2000 at 11:00:31 AM EST

I absolutely agree that content markers like the header tags are better than pure style tags like the FONT tag. I'd say k5 abuses the FONT tag way too much. In addition to the lack headers, it makes the mistake of setting the font size to be "2" and not "smaller" or "small" or "80%" or something relative.

As I understand it, abusing FONT tags degrades accessibility. But k5 still gracefully degrades reasonably in Lynx, which is a big thing, so I wonder how bad it really is.

Maybe someone with special requirements can speak up. Note that we already have had an article in Meta about display colours.

"Look! You're free! Go, and be free!" and everyone hated it for that. --r
[ Parent ]

Useit.com (3.00 / 1) (#32)
by skim123 on Tue Oct 24, 2000 at 09:38:24 PM EST

Since it sounds like you are familiar with HCI, I assume you're also familiar with Jakob Nielsen's site UseIt.com... while it doesn't offer any sort of "usability checker," it does have some great articles on Internet usability. Some of the better articles from there include: Writing for the Web and Top 10 New Mistakes of Web Design.

The best way to see if you have a good, usable interface, of course, is to test. Sit down people who fall into the category of your user-base and ask them to perform common tasks that you'd like you app to be able to do. Ask them to discuss aloud their progress, their thoughts, what they're looking for, why their clicking what they're clicking, etc. Take notes and take it back to the drawing board!

Money is in some respects like fire; it is a very excellent servant but a terrible master.
PT Barnum


Usability and the Internet | 38 comments (38 topical, 0 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!