You decide

3 Jan 2005 - 7:05pm
9 years ago
88 replies
1075 reads
Listera
2004

Every year I make a resolution to be kindler and gentler. :-) So it was in
that spirit I read the latest Nielsen Alertbox when someone asked me to
comment on it:

Reviving Advanced Hypertext
"To manage a huge, worldwide information space, users need proven features
like fat links, typed links, integrated search and browsing, overview maps,
big-screen designs, and physical hypertext."

<http://www.useit.com/alertbox/20050103.html>

OK, whatever.

What did attract my attention, though, was this gem:

"The benefit of explicit structural commands is that they FREE USERS FROM
SLAVERY TO INDIVIDUAL SITE DESIGNS. Users need no longer suffer under bad
sites."

Slavery. Individual. Design.

Is this the usual Nielsenian poor communication skills showing or a sign of
design fascism rearing its pretty head, again?

Ziya

Still waiting for machine-generated, anatomically-correct, medically-sealed
site design

Comments

4 Jan 2005 - 9:40am
Gerard Torenvliet
2004

Ziya:

I think you're overdoing your response to Jakob Nielsen a wee bit. I
don't think Nielsen was brilliant in what he said, but the quote you
gave at least wasn't awful.

I don't know that I've ever used a site that you've designed. I have,
however, used sites where I couldn't easily link back to a home page.
If my browser provided me the option to get around poor design (which
abounds) that would be great.

If I happened to browse to a good design, then I'd be happy to
leverage the good design instead of having to use my browser controls.
But for those times when I come up against a poorly designed site (in
my perception, even if it is a great deisgn) Nielsen merely suggests
that:

a. The internals of a site are coded in standard ways.

b. Browsers make use of this where possible to offer synergistic functionality.

Unfortunately, one of the phrases in Nielsen's "gem" of a statement
that you didn't latch on to was: "Users need no longer suffer under
bad sites." Nielsen's target here is sites that users perceive to be
poorly designed, even if they are well designed. Some users will
always have difficulty, even with world-class sites. If Nielsen's
suggestions help them, isn't that a good thing?

I missed the design fascism. Sorry!

-Gerard

--
Gerard Torenvliet
g.torenvliet at gmail.com

4 Jan 2005 - 9:56am
Dave Malouf
2005

Who decides what "bad sites" are? I think that's the facism piece.

I happen to agree pretty strongly w/ Ziya on this one.

The browser should be a non fixture of the solution. Browsers should be like
a platform. Manipulatable through good design, not a lockdown system that
people have to mold their design concepts to. Hard-wiring the browser while
"good intentioned" will lead us to a world of mediocrity.

Think of the analog of freedom of speech. Yes, a powerful concept. But to
make sure great expression is allowed we need to make sure that bad
expression is equally allowed and to tolerate bad expression is the real
sacrifice we all make towards supporting freedom of speech.

The same holds true here. You can be upset with "bad sites", but won't the
market forces take care of that? This is what its all about. I mean even MS
w/ its huge monopoly is making usability a bigger and bigger priority every
year. The market requires it.

-- dave

4 Jan 2005 - 9:59am
James Melzer
2004

My issue with Nielson's article is that badly designed sites will
continue to ignore these standards just like they ignore others. So
building in browser functionality that only works on well-designed
sites is inherently redundant, from the perspective of trying to
improve usability. Well designed sites already offer a consistent
link to the homepage. They already offer internal/external link
types. How would a browser standard impose those complex design
decisions on poorly designed sites? You can't legislate good design.

~ James

--
James Melzer
SRA International

On Tue, 4 Jan 2005 09:40:04 -0500, Gerard Torenvliet
<g.torenvliet at gmail.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
>
> Ziya:
>
> I think you're overdoing your response to Jakob Nielsen a wee bit. I
> don't think Nielsen was brilliant in what he said, but the quote you
> gave at least wasn't awful.
>
> I don't know that I've ever used a site that you've designed. I have,
> however, used sites where I couldn't easily link back to a home page.
> If my browser provided me the option to get around poor design (which
> abounds) that would be great.
>
> If I happened to browse to a good design, then I'd be happy to
> leverage the good design instead of having to use my browser controls.
> But for those times when I come up against a poorly designed site (in
> my perception, even if it is a great deisgn) Nielsen merely suggests
> that:
>
> a. The internals of a site are coded in standard ways.
>
> b. Browsers make use of this where possible to offer synergistic functionality.
>
> Unfortunately, one of the phrases in Nielsen's "gem" of a statement
> that you didn't latch on to was: "Users need no longer suffer under
> bad sites." Nielsen's target here is sites that users perceive to be
> poorly designed, even if they are well designed. Some users will
> always have difficulty, even with world-class sites. If Nielsen's
> suggestions help them, isn't that a good thing?
>
> I missed the design fascism. Sorry!
>
> -Gerard
>
> --
> Gerard Torenvliet
> g.torenvliet at gmail.com
> _______________________________________________
> Interaction Design Discussion List
> discuss at ixdg.org
> --
> to change your options (unsubscribe or set digest): http://discuss.ixdg.org/
> --
> Questions: lists at ixdg.org
> --
> Announcement Online List (discussion list members get announcements already)
> http://subscribe-announce.ixdg.org/
> --
> http://ixdg.org/
>

4 Jan 2005 - 9:54am
Elizabeth Buie
2004

Gerard Torenvliet writes:

<<Nielsen's target here is sites that users perceive to be
poorly designed, even if they are well designed.>>

I'm curious about this statement; it makes me wonder what "well designed"
means, if not that it supports people in achieving their goals and
objectives. What is the purpose of design, if not that?

I would ask: How can a site be well designed if a substantial number of
its users perceive it to be poorly designed?

On the other hand, you didn't say *how many* of its users perceive it to
be poorly designed; and knowing you to be a thoughtful kind of guy,
Gerard, it occurs to me that maybe you meant that even if a site is borne
out to be "well designed" for the vast majority of its users, there will
always be some who don't "get it" and will perceive it to be badly
designed. I could go along with that, I think.

Elizabeth
--
Elizabeth Buie
Computer Sciences Corporation
Rockville, Maryland, USA
+1.301.921.3326

----------------------------------------------------------------------------------------
This is a PRIVATE message. If you are not the intended recipient, please
delete without copying and kindly advise us by e-mail of the mistake in
delivery. NOTE: Regardless of content, this e-mail shall not operate to
bind CSC to any order or other contract unless pursuant to explicit
written agreement or government initiative expressly permitting the use of
e-mail for such purpose.
----------------------------------------------------------------------------------------

4 Jan 2005 - 10:17am
Alain D. M. G. ...
2003

--- James Melzer <jamesmelzer at gmail.com> a écrit :

> My issue with Nielson's article is that badly designed sites will
> continue to ignore these standards just like they ignore others. So
> building in browser functionality that only works on well-designed
> sites is inherently redundant, from the perspective of trying to
> improve usability. Well designed sites already offer a consistent
> link to the homepage. They already offer internal/external link
> types. How would a browser standard impose those complex design
> decisions on poorly designed sites? You can't legislate good design.

Yes, but that does not mean you should never try to legislate or give
regulations to some design elements and some related elements that
influence good design. The Government of Canada has enacted Common
Look and Feel (CLF) rules for all of its Web sites. Many of those
rules pertain to Metadata, and some of them give strict definitions of
the look of certain sections of any Web page. Theres is an enormous
amount that is missing in the rules, because, as you say you cant
legislate GOOD design. But the regulations force you to think about
all of the aspects in a more structured when the time comes to roll out
the Web pages.

Alain Vaillancourt

__________________________________________________________
Lèche-vitrine ou lèche-écran ?
magasinage.yahoo.ca

4 Jan 2005 - 10:21am
Gerard Torenvliet
2004

OK, a mega-response here.

Dave Heller points out that the fascism could lie in who it is making
the determination that a site is bad. I don't think a determination
needs to be made. If an individual user doesn't understand that design
of a site, no matter how good that design is, they use the
functionality in their browser. No explicit judgment is being made,
and even implicitly, if said user is an outlier who only ever reads
alertbox (grin) and so doesn't really understand design, the site
won't need to change. The user just gets results.

Between Dave Heller and James Melzner I get the idea that browser
functionality will mean that sites need to be coded a certain
browser-specific way to work in that browser. I agree that returning
to that sort of world would be a very bad thing. However, if browsers
can leverage defacto standards (like the fact that if you are on
www.a.com/b/c/d/e/f.html, www.a.com is probably the home page) and so
make reasonable assumptions to help a user's experience, why not?

In reality, I advocate coding to XHTML 1.0 Strict and separating
presentation from content as much as possible. (I'm glad to be in the
company of Tim Berners-Lee on this one. I think he's the uber design
non-fascist.) If site designers / authors do this, then there is a ton
of stuff that browsers can leverage (i.e., style-sheet switching for
accessibility concerns, etc.). That's the extent - let's choose a
non-proprietary standard and make it a good community practice to use
it.

So, I think I agree with Dave Heller / James Melzner that the type of
design fascism inherent in browser-specific standards is a bad thing.
However, I don't think this is what Nielsen is saying.

And, Elizabeth Buie is always charitable. Of course I was talking
about outliers who can't use, or don't take the time to use, a proven
good design. If more than a handful of your target audience can't use
your site, that is an important message that you should attend to, no
matter how well designed the site is in the eye of the designer. To be
even more specific, proving of a site could be done either by testing
on your target audience, or by revenue- or satisfaction-specific
metrics.

Phew.

I hope I haven't misconstrued anyone. I think we're in violent
agreement, except for on the single point of the offensiveness of
Jakob Nielsen's advice.

Regards,
-Gerard

P.S. More generally, I think if we as a community stopped reacting to
Jakob Nielsen (either because we disagree with him, or just because he
has a staid viewpoint) he would be a problem that might just go away.
Reacting to him might be part of the problem.

--
Gerard Torenvliet
g.torenvliet at gmail.com

4 Jan 2005 - 10:35am
Schlatzer, Kurt
2004

Gerard Torenvliet wrote:
> I hope I haven't misconstrued anyone. I think we're in violent
> agreement, except for on the single point of the offensiveness of
> Jakob Nielsen's advice.

Jakob has to be offensive/controversial to remain relevant. If we all
ignored him, he would eventually go away. What fun would that be?

Kurt

4 Jan 2005 - 11:01am
ralph lord
2004

Gerard wrote:

"The Government of Canada has enacted Common
Look and Feel (CLF) rules for all of its Web sites. Many of those
rules pertain to Metadata, and some of them give strict definitions of
the look of certain sections of any Web page"

There's a government in Canada? (Aw, I'm just a dumb red-stater)
Seriously, don't miss the matinee show on Saturday.

That metadata thing rings a faint bell and as I've been ripping and
burning some LEGALLY PURCHASED AND PERSONALLY OWNED CD's lately puts me
in mind of a wsdb (web site database). I'm thinking (ouch) that a
strict definition of look is certainly a bad (as opposed to good) idea,
but maybe what Yah-koab and our neighbors in the great white north
really want is a kind of standard web-site tag with all that hidden and
obscure and dreadfully important information in it.

Yer fav-o-rite browser might then have a button/pane/window/sumpthin
that would display all that stuff if you wanted it.

Or maybe you could just have big old database with all that stuff in it
and make it available on mr. gore's wild internet.

Or wait, maybe ONE PERSON could just do everybody's site so we wouldn't
have all this confusion. I think it works like:
www.drbronnersalloneworldsite.all/mysite/yoursite/anothersite...

Mmmmm. good times.

(this rambling post has no connection with any US government body of any
kind)

4 Jan 2005 - 11:09am
Elizabeth Bacon
2003

-----Original Message-----
From: Gerard Torenvliet
Sent: Tuesday, January 04, 2005 6:40 AM
To: discuss at ixdg.org
Subject: Re: [ID Discuss] You decide

<snip>
I have,
however, used sites where I couldn't easily link back to a home page.
If my browser provided me the option to get around poor design (which
abounds) that would be great.

~~~~

We've already got this tool in the browser: the address field.

Jakob speaks from a very narrow perspective: internet usability researcher.
All of his points in the last Alert Box express desires from that user
perspective. That's not to say anything about his ability to design or
communicate - we should recognize that he's speaking as a specific,
individual user, who just happens to have a very tall soapbox.

Cheers,
EB

4 Jan 2005 - 11:23am
Dave Malouf
2005

I'm not arguing w/ Gerard in my next response here, just pointing out an
interesting sentence:

"If an individual user doesn't understand that design of a site, no matter
how good that design is, they use the functionality in their browser."

I would say that if an individual user doesn't understand then the design is
not good.

Now it could be said that we will never be able to design to the individual
level, but then again, neither can the browser.

Back to the beginning.

Again, I would say the browser is an enabler technology for designers to
design w/in. Building in functionality for "home" for example assumes that
what is inside the browser is a site ... Can we really say that this is a
"standard" anymore? I haven't designed a website in 5 years, so this would
be a complete distraction to me and my users.

-- dave

4 Jan 2005 - 11:07am
Gerard Torenvliet
2004

Ralph:

Allain Vaillancourt wrote about the Government of Canada CLF rules, not me.

Note that these rules only apply to websites of the Canadian
Government and it's various departments, not to all of Canada.

-Gerard

P.S. A dumb red-stater? Sorry to show my colours, but that's a
contradiction in terms. :-)

--
Gerard Torenvliet
g.torenvliet at gmail.com

4 Jan 2005 - 11:30am
Dave Malouf
2005

Jakob as "individual" ...
I think some people bear the brunt of their fame.
There are some people who are not allowed to speak for themselves unless
they say they are speaking for themselves. When Jakob speaks in terms of
generalizations, I do not think of him as one man speaking for himself. His
job is to speak for others and I think it is fair to say that he is doing so
now by generalizing off of himself in this case.

-- dave

4 Jan 2005 - 11:32am
Dave Malouf
2005

Regarding the Canada thing.

This sounds like a single enterprise-wide guidelines. This is different than
something like 508c (accessibility guidelines in the US) which tells vendors
(not just internals) what they have to do to sell to the US gov't.

-- dave

4 Jan 2005 - 1:04pm
ralph lord
2004

Gerard wrote

"Allain Vaillancourt wrote about the Government of Canada CLF rules, not
me.

Note that these rules only apply to websites of the Canadian Government
and it's various departments, not to all of Canada."

Oops- Sorry Gerard! (and Allain)

And thanks for clarifying, I thought it was another socialist (did I
type that out loud?) thing applying to the whole nation from sea to
frozen sea.

RL

4 Jan 2005 - 1:16pm
Elizabeth Buie
2004

Ralph wrote:

>another socialist

That would be "totalitarian". Totalitarianism can be either left-wing or
right-wing economically, and has nothing to do with socialism.

Not that I want to get into a political discussion; I think it's
inappropriate here. But I did want to clear up a misconception.

Elizabeth
--
Elizabeth Buie
Computer Sciences Corporation
Rockville, Maryland, USA
+1.301.921.3326

----------------------------------------------------------------------------------------
This is a PRIVATE message. If you are not the intended recipient, please
delete without copying and kindly advise us by e-mail of the mistake in
delivery. NOTE: Regardless of content, this e-mail shall not operate to
bind CSC to any order or other contract unless pursuant to explicit
written agreement or government initiative expressly permitting the use of
e-mail for such purpose.
----------------------------------------------------------------------------------------

4 Jan 2005 - 1:49pm
Listera
2004

OK, now that we're sufficiently sober for the new year, let's forget the
principle design notion of scalability and review the absurdity of
hard-coding the web browser, you know that thing over half a BILLION people
use DAILY to access the net.

> I have, however, used sites where I couldn't easily link back to a home page.
> If my browser provided me the option to get around poor design (which abounds)
> that would be great.

If the home page is great how about a dedicated button for "Help System"?
"Category Listings"?
"Archives"?
"Product Lists"?
"Feedback/Contact"?

I'm not the one who's making these up. But I also want dedicated buttons for

Sitemap
Terms of use
Webmaster's phone number
Corporate org chart
Branch offices

What's your favorite page, Glen? How about a nice little green button for
the company physical stores, that would be handy, wouldn't it? Well, then,
don't you always want to quickly get at their store hours? How about a shiny
silver dedicated button with a clock icon on EVERY BROWSER HALF A BILLION
PEOPLE USE DAILY TO ACCESS THE NET.

Now imagine using this magnificently bedecked browser with 128 dedicated
buttons that so smoothly bypasses all 'badly' designed sites that, golly,
you DON'T EVEN NEED DESIGNERS ANY LONGER. Press a button, you are there. No
more "slavery"!

> Nielsen merely suggests that:

Glen, I can't even begin to fathom what Nielsen might be suggesting. I've
fallen down on my dedicated browser buttons and I can't get up.

> "Users need no longer suffer under bad sites."

May I suggest taking this gem of a thought and applying it everywhere:
universal rules of playbook efficiency, no more bad basketball teams.
Unbreakable rules of culinary efficiency, no more bad meals for anyone.
Ironclad rules of grammar, no more badly written books...

> I missed the design fascism.

...while we're at it, let's get the trains to run on time and call it a day.

Ziya
Nullius in Verba

4 Jan 2005 - 3:35pm
Marcin Wichary
2004

> I'm not the one who's making these up. But I also want dedicated
> buttons for
>
> Sitemap
> Terms of use
> Webmaster's phone number
> Corporate org chart
> Branch offices

Just wanted to point out that both Opera and Mozilla (NOT Firefox)
already had this feature at some point. Probably still have. They used
the information provided explicitly by webmasters in the <link> tags.
Opera 7 had separate buttons for: Home, Index, Contents, Search,
Glossary, Help, First, Previous, Next, Last, Up, Copyright and Author.
Both called it (Site) Navigation Bar.

Marcin Wichary
e:\> mwichary at usability.pl
w:\> www.aci.com.pl/mwichary >> Attached
w:\> www.aci.com.pl/mwichary/gui >> Graphical User Interface gallery
w:\> www.10yearsofbeingboring.com >> 10 years of Being Boring
w:\> www.usability.pl >> Usability.pl

4 Jan 2005 - 3:56pm
Alain D. M. G. ...
2003

--- David Heller <dave at ixdg.org> a écrit :

> Regarding the Canada thing.
> This sounds like a single enterprise-wide guidelines. This is
> different than
> something like 508c (accessibility guidelines in the US) which tells
> vendors
> (not just internals) what they have to do to sell to the US gov't.
> -- dave

Exactly! The guidelines apply only to departments and agencies of the
federal government, from sea to unfrozen sea. The Pacific does not
freeze up, or at least not the parts on the shores of the inhabited
areas of British Columbia. The Atlantic does not freeze up either but I
have to admit there are some mighty big icebergs in it.

By the way, Jakob Nielsen praised the Common Look and Feel and
everything else the Government of Canada's Treasury Board (they have
functions similar to the OMB and the GAO in the US) did to implement
the GOL (Government On Line) rules in his Alertbox column of June 2004:

http://www.useit.com/alertbox/20040621.html

Alain Vaillancourt

__________________________________________________________
Lèche-vitrine ou lèche-écran ?
magasinage.yahoo.ca

4 Jan 2005 - 4:34pm
Listera
2004

Marcin Wichary:

> information provided explicitly by webmasters

That sounds suspiciously like, gasp, a conscious act of design to me and
should thus be thoroughly expunged forthwith. We can surely agree on 12 (64,
78 or 324 at most) buttons that would eliminate that most barbaric and
unreliable thing call web navigation altogether. That level of global
uniformity should make us all model citizens and efficient surfers to boot.
Free the people!

Ziya
Nullius in Verba

4 Jan 2005 - 4:53pm
drewbam
2004

It seems to me that Jakob is merely arguing for giving Web users the
tools to browse Web content using alternate mechanisms, if they so
choose. Nowhere does he state that these alternate mechanisms should
replace a site's custom UI; they merely supplement it. As a designer
and a Web user, I wholeheartedly support this idea. Why not include
navigation tags that enable users to map keyboard commands or toolbar
buttons to commonly-used page types or information elements? I don't
see how the addition of invisible hooks to the code of a Web page is
in any way detrimental to whatever custom visual/interaction design a
site chooses to adopt. On the contrary, these hooks could be very
helpful to users who choose to employ them.

As an example of the success of presentation-independent structured
information, look no futher than the ever more popular RSS. How many
readers of this list use an RSS reader? Well, I do. And I see one of
the primary advantages as the fact that I can dispense with whatever
hare-brained navigation and layout scheme has been dreamed up by
"creative" blog-masters with too much time on their hands.

Design, with a capital D, is about giving people tools that they can
use, not about cramming the designer's fixed mental model down the
throat of every user. Yes, of course, 80% (or more) of users will use
whatever is the default (e.g. the links and buttons that show up when
the page is rendered). But, why not give more control to those users
who take the initiative?

d|b

4 Jan 2005 - 4:59pm
bill pawlak
2004

I think a major part of the problem is that the "web" is so
far-reaching in its scope... Of course, having a dedicated button that
says "Product Listing" only works if the designer/coder/whatever
executes whatever explicit code/method is needed to activate that
button. But a bigger issue is... what if the site isn't selling
products?

Ziya, your flair for drama in this case (128 buttons? Surely only 97
would do ;) is really what gets at the point of the issue. On a news
site, "Top Stories" as a dedicated button would be just as valid as
"Discography" would be on a recording artist's site... but isn't that
why we have different navigation schemes for different types of sites?

Independent of one's feelings about Nielsen, he has the exposure to
get people talking about the issues/questions/concerns he raises.
Look at this thread as a perfect example. And there's probably 2
dozen other similar threads on other lists going on right now about
these same topics.

Sometimes has some decent ideas, sometimes he showboats some real
clunkers - especially from a designer's perspective. Its unfortunate
that he doesn't participate in any email lists (that I know of) to
further explain his thoughts. I think there's been a lot of
assumptions made about what he means by "good design" or "bad design"
and he's not here to use winking smiley faces to let us know he's not
*really* suggesting 128 dedicated buttons on a browser.

bill

4 Jan 2005 - 5:01pm
Dave Malouf
2005

Ok, best example I can think of as to why adding functionality like this to
a web browser hurts experience design.
The back button!!!!

This is the arch nemesis of web-application design. It has no place in
application design, yet we can't get rid of or disable the darn thing.

Any one of these new browser add-ons will have a similar effect of becoming
ubiquitously relied upon.

The solution is demand good design (oh my G-d! I've become Ziya!)
Ya don't like the design of X store, go to a store that works for you. This
is what we do in real life.

How many times have you used a car (maybe you rented it, maybe you bought
it) and a feature was just off or just right and you use that as new
criteria when deciding whether or not to buy the next car. These subtleties
are not show stoppers, but they are selling points. Technology should not
make up for bad design, good design should. Stop using technology to do
things it should just stay out of the way from.

There is nothing in HTML that should limit us from good design and even if
that wasn't the case, add in Flash and Java and well, you can create
everything you need to make a good web site/application. Bad design has no
excuses in this day and age and while my stuff is far from perfect (shoot, I
know it is flawed) I never blame the web browser for the bad designs. It's
all my fault and well the constraints of life and business. ;)

-- dave

4 Jan 2005 - 5:34pm
drewbam
2004

On Tue, 4 Jan 2005 17:01:38 -0500, David Heller <dave at ixdg.org> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
> Ok, best example I can think of as to why adding functionality like this to
> a web browser hurts experience design.
> The back button!!!!

OK, if you don't want your Web application to support the "Home
button," then don't include the Home link in the code on your site.
That ought to disable this hypothetical feature just fine.

> The solution is demand good design (oh my G-d! I've become Ziya!)
> Ya don't like the design of X store, go to a store that works for you. This
> is what we do in real life.

Sure we do. But, I don't see any reason why the Web experience
shouldn't be better than "real life." If I want the ketchup to be in
the same exact place in every digital store that I enter, why
shouldn't it be? All those ketchup bottles in various stores are just
entries in databases, anyway. Yes, they are in different databases,
but, if the standardized language (e.g. XML) of the Web can make those
databases speak the same language so that my custom shopper
application can present the ketchup bottles in the particular way that
I want them, as a user, then why shouldn't we enable that? Most users
will still go through the front door, and experience the store exactly
as the designers had intended. Others, who are more interested in
finding the cheapest ketchup than in listening to the store's
custom-programmed Musak, can use powerful tools that enable
standardized navigation, aggregation, and analysis of information.

d|b

4 Jan 2005 - 5:41pm
Listera
2004

David Heller:

> The solution is demand good design

Sweet magnolia! That wasn't so hard, was it?! :-)

How do we save the village? Burn it.
How do we protect our freedom of speech? Silence the dissenters.
How do we create good designers? License them.
How do we tame unruly design? Standardize it.

This notion of design 'efficiency' often comes in the disguise of
'standards.' Needless to say not all promotion of standards is evil.
This one comes really close.

Ziya
Nullius in Verba

4 Jan 2005 - 5:27pm
Jack L. Moffett
2005

Dave beat me to the punch on this one. I wish I had a record of the amount
of time and effort my company spent to get our applications to work with the
back button.

This issue prompted another thought, for what it's worth. If a browser
developer were to implement such site navigation functionality, the "good
design" would be one in which the buttons would disappear/disable when the
target page for the button didn't exist. This, of course, would then lead to
further debate about the users' expectations for the existence of these
pages: food for another Jakobox.

Jack

> Ok, best example I can think of as to why adding functionality like this to
> a web browser hurts experience design.
> The back button!!!!
>
> This is the arch nemesis of web-application design. It has no place in
> application design, yet we can't get rid of or disable the darn thing.
>
> Any one of these new browser add-ons will have a similar effect of becoming
> ubiquitously relied upon.

Jack L. Moffett
Interaction Designer
inmedius
412.690.2360 x219
http://www.inmedius.com

Questions about whether design
is necessary or affordable
are quite beside the point:
design is inevitable.

The alternative to good design
is bad design, not no design at all.

- Douglas Martin

4 Jan 2005 - 5:41pm
Nathan Vincent
2004

d|b said:

> Most users will still go through the front door,
> and experience the store exactly as the designers
> had intended. Others, who are more interested in
> finding the cheapest ketchup than in listening to
> the store's custom-programmed Musak, can use powerful
> tools that enable standardized navigation,
> aggregation, and analysis of information.

Fair enough, but I motivation behind the point Jacob is making is to
help my mum find the ketchup, not an advanced user who wants to access
powerful tools and such. Adding crap to browsers that will help advanced
users to do some powerful stuff, is pretty much the opposite of adding
standard buttons to help everyday users do simple stuff.

Nathan

4 Jan 2005 - 6:02pm
Listera
2004

d|b:

> It seems to me that Jakob is merely arguing for giving Web users the
> tools to browse Web content using alternate mechanisms, if they so
> choose.

Jakob may or may not understand design but he never "merely argues". He
first advances notions based on pseudo-scientific observations. Then he
argues for them to become standards, rules, best practices, etc. Finally, he
issues his "99% bad" fatvas to pressure companies into reading his books,
attending his seminars and purchasing his consultancy services, etc. to get
in line with what were once "mere suggestions."

I'm not amused.

Ziya
Nullius in Verba

4 Jan 2005 - 6:09pm
Peter Bagnall
2003

On 4 Jan 2005, at 22:01, David Heller wrote:
> Ok, best example I can think of as to why adding functionality like
> this to
> a web browser hurts experience design.
> The back button!!!!
>
> This is the arch nemesis of web-application design. It has no place in
> application design, yet we can't get rid of or disable the darn thing.

No, it's the other way round! The best reason why you shouldn't use a
browser as an application platform is the back button. The browser was
designed as a document reader - and in that context the back button
makes total sense.

The problem is not the back button, it's shoe-horning applications into
a browser. We need a more appropriate platform for networked apps that
is separate from the browser. As app complexity increases this is more
and more important. There are very strong UI and engineering reasons
why browsers are not appropriate for this. UI richness (the widgets
available, active elements), and server and network efficiency are just
a few of the reasons.

So what I'd suggest is the browser should split into two apps, one for
documentation, and one for functionality. They'd probably reference
each other a great deal, but they should be separated.

The best thing about doing this is it would free design to be more
appropriate in each domain. You would be free of the back button as an
app designer, and you'd have more flexibility is how you display/update
and so forth.

--Pete

----------------------------------------------------------
Peace and friendship with all mankind is our wisest policy,
and I wish we may be permitted to pursue it.
- Thomas Jefferson, 1743 - 1826

Peter Bagnall - http://people.surfaceeffect.com/pete/

4 Jan 2005 - 6:29pm
Andrei Herasimchuk
2004

Peter Bagnall wrote:

> The problem is not the back button, it's shoe-horning applications into
> a browser. We need a more appropriate platform for networked apps that
> is separate from the browser.

We've had an appropriate platform for many years already... it's called
a desktop application.

Now why the "desktop application" got a bad rap (it needed to be
installed, was too complex to code from scratch, IT couldn't keep
multiple machines and installations in sync, no way to remote log-in,
etc etc etc) is an entirely different matter, but let's not kid
ourselves... The proper platform has always been there. It's whether you
had the right set of designers and engineers around who could make
applications do what it needed to do, which these days means needing a
networking component that many preferred not to write themselves.

> So what I'd suggest is the browser should split into two apps, one for
> documentation, and one for functionality. They'd probably reference each
> other a great deal, but they should be separated.

Once you make that mental leap, you'll slowly work yourself back into
the idea that it's actually ok to have multiple applications to do
different major tasks. There's no need for uber-apps that try and do it
all at some level.

Hey! It's 1980 all over again! Except this time we now have the network
infrastructure in place, more processing power and more experience in
app design.

iTunes only works as well as it does because Apple designed the right
app for the right set of tasks. And since it does what I need, I now use
it for all my music related stuff. As a user, I'm perfectly ok with
that. In fact, I only use a browser when forced to, not because I like to.

It's like your kitchen. You have many different appliances to cook and
prepare food. Why on earth should we expect to have uber-apps that try
and do it all only to do it all in a crap way? It makes no sense, and
has made little sense to me for some time now.

Andrei

4 Jan 2005 - 6:34pm
Listera
2004

Andrei Herasimchuk:

> We've had an appropriate platform for many years already... it's called
> a desktop application.

Like the web browser? :-)

Ziya
Nullius in Verba

4 Jan 2005 - 6:43pm
Wendy Fischer
2004

I will second this. I've seen this in action when I worked at Excite at Home.

Nielsen wrote a critique that liberally wounded the @Home ClickVideo product. The PM's and execs were up in arms wondering who this Jakob Nielsen person was and if they should bring NN Group aka Nielsen in as consultants or issue press releases to counter Jakob Nielsen's fatva regarding the bad usability of the video product.

I laughed when they asked me who he was and if they should be worried. I said to get a UI designer to work on it and quit worrying about Nielsen.

-Wendy

Ziya quoted:
"Jakob may or may not understand design but he never "merely argues". He
first advances notions based on pseudo-scientific observations. Then he
argues for them to become standards, rules, best practices, etc. Finally, he
issues his "99% bad" fatvas to pressure companies into reading his books,
attending his seminars and purchasing his consultancy services, etc. to get
in line with what were once "mere suggestions.""

4 Jan 2005 - 6:56pm
Andrei Herasimchuk
2004

Listera wrote:

> Like the web browser? :-)

I know you're being coy, but for the sake of discussion:

Do this mental exercise: Open your refrigerator, and now ask yourself
why the meats drawer doesn't also have a microwave oven built into it,
and why the egg container doesn't boil my eggs with a timed setting I
can enter the night before, or why my freezer doesn't have an
auto-thawer and meat cleaver to chop up my steaks based on knowing what
my dinner menu is going to be stir-fry for the week.

Everyone has a refrigerator, right? Why isn't everything I do in my
kitchen attached to the frig? It's the perfect entry point into people's
homes. Everything you design and build for the kitchen should be
intricately linked and attached to the refrigerator, right?

If the refrigerator was designed by most web companies, they'd spend a
lot of time and money trying to make the refrigerator do a bunch of
stuff for which it was not designed, instead of attacking the root
problem: the tasks are sufficiently different that they require
different appliances.

Combined, all those appliance make up your kitchen. Well designed
appliances that take their environment into context make a great
kitchen. But they are separate appliances.

Companies need to invest in building the right applications to solve the
right tasks. One can even make the argument that good design requires
ending attempts at making the web browser do things for which it was not
designed.

Stop building email clients inside a web browser, Ziya. It's a waste of
time. Especially when I have plenty of robust desktop clients that
already solve that problem more than sufficiently for me.

Andrei

4 Jan 2005 - 7:00pm
Clay Newton
2004

> So what I'd suggest is the browser should split into two apps, one for
> documentation, and one for functionality. They'd probably reference
> each other a great deal, but they should be separated.

Isn't this one of the many benes Firefox (and for that matter any XUL
engine) affords?

Example: the Mozilla Amazon Browser
http://tinyurl.com/5be5f

Example: Mozilla Calendar
http://tinyurl.com/2oold

XUL is amazing and it would be great if we could rely on it as a
deployment option. The staying power of the browser as a web app
platform can be reduced to one thing: ubiquity. It is one of those
tired terms that stopped being buzzy awhile ago, but is still an
issue. Web apps need some ubiquitous tool to launch them at the very
least. Maybe I am a pessimist, but we are going to be stuck with them
for some time.

IMO, adding XUL support to IE would probably be a much better option
than the people at Microsoft reinventing the wheel with XAML. Oh well.

-Clay

4 Jan 2005 - 7:02pm
Wendy Fischer
2004

Andrei said:

Stop building email clients inside a web browser, Ziya. It's a waste of
time. Especially when I have plenty of robust desktop clients that
already solve that problem more than sufficiently for me.

Andrei
_______________________________________________

Sorry, I think somebody beat Ziya to the punch.....

Introducing Lazslo Mail....

http://www.laszlosystems.com/products/modules/mail.php

Laszlo Mail* provides a breakthrough Web email user experience, delivering the functionality and responsiveness of desktop email without requiring any client software install. Laszlo Mail enables mail service providers to deploy a scalable, standards-based rich Internet application that runs in virtually any Web browser across Windows, Mac and Linux while leveraging existing IT infrastructure.

-Wendy

4 Jan 2005 - 7:13pm
Peter Bagnall
2003

On 4 Jan 2005, at 23:29, Andrei Herasimchuk wrote:
> Peter Bagnall wrote:
>> The problem is not the back button, it's shoe-horning applications
>> into a browser. We need a more appropriate platform for networked
>> apps that is separate from the browser.
>
> We've had an appropriate platform for many years already... it's
> called a desktop application.

For the most part, but I'd like to see support for auto-updates and
code portability. If you're going to deploy things like banking apps
over the net then you need to make sure that everyone will be able to
use them and preferably not have to write one for each platform. The
Web did well because it delivered this. Windows doesn't offer this
quite yet, although .NET is heading that way.

Mac's offer this through Java Web Start (which has been around for
years now), which is a great solution, since it does auto-updates and
portability very well, but it's nowhere near critical mass.

> Isn't this one of the many benes Firefox (and for that matter any XUL
> engine) affords?

XUL looks like a really good thing. It's been needed for some time.
It's not really part of the browser though as I understand it. It's
just a UI description language which makes building your UI simpler.
That there are XUL engines for various platforms from the Mozilla
effort is great, and could easily be part of the solution I'm looking
for.

Adding it to IE wouldn't help I don't think. Adding it to Windows would
though. I'm still trying to reduce the role of the browser to the
things it was really designed for - ie reading stuff.

But XUL doesn't of itself solve the configuration management problem
though. That why it's only part of the answer. And you won't dislodge
the browser until you convince IT departments that there is a better
way of solving the configuration problem.

--Pete
----------------------------------------------------------
Manners are a sensitive awareness of the feelings of others.
If you have that awareness, you have good manners, no matter
what fork you use.
- Emily Post

Peter Bagnall - http://people.surfaceeffect.com/pete/

4 Jan 2005 - 7:19pm
Andrei Herasimchuk
2004

Wendy Fischer wrote:

> Sorry, I think somebody beat Ziya to the punch.....

Woo hoo! Yet another email client inside a web browser that does what
I've been able to do with a "real" app since I started using email back
in 1991.

(I love that their website uses a Flash demo to showcase the product
offering, and that many of their product offerings use Flash
exclusively, only using the browser as a means to get networking
functionality and a container window for the UI.)

Andrei

4 Jan 2005 - 7:23pm
Listera
2004

Andrei Herasimchuk:

> If the refrigerator was designed by most web companies, they'd spend a
> lot of time and money trying to make the refrigerator do a bunch of
> stuff for which it was not designed, instead of attacking the root
> problem: the tasks are sufficiently different that they require
> different appliances.

I guess you just haven't heard about 'smart refrigerators' like the
Electrolux one that can take a picture of its contents from a remote
location (the supermarket?) with cell phones with MMS. It comes with a
webcam to capture images of all the items inside, plus of course, various
sensors, web access, TV, email, you know, all the kitchen essentials. :-)

> Stop building email clients inside a web browser, Ziya.

Look, how else Hotmail and GMail developers are supposed to buy their
Ferraris, huh? :-)

> Especially when I have plenty of robust desktop clients that already solve
> that problem more than sufficiently for me.

You are preaching to the wrong congregation here (me). Like I said, I was
doing desktop apps long before web came around. I love'em.

My issues with your argument are narrow and two-fold:

A. Web (as in DHTML/HTTP) is not dead. No reason to kill it either.
B. I don't want to relive the OS-wars again. Web has been a powerful
ecumenical force, I don't want OS-specific binaries and dependencies to kill
it off.

Of course, I don't care what the client is (desktop/web/RIA/C-S) as long as
there's access to the data-stream. But I'm not naïve enough to kid myself
that there are certain companies who are not interested in lock-in through
desktop dependencies.

Ziya
Nullius in Verba

4 Jan 2005 - 7:25pm
Andrei Herasimchuk
2004

Peter Bagnall wrote:

> For the most part, but I'd like to see support for auto-updates and code
> portability. If you're going to deploy things like banking apps over the
> net then you need to make sure that everyone will be able to use them
> and preferably not have to write one for each platform.

Nearly every new modern app these days is doing this now, especially
considering that the major OS also rely on this to keep desktops current.

> Windows doesn't offer this quite yet,

Windows has been doing the auto update thing for some time now. The
issue is that MS has a longer lag time to check for QA, but the OS has
been doing auto-updates for almost five years or so. I have no idea what
you mean here.

> Mac's offer this through Java Web Start (which has been around for years
> now), which is a great solution, since it does auto-updates and
> portability very well, but it's nowhere near critical mass.

You've lost me. Mac OS has been doing auto-updates for some time now,
and has deep penetration. I constantly get updates to my machine as they
come in and have for at least 3 or 4 years now. For not just the OS, but
iTunes, iPod, Calendar, Mail, iMovie, iChat... the whole lot.

Andrei

4 Jan 2005 - 7:27pm
Listera
2004

Andrei Herasimchuk:

> I love that their website uses a Flash demo to showcase the product
> offering, and that many of their product offerings use Flash
> exclusively..

It couldn't be otherwise, Lazslo runtime currently *is* based on the Flash
5+ player .

Ziya
Nullius in Verba

4 Jan 2005 - 7:43pm
Andrei Herasimchuk
2004

Listera wrote:

> I guess you just haven't heard about 'smart refrigerators' like the
> Electrolux one that can take a picture of its contents from a remote
> location (the supermarket?) with cell phones with MMS. It comes with a
> webcam to capture images of all the items inside, plus of course, various
> sensors, web access, TV, email, you know, all the kitchen essentials. :-)

I have.

I picked my specific examples for a reason. They are tasks not suited
for refrigerators in that they have little to do with the refrigerator's
main purpose: storing food.

The example you cite is at least in relation to that purpose, so adding
more functionality in the ways a refrigerator can store food or help you
to store more food can be somewhat useful.

I'm not opposed to stretching boundaries, but at some point, it becomes
a bit ridiculous to expect one, or two or even three appliances to do
everything you need to do in the Kitchen.

> My issues with your argument are narrow and two-fold:
> A. Web (as in DHTML/HTTP) is not dead. No reason to kill it either.

I never claimed it was, nor did I ever claim it needed to be killed. But
given an offering of DHTML/HTTP versus a robust desktop client app, I'll
take the thing that works better any day of the week. All things being
equal, and with a good designer behind both system, the client app will
always be able to do more and operate more robustly than the DHTML/HTTP
solution, due to inherent limitations of DHTML/HTTP.

> B. I don't want to relive the OS-wars again. Web has been a powerful
> ecumenical force, I don't want OS-specific binaries and dependencies to kill
> it off.

The web has? By what? Reducing Microsoft's dominance to 93% instead of
96%? By getting 90% of the tech companies out there to not designed
products almost specifically for Windows? By creating coding and
development standards that work for multiple platforms consistently?

Please pass the pipe, I want whatever you are smoking. The world I live
is nothing like what you speak of at all. (And I live Apple Valley, the
heart of the other 2% on the planet.)

In fact,you can make the argument that the "web" has done nothing but
re-enforce Microsoft's power. Look at Windows Media Player or Internet
Explorer. How many companies now design for Windows and/or IE only after
the equalizing force of the "web"?

Andrei

4 Jan 2005 - 7:48pm
Dave Malouf
2005

> No, it's the other way round! The best reason why you shouldn't use a
> browser as an application platform is the back button. The
> browser was
> designed as a document reader - and in that context the back button
> makes total sense.
<snip>
> So what I'd suggest is the browser should split into two
> apps, one for
> documentation, and one for functionality. They'd probably reference
> each other a great deal, but they should be separated.

Halleluja ... Got an environment I can work w/ that is installed on every
desktop in the world, runs a distributed GUI interface with the same
secruity as the browser and ya got a deal. Till then, I'm stuck. :(

Cant' even use Flash in the x-enterprise app environment b/c well, can't
install Active X controls so to update to Flash MX 2004 can't be done. :(

But we digress.

-- dave

4 Jan 2005 - 7:54pm
Listera
2004

Andrei Herasimchuk:

> I picked my specific examples for a reason. They are tasks not suited
> for refrigerators in that they have little to do with the refrigerator's
> main purpose: storing food.

I don't want to belabor this narrow example but I don't mind if the
refrigerator's main purpose drastically changed. I'm not married to the
notion of refrigerator-as-dumb-cold-storage. If the new combo-appliance
works, great.

> I never claimed it was, nor did I ever claim it needed to be killed.

Another point of agreement then.

> In fact,you can make the argument that the "web" has done nothing but
> re-enforce Microsoft's power.

We can open up that argument in much greater depth but this isn't the forum.
You'll have to wait for our sit-down for it. :-)

> Look at Windows Media Player or Internet Explorer. How many companies now
> design for Windows and/or IE only after the equalizing force of the "web"?

There are many, many, many apps out there that I am convinced would never
ever been made available to non-Windows platforms had there been no way to
do them in HTML/HTTP on the web.

Ziya
Nullius in Verba

4 Jan 2005 - 7:56pm
Wendy Fischer
2004

The real history behind Lazslo team is that made of a lot of people from the Excite at Home TV settop team.

They were doing funky things with Flash way back in the day and building custom dev tools around the Flash player at Excite at Home in order to have the set top UI built on flash. My personal opinion is that Lazslo technology is a descendent of the technology from Excite at Home.

I worked on messaging products at Excite at Home. The team was already extolling the virtues of email in Flash, except the Flash technology wasn't quite ready for it. It still isn't ready though.....

-Wendy

4 Jan 2005 - 8:15pm
Dave Malouf
2005

Wow! This thread has gone from the superb to the sublime.

Couple of hopefully moderate throughts:

No technology is where we need it to be.
Technology is not a solution for bad design. Technology will just be used as
part of the bad design but in a different way.
Design is the solution to bad design.

Usability is a real problem in the world as design is not a science and no
matter how many times you test an solution you are still going to have a
user that is not satisfied. Shit, I bet you can put a driver in a
Mercedes-Benz and they'll find something "wrong" w/ it.

Usability testing, and HCI research are useful tools in the world and very
informative to help craft a good final solution through design methods.

To me this discussion happens often on this list and it seems to always
start to spiral into some tit-for-tat type arguments by people who generally
agree, but start nit-pickin' on the details.

How can we push this discussion into a more productive level?

Things I've noticed. People tend to argue from the position of their
experience + the current problems they are facing. Not everyone is ever on
the same exact page based on that equation. I've also noticed that people
from specific backgrounds tend to take the comments from others to be
absolutist and then argue with a full pendulum swing instead of a more
corrective fashion.

Issues that I know I'm facing related to this discussion:
1. The need for a platform that is OS independent, that doesn't require an
"install" by the end user. That is to mean that the user doesn't need to be
an administrator, or the install of such items is not considered a security
risk by the "owner"/administrator of the computer itself.

I don't see this changing any time soon and I only see solutions coming
through the OS Level + runtime engines that are part of those OSes, but the
conflict's of interest and upgrade rates of OSes get in the way even
further.

2. As an organizer in the UX community I'm really struggling w/ this us vs.
them dialog going on. On the one hand I get it. We do different things. On
the otherhand, we are teammates trying to solve the same issues. We just
have different areas of expertise and the overlap I guess is what is killing
us. I really want to get past this.

-- dave

4 Jan 2005 - 9:01pm
Listera
2004

Wendy Fischer:

> The real history behind Lazslo team is that made of a lot of people from the
> Excite at Home TV settop team.

Any insight into why they have open sourced Lazslo? (Other than the usual
reasons.)

Ziya
Nullius in Verba

4 Jan 2005 - 9:09pm
Daniel Harvey
2004

It is absolutely possible to disable the back button but honestly if you're working in a full-blown web-application (as opposed to web site) and you're finding that your users are using the back button as opposed to focusing their attention on internal controls to resolve their issue then there's probably something lacking in the design.

Further, most web-apps have a limited, focused audience. They should have some training in the web-app before being put in front of it in the first place.

Additionally, given the limited audience, you'll find that the web-app can and is likely designed and built to work within a specific browser/os setting. That gives you even more opportunity to strip out whatever you want from the browser.

-----Original Message-----
From:
discuss-interactiondesigners.com-bounces at lists.interactiondesigners.com
[mailto:discuss-interactiondesigners.com-bounces at lists.interactiondesign
ers.com]On Behalf Of David Heller
Sent: Tuesday, January 04, 2005 5:02 PM
To: discuss-interactiondesigners.com at lists.interactiondesigners.com
Subject: RE: [ID Discuss] You decide

[Please voluntarily trim replies to include only relevant quoted material.]

Ok, best example I can think of as to why adding functionality like this to
a web browser hurts experience design.
The back button!!!!

This is the arch nemesis of web-application design. It has no place in
application design, yet we can't get rid of or disable the darn thing.

4 Jan 2005 - 9:28pm
Daniel Harvey
2004

Andrei wrote:

"In fact,you can make the argument that the "web" has done nothing but
re-enforce Microsoft's power. Look at Windows Media Player or Internet
Explorer. How many companies now design for Windows and/or IE only after
the equalizing force of the "web"?"

I'd say you're correct Andrei... if this was still 2000. Versions of Flash also an issue (often times moreso than os/browser) and that helps take away some of MS's stranglehold. Additionally -- and this is purely anecdotal so take it as you will -- more and more designers and builders taking into account other browsers now more so than even during the 90s browser wars. I'm delighted to see bugs so up that cite safari or firefox. I think part of this also has to do with increased reliance on CSS and xhtml standards and other similar impulses that suffer more in IE than in other browsers.

4 Jan 2005 - 11:31pm
Peter Bagnall
2003

On 5 Jan 2005, at 00:25, Andrei Herasimchuk wrote:
> Peter Bagnall wrote:
>> For the most part, but I'd like to see support for auto-updates and
>> code portability. If you're going to deploy things like banking apps
>> over the net then you need to make sure that everyone will be able to
>> use them and preferably not have to write one for each platform.
>
> Nearly every new modern app these days is doing this now, especially
> considering that the major OS also rely on this to keep desktops
> current.
>
>> Windows doesn't offer this quite yet,
> Windows has been doing the auto update thing for some time now. The
> issue is that MS has a longer lag time to check for QA, but the OS has
> been doing auto-updates for almost five years or so. I have no idea
> what you mean here.

For the OS, sure. But I mean the OS should offer services which help
applications update themselves. MacOS provides this for Apple's apps
but I've yet to see and 3rd party apps use it. Windows likewise doesn't
offer this as a service for 3rd party apps. At the moment, while many
apps to auto-update they each do it differently which means no solution
is being tested as deeply as it should be. I'm not talking about user
testing here, but testing with regard to security for example.

As a user I want one place where I can update everything (and I want it
to take care of itself), I don't want to have each app doing it it's
own way.

Linux can do this, but the mainstream OS's can't to the best of my
knowledge.

--Pete

----------------------------------------------------------
How much trouble he avoids who does not look to see
what his neighbour says or does or thinks, but only to
what he does himself, that it may be just and pure.
- Marcus Aurelius, 121 - 180

Peter Bagnall - http://people.surfaceeffect.com/pete/

4 Jan 2005 - 11:53pm
Listera
2004

Peter Bagnall:

> As a user I want one place where I can update everything (and I want it
> to take care of itself), I don't want to have each app doing it it's
> own way.

There are certain technical and legal barriers.

There's no amount of contractual cover the OS vendor can have third parties
sign that could mitigate the adverse reaction towards the OS vendor that
users will inevitably have should problems arise as a result of
auto-updates. As the sole conduit, the OS vendor will ultimately be seen as
the responsible party (by users). To protect against abuse or negligence by
its partners, the OS vendor would have to institute OS-wide DRM-like
measures/digital signatures/managed code to maintain the integrity of the
whole process. There's also the issue of updating/overwriting shared
libraries/resources, forced rebooting, kernel-level stability, timely patch
tracking/management, etc for literally thousands of third parties. It's a
lot of headaches.

Ziya
Nullius in Verba

5 Jan 2005 - 4:21am
Manu Sharma
2003

I believe Nielsen is entirely inaccurate when he says, in effect:
"adding explicit site structure in the browser would free users from
bad sites and that Opera already has this feature."

I have two things to say to this. One, it's currently not possible to
implement this feature in today's browser as he implies it. Two, Opera
doesn't have a feature that frees users from "bad sites".

With so much discussion on the usefulness of this feature, it surprises
me that no one noticed this little implementation detail. Marcin
Wichary came close but not far enough to join the dots. Perhaps because
the Navigation bar in Opera is a little used. I believe most users
would have it disabled in their browser to increase browsing space and
because also it's rarely of any use.

Anyway, my point is this - it's not possible for the browser to
understand site structure *unless the designer wants it to*. In Opera,
for example, the button leading to a site's homepage, help system and
other pages only work if the designer has added link tags for these
pages inside the header of each page. If no such tags are present the
buttons do not work even if the pages are explicitly named and titled
say, as "Home" or "Help".

Nielsen is wrong when he says explicit structure representation is
implemented in Opera. "Bad sites" remain bad even in Opera. If a
designer does add link tags, I'd call him/her an exceptionally aware
designer. I'd not expect his site to be poorly designed. Only a
miniscule percentage of designers use link tags. It's not consistently
implemented even on nngroup.com. Only one of the 13 buttons remain
active only on some of the pages. Even on Useit.com 6 of the 13 buttons
do not work including search, glossary and help.

Nielsen is wrong again when he implies that such a representation is
possible to implement in today's browsers. For this to be possible
without help from the designer, the browser would have to decide the
site structure on its own. Perhaps much like a search engine - from
directory structure, labels, page titles and keywords. Links at at the
bottom of the page? Are they always there? Are they correctly labelled?
What if there are two dozen of them? Remember we are talking about "bad
sites" here which, by definition do not follow standards. Even well
designed sites often have unique ways of doing things.

Take directory structure. How should the browser decide, for example,
whether the homepage for http://subdomain.domain.com/section/page.html
is http://subdomain.domain.com/ or http://domain.com/? Or that when the
user clicks the home button on the location
http://kinja.com/favorites/guys he wants to go to
http://kinja.com/user/guys and not http://kinja.com/ ? What about
student homepages on university sites such as http://mrl.nyu.edu/~alex/
and geocities homepages such as
http://www.geocities.com/Nashville/6555/ ?

The 13 Opera buttons are based on the "link element" as defined by the
W3C. This is perfect example of a bad standard. The different link
types described clearly indicate where they originated and how
inapplicable they are on the web. These are: Start, Next, Prev,
Contents, Index, Glossary, Copyright, Chapter, Section, Subsection,
Appendix, Help, Bookmark. The thirteen Opera buttons are: Home, Index,
Contents, Search, Glossary, Help, First, Previous, Next, Last, Up,
Copyright and Author.

Contents? Where should that lead... to sitemap? Do "bad sites" have
those? What about index? And Last? Previous and Next? Do individual
webpages have previous/next pages? What's next if you're reading about
shipping costs on a shopping site? And whose previous/next page do they
mean...the designer's or the user's?

In view of all the above, the question whether even well labelled
buttons if implemented, will be deemed useful by a user remains a
purely hypothetical one. Unless one can demonstrate how such buttons
would be implemented in a browser, to be of any use on a "bad site".

Also want to add that despite this, overall I found Nielsen's piece to
be pretty good. Although much of the criticism that we heap upon him is
justified as well, I think most of us cling on to it and ignore
everything else.

And oh, my theory why his own site is so poorly designed, even when
nngroup.com fares far better -- I believe it is *intended* to be
visually offensive [though only so long as it doesn't affect
usability]. It's an individual expression of his revolt against those
who put aesthetics above everything else. Remember the graphic
designers who ruled "web design" from 1996-99. They came with
background in print advertising and thought knew exactly what design
was.

Manu Sharma
http://orangehues.com/blog/

Syndicate content Get the feed