Site Ratings & Reviews - www.atlis.nl

Author

www.atlis.nl

Stephen Poley

2004-09-04, 4:19 am

A little while ago I was asked to tidy up the English translation of a
site. It didn't take me long to realise it had more problems than the
English grammar: frames, Javascript-dependent, mediocre when printed,
slow for the content etc. After some discussions with the owner I got
permission to restructure it, though with strict instructions to change
the actual page layout as little as possible. The new version is now
online at http://www.atlis.nl/

Your comments would be welcome.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

Deryck

2004-09-04, 4:19 am

"Stephen Poley" <sbpoleySpicedHamTrap@xs4all.nl> wrote in message
news:c8jhj09gfrd9uc0ikjtpifk16pmt07hq98@4ax.com...
> A little while ago I was asked to tidy up the English translation of a
> site. It didn't take me long to realise it had more problems than the
> English grammar: frames, Javascript-dependent, mediocre when printed,
> slow for the content etc. After some discussions with the owner I got
> permission to restructure it, though with strict instructions to change
> the actual page layout as little as possible. The new version is now
> online at http://www.atlis.nl/
>
> Your comments would be welcome.
>

Looks a nice clean site.

I use a 1440x900 laptop and the pages spreads nicely over the full window.

There is a minor validation error on
http://www.atlis.nl/en/pm/services/project_start.html and there doesn't
appear to be an English version of the pyramid image on that page which is a
shame.

I havent checked all the pages in the validator but with the exception of
the one above all those that I have checked do validate.

I like the fact that even with Javascript disabled I can still get to the
"top" level of each menu and get the "lower" menu options from the text on
that page. I mean, I can navigate the site without JS.

I tried it with Opera, FF and IE6 and it looked consistent in all browsers
with no obvious problems (I'm using windows XP BTW).

I feel like that bot/twat Ginae but it does look very nice
indeed...doubtless a better person than me will pick it to bits though :)

HTH

Deryck

Spartanicus

2004-09-04, 4:19 am

Stephen Poley <sbpoleySpicedHamTrap@xs4all.nl> wrote:

>http://www.atlis.nl/
>
>Your comments would be welcome.

1) The use of national flags for language selection is not smart.
2) It's currently a strange mix, some elements adapt to my window width,
but the important bit (the content) doesn't.
3) You've not specified a background colour, that's only ok if you don't
specify a foreground colour. Currently it breaks with certain user
specified preferences.
4) The blue'ish image on the top left is strictly decorative, hence it
should disappear when viewed sans css (including the space it occupies).
5) Page linearity is a real mess, look at it sans css.
6) Good coding style: quote all attribute values.
7) List of links (nav) not marked up as such.
8) Company logo appears twice sans css, one should not.
9) "Submenu's: rest van de pagina negeren." makes no semantic sense,
read the page content out loud.
10) No form mail fallback.
11) More good coding style: choose lower or upper case, stick to it
(currently it's a mix).
12) The html doesn't validate.
13) The "submenu" at the end of the index page is superfluous, these sub
links are present on the pages that can be reached via the "main" menu
(I'm looking at it sans css and sans js).
14) The breadcrumb bar is orphaned off to the right due to it being a
part fixed width site.
15) The company address, email, phone #, main navbar and breadcrumb bar
are not properly structurally separated sans css.

--
Spartanicus

jake

2004-09-04, 4:19 am

In message <c8jhj09gfrd9uc0ikjtpifk16pmt07hq98@4ax.com>, Stephen Poley
<sbpoleySpicedHamTrap@xs4all.nl> writes
>A little while ago I was asked to tidy up the English translation of a
>site. It didn't take me long to realise it had more problems than the
>English grammar: frames, Javascript-dependent, mediocre when printed,
>slow for the content etc. After some discussions with the owner I got
>permission to restructure it, though with strict instructions to change
>the actual page layout as little as possible. The new version is now
>online at http://www.atlis.nl/
>
>Your comments would be welcome.
>
English Version:

(1) The <title></title> contents on the English home page is in Dutch.

(2) The menu needs a separator between each item e.g. "|".

(3) The 'Home' link on the Home page is still active, as is the 'Project
Management' link on the Project Management page, etc.

(4) Most pages could do with some headings ( <h1><h2><h3> etc.)

(5) It could do with a 'skip to main content' link at the start of each
page.

(6) Dutch annotated diagram on English page.

(7) 'You are here' would be nice to announce the bread crumb.

(8) On the Atlis page, 'This Site' points to the same page as
'Customers'.

regards

--
Jake

jake

2004-09-04, 4:19 am

In message <JzqEwwe$eOOBFwtO@gododdin.demon.co.uk>, jake
<jake@gododdin.demon.co.uk> writes
>In message <c8jhj09gfrd9uc0ikjtpifk16pmt07hq98@4ax.com>, Stephen Poley
><sbpoleySpicedHamTrap@xs4all.nl> writes
>English Version:
>
[snip]

>(4) Most pages could do with some headings ( <h1><h2><h3> etc.)
>
Forget this one; I didn't have my stylesheet switched in when I looked
at the pages.

[snip]

>regards
>

--
Jake

The Doormouse

2004-09-04, 4:19 am

Stephen Poley <sbpoleySpicedHamTrap@xs4all.nl> wrote:

> http://www.atlis.nl/

Well, it does not validate.

Besides that, and the ugly dividers on the menu at top, and the meaningless
graphic element on the left, and the odd way the site is structured on the
server (../../../) ... um, okay. Maybe it is not that good.

The Doormouse

--
The Doormouse cannot be reached by e-mail without her permission.

Dave Patton

2004-09-04, 4:19 am

Chris Beall <Chris_Beall@prodigy.net> wrote in
news:YB8_c.10894$Y94.8597@newssvr33.news.prodigy.com:

> First, how do you view a site without the associated CSS? (Um, as a real
> user, not a hacker.)

In Firefox with the Web Developer extension loaded,
Disable > Disable Styles, or else Ctrl+Shift+D.

No matter what your 'regular browser' is, it's worthwhile
getting Firefox and the Web Developer extension(and the
DOM Inspector, and LiveHTTPHeaders, and Google Pagerank,
and probably others ;-)

--
Dave Patton
Canadian Coordinator, Degree Confluence Project
http://www.confluence.org/
My website: http://members.shaw.ca/davepatton/

Spartanicus

2004-09-04, 7:15 am

Chris Beall <Chris_Beall@prodigy.net> wrote:

>First, how do you view a site without the associated CSS?

In my UA I use Ctrl-G.

>
>It adapts fine in my environment.

The text doesn't flow in the available space:
http://www.spartanicus.utvinternet.ie/test/atlis.nl.png

>
>I've never heard the term 'page linearity' before. Google didn't help.
>Can you tell me what you mean by this?

>
>Does this imply rethinking his entire drop-down menu implementation?

I have js disabled, I didn't look at how the drop down menu is
implemented, at any rate currently it's wrong because it disassociates
the main menu items from the sub menu items.

>
>I didn't understand this one.

It appears on the right in a otherwise blank space:
http://www.spartanicus.utvinternet.ie/test/atlis.nl.png

--
Spartanicus

Jeffrey Silverman

2004-09-04, 12:15 pm

On Sat, 04 Sep 2004 01:10:16 +0000, Chris Beall wrote:

> OK, this isn't my site, but I think there's a chance here for me to
> learn something important.
> First, how do you view a site without the associated CSS? (Um, as a real
> user, not a hacker.)

Use Opera.

--
Jeffrey Silverman
jeffreyPANTS@jhu.edu
** Drop "PANTS" to reply by email

Jeffrey Silverman

2004-09-04, 12:15 pm

On Sat, 04 Sep 2004 10:18:41 +0100, Spartanicus wrote:

> Page/content linearity means placing the content in a logical order.
>
> http://www.atlis.nl/nl/atlis/ has a decorative image, a postal address,
> phone numbers, a nav section and a bread crumb bar before the actual
> page header (which should be "About Atlis" btw), all this preceding
> stuff should be positioned *in the source* below the "Atlis" header and
> it's following paragraph.

I'll extend this definition a smidge by saying that, since CSS positioning
really lets you move bits anywhere on the page you want, the actual
position and ordering of the content in the HTML code can be in any order.

So, to help with accessibility and improve cross-browser or CSS-disabled
readability, it is good to actually order <div>'s in the HTML source in a
way that makes sense.

--
Jeffrey Silverman
jeffreyPANTS@jhu.edu
** Drop "PANTS" to reply by email

Spartanicus

2004-09-05, 4:16 am

Jeffrey Silverman <jeffreyPANTS@jhu.edu> wrote:

>I'll extend this definition a smidge by saying that, since CSS positioning
>really lets you move bits anywhere on the page you want

Certain elements cannot be positioned anywhere in the source (floats
e.g.).

>the actual
>position and ordering of the content in the HTML code can be in any order.

Absolute positioning should not be the first choice, it throws up new
issues, most can be prevented by skillful implementation, but not all.

Often it's better to remove the cause, given that Stephen has to keep
the look close to the original site, that may not be an option though.

--
Spartanicus

Chris Beall

2004-09-05, 4:16 am

Spartanicus wrote:
> Chris Beall <Chris_Beall@prodigy.net> wrote:
>
>
>
>
> In my UA I use Ctrl-G.
>
>
>
>
> The text doesn't flow in the available space:
> http://www.spartanicus.utvinternet.ie/test/atlis.nl.png

OK, got it. You're using a much higher window size than I do. Note,
however, that there is a point of view that says that long text lines
are hard to read ('cause your eyeballs get confused when they do a
carriage return). One recommendation that I once saw was that a line of
text should not exceed the combined length of the upper and lowercase
alphabet in the font being used.

>
>
>
>
>
> I have js disabled, I didn't look at how the drop down menu is
> implemented, at any rate currently it's wrong because it disassociates
> the main menu items from the sub menu items.
>
>
>
>
> It appears on the right in a otherwise blank space:
> http://www.spartanicus.utvinternet.ie/test/atlis.nl.png
>

OK, I understand. Another consequence of your larger window size. This
seems to be a general problem with a 3-column design in which the widths
of the left and right columns are fixed. When the window size gets
large, all of the increase is applied to the center column, which then
gets longer than a 'good' line of text should be (see above), resulting
in lots of blank space. This could be made more attractive by centering
the paragraph (still left-justified) within the center column, thus
splitting the empty space. I think I'd prefer a solution in which the
side columns expand, so the proportions remain the same at all window sizes.

Thanks,
Chris Beall

Chris Beall

2004-09-05, 4:16 am

Spartanicus wrote:

> Chris Beall <Chris_Beall@prodigy.net> wrote:
>
> Forgot to answer this one in the previous post:
>
>
>
>
> Page/content linearity means placing the content in a logical order.
>
> http://www.atlis.nl/nl/atlis/ has a decorative image, a postal address,
> phone numbers, a nav section and a bread crumb bar before the actual
> page header (which should be "About Atlis" btw), all this preceding
> stuff should be positioned *in the source* below the "Atlis" header and
> it's following paragraph.
>

I understand. This goes along with the ability to disable CSS.

Regards,
Chris Beall

Chris Beall

2004-09-05, 4:16 am

Jeffrey Silverman wrote:

> On Sat, 04 Sep 2004 01:10:16 +0000, Chris Beall wrote:
>
>
>
>
> Use Opera.
>

OK, so some browsers allow you to disable CSS (actually, I assume, Style
of any flavor).

I understand some reasons to disable cookies and JavaScript.

Why would you ever WANT to disable CSS (other than during development to
test a page in that environment)?

Thanks,
Chris Beall

Stephen Poley

2004-09-05, 7:16 pm

On Fri, 3 Sep 2004 22:53:03 +0100, jake <jake@gododdin.demon.co.uk>
wrote:

>In message <c8jhj09gfrd9uc0ikjtpifk16pmt07hq98@4ax.com>, Stephen Poley
><sbpoleySpicedHamTrap@xs4all.nl> writes
>English Version:
>
>(1) The <title></title> contents on the English home page is in Dutch.

Oops! Thanks.

>(2) The menu needs a separator between each item e.g. "|".

For AAA compliance, yes. I'm only aiming at AA. AFAICS AAA compliance is
aimed at work-arounds for some buggy assistive technology.

>(3) The 'Home' link on the Home page is still active, as is the 'Project
>Management' link on the Project Management page, etc.

True, but not easily addressed as this is an identical 'include' in
every page. The alternative would make maintenance much harder.

>(5) It could do with a 'skip to main content' link at the start of each
>page.

I'll think about that one.

>(6) Dutch annotated diagram on English page.

Yes, the diagrams have to be redrawn unfortunately.

>(7) 'You are here' would be nice to announce the bread crumb.

I'll ask the site owner about that one.

>(8) On the Atlis page, 'This Site' points to the same page as
>'Customers'.

Oops again. Thanks.

I'll fix the errors when I'm back in the office next week.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

Stephen Poley

2004-09-05, 7:16 pm

On Fri, 03 Sep 2004 22:39:29 +0100, Spartanicus <me@privacy.net> wrote:

>Stephen Poley <sbpoleySpicedHamTrap@xs4all.nl> wrote:
>
>
>1) The use of national flags for language selection is not smart.

Perhaps not very, but I didn't feel it worth arguing about with the site
owner. Other things were more important.

>2) It's currently a strange mix, some elements adapt to my window width,
>but the important bit (the content) doesn't.

I've placed a max-width on the content, so it should always adapt to
smaller windows, but doesn't produce very long text lines unless you
choose a large font size. Which browser / window-size are you using? Are
you suggesting the max-width value is too small?

>3) You've not specified a background colour, that's only ok if you don't
>specify a foreground colour. Currently it breaks with certain user
>specified preferences.

Good point.

>4) The blue'ish image on the top left is strictly decorative, hence it
>should disappear when viewed sans css (including the space it occupies).

Why do you think that?

>5) Page linearity is a real mess, look at it sans css.

I don't really understand the problem. What do you think doesn't
linearise? I tried a few sample pages in a text-browser emulator -
http://www.delorie.com/web/lynxview.html - and they seemed OK to me.
(Well, apart perhaps from the submenus.)

(OK, having seen your later post, explaining what you meant: I tend to
avoid absolute positioning as I've seen it cause more problems than it
solves. In this case perhaps it would be usable to make the pages more
text-browser friendly, but whether I'll have time in the near future to
change 100+ pages is doubtful. Would a 'skip navigation' link satisfy
you?)

>6) Good coding style: quote all attribute values.

Unnecessary in HTML for attributes which are strictly alphanumeric.

>7) List of links (nav) not marked up as such.

I have previously tried marking up navbars as lists; for vertical bars
it works fine, but there seem to be just too many browser bugs still
around for horizontal bars.

>8) Company logo appears twice sans css, one should not.

>9) "Submenu's: rest van de pagina negeren." makes no semantic sense,
>read the page content out loud.

Not ideal perhaps: what wording would you suggest? I just want users of
text browsers to know they don't need to read any further.

>10) No form mail fallback.

Not yet. It's on the to-do list.

>11) More good coding style: choose lower or upper case, stick to it
>(currently it's a mix).

A bit harsh - on a quick grep I managed to find two files (out of 100+)
in which some upper-case tags had crept in.

>12) The html doesn't validate.

I validated a sample of the pages, which were all valid. (A bit
embarassing I missed the opening page!) Do you know an easy way to
validate all pages on a site?

>13) The "submenu" at the end of the index page is superfluous, these sub
>links are present on the pages that can be reached via the "main" menu

Not superfluous exactly; these are the drop-down submenus that appear
when Javascript is enabled and allow the reader to bypass one level of
index page if desired.

>(I'm looking at it sans css and sans js).

I gathered. ;-)

>14) The breadcrumb bar is orphaned off to the right due to it being a
>part fixed width site.

On very large windows, or for largish windows and the shorter
breadcrumbs, that is true. But it looks OK at common window sizes and
doesn't become unreadable at large window widths.

>15) The company address, email, phone #, main navbar and breadcrumb bar
>are not properly structurally separated sans css.

What would you suggest then?

Thanks for your comments.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

jake

2004-09-05, 7:16 pm

In message <vablj014kg8bg9g0bi9dtb689eliq4dt2d@4ax.com>, Stephen Poley
<sbpoleySpicedHamTrap@xs4all.nl> writes
[snip]

[snip]
[color=darkred]
>
>
>For AAA compliance, yes. I'm only aiming at AA. AFAICS AAA compliance is
>aimed at work-arounds for some buggy assistive technology.
>
No, it's not just for compliance. It's just that without a
(recognisable) separator, the words will run together.

[snip]

>
>
>I'll think about that one.
>
Well, don't think for more than 10 seconds before doing it ;-)

[snip]

>
>I'll fix the errors when I'm back in the office next week.
>
An excellent idea.

regards.

--
Jake

Spartanicus

2004-09-05, 7:16 pm

Stephen Poley <sbpoleySpicedHamTrap@xs4all.nl> wrote:

>
>I've placed a max-width on the content, so it should always adapt to
>smaller windows, but doesn't produce very long text lines unless you
>choose a large font size. Which browser / window-size are you using?

Did you really just ask that?

>Are you suggesting the max-width value is too small?

I'm suggesting that it's up to the user to define the window width he is
comfortable with and that an author should not try to restrict a width.

>
>Why do you think that?

See my reply to Chris Beal.

>I tend to
>avoid absolute positioning as I've seen it cause more problems than it
>solves.

People often introduce new issues when they use absolute positioning,
but if implemented wisely it works ok.

>Would a 'skip navigation' link satisfy you?

Yuck no, that's an ugly kludge.

>
>Unnecessary in HTML for attributes which are strictly alphanumeric.

I assume you meant "numeric", it's not required in html, but I said it
was "Good coding style", note that xhtml does require it. Xhtml is imo
pointless, but I do favor using some of the things that are mandatory in
those specs in html, such as all lower case code, quoting all attribute
values etc.

>
>I have previously tried marking up navbars as lists; for vertical bars
>it works fine, but there seem to be just too many browser bugs still
>around for horizontal bars.

Google for "ALA taming lists".

>
>Not ideal perhaps: what wording would you suggest? I just want users of
>text browsers to know they don't need to read any further.

<suitable header level>Site navigation:</suitable header level> Why
would you want to tell people who use talking browsers to "ignore the
rest of the page"? Inform them about what follows, let them decide if
they want to go there.

>
>A bit harsh - on a quick grep I managed to find two files (out of 100+)
>in which some upper-case tags had crept in.

I only looked at the index page code.

>Do you know an easy way to validate all pages on a site?

Online: http://www.htmlhelp.com/tools/validator/ (has an "entire site"
option)
Offline: A batch offline validator (e.g. http://arealvalidator.com/ )

>
>Not superfluous exactly; these are the drop-down submenus that appear
>when Javascript is enabled and allow the reader to bypass one level of
>index page if desired.

Provided that all sub links are present on sub pages the entire site is
navigable sans js, so you can freely use "js links" for the drop down
menu and dispense with the currently inappropriately marked up and
disassociated menu links at the bottom.

>
>What would you suggest then?

The navbar should be marked up as a list, the postal address should be
structurally separated from the phone #'s (I'd mark it up as a paragraph
with line breaks), both phone #'s belong together.

Some favor marking up bread crumb bars as nested lists and then styling
them to display horizontally, the ALA "taming lists" article contains a
demo like that. This is a semantically sound approach, personally I
think some have gone a bit "mark it up as a list" mad. I mark bread
crumb bars up as a div with a "You are here:" prefix, and then position
it in the source so that it follows the navbar list. That imo also
provides a semantically sound approach without the imo rather convoluted
nested lists.

--
Spartanicus

kchayka

2004-09-06, 4:16 am

Chris Beall wrote:
>
> Why would you ever WANT to disable CSS (other than during development to
> test a page in that environment)?

When a clueless (or misguided) author uses CSS in a way that creates an
unusable page. For example, absolutely positioned text elements when
different text sizes are not accounted for can cause overlapping, and
thus unreadable, text. Disabling CSS may be the only way to make that
content accessible.

But then, there's always the browser's back button, too. :-)

--
Reply email address is a bottomless spam bucket.
Please reply to the group so everyone can share.

kchayka

2004-09-06, 4:16 am

Spartanicus wrote:
>
> the postal address should be
> structurally separated from the phone #'s (I'd mark it up as a paragraph
> with line breaks),

Wouldn't <address> be more appropriate than <p>?

--
Reply email address is a bottomless spam bucket.
Please reply to the group so everyone can share.

Chris Beall

2004-09-06, 4:16 am

Stephen Poley wrote:

> On Fri, 3 Sep 2004 22:53:03 +0100, jake <jake@gododdin.demon.co.uk>
> wrote:
>
>

(snip)[color=darkred]
>
>
> True, but not easily addressed as this is an identical 'include' in
> every page. The alternative would make maintenance much harder.

If the process that includes the navigation also knew what page it was
putting it into, it could suppress the link for that page and just
include the text. An HTML preprocessor can do this. Here's a code
fragment from a site I'm doing. The $PageName variable is uniquely set
in a header on each page.
; Common footer on most pages.
#if '<$PageName>' <>'Home'
<div class="navbar">
<ul>
<li>
#if '<$PageName>' = 'Styles'
<span class="button"> Styles </span>
#elseif
<a class="button" href="styles.html"> Styles </a>
#endif
 
</li>
<li>
#if '<$PageName>' = 'Materials'
<span class="button"> Materials </span>
#elseif
<a class="button" href="materials.html"> Materials </a>
#endif
 
</li>
<li>
#if '<$PageName>' = 'About'
<span class="button"> About </span>
#elseif
<a class="button" href="about.html"> About </a>
#endif
 
</li>
<li>
#if '<$PageName>' = 'Process'
<span class="button"> Your ring </span>
#elseif
<a class="button" href="process.html"> Your ring </a>
#endif
<br>
</li>
<li>
<a class="button" href="index.html"> Home </a>
</li>
</ul>
</div>
#endif

The preprocessor I use is PPWIZARD, which has two strong points: 1. It
works. 2. It's free. See http://www.labyrinth.net.au/~dbareis/ppwizard.htm

Regards,
Chris Beall

Spartanicus

2004-09-06, 7:15 am

kchayka <usenet@c-net.us> wrote:

>
>Wouldn't <address> be more appropriate than <p>?

The w3c definition of address is that it's for contact info for the
page's *author*.

Stephen is the author of the page, it's the company's address.

--
Spartanicus

Stephen Poley

2004-09-06, 12:17 pm

On Sat, 04 Sep 2004 00:41:29 GMT, The Doormouse <doormouse@att.net>
wrote:

> and the odd way the site is structured on the
>server (../../../)

They're called relative paths. I'm sure a rodent of your calibre can
think of a reason for using them.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/

Spartanicus

2004-09-09, 7:22 pm