David Abrahams wrote:

"Pavel Chikulaev" <pavel.chikulaev@gmail.com> writes:

...

* IMO combination of Search by Google and Go Search button is not good idea.

My biggest problem with it is that it doesn't look like a button at all.

Well there are only two possible choices then.. 1. Have both the Google logo and a GO/SEARCH button. 2. Make a custom Google logo + button. The second is likely not possible given the rules of the Google search service.

...

* Links in the body are underlined, but on the right side are not.

They are except for the ones that are obviously menu items (look at the bottom). I appreciate the fact that there aren't lots of underlines in those things that are clearly going to be links. That said...

Yes. Exactly the same conclusion I came to when I switch to have the links underlined.

...

* IMO Links in the body don't look good. (I mean dash underlining).

I agree with this. If you want a de-emphasized indicator some kind of pale solid underline would work. I suppose the dashes are supposed to work with any background color the user chooses? Even a dotted (alternating pixels) line would be better.

I actually agree.. My first attempt was to use the "dotted" underlines and the do look much better than the "dashed". But that brings up another unfortunate browser issue: IE doesn't support the "dotted" borders and displays "dashed" borders instead. Guess I'll have to try and draw my own borders for this :-\ -- Or just make IE users suffer the less pleasing look. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com - 102708583/icq

Miro Jurisic wrote:

In article <424ECCC4.1090704@redshift-software.com>, Rene Rivera <grafik.list@redshift-software.com> wrote:

...

...

...

* IMO Links in the body don't look good. (I mean dash underlining).

I agree with this. If you want a de-emphasized indicator some kind of pale solid underline would work. I suppose the dashes are supposed to work with any background color the user chooses? Even a dotted (alternating pixels) line would be better.

I actually agree.. My first attempt was to use the "dotted" underlines and the do look much better than the "dashed". But that brings up another unfortunate browser issue: IE doesn't support the "dotted" borders and displays "dashed" borders instead.

Guess I'll have to try and draw my own borders for this :-\ -- Or just make IE users suffer the less pleasing look.

Several browsers use non-solid underline as the default appearance of elements which display a pop-up clarification on mouseover. See <http://tribblescape.com/archives/20030228_html_acronym_tagging.php> for an example. For this reason, I would be extremely hesitant to use non-solid underlines for links.

1. Why would you be hesitant? Is there some usability issue with using a similar, or even same, indicator for both elements? 2. Perceptually speaking it's not a problem to use the same (or similar) visual cues to decorate different elements all of which are active. The decoration is indicating that they are active not what the activity is. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com - 102708583/icq

On Sat, 02 Apr 2005 23:33:31 -0600, Rene Rivera wrote

Allen wrote:

...

The width is fixed, I think that's bad.

Why?

I agree with Allen -- fixed width is bad. Basically it wastes space on my monitor and forces me to scroll when I shouldn't have to. Jeff

<caj@cs.york.ac.uk> wrote in message news:33261.213.107.104.155.1112527137.squirrel@213.107.104.155...

...

On Sat, 02 Apr 2005 23:33:31 -0600, Rene Rivera wrote

...

Allen wrote:

...

The width is fixed, I think that's bad.

Why?

I agree with Allen -- fixed width is bad. Basically it wastes space on my monitor and forces me to scroll when I shouldn't have to.

Also some people use quite low resolutions and large fonts, and don't want to have to scroll left and right.

I've got a new Dell Inspiron 700m. The resolution is 1280 X 800. With that wide aspect ratio, I want a web page to use the full width. --Beman

Jeff Garland wrote:

On Sat, 02 Apr 2005 23:33:31 -0600, Rene Rivera wrote

...

Allen wrote:

...

The width is fixed, I think that's bad.

Why?

I agree with Allen -- fixed width is bad. Basically it wastes space on my monitor and forces me to scroll when I shouldn't have to.

I agree with this also. Cheers Russell

On Sun, 03 Apr 2005 17:35:02 -0400, David Abrahams wrote

"Jeff Garland" <jeff@crystalclearsoftware.com> writes:

...

On Sat, 02 Apr 2005 23:33:31 -0600, Rene Rivera wrote

...

Allen wrote:

...

The width is fixed, I think that's bad.

Why?

I agree with Allen -- fixed width is bad. Basically it wastes space on my monitor and forces me to scroll when I shouldn't have to.

But without a fixed width the lines of text tend to become unreadably long.

Unreadable for you, but maybe not for me -- if the page is designed to just flow then everyone can set the browser size however they wish and the page will wrap to fit. This is how the current pages work. Jeff

In article <uhdin48mv.fsf@boost-consulting.com>, David Abrahams <dave@boost-consulting.com> wrote:

"Jeff Garland" <jeff@crystalclearsoftware.com> writes:

...

On Sun, 03 Apr 2005 17:35:02 -0400, David Abrahams wrote

...

"Jeff Garland" <jeff@crystalclearsoftware.com> writes:

...

On Sat, 02 Apr 2005 23:33:31 -0600, Rene Rivera wrote

...

Allen wrote:

...

The width is fixed, I think that's bad.

Why?

I agree with Allen -- fixed width is bad. Basically it wastes space on my monitor and forces me to scroll when I shouldn't have to.

But without a fixed width the lines of text tend to become unreadably long.

Unreadable for you, but maybe not for me -- if the page is designed to just flow then everyone can set the browser size however they wish and the page will wrap to fit. This is how the current pages work.

True. It's just a shame that browsers maximize to this silly horizontal aspect ratio all of our screens use.

Fixed width is not bad, if it is fixed in ems (rather than pixels). The reason why newspapers are set in columns and books are printed in the line-length-to-letter-width ratio that they are printed in is that human eye has a very hard time scanning from end of one line to beginning of next line if lines are too wide. This is a well-understood principle in typesetting and is used by numerous tools (such as LaTeX) to achieve typesetting output that is comfortable to read. People who insist on saying that fixed width wastes spaces are usually trying to solve the wrong problem. If you are reading text, how much space is wasted is usually irrelevant -- the relevant qualities are your ability to quickly scan for important information and to read the content efficiently. Making lines arbitrarily long compromises your ability to read efficiently even though it makes more text fit on your screen. There are times when making lines arbitrarily long is a good thing; they generally involve cases when the ability to scan from the end of one line to the beginning of next is not important or is supported by some other visual cue (e.g., alternate shading of table rows). However, text should generally not be set in arbitrarily wide paragraphs. My opinion is that the text should by default be set to a maximum width, because I believe that the default layout should emphasize readability, not spatial efficiency. meeroh

In article <20050404045007.M6858@crystalclearsoftware.com>, "Jeff Garland" <jeff@crystalclearsoftware.com> wrote:

I favor user control over publisher control.

This can be designed so that you can use a custom style sheet to override.

ps: In case you can't tell this one is a pet peeve of mine -- I really, really dislike fixed width sites. Let's not make Boost one of them.

I don't think that "this is a pet peeve of mine" is a valid argument. You are basically saying that we should sacrifice readability for the majority of users who will come to the site and not even be aware of the fact that the reason that the web site is hard to read is that their window is wide. I don't think that's the right tradeoff here. meeroh

In article <20050404125730.M7937@crystalclearsoftware.com>, "Jeff Garland" <jeff@crystalclearsoftware.com> wrote:

...

You are basically saying that we should sacrifice readability for the majority of users who will come to the site and not even be aware of the fact that the reason that the web site is hard to read is that their window is wide. I don't think that's the right tradeoff here.

The assertion that the site is hard to read for the 'majority of users' really isn't backed up by any facts. You're assumming everyone has a wide display, small font sizes, etc. Not everyone does. I say the site is perferctly readable and that a non-fixed design allows users easy control the parameters.

It's backed by my understanding of the issues involved in designing for readability, which I continue to think should be more important than spatial efficiency. Your assertion that the site is perfectly readable at arbitrary widths disagrees with what I believe is credible research on this topic. I don't have the time right now to look for references, but <http://hubel.sfasu.edu/research/textmargin.html> is a good start. The salient quotes from it are: "Results indicated that, by itself, text width does not influence readability; however, there was a significant interaction between text width and margin width. The most efficiently read conditions were those with small text width (4-inch) and large margins, or the largest text width (8-inch) and no border." and "The lack of correlation indicates that preference does not necessarily lead to optimal processing. Some efficiently-processed conditions are liked by participants while others are not. "

I guess we'll have to do a survey...

A survey is not necessarily relevant, unless you are willing to (IMO wrongly) assume that preference for a layout is correlated to its adequacy. HCI studies find time and time again that people don't know what's good for them. meeroh

This line of argument is getting a little silly but... Jeff Garland wrote:

I was just using myself as an example and stating my preference. I didn't really want to start a research war here, but if you want to go that way see: Designing Web Usability, Jakob Nielson p 174. Key quotes under the topic How Wide Should the Page Be: "you shouldn't design for any standard width...Users who have large screens should be allowed to benefit from their investment".

Benefits come in different forms. As Dave pointed out having a "larger" monitor has the benefit of allowing you to place multiple narrower windows side by side. Which is something I do as I have large, 20in and 21in 1600x1200 dual monitors. So I don't see how the current constrained design prevents you from getting benefit.

It's not just about me and my big monitor, it's about the small monitors and resolutions too.

How is it about small monitors and resolutions? Because you might have to scroll left and right? The design can be fixed to allow for that.

This issue is much more complex than a single study.

Yes it is and I think there was a previous post which quoted this summary: http://www.humanfactors.com/downloads/nov02.asp Past Issues - UI Design Newsletter

Nielson is an acknowledged expert in usability and web design.

But he is only one expert, as you said it's more complex than a single study. Hence by extension it must be more complex than one persons study.

I'm sure he is fully aware of the research you cite. Take a gandor at his site http://www.useit.com/. Tell me what you find regarding resizeability and use of space.

I find that his front page is split into two narrow columns.

Look at an article on the site:

http://www.useit.com/alertbox/20031110.html

Note that the lines take the entire width of the browser. Also note most violated home page guideline #2:

"2. Use a liquid layout that lets users adjust the homepage size Compliance rate: 28% Guideline number in Homepage Usability book: 67

Fighting frozen layouts seems a lost battle, but it's worth repeating: different users have different monitor sizes. People with big monitors want to be able to resize their browsers to view multiple windows simultaneously. You can't assume that everyone's window width is 800 pixels: it's too much for some users and too little for others. "

That is an argument which does not apply to my design. I do not assume any particular pixel size! I only assume that 10 to 12 words per line is the most pleasing line length. That is what centuries of research and experimentation have shown. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - Grafik/jabber.org

from my experience, people with larger monitors buy them for a reason, generally that reason being they want to open more windows without as much clutter. Rarely do they open pages a their max monitor resolution, and let me say nothing makes more more mad when popups use javascript to resize themselves to 3200 x 1200, my desktop resolution. Generally about 12 words per a line is right for normal text. The best way round this is to use em's where possible to dictate the layout by default. The option to go liquid or fixed with would be a user option. This way people can resize the text without break ups. A lot of people who use high res screens seem to run using a very high dpi, and therefore tend to shunt the font size up, which can break layouts the work to an absolute pixel based/fixed layout. Another thing to bear in mind, a good web design for a site like the boost site should be designed so that the printed version of the page uses paper efficiently. If I make a print style sheet, I normally disable all navigation elements, and keep headers / design as simple / minimal as possible. Too many sites produce bad print outs where there is a huge waste due to a left margin caused by a nav bar, and then the body of the text being cut off on the right because it doesn't fit the page width. This is typical of pixel based layout that are fixed to a set width. The print issue is more important with something like Boost than your average company site, because things like API documentation is often printed for later reference. many find it easier to read paper, not to mention not many people have big dual screen layouts to place the IDE on one window and the documentation etc on the other screen. If the site was made with CSS, you wouldn't need to concern yourself should the site be fixed / elastic / liquid as the user can choose. By default liquid or elastic design is best. I agree with the idea of let the user resize the screen, as many people will do that to their taste. From what i can see, is the whole documentation in docbook format? in that case, it would be nice to see a CHM version for Windows users, as the indexed search system is very handy, and is ideal for people who have laptops and might not be online all of the time. Single HTML / PDF docs are too bulky for many. If everything is in docbook format as well, it should be easy enough to use a single layout to redo all the docs should the website's design change a little as everyone is probably aware. Jason On 5 Ebr 2005, at 09:17, Paul A Bristow wrote:

| -----Original Message----- | From: boost-bounces@lists.boost.org | [mailto:boost-bounces@lists.boost.org] On Behalf Of Rene Rivera | Sent: 05 April 2005 05:36 | To: boost@lists.boost.org | Subject: Re: [boost] Re: [website] More "improvements"..

| I only assume that 10 to 12 words per line is | the most pleasing line length. | That is what centuries of research and experimentation have shown.

But that is only for plain text, and nearly all of Boost is broken up with lines of code and other aids to the eye that the short line length provides, so I don't think it is safe to extrapolate this our use.

(I also note that 'ragged right' has been shown better for reading, but when did you last read any printed stuff like that? Graphic and web designers are obsessed with overall apperance!)

So I strongly support Jeff Garland on this -

"Let the user adjust the width of the browser".

Paul

Paul A Bristow Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB +44 1539 561830 +44 7714 330204 mailto: pbristow@hetp.u-net.com

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Rene Rivera <grafik.list@redshift-software.com> writes:

This line of argument is getting a little silly but...

Jeff Garland wrote:

...

I was just using myself as an example and stating my preference. I didn't really want to start a research war here, but if you want to go that way see: Designing Web Usability, Jakob Nielson p 174. Key quotes under the topic How Wide Should the Page Be: "you shouldn't design for any standard width...Users who have large screens should be allowed to benefit from their investment".

Benefits come in different forms. As Dave pointed out having a "larger" monitor has the benefit of allowing you to place multiple narrower windows side by side. Which is something I do as I have large, 20in and 21in 1600x1200 dual monitors. So I don't see how the current constrained design prevents you from getting benefit.

I did point that out, but FWIW, I agree with those who say the page should respect the size they choose to make their browsers. Eliminating that choice just doesn't make any sense to me. What about using multiple columns like useit.com does? That would mitigate line length issues (at least until we start trying to use handhelds to browse Boost.org) -- Dave Abrahams Boost Consulting www.boost-consulting.com

Coming from a web design / web programming background (PHP), if there is anything i can do towards the design / interface / back-end, I would would be happy to put some time into site if there is anything I can do. There are pro's and cons with fixed width design. Fixed width can make things a lot more readable as you don't have excessively long lines to read on computers with hi-res screens. A lot of people do find this is true. On the other hand for a site involving a lot of code snippets, you will have a lot of <code> and / or <pre> blocks that have code that need to be non wrap-able and monospace fonts, so liquid layouts are a bonus there, as these code snippets can often break a fixed width design. Using CSS / XHTML it is possible to allow the user have choice over fixed / liquid / elastic layouts (This way the user can have the choice what layout they want), and I have experimented with this personally. I would be happy to contribute my skills in this direction. I have decent design skiils and could probably come up something design wise a little more interesting without going too OTT (nothing worst than sites that have that nasty generic / busy Postnuke / Tikiwiki feel to them). Anyway thats just a bit of food for thought. Jason On 4 Ebr 2005, at 13:35, Miro Jurisic wrote:

In article <20050404045007.M6858@crystalclearsoftware.com>, "Jeff Garland" <jeff@crystalclearsoftware.com> wrote:

...

I favor user control over publisher control.

This can be designed so that you can use a custom style sheet to override.

...

ps: In case you can't tell this one is a pet peeve of mine -- I really, really dislike fixed width sites. Let's not make Boost one of them.

I don't think that "this is a pet peeve of mine" is a valid argument. You are basically saying that we should sacrifice readability for the majority of users who will come to the site and not even be aware of the fact that the reason that the web site is hard to read is that their window is wide. I don't think that's the right tradeoff here.

meeroh

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

David Abrahams wrote:

Jason Earl <Jason@hybd.net> writes:

...

Coming from a web design / web programming background (PHP), if there is anything i can do towards the design / interface / back-end, I would would be happy to put some time into site if there is anything I can do.

There are pro's and cons with fixed width design. Fixed width can make things a lot more readable as you don't have excessively long lines to read on computers with hi-res screens. A lot of people do find this is true. On the other hand for a site involving a lot of code snippets, you will have a lot of <code> and / or <pre> blocks that have code that need to be non wrap-able and monospace fonts, so liquid layouts are a bonus there, as these code snippets can often break a fixed width design.

Of course we really should keep our code snippets reasonably narrow for the same reasons, but I doubt we can actually get buy-in and disciplined attention from everyone who's writing docs.

With my design I wasn't suggesting that it be used for anything but the front page. And it's specifically design for the news style content of the front page. So I'm not sure concerns about code snippets applies. Unless you are thinking of applying the front page design to the rest of the site? I which case I would have to take that into consideration. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com - 102708583/icq

On Sun, Apr 03, 2005 at 09:45:44PM -0400, David Abrahams wrote:

True. It's just a shame that browsers maximize to this silly horizontal aspect ratio all of our screens use.

Use a window manager that lets you maximise a window vertically only, or horizontally only. It's not the browser's job really. jon