Forum Moderators: phranque
1. Unclear Statement of Purpose
2. New URLs for Archived Content
3. Undated Content
4. Small Thumbnail Images of Big, Detailed Photos
5. Overly detailed ALT Text
6. No "What-If" Support
7. Long Lists that Can't Be Winnowed by Attributes
8. Products Sorted Only by Brand
9. Overly Restrictive Form Entry
10. Pages That Link to Themselves
making a trivial point like "Pages That Link to Themselves" one of the top 10 web design mistakes just makes him sound like a crank.
OK, I'm a crank, well cranky anyhow.
I was on a page today that had 4 links that I assumed would take me to more detail. Nope, all 4 simply link to the page I was already on.
I assume there is a point to making links like that, but haven't a clue as to what the point is.
Given loose rein, many of the people we're supposedly making the site more usable for plow through forms putting the wrong info into form boxes. Because much of the form data is sent to a database, it IS important to screen data from user error.
Unless, of course, Mr. Nielson wants to spend nights and weekends with coding crews writing regexes to anticipate endless bad data scenarios, he might think twice before condemning such restrictions as to cause sites to be less usable.
The fact is, the data entered must ALSO be usable, and that means there must be some kinds of restrictions on form entry for the same (seemingly helpless) minions we're supposed to be building sites for.
You want to see cranky? Spend some quality time looking at a database full of nothing but slop entered by users left to wander freely from one form field to the next.
Give it a rest, Nielson. And have an eggnog.
So, I propose that No.11 mistake would be to pontificate about web design whilst owning that 'flagship' web site.
Did anyone notice that Nielson is on Google's "Technical Advisory Council"? When did that happen!
Jakob Nielsen was according to this page [discuss.washingtonpost.com] already in June 1999 at the advisory board for Google. Then Google was still in Beta (official start: September 21, 1999).
I think this one is quite weak
"Text only sites"
There always will be the die hards who love text only sites and browsers, are good at programming but can't do site layouts and graphics
"Small Thumbnail Images of Big, Detailed Photos"
This guy should try doing a few sites on a tight budget and timescale and then start writing comments like he has.
Just add text under the thumbnail hyperlinked to the large graphic
"Overly detailed ALT Text"
This is OK only if it's full of keywords hehe :)
ncw164x
1) Floating advertising.
This new kind of ad that's not integrated into the page but covers the actual page content. Super annoying. Even wired.com is doing it now. Stupefies me.
2) Animated banner ads.
I'm willing to take some advertising, but some pages look like pinball displays. I go bonkers trying to read an article sometimes with all the blinking and whatnot going on.
3) Disguised mailto links.
Make it clear it's a mailto link please.
4) No link to homepage on sub page.
Unbelievably many people seem to forget about this basic thing.
5) Disabling right-click.
The fact that it's so utterly and completely pointless adds to the aggravation.
6) Pdf files when html would do perfectly.
I hate slow loading pdf files. They're a **** to read too.
7) Title attribute set badly
Either too long and/or pointless or even missing.
Not a big gripe but some of my bookmarks (when forgot to rename) are mysteries now.
And it's just annoying having to rename only because the webmaster chooses to cater to Google instead of his users.
8) Horizontal scrolling.
Weird but doable if there's no vertical scrolling. Both is just madness. (Luckily very exceptional.)
9) All Flash sites, and of course ;)
The fact is, the data entered must ALSO be usable, and that means there must be some kinds of restrictions on form entry for the same (seemingly helpless) minions we're supposed to be building sites for.
True, that a database usually needs to hold only validated data in a predetermined format.
But it is not necessarily good practice to assume that a user will spend the time to learn the precise format you want. They may give up and go to a site that is less finicky.
If a user enters a date on one of my sites, I'll take any of these (and more besides) and convert it to the format I want internally:
If they enter something ambiguous (eg 4/5/6), I highlight it and show them how I've interpreted it.
I also write all cgi validation messages and the field values that prompted them to an application log file. I analyze that, and if I can see mistakes being made, I improve the validation and/or make the explanatory text clearer.
I suspect a great many forms (and thus sales) get abandoned because people get frustrated at having to contort their natural way of doing things to the way a particular site wants it.
It is not hard work at all to write bullet-proof validation and conversion routines. And, I reckon, there is a commercial advantage to doing so.
That's easy to confirm by checking the number of errors you throw out before a form is abandoned.
Content and ease-of-use is all well and good and necessary, but in a marketplace with very little barriers to entry and increasing commoditization, (such as the PC market) aesthetics can mean the difference no? Content is part of a package.
I want a Mac. Any Mac.
As for self-referencing links, I think this happens when the same navigation bar is used on every page, and when the link for the current page is also enabled. I regard that as a toss of a coin for usability -- some users will want to click on the nav bar button for the displayed page, presumably because they don't realise it. Should they be given an inert link that looks like other nav links, or do we let them refresh the page by making it clickable? I prefer the latter -
I once experienced this confusion. It was something like this:
Home page > Company, Products, Support etc
Products page > Product X, Product Y, etc
I was on the Product X page and I found that I could not click on the Products "link" so that I could navigate to another product, but I was forced to go back to the Home page before being allowed to see the Products page. Yes, I could hit the backspace key, but IMO good design should not draw attention to navigation quirks.
My #1 Design Hate:
Small fonts that are set with CSS and override anyone's attempt to view -> text-size -> larger. I'm 25 with 20/20 vision, and reading 9 pt font ticks me off. I can't even imagine what older visitors think. They probably just leave... I don't know where this cram-it-all-on-a-screen-so-no-one-has-to-scroll junk came from but... annoying.
Small fonts that are set with CSS and override anyone's attempt to view -> text-size -> larger. I'm 25 with 20/20 vision, and reading 9 pt font ticks me off.
I agree - the fix is to use em instead of px or pt for the font size.
I learned this when we made a really sharp site for a consultant who sold mostly words - he pointed out that some people really do have failing sight, especially his clients - once again visual design simply repelled the target user group.
nowadays not only do I check a site in Lynx, but I also look at various pages with the View -> Text size made small and large.
hmm.. this chappie has it both ways he's got the name that always gets comments
he has a text only website which is now wearing thin.. but it has served its purpose well has it not ;)
he was the subject of on unofficial CSS design contest this year.. i.e. he got his whole site and NNC designed for him for free!
I wouldn't wanna bet he doesn't realise what his next move is? (not necessarily CSS ;)) but he must know that in order to keep up his "accessibility guru and/or web guru" status he has to explore all possibilities, text only sites are available at the click of a button in some browsers it's just unfortunate for him that he doesn't have a "pretty" version for those who care!
>>Pet Hates/ Font size
Don't go there please the world isn't perfect but it's a heck of a lot better than it was.. there is a way that doesn't mean you have to build a text only website and there are even designers out there who now know this!
Nielson = "love him/hate him" : My View - he's not going away! more's the pity but realism kicks in!
Suzy
10: Not enough Flash!
9: Nothing happens when you move your mouse over stuff!
8: No hotlink to the Victoria's Secret web site!
7: Visited links an annoying shade of lavender.
6: It's not as cool as Neostream
5: Shouldn't this headline blink, you know - to get people's attention?
4: The colors don't match!
3: What's with that dude's hair?
2: No pictures of author's pets!
1: Doesn't give weather forecast for our zip code!
If you traffic is comming then going, if your customer life cycle is stagnating, then there is some lack of balance.
Find out what your customers want,and what your competitors are doing to meet those wants.
Even ask friends, relatives, to identify problems.
I like and respect Jakob Nielsen, because he splashes fresh water on our preconcieved ideas that design and content alone are enough.
When none of it matters, unless it's fulfilling a new the customer has.
The internet grew on a diet of freedom of expression, discovery and pushing the envelope. Why should we let that spirit stagnate and turn the internet into one large and very bland data repository with no identity?
"Pages That Link to Themselves"
At least that was number 10 on the list. When I designed my little personal geocities page way back when I was really careful about not having any page link to itself. What a pain, but I'm sure the effort is worth the result.
But frankly, a few customer comments of late have me re-considering this view. Every page on my site has a standardized menu at the top and at the bottom of the page. It makes maintenance a breeze if I have make any changes. And as far as I'm concerned, based on my customer comments, it makes the navigation much easier. Think about it for a second... would you rather see familiar navigation or have to scour each page just to get to the next? We're all probably pretty good at navigation, but I was on the phone today with someone who was having trouble getting to my checkout - never mind that it was right in front of them.
It's a toss-up I suppose, but since I have happy customers I won't lose any sleep worrying about what some know-it-all-who-has-probably-never-spent-a-dime-on-my-site thinks.
grandpa
About having no self-linking pages, it's not all that hard to have the same text in the same menu position, but no active link. Instead, just have some change in display -- a different color, all caps, something to show that inh this case it is NOT a link.
By doing that you create a location cue. Having a number of location cues on a page may seem redundant but I've seen many times that it generates a stickier website because visitors feel more comfortable.
In fact, one of my design pet peeves falls right in this area - when you click on a link that says "wordA wordB', you should arrive at a page that clearly says "wordA wordB" somewhere near the top and very obvious - a heading tag is often best. If the top of the page says "Section 2 - wordX" then even if those words are synonyms for the link text, my first reaction is "Maybe I didn't click on the link I intended to."
It's widespread and it's a usability error. Similarly, it's helpful if the first word or two in the window title tag mirrors the same language. Then when you have several tabs or windows open, you can easily pick out the page you want.
It makes maintenance a breeze
I sure do sympathize, grandpa, but I also think that some of the most atrocious usability gets generated by designers thinking about themselves first and their users second. If your attitude is really service, then you care more about site usability being a breeze than about making site maintenance a breeze.
Location-cues. That's what I have in mind with a standard menu. But honestly, if a fellow keeps clicking a link on Page 2 that clearly says "Link to Page 2", should he even have a credit card? (Just a rhetorical Q.) I do keep my customers in mind, and made some changes recently that I personally didn't really want to make. The issue there was the length of time for the page to load. I was afraid that by breaking the page into (4) smaller faster loading pages I would give up the flexibility of ordering many different (but similar) widgets from the one big page. I was wrong. I actually had those pages ready to go back in October, but it wasn't until I saw that my PPC hits on that page were leaving before it loaded that I made the change. (See, some good DID come out of Florida - I was forced into PPC and a resultant 'good and necessary' page change.)
But back to the location-cues. If it were me, I would not like having what appeared to be a link simply not work. I could be in the minority here, and have been persuaded to make a change thru logical discussion. (Disabling Right Click is what I'm thinking about) But I don't see it happening with self link, not yet, not on my Top Menu.
I mentioned before that I have a menu at the top and bottom of my pages. For this discussion, I'm not talking about my index, but all of my product pages. That top menu is grouped by product categories - Red Widgets, Blue Widgets, Green Widgets. Once a user is in a category, there may be several pages of, say, Red Widgets - so I have more menu options for Fuzzy Red Widgets, Balding Red Widgets, etc. I don't allow self-linking within the category of widget types, but that Top Menu still has Red Widgets. It is, as you say, a visual cue. You can find it in the same place on every product page. I just believe that helps, and if it happens to be self linking on a particular page, well:
1) I'm not a very good designer yet (a real possibility)
2) It should not matter to the vast majority who *know* they are on the Red Widgets Page.
My server logs don't reflect that my users are clicking these links, stuck in an endless loop of Red Widgets.
Having said all of that, my bottom menu *is* a place where I have begun to remove those links. You won't find a link to my comments page on the comments page. (Maybe you will, but I *am* addressing those as I come to them. In fact, I know right now that my site map has a link to itself - and that was simply a matter of convenience at the time.)
Now that it's Post-Florida, Post-Christmas I've found time to actually get back to some sorely needed page design. I'll look at those Top Menu links again, and again, and maybe one more time too.
grandpa
10. Pages That Link to Themselves
Let's think of a website - say webmasterworld.com. I don't know if you've ever done what I do, but click on recent posts - wait for 30 seconds and click on recent posts (again). There - a page that links to itself - perfect in the situation - used as a refresh button. Every case has it's exception. And it makes it a nightmare for those who use templates/parsing/SSI/Dynamic(using modular programming design) navigation sites to actually implement.
Every case has it's exception - what this guy is talking about is a generalisation. Some of the points he makes are valid, but may not be the best for the site you are working on. It is up to you to decide if you follow his guidance (not rules) and if you decide that you are not going to follow his guides, then make sure you have a valid counter-argument, relevant to your site, before dismissing the ideas.
P.S. Hey Mr UseIt.com, at least we have decent navigation!