Ads

mona-lisa-caribb.jpg

Famous painting, image by caribb, but no clue in the URL: http://flickr.com/photos/caribb/2355878576/

Do you know what the New York Times, the World Bank, Wordpress.com, PHP.net and others have in common?
Their URLs suck!

A few days ago my list of the top 10 fatal URL design mistakes has been hugely popular:

To prove how messed up URLs, these most important guiding units on the Internet, still are, I made a list of renown sites using completely inappropriate Internet addresses, directory and other URL structures.

You’ll be surprised to recognize some of the top 10 url design failures out there. I listed the examples accordingly to my original URL design mistakes list:

1. Bloomberg.com, renown news outlet: Session Ids (+ multiple random URLs for each page #5). Example: http://www.bloomberg.com/apps/news?pid=20601087&sid=aH5xJRoWZFOU&refer=home
Also try
http://www.bloomberg.com/apps/news?pid=20601087&sid=aH5xJRoWZFOU&refer=spam
http://www.bloomberg.com/apps/news?pid=20601087&sid=aHhgZh8jHAs02
http://www.bloomberg.com/apps/news?pid=20601087

2. Inhabitat.com, Technorati Top 100 blog: Mangled apostrophes in URL (+ date based URLs for timeless information #9): http://www.inhabitat.com/2008/07/02/philippe-starck%E2%80%99s-designer-windmill-for-all/

3. Fox News, infamous war propaganda machine: Numbers instead of speaking URLs
http://www.foxnews.com/story/0,2933,308077,00.html
What’s wrong here? Consider the headline: “Pop Tarts: Angelina Freaks Out Seeing Herself Naked in ‘Beowulf,’ Calls Home to Explain”

4. PHP.net, homepage of the world’s most popular server side script language: Multiple canonical URLs
http://php.net/
http://www.php.net/
http://www.php.net/index.php
etc.

5. New York Times, most renown US newspaper: Too many parameters which also change randomly, this example is so horrible it mus be repeated.
http://www.nytimes.com/2008/06/27/technology/27google.html?_r=3&adxnnl=1&oref=slogin&ref=business&adxnnlx=1214553738-5Jvl01JfMCKLx5duMGRv9g&oref=slogin&oref=slogin
Also try:
http://www.nytimes.com/2008/06/27/technology/27google.html?_r=3&adxnnl=1&oref=slogin&new-york-times-urls-suck
http://www.nytimes.com/2008/06/27/technology/27google.html?_r=3&adxnnl=1&oref=slogin
http://www.nytimes.com/2008/06/27/technology/27google.html

6. New York Times and multiple bloggers: Only one very broad and boring keyword in URL:
http://www.nytimes.com/2008/06/27/technology/27google.html
If you’re eager to know what Google did on this date check out also this blog:
http://julielemonde.com/2008/06/27/google/
In fact you can find such intriguing URLs for almost any date.

7. World Health Organization (WHO): Too many useless subdirectories
http://www.who.int/csr/don/archive/country/arg/en/
Also make sure to check out the “killer” content of this page!

8. Universities: UUCP, Berkeley, UCN and the World Bank, the world’s most hated bank: Check out these Joomla! crap URLs, don’t they have some smart computer science students to fix that?
http://www.uccp.org/index.php?option=com_content&task=view&id=29
http://iurd.berkeley.edu/index.php?option=com_content&task=view&id=173&Itemid=164
http://global.unc.edu/index.php?option=com_content&view=article&id=75&Itemid=81

Remember those black clad anarchist in Seattle 1999? Yes, one of them apparently infiltrated the World Bank’s computer department to sabotage their URLs, this is one of the worst examples of URL crap:
http://web.worldbank.org/WBSITE/EXTERNAL/NEWS/0,,contentMDK:21828803~pagePK:34370~piPK:34424~theSitePK:4607,00.html

9. Wordpress.com: Blog service and Smashing Magazine, Technorati top 10 blog:
Now tell me, is the date the most important and first to be seen info for this post here?
http://princessofsomething.wordpress.com/2008/07/06/where-the-heart-is/
Is this resource’s most important factor the the date when it was published, like it’s a 4th of July celebration or something?
http://www.smashingmagazine.com/2008/07/04/web-form-design-patterns-sign-up-forms/
Also consider this article, would you still read it after seeing the date?
http://www.webpronews.com/insiderreports/2005/06/27/google-video-to-launch-video-playback-service

10. SEO 2.0, blog dominating the global SEO 2.0 market: Yes, I failed here recently when I renamed my categories
http://seo2.0.onreact.com/category/reputation-building/
This will result in an error. I could have used this WordPress SEO plugin instead to prevent this error.


So you see the Web is full of broken URLs and there must be much work done before this mess is cleaned up. In 2008 we still face even huge sites which get the most fundamental findability and SEO basics wrong.

These top 10 URL failures prove that point. Contact their webmasters and make them aware of these issues. they can save thousands of dollars or even lives in the case of the WHO.

del.icio.us StumbleUpon Facebook Google Mixx Sphinn TwitThis
July, 2008 | You can follow comments through the RSS 2.0 feed. You can leave a comment, or trackback.

This thing has 11 Comments

  1. Posted July 7, 2008 at 6:17 pm | Permalink

    Amen. It is so simple to set these up in the most optimized format to begin with. Whether this impacts these sites is debatable, but if this was my premier property with lots of love, I’d want to crush the competition instead of leaving the door open.

  2. Posted July 7, 2008 at 7:13 pm | Permalink

    Basically the NYT e.g. is really struggling online. This is one of the main reasons I guess. The Web or hypertext consists of links, if people can’t link you properly you’re doomed.

  3. Posted July 7, 2008 at 7:20 pm | Permalink

    I think dates in URLs are more likely to make pages seem current than dated. If the year in the URL is 2008, readers will at see that the site has been updated at least once this year. Also, I realize this is an SEO blog and all, but a date in the URL does provide context and to your reader. If an in-site search returns multiple URLs for a search term, the reader would be able to easily pick the most recent article.

  4. Posted July 7, 2008 at 7:27 pm | Permalink

    Dan: Think about when you are looking at the URL at all. Do you see it when following a link? Or in your RSS reader? No, when they are current you don’t see them. You see them in the Google results though, or when you arrive from Google. Then you bounce because the “news” is too old. The date might provide context, but is almost never the single most important part of the content. You don’t make the date the h1 headline either, do you? Why do you force the readers then to read the date first in the URL?

  5. Posted July 8, 2008 at 11:38 am | Permalink

    Complaining about unreadable/unintelligible URLs might be justifiable if you’d bothered to proofread your post. Bloomberg is, I suspect, a renowned news outlet - you want an adjective, not a noun.

  6. Posted July 8, 2008 at 12:05 pm | Permalink

    Alex: You are complaining. I am showing webmasters how they can avoid pitfalls of URL design. Are you the webmaster of Bloomberg? Therefore the grudge? I noticed that Bloomberg has partly removed the issues already.
    On a side note, I’m a non-native speaker of English, it’s my third language of 5 so I sometimes make mistakes especially in a hurry while blogging. What about your Polish, German, Spanish and French? Thanks for the tip in any case.

  7. Posted July 8, 2008 at 7:45 pm | Permalink

    Fox news is just too cheap to upgrade from their ancient Vignette StoryServer 5.0. Those are old Vignette URL tags.

    0,2933,308077,00.html

    0 = cached page (1 is not cached)
    2933 = template id #
    308077 = database record number
    00 = No browser variations (FF has browser variations)

    Wanna have some fun - advance the third number to see fox stories directly - even ones that are “pre-launch” on occasion. Fox is proxied, but some other sites (like iVillage) will still bypass the cache if you change the first number, and give different layouts if you change the template id, etc

    Of course, you have to be REALLY bored . . .

  8. Posted July 16, 2008 at 2:28 pm | Permalink

    Re: #9 — true, the date might not be the most sought after nugget of info. But in appreciation for the ever present battle between human eyeballs and the robots… yes, sometimes that date is useful. When reading an article about SEO, for instance, I often remind myself about how fast things change in the industry, and read the date before I read anything else. That way, if I see that the post is from 2006, I know to read it with a certain discerning eye.

    So in that case, perhaps it’s a design failure from a robots point of view — but overall I think Wordpress is doing its human readers a favor.

  9. Posted July 16, 2008 at 3:01 pm | Permalink

    Yeah Paul, exactly. That’s why I don’t read these posts at all after seeing the date. Otherwise I would read it first and then due to the date take the news with a grain of salt. I would read it though!

  10. Posted July 23, 2008 at 10:49 am | Permalink

    Glad and sad about this post. Glad you wrote it and sad ‘cos I’ve now got a small mountain of work to fix my - and a few clients’ - urls :)

  11. Posted November 11, 2008 at 6:39 pm | Permalink

    there are those that say your should always hide the urls so it can’t be manipulated or critiqued

This thing has 2 Trackbacks

  1. Posted July 9, 2008 at 10:29 pm | Permalink

    […] Design aus SEO-Sicht nicht perfekt. Dies wird in einem Post im SEO 2.0 Blog mit dem Titel “Top 10 URL Design Failures of Famous Websites” […]

  2. Posted August 8, 2008 at 2:18 pm | Permalink

    […] url illisible pour un moteur ralentit (voire bloque) l’indexation d’un site internet. SEO Blog a listé un top 10 des sites anglophones dont la structure d’url est catastrophique. Je vous […]

Post a Comment

Please mind the commenting netiquette, most notably:

  • A "name" is a real name or nick name, not a keyword! SEO Company is wrong. John Doe of Google is OK.
  • For the "website" URL: No deep links allowed unless it's your "about" page.
  • No extra signature allowed, one "website" link is enough.
  • No bot-like "Thank you" comments with no context or added value to the post.

Your email is never published nor shared. Required fields are marked *

*
*