Tuesday, 1 January 2013

Yearly statistics 2012

The website of the Helenahoeve is that stable, and there isn't that much new to tell, that the site didn't change much the last year. May be that's the reason the number of visitors is 11% lower than in 2011. See the graph below.

Sunday, 1 January 2012

Yearly Statistics

It's amazing, the number of unique visitors was growing spectucalar last year. Plus 40% compared with 2010. See the following graph:

The average number of unique visitors per day for 2011 is: 393

Tuesday, 12 July 2011

Troubles with the Google Plus One button

As always external content on your website is difficult to embed. This even applies for Google's Plus One button.

Invalid HTML code
The initial problem is that it is not valid HTML code. The g:plusone tag isn't a valid tag. The alternative is to use a div tag with the class g-plusone. That's better.

Invalid Javascript code
The second problem was with the Javascript code. For the Dutch website I need to include the script tag parameter {"lang":"nl"}. Again this is incorrect code according to validator.w3.org.

The solution for this problem was found at kompoos.nl. A Dutch site dedicated to show correct and fast working code. I love it. There I found a page about the plus one button where a trick was described to insert the required Javascript code with Javascript.

Nice code, but the result is of course invalid code, allthough the validator of w3.org is not able to detect this. However I have decided to use this trick. For the parameter {"lang":"nl"} I have added the this line of code:

google_plus.appendChild(document.createTextNode('{"lang": "nl"}'));

The result is a +1 button with Dutch text at every Dutch page of www.helenahoeve.nl.

Where to place the button
Google states that "You know your page and your users best, so we recommend putting the button wherever you think it will be the most effective. Above the fold, near the title of the page, and close to sharing links is often a good location. It can also be effective to place the +1 button at the end of an article or story as well as the beginning.". I decided to place the button at the bottom of the page, because it is not the main goal of the visitor to give a +1.

That decision was the cause of the third problem. The button was partly visible, because it was placed in the lower left corner. Yes, that's the area where your browser shows some additional information. A little CSS tweaking was necessary to prevent this.

CSS is useless
The fourth problem is that it is impossible for me to lay-out the Plus One button. The initial thought is that this should be possible by using the class g-plusone in the CSS code. Whatever I tried, it seems not possible.

Parts of the hidden button are visible
The fifth problem is that when you hide the surrounding div {display:none} there are some lines visible for the number of +1's. Due to the fact that I was not able to manipulate the Plus One button with CSS, I decided to choose for a button without the counter.

The result so far is neat. The only thing I have to do now is watching my stats.

Testing required
Dammit I should have tested this code better. It isn't working in Internet Explorer 8, or the latest version of Opera. So I fixed the problem as soon as possible by checking the availability of a script function before I execute this function.

Now I can say: The result so far is neat. The only thing I have to do now is watching my stats.

Oeps, still not working in earlier versions of Internet Explorer. It is working in Opera now, but not in, for example IE8. Due to the Javascript availability check I don't get an error message. However I don't get the button. IE8 is important because the market share of IE8 is still 35% in the Netherlands.

Wednesday, 8 June 2011

W3Fools - Bad value for attribute name on element meta

Since early June 2011 the validator of validator.w3.org works different with respect to meta and link tags. Before june 2011 almost all attributes where possible as an attribute. For example www.helenahoeve.nl uses:
  1. name="language" content="nl"
  2. name="WGS84" content="51.347669, N, 3.67498, E"
  3. name="ICBM" content="51.347669, 3.67498"
  4. name="DC.title" content="Boerderijcamping de Helenahoeve"
  5. name="geo.position" content="51.347669;3.67498"
  6. rel="P3Pv1" href="http://www.helenahoeve.nl/p3p.xml"
The website of the Helenahoeve is coded as a HTML5 website and used to validate. However, when you validate this website now, you will get 6 Errors. Check it yourself at:

In my opinion this might be an error in the specification of HTML5 or an error in the experimental validator. When you take a look at the specs
you will find that the bold keywords in the numbered list above are not mentioned in these specs.

So I think it is time to make use of the last call for the HTML5 specification. I think that the lists mentioned in the specs above should be read as containing suggestions of normally used keywords instead of an exhaustive list of allowed keywords.

My suggestion is logged as Bug 12918. When you take a log at this bug in bug / issue tracking service of the W3C you will learn that I made a mistake. I thought that http://www.w3schools.com/ was part of the World Wide Web Consortium but that's not the case. The only thing I could say was that I was fooled. So the only thing I have to tell you now.

Don't get fooled, take a look at http://w3fools.com/

There they state that: (...) W3Schools.com is not affiliated with the W3C in any way. Members of the W3C have asked W3Schools to explicitly disavow any connection in the past, and they have refused to do so. (...)  W3Schools frequently publishes inaccurate or misleading content. We have collected several examples illustrating this problem below. (...)

BTW I have updated the website of the Helenahoeve, it is now validating again.

Thursday, 26 May 2011

Usability Award 2011

In 2010 the helenahoeve was a candidate for the Usability Award. This year there are 50 nominees, in 5 categories. This time I check the sites with ZOMDirs Website Quality at a Glance.

The results for the categorie "E-commerce" are:
  1. http://www.sunweb.nl/ 4 stars
  2. http://www.vd.nl/ 4 stars
  3. http://www.zalando.nl/ 4 stars
  4. http://www.mycom.nl 3 stars
  5. http://neckermann.nl/ 3 stars
  6. http://www.bol.com 2 stars
  7. http://www.debijenkorf.nl/ 1 star
  8. http://www.esprit.nl/ 1 star
  9. http://www.greetz.nl 1 star
  10. http://www.wehkamp.nl/ 1 star Winner and overall winner
The results for the category "Government / non-profit" are:
  1. http://www.rijksoverheid.nl/ 5 stars
  2. http://www.wnf.nl 4 stars
  3. http://www.dance4life.nl/ 3 stars
  4. http://www.denhaag.nl/ 3 stars
  5. http://www.jantjebeton.nl/ 3 stars Winner
  6. http://www.consumentenbond.nl/ 3 stars
  7. http://www.rotterdam.nl/ 3 stars
  8. http://www.1procentclub.nl/ 3 stars
  9. http://www.nibud.nl/ 3 stars
  10. http://www.energiebespaarwijzer.nl/ 2 stars
The results for the category "Company website" are:
  1. http://www.nn.nl/ 4 stars Winner
  2. http://www.t-mobile.nl/ 4 stars
  3. http://www.unilever.nl/ 3 stars
  4. http://www.manpower.nl/ 3 stars
  5. http://thuis.eneco.nl 2 stars
  6. http://www.ns.nl/ 2 stars
  7. http://www.vodafone.nl/ 2 stars
  8. http://www.philips.nl/ 2 stars
  9. http://www.ing.nl/ 2 stars
  10. http://www.heineken.nl/ 1 star
The results for the category "Communities" are:
  1. http://www.goeievraag.nl/ 3 stars
  2. http://www.iens.nl/ 3 stars
  3. http://www.funda.nl/ 3 stars
  4. http://www.beslist.nl/ 3 stars
  5. http://www.hyves.nl/ 3 stars
  6. http://www.independer.nl/ 3 stars
  7. http://www.micazu.nl/ 2 stars
  8. http://www.vliegtickets.nl/ 2 stars
  9. http://www.zoover.nl/ 1 star Winner
  10. http://www.telecomvergelijker.nl/ 0 stars
The results for the category "News and entertainment" are:
  1. http://www.nrc.nl/ 4 stars
  2. http://www.ondertussen.nl/ 4 stars
  3. http://www.nederland24.nl/ 3 stars
  4. http://www.styletoday.nl/ 3 stars
  5. http://www.trouw.nl/ 3 stars
  6. http://www.bnr.nl/ 2 stars
  7. http://www.z24.nl/ 2 stars
  8. http://www.nu.nl/ 2 stars Winner
  9. http://www.vi.nl/ 1 star
  10. https://www.ziggo.nl/ 1 star
Strange enough the score varies in time. However the first conclusion is that there is no relation between website quality and usability, because the website quality is not that good. Hmmm. 

Lets check the winners with http://www.qualidator.com:
  1. wehkamp.nl overall score 75.4%
  2. jantjebeton.nl overall score 76.3%
  3. nn.nl overall score 77.0%
  4. zoover.nl overall score 73.6%
  5. nu.nl overall score 73.1%
Again, no relation. May be the Quickscan of the webguidelines (max score is 47)?

  1. wehkamp.nl score 32 
  2. jantjebeton.nl  score 31
  3. nn.nl  score 39 
  4. zoover.nl  score 30
  5. nu.nl  score 27
Again, no relation. Last test. The W3C validator

  1. wehkamp.nl invalid 48 errors
  2. jantjebeton.nl  invalid 3 errors
  3. nn.nl invalid 3 errors
  4. zoover.nl  invalid 89 errors
  5. nu.nl  invalid 147 errors
Again, no relation. It's a pity.

Monday, 9 May 2011

Hey Lipperhey

This year lipperhey.com celebrates her birthday with a free account for one month. I have tested Lipperhey with the website www.helenahoeve.nl and learned the following:
  1. As Lipperhey user you are allowed to export a rtf report and a csv-file. This way all necessary information is available even if your account is stopped. That's great;
  2. Lipperhey analyses all your webpages, allthough sometimes they seems to make some small mistakes. For example they state that a webpage should load in 500 milli-seconds. They marked two websites as loading to slow. However these webpages loads as fast as other webpages;
    Note: Googles Webmaster Tools stated that "On average, pages in your site take 0.5 seconds to load (updated on May 7, 2011). This is faster than 98% of sites."
  3. Lipperhey checks all webpages, even the webpages without incoming links. In my opinion this  contaminates the results. I have double checked this and indeed there are pages without incoming links.
    Note: These pages are still alive to give a good user experience to search engine users;
  4. The code of a webpage should have a maximum of 70%, while almost all pages of the Helenahoeve 174 of the 200 checked pages seems to use more code than suggested;
  5. Lipperhey checks also non-html files like a KML file and a VCF file, which is in my opinion incorrect;
  6. There is a check on comments, which are useless for the avarage website visitor. Lipperhey states that a short comment is reasonable. Unfortunately there are some webpages with local javascript to calculate the date of an event. This javascript is hidden as comment for older (very old) browsers. I don't think this should be marked as incorrect;
  7. 33 webpages have a meta description which is too long (longer than 150 characters);
  8. 140 webpages have at least one keyword in the meta tags which is not available in the text of the webpage. Oops that might be fatal for top results in search engines;
  9. There seems to be webpages (2%) which use a capital in the location, allthough I thought that all pages use lowercase characters;
  10. Lipperhey checks also for deprecated tags, even for tags I have never heard of, like: "staking". Hmmm, what will the effect of "staking" be? Anyone a clue?
Some recommendations don't make sense in my opinion, like:
  1. An URL should not contain the document type extension .htm (or another extension) to make it easy switch from technology. So in theory it should be easier to switch from index.htm to index.asp or index.php. Hmmm, I doubt how often you switch from technology and if a redirect via .htaccess should not be sufficent;
  2. The use of mailto:e-mail adresses should be avoided to hinder spam-robots. Hmmm, I thought that an user should be able to mail with her own e-mail program. Even the webguidelines suggest not to block the mailto adresses;
  3. The recommendations for the keywords, I thought that they are almost useless.
Overall Lipperhey seems to be a good tool to use. I don't know if it's worth the money, because there are free tools like Googles Webmaster Tools and the Webguidelines Quick Scan which give you also al lot of detailed information about the quality of your site. 

For the Helenahoeve is seems that I have to check duplicate URL's (content) for the pages index.htm and interactievekaart.htm, something I already knew but never got the right priorities. Another thing to do is to improve the meta descriptions (and keywords).

Note: With Website Quality at a Glance you get a quick rough indication of the quality of a website

Saturday, 1 January 2011

Yearly Statistics

The number of visits at the website www.helenahoeve.nl is growing every year. See this graph:

The average number of visits per day was 279 for 2010.