Jeroen Sangers

Today we turned on the heater for the first time, since the temperature dropped below 20°C.

Summer’s over…

There are some sites, whose content only consists of answers to Which XXXXX are you? contests.

To me it looks as a very cheap way to generate content, and tell something about yourself without telling anything concrete.

I never added this kind of stuff to my site, untill I found What Pixies song are you? Of course, if they mention my favourite band, I’ll have to do it:

In Heaven, everything is fine In Heaven, everything is fine In Heaven, everything is fine You got your good thing, and I've got mine In Heaven, everything is fine In Heaven, everything is fine In Heaven, everything is fine You got your good thing, and you've got mine

In Heaven

You’re mysterious, picky, and a bit aloof. Some people say you have a very cheery personality, but it’s hard to say because you mask your emotions behind dark pretenses. You enjoy watching obscure movies and going to plays, but by the end of the day you’re usually too wound-up by the ironies of your life to relax.

Mmmm… that doesn’t sound like me. Guess this will be the last post of this kind of contest results.

[Update 2004.01.21] Removed broken link

Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.

In het Nederlands:

Vlgones een oznrdeeok op een Eglnese uvinretsiet mkaat het neit uit in wlkee vloogdre de ltteers in een wrood saatn, het einge wat blegnaijrk is is dat de eretse en de ltaatse ltteer op de jiutse patals saatn. De rset van de ltteers mgoen wllikueirg gpletaast wdoren en je knut vrelvogens gwoeon lzeen wat er saatt. Dit kmot odmat we neit ekle ltteer op zcih lzeen maar het wrood als gheeel.

I guess everybody has received the message above, but it is not true.

I just love numbers, and check the access statistics of my sites almost on a daily basis. Sometimes when I visit a site, I also check the statistics of that site. What kind of people are reading that site? Where do they come from? What were they looking for? All very interesting information…

My own visitors could not have a look behind the screens of this site, since my access statistics are located on the management control panel, safely protected by a password.

The application to generate these statistics is called AWStats, which I was using already before I moved to my current hosting provider. Since I knew the structure of the application, I figured that it should be possible to install AWStats myself, and have my copy point to the data files that are generated automatically by my provider's copy of AWStats.

And that is exactly what I've done. I only needed to copy the main awstats.pl file, the lib directory and the icons directories, and could use the existing configuration file (after changing the DirCgi parameter) to have everything working.

So now everybody can check my access statistics!!!

Maybe someday I will create an About this site page, on which I will put a permanent link to my statistics. Some people put this link on their home page, but in my opinion this feature is of no use to the general visitor, and only adds bloat to the home page.

There are many ways to keep contact with the rest of the world.

By far the best way is to meet them in a bar, drinking a beer together. Unfortunately, this is not always a viable solution, so I also make use of electronics to keep contact.

Number two on my list is the good ol' phone, which has the benefit of meta-communication (emotions, emphasis,…) over the other electronic methods.

Then I use of course the internet to communicate, I am using e-mail (more personal), Usenet newsgroups (good for technical problems) and this site.

However, it might take a while before I receive feedback to the things I express using these media. Then we have the two I’s: IRC and IM. The first I use when I have a technical problem that needs to be fixed urgently, the second for all personal communication for which using the telephone would be overkill.

Considering IM, at work I use Microsoft Messenger a lot. We have a lot of contacts abroad, and small companies are not able to hang on the phone to talk with the other side of the world for half the day.

Since we are basically a Microsoft shop, and most people have Messenger already installed on their machines, we are using this.

Personally, I am not a big fan of Messenger, mainly because you have no option of connecting without giving away your e-mail address. For business use this is no problem, but for my personal use, I prefer a solution that does not give away more than strictly necessary.

Until I moved to Spain I have been using ICQ at home, and today I started using it again to be able to solve some problems on the fimcap site.

So back to ICQ! I installed the Lite version at home, and everywhere else I can use the web version. For those people who also use ICQ: my number is 11807773. Everybody else can use my Messaging Center to send me a message when I’m on-line.

Visitors using older browsers like IE might have noticed that I recently started serving advertisements through Google.

People using more modern browsers as Mozilla or Firebird might not have noticed this, since these advertisements are not visible for them. This is not because I want to tease IE users (everybody is free to use any software to visit my site), or give compliments to Moz users (their clicks are also worth some money), but only because Google’s scripts are not working when send as application/xhtml+xml. I found the following errors in my Javascript console:

Error: w.google_ad_client has no properties Source File: http://pagead2.googlesyndication.com/pagead/show_ads.js Line: 37 Error: google_ad_width is not defined Source File: http://pagead2.googlesyndication.com/pagead/show_ads.js Line: 128

Of course I reported these errors to Google, but all I received was the following reply:

Hello Jeroen,

The errors that you were experiencing last night were due to an AdSense system upgrade yesterday evening. This upgrade unfortunately created some technical difficulties and were temporarily unable to serve ads to sites in certain browsers. Not all browsers were effected.

However, our engineers have identified the problem, and I am happy to inform you that the issue is now resolved. Ads are now running on AdSense websites and you should not be seeing these errors anymore.

We value you as an AdSense publisher, and sincerely apologize for any inconvenience we may have caused you. If you are continuing to have problems with this or have any other questions or concerns, please let us know by replying to this email. One of our AdSense specialists would be happy to assist you. Thank you for your patience in this matter.

Sincerely,

The Google Team

Of course I answered them that the problem still exists, but until now I haven’t received an answer.

[Update 2003.09.26] Answer from Google:

Hello Jeroen,

Thank you for writing us about this issue you are experiencing with your AdSense code. Unfortunately we are unable to replicate this problem. We taking your page, http://braintags.com/archives/2003/08/applicationxhtmlxml/, and running it through the W3C’s XML validator tool ( http://validator.w3.org/), we find that the entire page including our JavaScript code is currently XML complaint. (See Results: http://validator.w3.org/check?uri=http%3A%2F%2Fjeroensangers.com%2Farchives%2F2003%2F08%2Fapplicationxhtmlxml%2F). Unfortunately we are unable to replicate this problem you are experiencing.

From previous emails it appears you state you are having this problem in Mozilla 1.4. Here is a list of browsers that supports our code https://www.google.com/adsense/faq-tech#q10. Are there other browsers on this list that you are having difficulty with? Also can you include a screenshot of the error since we are unable to replicate this problem on our end. Please include any additional information you may find may be useful in helping to resolve this issue.

Sincerely,

The Google Team

It looks like they don’t understand anything about specifying the MIME type when sending a page, so I send the following reply.

Hello,

I know that my page is XML compliant, that surely is not the problem.

You have not been able to replicate the problem, because you have not used a gecko-based browser. Normally I send my pages with MIME type text/html. This is what IE and the W3C validator get. Only to browsers which tell me in their request headers that they accept the MIME type application/xhtml+xml (like Mozilla and Firebird), I use this MIME type to send my files.

Note that even though the validator did not complain, I did not send it a valid page, since the specification tells that XHTML1.1 should be delivered as application/xhtml+xml.

To reproduce the error, just visit my page with Mozilla or Firebird, and you will see that the AdSense box will not appear. The only screenshot I can send you is the JavaScript console, but it only contains the two messages I have send before.

Kind regards,

Jeroen Sangers

[Update 2003.10.01] And finally Google realises that their script is not working:

Hello Jeroen,

We appreciate your taking the time to offer us this feedback and encourage you to continue to let us know how we can improve Google AdSense. As AdSense is still a young program, new features are under consideration and your feedback is very helpful. This information has been passed on to our engineers and they will look into allowing our ads to be served in xhtml+xml pages.

Sincerely, The Google Team

It took them almost a month (I reported this problem on September 4^th) to realise that something is wrong. Let’s see how long it will take them to fix this problem.

How could I have missed the news! One of my favourite bands is getting together again.

Although the Pixies are not well-known by the general public, they have greatly influenced many big bands as Nirvana and the Distillers.

I was lucky enough to see them perform life in Vredenburg, Utrecht during their Doolittle tour, and later have seen spin-offs Frank Black & the Catholics and the Breeders; all very impressive concerts.

I surely hope that they will cross the Atlantic with their new tour.

So many things to do, and only 24 hours in a day. I guess everybody sometimes has the feeling that too many things lie around unfinished.

For me, these are some small, usually technical, things as fixing the stylesheet of this site so the menu appears on the right place in IE, converting this site to PHP (so I can degrade pages to XHTML1.0 for IE), fix the light in the stairway, reading hundreds of pages I have bookmarked lately, look for a new computer for my in-laws, build a wireless network at home, put the latest Link on the Fimcap site, paint our front door, learn how to use ZOPE for future use for Fimcap, install freeBSD on my laptop…

Looking at this list, I wonder where all my time is going. Lets see: 9 hours sleeping, 9 hours working, 1.5 hours household (cooking, cleaning, ironing), 2.5 quality time (hours drinking beers on a terrace, watching TV on the sofa, reading the newspaper), 1 hour taking care about myself (eating & shower)… that leaves me 1 hour each day to be productive! I definitely need more hours in a day.

A good holiday is a holiday without internet connection. I had a quite good holiday, so you haven’t read any news from me lately. I did connect to the internet once in a while, but that was only to fix the computer of my parents and to install Mozilla Firebird (if you’re learning to surf the internet, you better do it right). I won’t bother you with my holiday stories, since we didn’t do that much. I was happy to see my friends and family again, and for those that I could not meet: sorry, maybe next time better.

For them moment I have plenty of things to arrange, so I won’t fix the menu bar on this site to work with IE and probably I won’t write a lot neither. That’s all for the moment, I better continue processing my overly full inbox…

I use this site a lot for testing out things. You could say that this site is Forever under construction. Today I have been testing again, and as a result, it looks more horrible in Internet Explorer. And since I am going on holidays, I am not going to fix it until next week. Sorry for the inconvenience.

I have been working on the menu. The whole page header is situated inside a <div>, with an inline ordered list with bottom border as menu. Until today, I did nothing to position the menu, which resulted that the menu walked outside the header if you changed the font size. Not too beautiful…

So today I relatively positioned the header, and specified an absolute position of the menu, so it is always at the border of the containing positioned element, the header.

It solves the problem perfectly, but not in IE. I tested the new stylesheet in IE6, and the menu appeared at the bottom of the screen instead of the bottom of the header. I don’t know yet why IE is behaving like this, but promise to look further into it next week.

Braintags is also available in WML format, so you can read your favourite 😉 site using your WAP enabled telephone. Just point your telephone to http://wap.braintags.com/.

As this is the first time I tried to create something in WML, it surely is full with mistakes. Just let me know, so I can change it and learn something more.

If you are creating (X)HTML code, you always have to make sure that you use the right elements to specify the meaning of certain parts of the text. After that you can use stylesheets to determine how these parts will look. This is what they call the semantics of the text.

A lot of web designers abuse tags to accomplish a certain look. If they want a bigger text, they use the <h2> element, while the text is not a header at all, if they want to indent a text, they use <blockquote>, even though they are not quoting. I won’t go into detail why this is a bad thing, there are plenty of resources explaining this better than I could. But a lot of designers like me who try to do it correctly still make some mistakes. Today I realised that my pages have two semantically errors:

For my navigation lists, I use—as many other sites—unordered lists ( <ul><li>XXX</li></ul>). But my navigation links do have an order! My archive links are ordered chronological, so the correct thing would be to use the ordered list (<ol><li>XXX</li></ol>) instead.

By default, an ordered list is shown with numbers in front of each item, but it is very easy to remove them, just as I removed the bullets from my unordered navigation lists. I have seen many sites doing this wrong, and most examples of you will find on internet also use unordered lists.

Another thing I thought about today is that when I display trackbacks (comments on another site about my text) I am quoting another site. Therefore, I should be using the <blockquote> element (or <q>). I have looked around at other sites, but nowhere have I found trackbacks marked up as quotes.

You might think that these are minor issues, but since I am rewriting my templates anyway, I better do it right, not?

You might already have noticed it: I added some advertisements to this site.

Since now I am paying a hosting provider to host this site, I thought that I might try to get back at least some of the money I spend for it.

The advertisements are related to the contents of the page, so I only added them the pages of the individual items. This makes the chance bigger that they may actually be useful for my visitors as well.

I decided to give it a try for about two months; if I notice at that moment that they are not working, I will remove them.

People using a modern browser won’t see the advertisements, since the advertisement script does not work when the page is processed ad XML.

One more week to go, and we have our second short vacation of this year!!!

After going to Croatia in July, we will visit the good ol' Netherlands this time. From September 9 until 16 you will have a bigger than normal chance of seeing us in the north of Europe.

We didn’t make an exact planning of what we are going to do, but we have made two short lists.

One with some practical things I have to arrange, and one with some places we want to visit. The first list has to be followed, and the second list will almost surely change on a day-by-day basis.

The moment I changed the looks of this site, I also changed the version of the code used for these pages.

My site is now made up in XHTML version 1.1. I only forgot to change one thing.

I did change some elements to their new variants, and of course I included the right DTD. But my pages were still served to your browser with the text/html MIME type, while the specification tells me that I should serve my pages as application/xhtml+xml.

The specification is quite clear in this, but as usual the reality is more complicated. I could simple change the file extensions to .xhtml, which are automatically send with the right MIME type. But at this moment a lot of browsers (including IE6) are not able to handle this kind of documents.

So I need to distinguish browsers that can handle application/xhtml+xml from those that cannot handle it. Fortunately, most browsers send a list of MIME types they can handle each time they request a file from the server; so all I have to do is look at this list, and change the MIME type for modern browsers.

Therefore I added the following lines to my .htaccess file:

RewriteCond %{HTTP_ACCEPT} application/xhtml+xml RewriteCond %{HTTP_ACCEPT} !application/xhtml+xml\s;\sq=0 RewriteCond %{REQUEST_URI} .html$ RewriteCond %{THE_REQUEST} HTTP/1.1 RewriteCond %{HTTP_HOST} !photos.braintags.com RewriteRule .* - [T=application/xhtml+xml]

The first line looks whether the MIME type is in the accepted list of the browser.

The second lines looks whether it does not have q=0 (q stands for quality), which means that the browser does not want to use this MIME type.

The third line limits the requests to .html files only, since we do not want to change the MIME type for images, stylesheets and other files.

The fourth line limits the requests to HTTP version 1.1.

The fifth line excludes my photo pages, since I haven’t updated them to XHTML1.1 yet.

And the last line finally specifies the right MIME type for requests that satisfy all conditions above.

Throught the wonderful Internet Archive Wayback Machine, I was able to visit my old site again.

My biggest proud of that site was the big list of all the CD’s I own. So here it is, the list of March 12, 2001!

Today I noticed something strange in the access report of this site. In the error codes list I found error 411.

Since I haven’t seen this error code before, I googled on internet for some more information.

The official RFC 2616 says the following: The server refuses to accept the request without a defined Content-Length. The client MAY repeat the request if it adds a valid Content-Length header field containing the length of the message-body in the request message.

It appears that a server is returning code 411 when a client (=browser) tries to send data to the server without specifying the size of the data. Normally a client only receives data from the server (reading pages), and all common modern browsers are programmed in such a way that they include the data length when sending data.

Therefore I suspected that somebody with a home-grown application had tried to mess with my site. I did a quick grep in my Apache log files and found out that somebody had accessed the file mt-xmlrpc.cgi with an application that identified itself as Java/1.4.1_01.

This file is a XML-RPC interface to Movable Type, allowing other applications to interact with MT, for example to publish entries on this site from a special desktop client.

Since I am the only responsible for this site, nobody but me has any reason to use this file. This visitor also triggered my Spam trap, and was already blocked from my system, but if not, his badly programmed application did ring the alarm bell.

I just found a small list on my desk, and remembered that I wrote it last Friday (or better said! Saturday morning) to publish it later on this site.

The aim was to make it clear that I had a great night, visiting my favourite club Pentagrama again.

These are the titles and artists of some songs they played, and people who know me and these songs can imaging that I was enjoying a lot.

• London calling — Clash
• Personal Jesus — Depeche Mode
• Black Betty — Ram Jam
• Tunnel of love — Dire Straits
• Should I stay or should I go — Clash
• Entre dos tierras — Héroes del Silencio

[Update 2003.08.28]: Coen van Kuijk sent me a list which contained only one mistake. After a small hint he was able to give me all the names of the artists.

[Update 2003.09.03]: Since I haven’t received any other messages, I decided to fill in the artists and close the competition.

When I designed this site, I of course tested the layout on my Windows computer. I know that it looks alright with IE6 and Mozilla, and I suspect that it looks acceptable in most other OS/browser combinations.

If I would be a professional designer, I would run it through the famous BrowserCam, but my little site is not worth the money.

Of course there are some emulators available, which might give you an idea of how your site looks, but remember that if something looks wrong you never know for sure whether it is your code or the emulator.

A new emulator I found today, shows how this site looks on an iPaq PDA.

Mmmm. Header to big, no skip navigation link, useless side column, font too big, … But still readable. I also loaded the Fimcap site, which still has a tables based layout, and that looked way worse. To read the text you have to keep on scrolling.

I guess I will have to make a new stylesheet, or maybe serve PDA’s the print stylesheet. But the print stylesheet removes all the navigational elements…

Anyway, the coming days I will tweak this site so it will look better for my mobile readers — do I have any?

First a question for the music industry: “What the heck are you doing?”

Yesterday MJ gave me a present; the CD Fleshwounds from Skin.

Of course I was really happy with it, but I immediately noticed a big label on both sides with the words COPY CONTROLLED. In the specifications I read that this CD would work in any CD-player, the latest Macs and PCs with minimal Pentium II, 233 MHz and 64MB RAM. I am perfectly happy with my Cyrix P133+ with 48MB RAM, and until now I was able to play any CD I liked.

Of course I tried to play the CD, but my computer could not find the CD. 🙁

I really like buying CD’s; I prefer to buy one to using a copied CD. Because for me the CD is more than just the disk with the bits and bytes. I like to look at the artwork in the booklet, and no inkjet printed paper gives me the same feeling as the original CD.

And because I usually buy older CD’s, I never feel that I pay a lot; of course I wouldn’t complain if they became cheaper.

So I buy the original CD’s because of the better quality of the total product.

But now the record companies are throwing in their own windows, because they decrease the quality of their product. And why would I want to pay a lot for something that is broken?

I don’t have any problem yet, since there are so many old CD’s out there that I still want to buy. But I can assure that I am never ever going to buy a copy protected CD anymore. Not because I want to copy it, which I feel is my right, but because it is essentially a broken product.

You can read more about copy-protected CD’s at Kuro5hin or if you truely want to know everything you can go to the UK campain for Digital Rights.

Most viruses that replicate themselves through e-mail forge the from: field of their messages. Usually they take a random address, or pick one from your address book or web cache. Therefore, a virus appearing to come from jeroen@example.com, rarely really originates from this address.

I can image that some users don’t know this, and when they receive a virus from me they send me a reply with a warning that I have a virus.

Usually I explain these people the story above and tell them, of course after checking that my anti-virus software is up-to-date, that I don’t have a virus. No problem for me.

What I don’t understand is that some people who are responsible for really big mail servers also don’t understand that viruses forge headers, and have their servers configured in such a way that they reply to me.

Of course they have the right to block a message with a virus to protect their customers. But please don’t bother me with these messages. I can assure you that I am not the one sending these messages; just check the mail headers and you will see that they originate from a completely different server.

This morning somebody who apparently had visited my site cached a virus and started sending out the virus in my name. Luckily for him, his provider scans all incoming SMTP mail and intercepted all messages.

Instead of simply not accepting these messages, they had configured procmail to send a warning to the address mentioned in the from-field (me) including the virus!!!!!

As a result, my inbox is flooded with viruses from this server (mail.infosys.tuwien.ac.at).

Once again, none of my computers is infected with a virus.

I have written before about the Movable Type plug-ins I use for this site; at this moment I am using seven different plug-ins.

The more plug-ins you use, the more difficult it becomes to maintain the site. Because once in a while new versions appear, and I don’t want to scan seven (eight if you include MT itself) sites to see if updates are available.

But now there is the MT Plug-ins Manager. I downloaded the files, placed them on the server and fired up the page. A neat list with all available plug-ins appeared.

Since I didn’t install my plug-ins with this tool, the plug-in manager did not know yet which plug-ins I already have installed. So I went to the Manual Registration section, where I could specify the version number and plug-in name of each file in my plugins directory.

After entering these details I went back to my main page, and noticed an update button next to the IfEmpty plug-in. I was using version 1.1, while the latest version is 1.11. I clicked the button, waited some seconds and my plug-in was updated.

This is great!!! No more downloading from different sites, unzipping, FTP-ing,… One click on the button does the job.

Tomorrow evening my parents will be flying back home.

They have been living with us for one week now, and it has been really wonderful. At this moment they are enjoying the weather in the swimming pool, just like most other afternoons.

Before going home, they will say goodbye to my in-laws.

They can get along very well, despite of the huge communication problems they have.

Tonight we will surely play some cards (for the last time) and have a small dinner. Tomorrow they will pack their backs, go one more time to the swimming pool, and catch the airplane.

We will surely miss them; our house will be more empty without them. But the good news is that we will see them again when we will visit the Netherlands next month.

I am having some problems in the comment form of my individual archives.

After entering a comment, the form correctly sets the mtcmtauth, mtcmtmail and mtcmthome cookies.

But if I view the page later, the contents of these cookies are not loaded into their respective fields (I’ve tried it with Moz and IE6). My browser tells me that document.comments_form.email is zero or not an object. Of course I checked my template, but I cannot find anything strange.

The <form> element has id=“comments_form”, and the field is also nicely called email.

I don’t know where to look further anymore. Can somebody have a look at this page, and help me with the solution?