Weblog: January 6, 2004

Entries

Weblog entries posted on this day.

New design, new ideas, new day

Posted
January 6 2004

Screenshot of this weblog in Mozilla.

Yup, I have done it again. I put up an all new style for this weblog right before I went to bed. Well, an all new style sheet, that is. The design didn't actually change that much. Can I even say design? I think the styles of this weblog are more and more CSS tests than anything else. I also added an automated blo.gs thingy, that shows some of my favorite sites, and when they were last updated. The technique might not be pretty (JavaScript document.write), but I like the result.

After my little update, I was preparing things to go to bed (it was around 6:30 AM by then). During these processes, certain things about the future of my personal website started to become clear to me. I grabbed my dummy book, and started writing and sketching (workaholic, anyone?). I think I have some great ideas about the title, domain, content, technologies, style, etc.

One big drawback —and huge annoyance— is that the two domain names that I want/need have already been taken. I knew of the first for some time, and just discovered the second one I thought of has been taken also. What is really sad about it is that both domains are not the primary domains of their respective owners. One is just a page that tells you that they have a new name and a new domain (ugh!), and the other one functions as a mirror for people that can't type three characters in the proper order (argh!). I think I'll contact the owners sooner or later, but I doubt they will be letting the domains go for anything less than a fortune.

Before I started writing my new style sheet, I dropped Bert Bos an email yesterday. I found some errors on the CSS: fonts example page. In the Font families, where the label says Arial it's actually showing the Helvetica, and where it says Helvetica it's actually showing the Verdana. First, I thought I was tripping out of my mind, but when I checked the source, I saw I was right. ;-)

I felt like complete shit yesterday. My head felt like it was going to explode, I was couching up things that looked like they could raise a family, and I experienced several anxiety attacks. I feel tons better today... I don't get it.

The quest for the ffi ligature

Posted
January 6 2004

Something that has been a personal annoyance factor in webdesign for many years now, is the fact that the fi, ffi, fl and ffl ligatures are not supported in HTML documents. Other ligatures, like the ae ligature (æ) have been properly supported, while the —for me— most important, have not. I had put it at rest, and got on with my webdesigning life, but I've never really got comfortable with it. The way the the first f seems to be hesitating to touch the next letter, and the eerie white space, are things I will just not take for granted. There has got to be a way.

Today, it really got on my nerves again, and I decided to act to it. The general idea was to add a module to the Document Type Definition. The XHTML1.x DTDs have three entity declarations, being xhtml1-lat1, xhtml1-special, and xhtml1-symbol. Needless to say, non of these contain character data of any of the f ligatures. The ISOPUB Entity Declaration, on the other hand, supports the ffi ligature (ffi), ff ligature (ff), etc. It only seemed logical to add the ISOPUB Entity Declaration to the list of Entity Declarations of the Document Type Definition.


<!ENTITY % isopub PUBLIC
  "ISO 8879:1986//ENTITIES Publishing//EN//XML"
  "isopub.ent"
>
%isopub;

&ffilig; kept being displayed as &ffilig;, so I decided to take a somewhat more drastical step. I decided to write my own tiny little markup language (called ACJ), and write a little DTD for it, with two elements (fi and ffi), and with ISOPUB as its only entity declaration. I then created HTML, XHTML, and XML files with xmlns:acj specified in the html element, and with the following code in the body:


<acj:fi xmlns="http://home.planet.nl/~acjbizar/tests/20040105/omg.dtd">&ffilig;</acj:fi>

Unfortunately, no luck, still. The only thing that seems to be working in at least some browsers is &#x0FB03;... but that appears to be some standard thing, rather than the actual ffi ligature from its parents font family. It causes more trouble than it's worth (it doesn't pay too much attention to CSS, etc).

Argh! I guess I will keep searching...

External

This day, elsewhere.