Branko’s First Law of Web Dynamics

Branko’s First Law of Web Dynamics is: “The less relevant a part of a web page is to the page’s subject, the less prominence it should have on that web page.”

Or, to put it differently: relevancy first!

Odd, that no-one should have formulated this law before. (Well, not that I know of anyway.)

I was thinking about this in the context of advertising on the web. Did you ever notice how the most annoying ads on the web often are the least relevant to what you were looking for? Is this because the ads are so annoying, that any relevance they might have had immediately dissipates? Or is it because the advertisers realize that this particular ad is so irrelevant that it is not going to bag them eyeballs, unless they display them over anything else? Or (my favourite) is it because the sort of advertiser that would place irrelevant ads on a website, is also the sort of advertiser that would use annoying (read: prominence-boosting) techniques? Or (equally likely) is it because annoying ads appear on high-traffic websites, where a tiny percentage of conversion is still better than a large percentage of conversion on a small website?

The oddest thing is of course that there are irrelevant ads in the first place. If I were head of a marketing department, I would be making sure that all my minions spent the bulk of their time ensuring that ads are targeted at readers who are interested in them.

It’s official: civil servants are nuts

The Dutch department for internal affairs has seen fit to publish a style guide for ministerial websites. It is full of sound and meaningful advice, such as:

  • Do not use upper case letters, spaces and special characters in file and path names.
  • Use logical and simple URLs.
  • Use unique URLs that will not change.

This style guide can be found at:

For now.

(Spotted in a posting by René Pijlman in nl.internet.www.ontwerp.)

Reporting bugs to FOSS projects

Graphic designer Jimmac, whose work you may be familiar with if you use GNOME or the GIMP (amongst others), juxtaposes two approaches to reporting bugs or feature requests: “This blurb has been inspired by a badly formulated feature request on the GIMP mailing list. Quite a contrast to a recent mail to the Inkscape mailing list which has been promptly implemented.

I am not sure I agree with his conclusion, which seems to be that the reporter needs to do as much work as possible before approaching the poor, overworked developers. For one thing, not every patch gets accepted. For another, sometimes it can be useful to knock an idea about with a couple of like-minded people (not necessarily the developers) before hashing it out.

Unfortunately, the tools with which developers are interacting with users (if I may suggest that a developer/user distinction exists) are often inadequate. The GIMP uses Bugzilla for reporting bugs. Bugzilla is a great tool: I couldn’t imagine life without it. But on the same hand, it has got a far way to go.

Some of the problems with Bugzilla (or at least the version the GIMP used when I last filled out a bug report):

Read the rest of this entry »

E-mail usability

There’s a discussion at Slashdot about e-mail usability. Actually, it is an Ask Slashdot thread, with a sysadmin wanting to know how to force users into responsible e-mail management; which to many means saving messages to a place where the client will not get at them automatically, and throwing messages away that you do not think are important enough to keep.

Both methods are so that your .pst file will remain manageably small, which apparently is nice for sysadmins. I know from experience that Outlook tends to have problems with large .pst files, so apparently this sort of management would be beneficial to users too.

There is a school of thought in interface design that says that whatever users come up with themselves is good, and I think this holds true for e-mail.

For instance, my experience is that users typically subdivide their Inbox into several layers of mailboxes, and transfer messages there accordingly. In other words, they manage their Inbox. Also, users use the search function of their e-mail clients. These are good things that do not transfer readily to other methods of managing e-mail. What’s more, throwing messages away, or stashing them in “archives” that the client will not access when the user is searching or browsing e-mail, breaks the user experience of being able to access everything they ever wrote or read using that particular persona.

Of course, from a system administrator’s point of view, e-mail clients could be much and much better. Backing up and restoring e-mail is currently far too error prone. But there is no reason to punish users for using a perfectly usable tool.

Guaranteed downloading

[cropped screenshot]

Linotype promises to replace any fonts that you bought and may have displaced or damaged. I thought this so remarkable, that I’d blog about. The truly remarkable thing is of course that this sort of service should not be remarkable in the first place, but standard practice.

There are other companies that replace the copyrighted works you license from them, but usually at a cost, and not so prominently through their website.

Some other features of the Linotype website that seems to indicate that the company, unlike so many others, think that they can engage with their customers in a friendly manner, rather than through all-out war: the download of the Font Explorer font manager (their iTunes of fonts) does not require you to register for sixty newsletters, and they link to font tools by third parties, meaning they trust outsiders.

These are little things, but it makes them look trustworthy in return.

Firefox to become “spyware” agent?

The Firefox developers have added support for a non-W3 attribute that allows webmasters to “spy” on you. This is already possible using cookies, redirects and JavaScript; the developers added the feature in order to reduce overhead that slows down the web browsing experience.

I have not yet made up my mind as to whether this ping attribute is desirable. It reminds me of giving methadon to heroin addicts (in order to ween them of the latter drug); a method that is debated up to this day. The idea that the developers have is that we, the web surfers, are going to get cheated anyway, so why not make the experience slighty less unpleasant for us?

I wrote the following as a comment at the blog entry that announced the feature:

The reasoning some folks employ here seems to be that if other folks do bad things, they themselves should be able to introduce entirely new bad things. The famous arms manufacturer argument: Firefox doesn’t hurt people, people hurt people. I hope I do not have to explain what’s wrong with that argument. (In case I do: think BLINK.)

If you decide to implement a feature according to spec, one would expect you to implement the entire feature, which includes the GUI. People who click links may keep an eye on the status bar because they have gotten used to getting a raw deal on the web (this should provide you with a hint about the importance of knowing what a click means!), so the status bar seems the logical location for any feedback on the type of link you’re clicking. But the address that typically appears in the status bar is situational knowledge; whereas the fact that a link leads to multiple addresses is functional knowledge, and should probably be part of the rendering of the link text or object itself.

Anyway, it is good to see that you are working on an advanced hypertext experience. Does this mean you are going to implement (or already have implemented) fat links too?


Desperately needing to reinstall the entire system, I decided to do a double back-up of all my data, just to make sure nothing vital gets lost. One back-up of important data to CD-ROM, another of most of my data to a second, smaller hard disk.

I have an old pc, and an old old old pc, which had an old old hard disk. The web is full of advice for the case that you wish to put an old old hard disk (what “they” call “new”) into an old old old pc (what “they” call “old”), and aeons ago when I installed the then new Western Digital Caviar 6GB hard drive and disk into my 486DX, that advice came in handy. But now when I wanted to put that drive into my Pentium II, advice came not at all.

Read the rest of this entry »

Just say when

A guy decided to pay his way through college by selling pixels of his homepage for a dollar a piece. An excercise doomed to end in tears of course, since you wouldn’t find enough idiots to even cough up ten bucks collectively. Did I just say cough?


Apparently the planet has enough idiots to pay up for 999,000 pixels. So I would have liked to tell my new best friend; just take the money and run. You already did do the impossible and netted 999,000 dollar.

If somebody did give him that advice, he ignored it. The last 1000 pixels are up for bid at eBay, and are currently going for 43,000 dollar. That’s 43 dollar per pixel.


I don’t think I ever quite got the blogroll concept; the list of links you see in the sidebar of this page are blogs that I try and read daily.

Some of the other sites that I visit from time to time:

Slashdot competitors Webwereld (Dutch), Heise (German) and The Register. Heise is probably the best of these four, but I rarely read it. I used to read Webwereld almost every day, but a recent site overhaul made it a much worse place to be.

GIMP people: Tigert and Carol.

For Teleread I try and keep track of what’s happening at Digibieb (Dutch), the digital library of the town of Oss, and the Digital Media Europe news site (not a blog).

Low-tech web bookmarking solution using browser-based situational knowledge

[slice of Yoho the Blog]

Do you, if you need to scroll a long web page up and back down, for instance because you need to look up the bloke’s name to decide whether it was Weinberg or Weinberger, select part of the text to remember where you were?

Because I do.

(Undoubtedly this is illegal in some countries, like the Directory Traversal Attack.)