Leesvoer van de dag

ie_icon.jpgMet Internet Explorer 7 en haar betere CSS ondersteuning in aantocht zou de vlag uitmoeten bij elke webdesigner maar helaas, zo simpel is het niet.

Dave Shea legt uit “With the imminent launch of IE7 your usual CSS hacking methods are going to fail. If you want to save web design, as we know it, it’s time to take some drastic action.”

De bottomline? Speel op safe en hoop op mooi dingen in de toekomst. Helaas, het is niet anders maar structurele oplossingen voor verschillen in weergave tussen browsers en -versies zijn op korte termijn nog niet in zicht.

More fonts but less choice?

Jason Santa Maria wrote a nice article about The font problems of designers. Yes, there are six new fonts to be released by Microsoft, but at the same time Mac users lose Georgia and Verdana, which are among the most widely used fonts on the web.

Interesting stuff, I like fonts to experiment with, but have never paid much attention to real typographic issues, but this is the stuff the professional graphic designers have to deal with.

Very good news for Drupal and WordPress!!!

On or around March 25, 2005 12:36 PM Chris Messina wrote this: #

I work for CivicSpace and helped create Spread Firefox. The choice to use CivicSpace for SFX was an important one and the right one because the tool fit the need.

In my discussions with Matt, we are working to come up with ways for open-source CMS’ like Drupal and WordPress to coexist and better comingle. It would be excellent, for example, if it were easier to port themes back and forth between the two so that you could run your own personal WordPress blog (as I do) but run your community with CivicSpace or Drupal and have the external (and internal) appearance be consistent for both admins and external visitors.

So Matt and I have been talking about how we can bring our user interfaces closer together so that the experience of moving between either tool is less jarring and more consistent. One real-world thing that’s happening as a result of our collaboration is that Drupal will be moving back to a separate admin UI, to be more consistent with WordPress. This will ultimately, or hopefully, lead to a more seamless upgrade path between WordPress and Drupal — so that you can start out with a personal blog and then scale up to a community when necessary, without having to unlearn or relearn a whole new UI. This improves the whole ecosystem of personal and community publishing.

Though the details are still hazy and is mostly talk at the moment, we are making progress. And since it’s my job to create a very positive user experience in Drupal (and by extension, CivicSpace), working with Matt completely makes sense. With the WP Foundation getting started up, I think that bringing our collective effort together will lead to a lot of really great things for the open source and blogging communities.

No follow, no interconnection

There is a lot of enthousiasm on the web about Googles steps to block comment spamming, by using the rel=”nofollow” attribute of links. That is, when you want to completely block any links originating from your website. And if spammers would stop posting their spam.

Sargasso is an informative site with a lot of commenters who nog only babble away, but also post valuable links and addons to our articles. Removing any links from the comments is a blow for the interconnected web. But there is another option, one that would be much more fun and that would leave the open structure intact. But it would require some manual labour from the search industry.

I think it would be much more efficient if spamsites were removed from the index directly. Set up a few hidden weblogs on several systems, watch what spam appears on those sites and remove the sites they link to completely.

You’d probably have to change sites every few weeks, but after a while, spammers will notice that plugging around their urls on weblogs will only result in removal. And therefor would become counterproductive.

Links in this article:

Google Blog: Preventing comment spam
Ben Hammersly: Let no fellow nofollow, lest we all lie fallow
Simon Collison: Spammers get spannered by Google

+Eric Meyer: More Spam to Follow

Murphy’s graphic design laws

If three designs are shown to a client, your least favorite will be chosen.
If two designs are shown, a third will be requested. If provided, then one of the first two will be chosen.

Hey… I know this sound… it is the sound of webdevelopment reality. The trick though, is not to consider it a problem but as a challenge of skills. I know that it can bequite frustrating, but in the end you go home, and your customer has a site that could have been much better.

There are some true remarks here: Murphy’s graphic design laws

Listit baby

Did you say list? This will help: Listamatic: one list, many options

Incredible collection of lists, all with clear code and easy to implement. Also has links to lists. A real list of links to list resources. I like this linklist a lot 🙂

The best thing about webstandards

The best thing about web standards is never having to say you’re sorry. (Joe)

Well, actually there are a lot of good things about webstandards, but besides that your documents are semantically correct, more accessible, faster to load and platform independent, we now have a way of creating webpages that is ‘the-right-way-to-do-it’ and that puts the burden of not being rendered correctly where it belongs: at the Browser vendor!

So yeah, you do not have to say sorry anymore. If somebody wants to visit valid pages with a poor browser, it’s their problem.

I don’t mean to say that proper rendering shouldn’t interest you though… just that when you have to choose between a hack or using standards… well, you figure it out 🙂

How we really see the web

We don’t read everything on the web. We scan, but we don’t scan everything. We zap from left to right in a ridiculous pace, searching for something that grabs our attention. We don’t really look at photo’s in news articles, unless they are big enough. The same for ads goes as well, size matters and we look longer, a lot longer, if it’s a text ad close to the main content of the page.

We have a way with bold intro paragraphs: we actually read them! Paragraphs’ best size is around 45-50 words, much longer will result in less attention. We also don’t read all the headlines and intros but spread our attention what results in a heatmaps of attention.

Eyetrack2004 is a loosely held investigation of what people watch when they visit newssites. Some of thse issues already surfaced in usability studies, but there are some pretty interesting findings in their public report. Which is also available as a 340 page downloadable pdf. One thing is clear: there definately needs to be more study by independent institutions of the way the user interacts with a website.

What we’ll find may very wel not please designers, and the final result may not be as pretty as it could be, but we’ll be sure that it will work.

The Hell they call Specs

I have seen the worst specs ever: A phonebook of descriptions, channels of content laid out in such detail that you would know exactly where which pixel should be. If you were able to read it of course.
I have also seen those very, very flexible specs. “Just make it send an e-card.. or something like that”.

The problem with specifications is that they can be too light or too heavy, and there is no way of knowing which is right for a project before you start. If you are really lucky, you work with thinking programmers who are able to correct mistakes, suggest improvements and fill in the gaps. They know the intention of the project and know what works and what doesn’t and most important: they aren’t afraid to speak up. Those people are rare, which is one of the reasons why a lot of websites and applications suck. In desktop software teams work for years on a piece of software, but on the web most projects are temporary and when it is done it is done. Not the best product matters but a cool look and acceptible coding in the shortest time possible.

The right specs differ from programmer to programmer, and one of the reasons for that is that we don’t all have the same imagination, (project) methodology, vocabulary, interest in users or even common knowledge.

Mark states it a bit bolder: there are two kinds of programmers… the assholes and the morons. Why specs matter. Some of the things he writes sound familiar, don’t they?