Archive for the ‘html’ Category

OpenID.si launched!

Monday, April 9th, 2007

I’ve launched a Slovenian OpenID site at openid.si. I’m looking for other Slovenian OpenID enthusiasts that would help build a meaningful center for all Slovenian OpenID users and developers.

It’s a long road that OpenID has to travel to become a serious player on the landscape of the internet. If you think it’s the right way to go and are willing to do something about it you’re welcome to join us – contact details at openid.si.

At the same time I would like to mention that we’ll be having (probably the first Slovenian) OpenID talk at Spletne urice on 18 April at 19:00. Sebastjan Trepca, the founder of Slovenian Orthodox Users of Python, developer of Marela and a web developer at Parsek, will be talking about what OpenID is and how to implement a client and a server.

If you want to read more about OpenID head over to openid.si to find a bunch of OpenID resource links. If you have more don’t hesitate to email.

WPF/E competition

Thursday, March 29th, 2007

Yesterday was a fine day that ended with a talk about the WindowsVista.si website (now offline). It’s made with WPF/E technology and is made to mimic the look & feel of the real Vista operating system. A great showcase of the technology.

There’s been much talk about WPF. The whole Windows Foundation Platform seems to be competition to the Adobe Apollo platform. They’re both made to create desktop applications. Apollo seems to be on top with the cross-platformness while Microsoft is putting its hopes on the size of the Windows developer community.

When we come ‘down’ to WPF/E (the E stands for Everywhere), the competitor everyone is talking about is Adobe (formerly Macromedia) Flash. They both solve a lot of common problems – animation, multimedia, drawing – stuff that you can’t do in HTML. But when you look under the hood of WindowsVista.si you’ll find there’s a bunch of JavaScripts that seem to do all the magic. And the code looks much like when you’re working with the infamous <canvas> tag.

This was also confirmed by the developer of the page – due to the limitations of the current plugin and it’s work with XAML everything on the page is dynamically created with JavaScript and is not present in the source XAML file. Since there are no components available he actually wrote all the controls himself – tabview, scrolling, panes, menus, windows,…

Two things come to mind:

  1. Direct3D vs OpenGL battle that went on about a decade ago
  2. document.layer vs document.all and the time of the DynamicDuo

Seems like a good idea would be to write a library that will seamlessly switch between WPF/E, and Flash whether they’re present or not. Especially since the tag will obviouslly never be trully cross platform (at least for a while) and that WPF/E doesn’t yet have a plugin for all platforms & browsers. Then again – what’s wrong with Flash?

Dear Apple

Friday, March 16th, 2007

Coolness is like trust. It’s hard to earn and easy to lose.

Update: Others seem to have finally gotten why this is more than just an IP issue.

The tables – part 1

Wednesday, March 14th, 2007

In this post I’ll go into the semantics of tables. Tables are still hated by many web developers and since sometimes a firm ‘NO’ is the best way of changing things some people are still scared of using them. I’ll say it out loud:

Using tables in HTML is OK.

That is if you’re using them for tabular data. Some things are a table no matter what. To rephrase – if it looks like a table it IS a table.

The beginning

The table is a block level element so it goes anywhere a block level element can. It can even fit inside the button element.

Every table begins with a table tag. Needless to say that the table needs to be opened and closed.

<table>

The table element can have attributes. One of them that should be user more often is summary that describes the contents of the table. There are a few attributes that actually control the display of the table and should probably not be used, or at least be used as little as possible. These are all still valid even in XHTML1 strict: frame, rules, cellspacing, cellpadding, width and border. Align and bgcolor attributes were dropped in strict and should not be used anymore. You can achieve almost everything that the last two groups of attributes are there for with CSS. You might stumble upon some problems though.

List of attributes:
  • summary
  • frame
  • rules
  • cellspacing
  • cellpadding
  • width
  • border
  • align
  • bgcolor

The regular attributes apply, a special case is the dir attribute that sets the direction of the table. With tables right-to-left (RTL) means the first cell will be on the right and the text will start on the right.

The header that’s not the head

Before we get to the content of the tables we need to check some other tags.

<caption>

The caption tag is used to “describe the nature of the table“. The caption element is visible to all users and should be rendered as wide as the table. With the summary attribute it should provide all the information about the table to people using non visual renderers. There can only be on caption element and it has to be the first child in the table element (ignoring whitespace nodes of course).

There are some issues with rendering the caption though:

  • Firefox renders the tables border differently that the captions border which creates a 1px difference – corrected with margin-left:-1px
  • Internet Explorer 6 ignores margines on the caption element entirely. It also ignores the border collapse between the table cells and the caption
  • caption-side property that defines if a caption is placed at the top or the bottom of the table is not really supported across A-grade browsers at the time of writing (Firefox supports it, but not changing top to bottom via scripting)

Read more about styling table captions.

Next time…

In the next edition of the tables we’ll go into rarely used colgroup and col tags.

Sources

What do you want in HTML?

Thursday, November 9th, 2006

Have your say about the future of HTML.

What I’d want:
– Better form controls. But only if they follow other computer conventions.
– Application markup. We already have most things for marking documents.

It’s hard thinking about this since what I want should really be written on small papers, put on a desk and then sorted into structure (HTML), presentation (CSS) and behaviour (JavaScript). But I don’t really have time for this. Do you?

Update: Another reason for the lack of ideas might be that I just learned to work with the limitations. They’re so built into my brain that I’d have to go check some of the pages I made to see where I might have been exploiting the current standard..

Triptracker

Saturday, October 28th, 2006

Spletne urice served its visitors a great talk from one of the founders of a Slovenian company called Klika. It’s primarily a software company that specializes in tracking software with GPS and then mapping the data to support sporting and other events (like the Ljubljana marathon that’s on tomorrow).

But the talk wasn’t about their programming but rather about the first global Slovenian Web2.0 website/service – the TripTracker. In short it’s a website that allows you to upload positional data along with pictures, matches them all up by time, draws a map and more or less creates the whole trip into a nice little presentation that you can use to show at meets or just send to your friends. It’s also a good resource for travelers that want to travel to places they know almost nothing about – you can check other people’s trips and see where they went and what they saw and see any comments they might have made.

The most fascinating thing about all this is that they created all their mapping data themselves – all the JavaScript and html for the mapping was written for this site exclusively and all the images converted from NASA sources (I hear Google has exclusive rights on HQ satellite photos so no other mapping service can use them). As the talk progressed questions were popping to my mind but were soon answered – the API is in progress, the service can use Flickr as the image resource and they might even switch their own mapping tool with Google Maps. One thing that came as a surprise is that they’re not using Microformats even though it clearly supports stuff they present on their site. They do allow export to KML though. As all true Web2.0 sites TripTracker also has a developer blog so check it out to see when they release the API to the public.

The slides will probably online soon.