Alors essayer le plugin Onswipe pour WordPress.

Celui-ci transforme la lecture de votre blog sur iPad. Les lecteurs découvrent alors une autre manière de parcourir le blog.

L’image du dernier post vient agrémenter une page de couverture. Les posts sont présentés comme une succession de dalles (“à la Flipboard).

Le seul point auquel il faut faire attention, c’est d’illustrer chacun des posts d’une image, sans quoi les pages paraissent un peu vides.

I finally decided to move from Apache Roller to WordPress.

As a Java aficionados, I usually prefer Java solutions to PHP ones, but in this case I should admit the superiority of WordPress.

The WordPress community is by far larger than the apache roller community and the number of WordPress plugins is not comparable to the Roller’s one. And the dashboard feature is really more complete.

Of course, I could have spend some time to implement such features in Apache Roller but why reinvent the wheel when WordPress seems a perfect fit.

To migrate the content of my previous blog, I followed from here and everything went properly.

Today, some of the players has already made their choice and are moving to a 100% digital media delivery. The others want to take advantage of both media but they have to face organizational issues. Should they merge print and digital editorial teams?

The Belgian Rossel press group has made his choice by creating a true bi-media editorial teams.

Next week, it’s the 2009 ez conference and award event. One of the keynote is presented by James Hewes, Head International Development at BBC Magazines. he will explain how the organizational aspects at BBC have influenced their digital publishing projects.

I’m wondering what has been the choice of BBC and if the weekly or daily characteristic of the print edition may have an impact on the origagizational decision.

Suppose you have a branch. You merge changes coming from another branch and you commit the changes. You test it and you decide that what you merge is not stable enough and you revert to the version previous to the merge.

Later on, you decide to try a new merge and that’s where the issue occurs.

Because Bazaar knows you have already merge the branch, the second merge will only be partial. It will not include the changes made before the first merge.

To do that, when you revert a merge, you neet to tell Bazaar to forget about the merge. The information below comes from https://answers.launchpad.net/bzr/+question/45980.

First tou need the sources you want to work with. Usually you will do that with a command like:

bzr revert -r X

X being the revision before the merge.

Then you need to uncommit the merge.

bzr uncommit -r X

X being the revision before the merge.

This does not change your working tree but will delete revisions older than X from the repository

If you do then:

bzr status

You will see the pending merge and you need to tell Bazaar to forget about the merge. To do that, just type:

bzr revert --forget-merge

This will not change your working tree but will remove any pending merge.

You can know commit your latest change and retry the merge later.

I think this was worth a post in the blog.

After a few hours googling around, I found the following post.

The issue was indeed due to the testAccessible function within a11y.js.

I patched a11y.js as follow:

dojo.a11y = {	// imgPath: String path to the test image for determining if images are displayed or not	// doAccessibleCheck: Boolean if true will perform check for need to create accessible widgets	// accessible: Boolean uninitialized when null (accessible check has not been performed)	//   if true generate accessible widgets	imgPath:dojo.uri.dojoUri("src/widget/templates/images"),	doAccessibleCheck: false,	accessible: null,		

and this solved my issue.

Here it is. TobFlow has now it’s own site.

It has taken some time before it went out. I started to work on a TobFlow prototype about 2 years ago. Since the early beginning, it was decided to release it as an open source project once it would be ready. Well, evertyhing is not fully ready and we still have to work on different points such as an installation procedure, some documentation but at I think it was time to let the baby grow. Now the project has some visibility and it’s the beginning of a new adventure.

I hope to have some free time in the coming months to work on points such as the wiki, the installation documentation and to start thinking about Tobflow 2.0 (more ajax, more modular).

I wish I could also work on the design tool with I think is essential for a large adoption of TobFlow.

TobFlow is hosted on Sourceforge and on Launchpad.

Therefor, I performed a quick install of Windows XP, and I tried to secure most of the thing based on the following tips:

• How to secure Windows XP/2000
• Disable the control panel
• Hide a user from the welcome screen

Despite all those securities, I wanted an easy way to create an image of the installation, so that I can quickly restore the PC if needed.

Looking around the web, I end up on a post from Habibbijan blog explaining how to perform an image of a windows NTFS partition and leave it on the disk for future restore. The solution uses Puppy Linux or SystemRescueCD.

I tried it for my own purpose and it was a success.

Not too bad !!

However, I was looking for something different, something closer to a TV channel.

A friend of mine shows me Mercedes-Benz TV  and Vpod.tv. Of course it’still a set of categorized video but the presentation is different and is closer to what I would expect from a web TV.

Everything is working properly. I just had an issue with the template for comment posting but some css tweakings bring back everything in line.