We tried on all browsers on our machines, and all images were okay except for IE6 on Windows 2000 on my VirtualPC. But IE6 on other machines was working, so we started thinking it was a proxy issue.

After digging the issue further with the customer, we discovered that not only images were not loading, but also that all sites visited after this one had problems downloading images until the browser was restarted.

I was scratching my head until Matt came out with this link (again, thanks Matt!).

It looks like PhotoShop 7 embeds some xml code in jpeg images when saving them using the default "Save as" menu and not the "Save for web" feature, and that this xml code breaks some builds of IE6 for Windows.

I find this totally amazing...

PS: digging a little further, this is what PhotoShop embeds in a jpeg image:

Dear Paolo, I can't buy music at Apple's store since I use a Sony laptop, not a Mac. [Scripting News]

Dear Dave, I can't buy music at Apple's store since I live in Italy, not in the US. Since none of us is likely to change anytime soon, let's hope that Apple will.

I did not spent much time in the store, but I did not find one single song that I would have liked to buy. 200.000 songs are a lot, and I'm sure that there's plenty of music I would like to have, it's just that I have not found it yet. Most probably somebody could help me finding something I like.

This leads to a major missing feature on the store: links. Apparently the store is not much more than a web site you have to navigate using a browser embedded in the new version of iTunes (maybe this is another good reason for Safari). Still there is no URL to copy and paste.

In today's "two-way,bloggin',social,you-know-what-I-mean" web this is simply crazy. Thousands of bloggers could start pointing to specific songs on the Apple Store, I would have expected a "Post This Song" button at the top of the page, but nothing, there's no easy way for me to publish a link to a specific song. I think that this should be addressed asap.

iTunes 4 comes with a bunch of new features, not only access to the Music Store. For example you can associate the album cover's image to a song. As usual it is beautifully implemented: all you have to do is find the image and drag and drop it from the browser window to iTunes window. Album covers images can be easily found on Amazon and Google Images. Since both sites have open APIs, I guess that there will soon be a script to automate this process (also if Apple is now technically an Amazon competitor and searching on google images does not guarantee 100% accurate results).

Also if it can, and it will, be improved, I see the Apple Music Store as a step forward. We all loved Napster, but it could not work for free. Apple is trying to find a solution and I think that if we want the music on-line business to go one step forward we need to support this effort. This is why we should buy some music on the Apple Music Store even if it's not exactly our favourite: if they are successful more majors and artists will join, possibly also our favourite ones.

Of course, if it will be succesful, if they will work out some copyright agreement with the majors, if they will decide that it's worth, then they will introduce it to the rest of the world.

Oh... well... whatever.

[later...] downloaded and installed iTunes. Much better than I thought, it only reads "The iTunes Music Store requires a credit card with a U.S. billing address". Now all I have to do is convince somebody to give me his credit card number

Food is a very important part of our coltures, we should definetly do more food blogging. I'll try.

PS: I guess it all started with spicy noodles.

I was doing the last check mail/check aggregator of the day and I decided to peek at the server's log files when, at 22:45:26, I saw a hit coming from an IP address that did not look familiar.

I pinged instantly Matt with: "Hey, they are watching the site right now". Seconds later in another window Mikel, who was at the conference, wrote "ben hammersly is demo-ing k-collector right now".

I called Matt on the phone and we kept receiving real time reports from Mikel, while we were trying to figure out what was going on in that room on the other side of the world. At some point we got "marc is going off, rallying the crowd", I have spent the whole summer of '98 making demos with Marc in Trieste, I could perfectly picture the scene. It went on until 23:11 when Mikel wrote "we're done", but not before saying hi to David Weinberger who was sitting right next to Mikel.

As stupid as it might seem... it has been exciting.

Dave Winer: RSS readers that work like Usenet readers are a waste of time, imho. Aggregators should not organize news by where items came from, just present the news in reverse chronologic order.

I totally agree, this is why after trying other aggregators I came back to Radio: it's just like reading a very large weblog updated by several people every hour, it's an activity we are familiar with and the ability that we have to scan titles and relevant information in a page makes this approach much more efficient than having to move from source to source.

If an aggregator is meant as a way to take a snapshot of what's going on on hundreds of sources and quicky present it to us, I believe that presenting news in reverse chronological order is the way to go.

But I also think that aggregators could be an interesting way to archive content, to let somebody quickly retireve something wrote sometime in the past.

Archiving by author, again, does not make sense: most weblogging applications already do that, if I'm looking for something and I know who wrote it, I can simply look on the author's site.

There are search engines, which are of course a good way to find information, but not always very efficient. There are cases when a directory might be more useful.

We believe that archiving by topic in a directory could be a solution, and this is what we are trying to do. It's not for daily instant reading, it's to archive content.

This post is actually just a test for the aggregator.

Topics abuse, just like metatags abuse with search engines, becomes relevant when topics from several bloggers or news sources will need to be merged and categorized by aggregators. The way we thought to partially address these issues is by adding the cloud element to the specs, which should specify whose topics are in use and, thanks to topicRolls, allow to flow topics lists in every direction.

Here's how I see possible topics usage in the near future:

There are 3 main ways topics could be used:

1. Intranet Application. Weblogs are a great way to collect information and share it with other people in real time within a working environment. In our new product, a centralized aggregator will extract topics from users' feeds (each user can create new topics) and redistribute them on the network via topicRolls.

2. Topics communities. I couldn't say way, but I see these communities gathered around publishers, which already have experience categorizing contents. I could, for example, decide to become part of the NyTimes cloud, and use the NyTimes categorization system to markup my posts. Of course, all contents coming from weblogs belonging to the same cloud will be easily organized.

3. Local users. It's the position current LiveTopics users are: they can create their own topics and use them on their weblogs. A new generation of smart RSS aggregators will be able to use these topics, but I don't think that these "local" topic maps will be easy to merge anytime soon.

Using the last version of myRadio by Mikel Maron you can now contents in your desktop aggregator using topics. Thanks Mikel.

Another small step in the right direction.

We are proud to announce the launch of the Social Software Alliance

The brainchild of the pioneers at SocialText the alliance is intended to be a place where the developers & users of social software can come together to create open standards, and, contribute industry best practices.  Our initial aims are:

• aid discovery of developers working on synergistic projects and standards
• assist in shaping open standards that mesh well with other alliance and Internet standards
• help promote each standard to gain wider adoption

Quoting from the call for discussion:

"The fast-paced nature of the social software space now argues for developing light-weight, easy-to-implement standards, following the Internet tradition of rough consensus and running code, but perhaps moving faster than the larger standards bodies. It is expected that those standards promulgated by the alliance which become widely adopted will be proposed to the appropriate general standards body or bodies: W3C, IETF, ISO, etc. "

There will be a SocialText sponsored Happening tomorrow (see the site for more details) and we hope that anyone who is interested in the development of social software will come and get involved.

During a historic ceremony in Athens, the leaders of 10 countries signed treaties today to join the European Union. [NyTimes]

For most of the world this is probably nothing more than "just another international news". Growing up for the last 32 years a few hundred meters away from a quickly vanishing border (the one with Solvenia, in my case) makes things a little different. Very soon there will be almost no border. Even if each country will maintain its own languages, traditions and identities, we will all belong to the same family. And it's a peaceful process.

For most of my life I've lived within walking distance from a border. First Yugoslavia, then Slovenia. Crossing the border has always been easy, but soon it will be much easier. It's quite a strange feeling.

New RSS features on topicexchange.com. A minor (but very handy) change to the Internet Topic Exchange today: it now supports ENT (spec), which means suitably equipped aggregators will be able to pull topic information straight out of the RSS feeds.

Making this actually useful is a new RSS feed: all posts on the site. If you want to keep track of everything, subscribe to that one (traffic on the Exchange is still not awfully high, so you won't find yourself overwhelmed). An aggregator which understands topics will be able to just pull down this one RSS feed instead of heaps of individual topic channel feeds.

The most interesting bit is yet to come: I've been contacted by Scott Johnston and Greg Gershman, who both seem interested in using Topic Exchange information to do some sort of classification of search results. Sort of like the way Google uses dmoz to give you links to relevant categories when you search. This functionality is yet to come, but the hooks are there in the Topic Exchange, so any developers are welcome to start using them from now on!

For people who are interested in using this, I've written a page to explain how to handle the data. Enjoy!

[Second p0st]

Much better than I hoped. Thanks Phil!

Jon Udell has posted the topic for his keynote talk at OSCOM. Perhaps I'll talk about OPML and directories. I want to zero in on something we can all do together to move Content Management forward. I want to see OPML directory renderers in all environments, not just UserLand's. What better place to spread the gospel than OSCOM? Maybe we can come out of the conference with a revolution in our pockets? That would make me happy. [Scripting News]

The way topics and types are currently distributed within ENT1.0 applications is OPML. Our own centralized aggregator will automatically create directories of posts using Types and Topics and, guess what? That directory is based on OPML and can be edited using a Radio tool.

Until now we had always simply used OPML because it made sense within our internal structure, it was not our intention to make the directory available to others in OPML format. But techically speaking it would be very easy to do and if this would contribute to the development of all directories it would be great to do it.

Now... what's next?

Well, of course, we are working hard to try to be the first ones to support the protocol that we designed.

Within a couple of days LiveTopics 1.1.3 will be available, so at least some RSS2.0/ENT1.0 feeds will be available to play with.

Next we are working on a new centralized aggregator. Murphy willing, there will be a first version available at the O'Reilly Emerging Technology Conference, while a final version should be ready for BlogTalk in Vienna.

What we are going to do is a relatively simple application, just to start showing the benefits of topics and types in RSS 2.0 feeds.

The other very important development front is enabling more weblogging tools to support ENT1.0. While Radio is going to be covered by LiveTopics, there's Manila, MovableType and other applications that, as of today, we don't even know how to approach. Another front which should be explored is supporting topics within personal news aggregators. I believe that some interesting news about this will come soon.

We have received quite some interest also by companies working on other kind of knowledge management systems. This is great, since as soon as they will start supporting the specs, they will also be instantly compatible with any aggregator.

Weblogs are first of all an incredibly powerful tool to share points of view. Weblogs allow people to become richer by sharing their views on the world surrounding them and by being able to see the world with other people eyes. This can be applied to the definition of new protocols or to an ongoing war.

The huge success that weblogs are having is creating an incredible amount of information. RSS feeds and aggregators, both desktop or centralized, provide an excellent and powerful tool to keep track of all these weblogs, but at they are starting to show their limits as the number of sources increase.

Currently news distributed trough RSS feeds can be organized by author or categories, which is definetly a good way of doing that, but we believe that some alternatives might prove useful too.

Traditional media has always organized their content using topics or categories. Look at this page on the BBC site to get an idea.

Of course, since weblogs are not published by a single editor, a topics system should be flexible and dynamic eough to be usable in any situation. While, for example, a news oriented weblog could use categories similar to the traditional media ones (Art, Business, Economy, etc.), a techinical weblog could use categories much more refined: a weblog discussing scripting tecniques could for example be organized using scripting languages.

To make this approach usable there's anyway the need to be able not only to define our own categories but also to subscribe to other editors' categories listings in order to be able to synchronize our contents. This is where the topics clouds concept comes from: anyone could both choose to use somebody else categories or/and to create his own.

Topic clouds can be published trough topicRolls, such as outlined lists of topics currently described using OPML. But the ENT specification allows the use of any format, such as XTM, to describe a topic cloud.

This very first step we are doing will shortly be supported by the release of a new version of LiveTopics, which will support the standard, and followed shortly by a new Evectors application which using ENT will create dynamic directories of contents created using weblogs for intranet applications.

The ENT1.0 specification is published under a Creative Commons License, as usual, any comment, contribution or critique is very welcomed.

Two weeks ago they were parading around kissing pictures of Saddam. Now they're kissing pictures of Bush. Is that love? [Scripting News]

If you, like me, have none of the above technologies, you can still take a picture every few miles and then upload them to your blog. Here's the tripBlog of the Gradisca d'Isonzo - Loreto trip I just did.

I'm not a programmer, but I'm not sure that this kind of approach could actually help development. What I saw is a very interesting collaboration tool. It's better than IM (anyone can edit text in any part of the documenet) and it's a very fresh and interesting new approach to collaboration. Build an outliner in the application (programmers should love writing code in an outliners, and if they don't they should try), and I'm sold.