We have just increased out bandwidth at the colocation facilty by 10 times. This page should now download much faster  (well, comments and google ads are slowing it a little bit, but we did it for our customers, not for this weblog ;-).

Tim Bray today links to different taxonomies by different bloggers:

• Bray
• Walsh
• Pilgrim
• Winer

Some of these taxonomies are nested, other are flat.

Looking at them I still think that we could easily use this metadata with the approach we are taking for K-collector. You can almost instantly notice how several identical topics used in more than one of the taxonomies linked above. A lot of them also appear on our k-collector site.

Where everything breaks is in the RSS feeds. Let's see how each blogger include (or doesn't include) topics/categories in his feed:

Bray:
Apparently there are no category/topic information in his feed.

Walsh:

Pilgrim:
Apparently there are no category/topic information in his feed, but there are per-category feeds.

Winer:

To make things worse, I use yet another method to describe topics in my feeds: ENT.

Bottom line? I guess that it all boils down to what we want to do with these taxonomies. As long as they are simply a tool used by every author to organize what he writes there are not going to be problems or conflicts.

The day we will want to be able to compare opinions about the same topics expressed by different authors, possibly in the same page, a solution will have to be found. 

During that first trip, Matt and I sketched an idea about how weblogs are influencing people while they work.

This essay is where we got from that idea.

I think it's interesting not only to explain how weblogs influence thinking and working within a company but also in the general blogosphere.

Almost exactly one year ago Matt came visiting us from London. It was another cheap RyanAir flight.

During that first meeting we envisioned a new kind of knowledge management application built using Topics (he had just released LiveTopics), an RSS aggregator and our content management system, IdeaTools.

One year later we are proud to officially announce K-Collector 1.0.

Even if we got to this point later than we were expecting (as we were expecting ;-), I'm very happy about our new product and very proud of my partners in this project: Simone Bettini and Matt Mower.

Here's a picture of us last June at BlogTalk.



There are many other people who helped or supported us during this year. We tried to list them all on a page in the application, and we are planning to add more.

Now that we have made this announcement, we are ready to take orders. If you don't feel like ordering yet, we'd be delighted to show you a demo. Write to k-collector@evectors.com to arrange one.

Jeremy Zawodny: Is Google The Next DNS?

It's already happening. Do this little exercise:

• Download the latest version of Mozilla Firebird
  
• Once installed, write in the Navigation Toolbar Dave and hit return
• Now try with "Paolo", "Scoble", "Searls", "Joho", "Zawodny"....

If it's not a URL, the string is routed to Google's "I'm feeling lucky" feature. Now, of course you still rely on DNS to get to the site, but if google would index IP addresses of sites we could probably live without DNS.

Now: do we want to give all this power to a company?

I have just met an old acquaintance whith a brand new Tablet PC. For the first time I've been able to hold one in my hands. It's kinda cool, the only wrong part is the price. I mean:

• If somebody should ever decide to give me a free Tablet PC I would very much appreciate it
• If it would cost EUR 750, at some point I could decide to buy it for my "sofa browsing"
• I would never spend EUR 3,500 for that toy, sorry

PS: I have just visited Microsoft site to check what's new in Tablet PC land. Only to find out that the Tablet PC Hardware Demo only works on Windows (it's a plain Flash presentation with nothing special as far as I can see). This is not about being the evil empire, it's simply bad marketing and sloppy programming. Whatever.

I was visiting the RyanAir site and found these prices:



Yup. 0.02 Pounds for a London/Trieste roundtrip. No tricks, it's true. Less than 20GPB including taxes. Cheaper than a phonecall.

You always realize how important tools are when they stop working. I have spent the weekend desperately reloading Technorati pages hoping to see some new links, but the blogosphere seemed to have stopped 4 days ago.

This morning I have learned from David Sifry that they are upgrading their systems. Now they are tracking 1.2 million weblogs, adding 8-9,000 new ones per day and finding a new post every 0.86 seconds. Whew!

Congratulations!

I'm so happy about this that I have just purchased two new RSS watch lists. :-)

I have just received the annoucement that BlogTalk 2.0 is taking off. Vienna... here we come!

For the last 24 hours we have been parsing Dave's new category items from his RSS feed within our k-collector w4 test server.

We have decided to split the category nesting and to match each category level to what we consider a topic. While topics should be unique and absolute within a cloud, nesting is subjective and can be defined both by the author and by the reader.

As an example, in Dave's taxonomy you can find:

- Politics
    - Presidential Election of 2004
        - Dean Campaign
        - Clark Campaign
        - ...

If Howard Dean would have his own taxonomy, it might contain something like:

- Howard
    - Presidential Election of 2004
        - Politics
        - Budget
        - Ideas
        - ...

Same topics, different nesting according to the different points of view.

After splitting Dave's categories we match them against our OPML topicRoll to see if any of our active users has already used that topic and we archive Dave's posts according to the matches we foun (guess what? a lot of categories Dave has been using already exist on our server: it's a small world ;-).

For example you can see here a post by Dave about Google which has been archived under the Search Engines topic thanks to Dave's category.

Topics on this test server require a classification (Person, Thing, Company or Who, What, Where) so we cannot automatically create new topics from a feed using categories as we currently do with ENT feeds.

But anyway it's a quite interesting first step.

The only reason for this post is creating a few new topics on our server that are matching categories Dave is adding to his RSS feed. Just an on-going experiment. Stay tuned.

Got an email this morning from a friend who has recently started a new service: authentication via caller ID.

Basically if you need to get to a private page of a site, you call a number from your cellphone when prompted and the system lets you in. Since the call is not answered (all the system has to do is check that a call was coming from your number) it's free.

Beautiful in its simplicity. More information on SAINTlogin.

As promised yesterday, here are my thoughts about Categories in RSS feeds, ENT, TopicRolls, etc.

The last changes to Scripting News RSS feed are quite interesting. I'm especially interested in the innovative use of the category element and of an OPML-based hierarchic set of categories since it's quite similar to what we've been working on recently.

I'll think about this while building the furniture of my walk-in closet and comment later of tomorrow.

Dave about the shape of weblog software

Weblog software is going to be like mail servers. Lots of ways to deploy, every niche filled. For the masses, services like Yahoo, MSN and AOL. Blogging servers for corporations, inside and outside of the firewall. For schools, for the military, specialized systems for lawyers, librarians, professors, reporters, magazines, daily newspapers. The next President will have a blog. Writing for the Web, the prevailing form of publishing in the early 21st Century, will come in many sizes and shapes, flavors and styles. It won't be one-size-fits-all. Open formats and protocols will make this possible. I'd bet on the formats and protocols we're using now, RSS 2.0, OPML and the Blogger API.

I completely agree on the fact that what we are currently calling "weblog software" is going to become a big and variegated business. But I also think that it it's not going to be only about publishing.

Not everybody needs or wants to publish on the web, but everybody will need to manage his "presence" on the web. Weblogs are revolutionary because they allow us to show our web side, but this won't be necessarily only about writing posts. What we are seeing today is the first step in this direction: cheap, powerful and scalable software that anybody can use and that allow to manage a wealth of information, input and output, on the web.

I also agree that this new reality will have to be made of standards a significant part of developers will agree upon and will rely on existing and new kind of services that will glue it all together.

A few weeks ago we switched email servers. To manage the transition I had created a new account in my email client (Apple's Mail.app) with the new server's data.

Yesterday I closed the account on the old server and decided to get rid of the account on my email client. Before removing the account I remembered to move all messages saved in the default inbox to another mailbox since I knew that the default inbox was actually part of the account and that deleting the account would have also deleted the received messages.

But I didn't think to backup messages contained in the default Sent mailbox.

So now I have lost all messages that I have sent in the last.... well, I don't know how much time. It looks like the mailbox is unrecoverable and I don't have a fresh backup.

Also if I think that a good application should have warned me ("You are going to delete 2.387 messages, are you sure?"), strangely I'm not particulary upset about the data loss. Some time ago I would have lived this as a tragedy.

But if you'll ever have to delete an account in Mail.app remember to back-up. I will.

... Poughkeepsie ...

Halley: I think we might be fooling ourselves. I think getting to know someone online creates a false sense of intimacy that can't compare to a real world connection.

I have met several bloggers in real life too and I agree with Halley's opinion that the real life experience is always different from the on-line one.

But overall I must say that in most cases it has been a very good experience. Maybe because even if we only knew the on-line side of our personalities it was more than nothing. Maybe because the reason I met fellow bloggers was because I choose to meet them after reading their blogs, thus creating a sort of filter.

I find it's amazing how business meetings change if you are meeting a blogger. Non-bloggers meetings usually start with at least 30 minutes of introductions, friends of friends talking, who is who investigations. Meeting a fellow blogger means talking 2 minutes of the last Technorati feature, 3 minutes of Dave (it always happens) and then you get to the core of the conversation. Cool.