Word on the street is that Ezproxy, the remote access solution for libraries, is to be acquired by OCLC. I predict a change in the licensing and potential litigation for users in the future. More to the point, this should be seen as a call for an open-source equivalent.

For years ezproxy has been an outstanding solution for providing library patrons with remote access to licensed web resources. It has been dirt cheap with excellent support. Yet, at the same time the documentation is poor and the configuration bizarre. I know, I administered a large, complicated, ezproxy for a library consortium for many years.

With all strengths and weaknesses combined, ezproxy was head-and-shoulders above any competitor. Ezproxy’s well-earned popularity meant that there was really no advantage for anyone to create an open-source equivalent. However, now that it is to be acquired by a large library vendor, we can expect many reasons to emerge. The quality of support will likely change over time. The licensing is likely to become complex.

I believe that either a Java application or a apache module with a management tool for building, managing, and distributing the rules-sets for specific web resources is in order. Apache would be interesting as it already has an API for doing authentication/authorization and could be integrated with federated identity through emerging projects that would be seperate (that is to say, ezproxy wouldn’t need to support these… apache will). This is also true of Java in some sense as well as many Java Application Servers have identity management components that would help with this. But they are not open-source (generally speaking… some specific ones are).

YouTube Canada is now available. I think having a Canadian version is fantastic: for example having a Canadian-content site works really well for Google News. It provides value to a specific audience instead delivery plain-vanilla to all audiences.

But why don’t they use an actual “.ca” domain. When you go to youtube.ca you get redirected to ca.youtube.com. I suppose there could technical reasons for this. Maybe their technology platform makes it challenging to use anything else but I highly doubt it. If they truly believe that Canadian audiences will value content specific to their community, then they should really celebrate the identity of those customers/users. A “.ca” domain is the way to do that.

I wonder how big a difference it will make for YouTube. Right now the visible differences are that the homepage contains different content and there is a little indicator in the top right of the screen indicating that I am on the Canadian YouTube site. If the existence of the Canadian site change the criteria for deciding which videos are related to each other that would be very cool. The CBC has more coverae on YouTube Canada.

To celebrate the launched of YouTube Canada, here is a video I took in 2006 of the Canada Day Fireworks in Millwoods Park, Edmonton, Alberta, Canada.

During the RIPE 55 meeting in Amsterdam last week, a community statement was issued on the future of IPv4. Basically they say that we will run out of IP addresses in two to four years and we need IPv6.

To drive the point home, Gary Feldman performed a song titled “The Day the Routers Died” (sung like “bye bye american pie”).

Via iLibrarian: JupiterResearch has released a report showing significant use of widgets and gadgets on social networks.

If only I had known about this last week before my presentation on widgets, gadgets, and badges I could have included some really nice statistics about how much they are used. While the stats are encouraging, they do show that most people do not use widgets. Based on feedback from my presentation, I would guess this is because people do not have a sense of how trivially easy it is to use them, and because they feel that do not have a place to use them. Widgets can be used in more places than just in blogs. The JupiterResearch report focuses on how they are used in social networks, but they can be put on websites, in signature lines and just about any place that you can cut-and-paste a little bit of HTML.

The report also shows that people usually get their widgets from a friend. This could be because of the impact of social network applications and the concentration of information sharing activities there. On the other hand it could be because, people don’t realize that using widgets is as easy as “cut-and-paste” until a friend makes that obvious to them.

Internationalized Top-level Domains (TLDs) are Internet addresses written in languages that do not use the latin (e.g. English) alphabet. For example, Chinese and Arabic TLDs would be “internationalized”. Associated Press (AP) reports that ICANN will be announcing a new system for developers to test support for internationalized TLDs in mid-October.

At this point, I would suspect that there are far more non-English speakers using the Web than English speakers due to adoption in India and China. There has been a long-running concern over how domain names (i.e. DNS: non-numeric Internet addresses) might be made more usable by people (and systems) that do not primarily speak English. For example, the goal is to allow a native Japanese, Chinese, or Arabic speaker to type a website address into their browser using the script that they are used to, and not have to use an English or “romanized” equivalent (e.g. pin-yin for Chinese speakers).

Part of the problem is that the all DNS names must be processed by the so-called “root level” DNS servers, and these servers currently handle latin characters only. ICANN propose to use an encoding scheme called Punycode to convert internationalized domain names to latin characters so that the root-level servers can process them. Users would type domain names in the manner that they expect to, and the servers would take care of the translation transparently.

During the test people, people will be able create websites to test the handling of Internationalized domain names but will not be able to register new domain names. A great deal of the testing will concern how existing software handles the internationalized domains. This includes popular and specialized browsers, but also a great deal of server-side software that most users are not aware of (e.g. DNS resolvers, proxy servers, and more).