I am attending a project proposal meeting in Stockhom, with partners from around Europe (Sweden, Germany, Spain and so on). The university here (KTH) provide a WLAN wireless network (802.11b) here on their campus, and elsewhere in Kista (a suburb of Stockholm) and the city centre itself, with Internet connectivity. So I am connected to the Internet via my laptop with WLAN wireless connectivity while sitting in the meeting on campus.
Now, many companies and universities have private WLAN networks, and some allow visiting guests access to these. What is very exciting about the deployment here is that the wireless network is shared between a number of different operators using the StockholmOpen.net software and infrastructure. Currently the university itself (who sponsor the experiment), Yanzi Networks, Lidnet, and Telia all share the same access network and provide Internet access via this network. The system allows users to select the operator of their choice (e.g. students use the university and their standard user accounts, shoppers in a shopping centre may buy a card at a counter giving a code which allows access via one of the other operators but both use the same network). The technical details of how this is achieved are available at StockholmOpen.net, and the software is open source and freely downloadable.
To users, once they are connected and using the system, the fact that they are using a multi-operator network makes little difference. It appears to you that the network is owned by your operator. However, it makes a big difference to the cost of deploying such a network, as it can be shared by all operators some of which may be commercial, and others community funded schemes. Potentially the cost could be so low that the network could grow organically as people deploy extra parts and join in to the network.
I will keep this weblog updated with progress on the project proposal we are building around this concept.
I came across (thanks to Andrew Betson) an excellent description of a commercial UK local access point which can integrate with others via Wi-Fi to share broadband upstream access in a local area Become a wireless ISP: for ｣300. It refers back to the author Guy Kewney's own weblog article on the same topic. The basic tenet is that many have argued that mesh computing (sharing wireless connectivity with multiple uplinks) is a complex task, but this UK company LocustWorld (with products like MeshAP and MeshBox) may have proved them wrong. they even have s/w for normal Wi-Fi access points (like US Robitics ones) to allow them to take part in the mesh.
I've just spent a few hours reading around about the use and misuse of metadata. I started with a an interesting article on XFML (a lightweight distributed categorisation DTD) XML.com: Introduction to XFML [Jan. 22, 2003] (via a mailing list digest I'm on, I know, I should switch to consuming it as an RSS feed but I haven't got there with everything yet). Anways, this led to a criticism of metadata (Metacrap) which reconfirmed my belief that automated reaping of implicit knowledge always beats manual metadata editing (excepting small subdomains where one has a budget to mark up things properly, such as library catalogues, MARC records, the OCLC and all that jazz). Undaunted I browsed on via multifarious links and ended up in the land of TopicMaps/XTM, the W3C Web Ontologies Working Group . Finally, I ended up reading about Dan Connolly's Semanitic Web Travel Tools. It is good to know that we live in a world where people are still working to create open flexible mechanisms to share and exchange structured information in useful ways. It is also interesting that people are using Perl and Python to create filters to exchange between mobile devices and fixed ones in meaningful ways. This mini-browse has reinvigorated me to keep abreast of the lastest in metadata and the semantic web.
Another interesting article from Jon Udell: The name game
This article Digital ID and eGovernment :: Digital Identity World :: Digital Identity, Digital ID by Phillip J. Windley discusses the issue of digital identify (in a US context). "Governments are vitally concerned with identity and yet, paradoxically, most governments have been largely unwilling to take a leadership role in the digital identity arena."
This article from SecurityFocus.com deals with the security issues relating to IM: Instant Insecurity: Security Issues of Instant Messaging.
This is an overview article on the current state of Projection Keyboards. It describes four products, gives links to their websites, and describes the companies and lists patents. All very interesting.
The recent Netcraft Web Server Survey (Jan 2003) listed a Windows 2000 server as one of the machines (it knew about) on the Internet with the top 50 longest uptimes (the machine was running without a reboot since November 2000).
In London this weekend (UCL) I met with a number of people from the StockholmOpen.net group. This is driven by KTH (The Technical University in Stockholm). The idea is to have a Wi-Fi network shared by multiple operators and to enable end users to choose which operator they wish to use. In Stockholm some WISP (Wireless ISPs) and the uiniversity share the access network so students and anyone else can get wireless access to the Internet in the parts of the city where they have coverage. They compare themselves technically to NoCat and NetLogOn, and their software is open source and avialable for download on their website.
A recent entry in Jon Udell's weblog referenced a whole series of interesting things: New York Times Article on who owns the internet (quoting Joseph Turow's suggestion that a small "i" was more suitable than a capital one so as to allow for global ownership); that William Gibson's statement that "the future is here, just not evenly distributed" (as cited in Tim O'Reilly's 2002 column on emerging technologies); and concept of active paper (as refernced here in an earlier weblog entry); and reference to an interesting book on W-Fi.