Month: August 2004

Apple-X.net has a post on how Google could overthrow AOL’s Instant Messenger with its own solution built around Jabber:

[Jabber] is an open-source messaging protocol that can do basically everything that AIM can do, except that no large companies have really endorsed it yet, so it hasn’t caught on that much (a mixed blessing). Like other ways of communicating online, like email and IRC, Jabber doesn’t strictly depend on a particular server. Most Jabber IDs are @jabber.org, but they don’t have to be.

An important feature of Joogle would have to be contact list portability and file transfer behind NAT, neither of which Jabber can really provide. Thus, Google could implement such features in their own client, which would also display a text ad in the contact list.
And it would work, even if users could choose to use a different Jabber client that didn’t display an ad and was even more customisable, for the same reason more AIM users don’t use Gaim.

What else would it need? Audio chats are important in modern IMing, so this is a must. Optimally, Joogle’s audio chats would be compatible with other VoIP software, like Skype and iChat. Also, since AIM maintains some public chatrooms, Google would be obligated to provide these as well. The difference? AIM’s are totally unmoderated, which allows bots to clog the rooms (there are often ten bots per user in these rooms). Google could moderate their own rooms, filtering out the constant droning of COME SEE MY WEBCAM, enabling users to carry on, at least, semi-intelligent conversations.

In the end, what would this accomplish? It’d hand the world of real-time messaging back to the public (there are dozens of different Jabber clients), end client incompatibility once and for all (assuming many people switched over to Jabber), and stop companies like AOL, which are arguably very evil, from monopolising a very fundamental ability and use of the internet.

WSJ writes about the increasing sue of virtual machines software:

VMware’s trick is to fool each operating system on a physical machine into thinking it is talking directly to the hardware, when it really is communicating with VMware. Each virtual machine is essentially a changing file that describes what it is doing at that moment.

This can yield dividends for corporate users. Servers are generally dedicated to one task — processing e-mail, for example — and in the era of fast processor chips, their full power often isn’t used. Some studies put the average utilization rate on Intel-compatible servers, which VMware works on, at about 15%.

So four or more virtual machines could run on one server with little performance penalty. That means buying fewer servers. Administrators also can create scores of virtual desktop machines for office workers that actually run on centralized servers.

Prudential UK, a unit of London-based insurer Prudential PLC, decided last year to put some call-center and back-office operations in Bombay, in part to cut costs. But Prudential (which is unrelated to the like-named U.S. insurer) realized that computers in Bombay, more than 4,000 miles from London, couldn’t maintain a quick-enough connection to the insurer’s databases in the home office, says Andy Ruby, head of infrastructure design.

So he left the hardware in Britain. Mr. Ruby set up 800 VMware virtual machines, acting like desktop PCs, running on about 60 servers in Prudential’s data center, where they can quickly connect to databases. The Indian workers sit in front of PCs that are essentially empty shells to display the far-away virtual PCs. While the time lag created by the distance is still there, the virtual machines’ proximity to the database lessens the effect.

Jon Udell writes that with the two services, “social networking goes beyond sharing contacts and connections.”

Both Flickr and and del.icio.us address specific activities that benefit from an informal, diverse network of people. Flickr, as I would explain it to my friends and family, is a way to easily upload and share digital photos. And del.icio.us does the same thing, only for Web bookmarks.

To CTOs, though, Id say that both are collaborative systems for building a shared database of items, developing a metadata vocabulary about the items, performing metadata-driven queries, and monitoring change in areas of interest. In the case of Flickr, an item is a photo; in the case of del.icio.us, its a URL. But the same methods could apply to any of the shared digital artifacts that we create, find, and use in the course of our daily work.

Abandoning taxonomy is the first ingredient of success. These systems just use bags of keywords that draw from and extend a flat namespace. In other words, you tag an item with a list of existing and/or new keywords. Of course, that ideas been around for decades, so whats special about Flickr and del.icio.us? Sometimes a difference in degree becomes a difference in kind. The degree to which these systems bind the assignment of tags to their use in a tight feedback loop is that kind of difference.

Feedback is immediate. As soon as you assign a tag to an item, you see the cluster of items carrying the same tag. If thats not what you expected, youre given incentive to change the tag or add another. If your items arent confidential and online-only access is sufficient, this can be a great way to manage personal information. But the real power emerges when you expand the scope to include all items, from all users, that match your tag. Again, that view might not be what you expected. In that case, you can adapt to the group norm, keep your tag in a bid to influence the group norm, or both.

These systems offer lots of ways to visualize and refine the tag space. Its easy to know whether a tag youve used is unique or, conversely, popular. Its easy to rename a tag across a set of items. Its easy to perform queries that combine tags. Armed with such powerful tools, people can collectively enrich shared data. But will they? The success of Flickr and del.icio.us wont necessarily translate to the intranet. You can import the global-hive mind, but you cant export the local-hive mind. That asymmetry defines the challenge we face as enterprise knowledge gardeners.