|
 |
Tuesday, July 16, 2002 |
Microsoft: .NET Architecture Center. "The .NET Architecture Center is a new site devoted to business, software, and infrastructure architects. The Center is a collaborative effort involving Microsoft product teams, MSDN, TechNet, and Microsoft's new Architecture Review Board, and it spans the boundaries of MSDN, TechNet, and other sources to serve multiple perspectives, or views, of enterprise architecture." [toolbox]
1:47:55 PM 
|
|
 Richard Florida
The Rise of the Creative Class: And How It's Transforming Work, Leisure, Community and Everyday Life
From his recent interview: Be Creative Or Die. "My theory uses the three T's: technology, talent and tolerance. You need to have a strong technology base, such as a research university and investment in technology. That alone is a necessary but not in itself sufficient condition. Second, you need to be a place that attracts and retains talent, that has the lifestyle options, the excitement, the energy, the stimulation, that talented, creative people need. And thirdly, you need to be tolerant of diversity so you can attract all sorts of people -- foreign-born people, immigrants, woman as well as men, gays as well as straights, people who look different and have different appearances." [Gurteen Knowledge-Log] [toolbox]
1:47:05 PM
|
|
Steve Pepper: The TAO of Topic Maps. "The generality and expressive power of the topic map model bring with it other advantages that go far beyond those traditionally associated with indexes. The close similarity to semantic nets gives an idea of how topic maps, even without any occurrences connecting them to an information pool, can become valuable resources in their own right. This in turn opens up new business opportunities for creating and selling “portable topic maps” that can be overlaid on multiple information pools. For traditional commercial publishers, producing well-crafted topic maps could be a new way of leveraging their existing knowledge and experience and combating the threat to their existence posed by the vast amounts of information now available for free." [Curiouser and curiouser!] [toolbox]
1:40:25 PM
|
|
Peer-to-peer air travel.
When my business partner (who is also my wife) and I moved to a small town in Minnesota in the early 1980s, ARTS & FARCES was a tiny independent video production company and we had enough raw footage in the can from several documentaries to be able to live and work outside of the coastal media hubs.
After all,we were only a half-hour from an international airport and could get to either coast or anywhere else in a few hours. Besides, we had this new-fangled 300-baud modem thing that would soon allow us to collaborate with clients and subcontractors just as good as being there. Or so we thought.
I’m sure lots of you had similar dreams back then. Probably still do. Broadband connectivity has finally made that dream a possibility, at least if you have enlightened clients (of which we could use a few more, by the way). The problem of getting from anywhere to somewhere else for independent creatives has become a nightmare of gargantuan proportions because the hub-and-spoke, centralized air travel system is clearly broken and beyond repair.
Jon Udell is clearly someone who had similar ambitions of living and working outside of a megalopolis and has written an excellent overview of one of the missing pieces: peer-to-peer air travel. Decentralizing air travel is certainly doable and now’s the perfect time. As Jon says, “nobody is going to miss visiting O’Hare.” [ARTS & FARCES internet]
1:31:33 PM
|
|
six/four: "It's at least a first crack at a working model of the next phase of the Internet -- secure communications by trusted peers over an untrusted network." [Adam Curry: Adam Curry's Weblog]
1:28:40 PM
|
|
Reading up a bit on peercast. The slashdot mention got lots of interest. I've been running the app on my windows machine for a few days now and it really works. According to their foum, the peercast network is intended to be infrastructure and can route not just mp3 streams, but any kind of data.
This idea of 'chaining' listeners together makes broadcasting affordable to anyone, because you only have to take care of your stream into the network once. From there it (theoretically) gets relayed by each listener. I hope they release information on the protocol soon.
This still isn't going to be a solution for broadcasting to dialup users, since you need to pass on lots of overhead information and preferrably route the data to the next peer from your own node. That seriously limits what you can do audio-wise over dialup. But it certainly will be a hit with cable and adsl. Perhaps some of those dedicated connections becoming 'neighborhood nodes' to non-relaying dialup listeners [Adam Curry: Adam Curry's Weblog]
1:28:13 PM
|
|
NewsNetWire is a newsreader for RSS feeds. Very slick. For OSX. You can subscribe to my weblog wih it as well, It's the first entry under 'Personal Weblogs'. Radio users note: you can also import your mySubscriptions and any other OPML file that contains subscrition information. [Adam Curry: Adam Curry's Weblog]
1:26:41 PM
|
|
This morning I renewed a few -about to expire- domain names and noticed verisign only lets you up to 9 years (at a 44% savings no less) but that appears to be the max.
I'm pretty sure I want my family to keep the curry.com domain name as long as it is relevant, which I bet will be more than 9 years. Should I write this in my will, or is there a registrar willing to quote me a price for a lifetime domain name? [Adam Curry: Adam Curry's Weblog]
1:25:23 PM
|
|
Joshua's Homemade Blog.
Joshua Allen goes through an interesting exercise of deciding what's necessary to replace Radio. Apparently, he decides that the most important features for him are: (1) avoiding 'procedural' code (he uses XSL-T), and (2) generating static content.
Not surprisingly, the end result is slow and cumbersome. He limited himself unnecessarily, and then makes the unfortunate assumption that, since his blog tool is unwieldy and slow, that therefore there's no reason to consider replacing Radio. What boggles my mind is when he extrapolates the experience to say that a blog tool made with C# would be no better! He points out that generating the calendar would've been exceedingly hard in XSL-T, which just goes to show IMO how the choice of a poor toolset can hobble you from the start.
So I ask: What's wrong with code? Why is static content holy? In fairness to Joshua, he's not alone in this regard. When I started talking about making a .NET-based blog tool to replace Radio, everyone assumed that I'd have a local web server, use an HTML UI, and upload static content via FTP -- myself included!
I started planning out what I needed to do, and how long it would take to accomplish it. It was just ludicrously complex, considering I already had ASP.net available on my web server. At least one major blog tool required specific server side software already, so why should I do things exactly -- or even remotely -- like Radio had?
So in I dove, and after about 15 hours I'd gone from a few hundred Radio XML backup files to a fully functional blogging system, using a database backend, using ASP.net to render pages, pre-seeded with my posts and existing comments, and partially implemented the Blogger XML-RPC API. Sure, I'm leveraging a very feature-shallow blog posting tool (which I'm work on replacing as we speak), but even so, it was pretty easy for me to replace Radio.
But... I had a motivation that Joshua didn't. I wonder if his perception of replacing Radio (and his choice of tools) will change the first time Radio stops publishing his content and starts eating his posts.
Link Discuss [The .NET Guy]
1:21:23 PM
|
|
Groove-ing in VS.net.
Sam's been off getting jiggy with Groove. He mentions that a tool he (and others) have developed will allow you to easily build P2P tools in .NET using their Groove toolkit that integrates with VS.net. Congratulations, Sam. :) You mentioned that the kit would be free to download, but I'm curious to know how they intend to recoup the costs of running the network (per-copy royalty fee, users must already be owners of a Groove licensed product, etc.).
--
To rant off a little on Groove (unrelated to Sam)...
I wish the Groove Networks web site had more technical information. I went there with the point of educating myself so I could speak semi-coherently on the subject in this web log, and came away with the sticky feeling of being "marketed at". :-p One of the things that always bothered me about working at a dot-com was that the corporate web site was always devoid of useful information -- it spoke in glossy terms and platitudes that seemed almost insultingly vague. It's like there's some global marketing rule somewhere that says "all corporate web sites must look like brochures". Yuck.
Unfortunately, the Groove web site makes me feel this way. I suppose I could download the test version of the tool, but it seems silly that I have to install the software just to figure out what it does! Clearly, the price of the tool combined with the fact that you can download a test version of it indicate that they're hoping for grassroots adoption of the thing instead of looking for that 1000-seat initial IT purcahse. That's a good move. However, it would be nice if there was a link that said "I'm not upper management in a Fortune 500 company. You can speak to me with real words, and they can be longer than one syllable. Take me to the web site with the screen shots and the technical specifications, please." :)
Link Discuss [The .NET Guy]
1:16:55 PM
|
|
A follow up on chat protocols. I downloaded and played around with the latest version of Proteus X 2.0b3 yesterday, and as I suspected, it cannot connect with the Yahoo Messenger chat protocol right now either. Trying to connect with Yahoo Messenger through Fire and Proteus stopped working a couple of days ago. I haven't seen anything definite mentioned about the change yet, but until either the third-party chat programs are updated to work with Yahoo or until the server that allows access to the Yahoo protocol is running again, the only reliable option for using Yahoo Messenger under OS X is to use the official Yahoo client...
I should offer a couple of comments on my impressions of the latest version of Proteus. It works well and it should appeal to people who want more bells and whistles with their multi-protocol chat program than they can get with Fire. It is much easier to set up Proteus to work the way that you want to work, but if you want a chat client that will make the most of your screen space (like I do, as a PowerBook user), Fire is still the better choice. Both Fire and Proteus can be used for free, but the author of Proteus is now seeking a small shareware fee to help support the development of the program. If you like Proteus, pony up the cash. It is growing into a great program! [Mac Net Journal]
1:12:18 PM
|
|
Microsoft patents may hamper OpenGL. It looks like the OpenGL graphics system used in recent Apple software and by other companies could be in trouble thanks to recent Microsoft patent claims. MacCentral has some details, but it boils down to the fact that Microsoft has been buying up intellectual property from other companies and cornering the market on some key computing technologies, and now the behemoth from Redmond is likely to start charging licensing fees for using what are now considered open technologies... [Mac Net Journal]
1:10:32 PM
|
|
Peer-to-peer air travel. "United 114 heavy, contact Minneapolis Center on 135.32."
"Thirty-five thirty-two, United 114."
"HIIIIISSSSS GARBLE GARBLE stuck mike GARBLE GARBLE HIIIIIISSSSSSSSSS"
"Minneapolis Center, United 114 heavy with you at four-one-thousand."
"United 114, maintain four-one-thousand."
"HIIIIISSSSS still GARBLE that GARBLE stuck mike GARBLE HIIIIIISSSSSSSSSS"
"Four-one-zero, United 114."
"OK, everybody, make room for Chicago sequencing. United 114, what is your Mach number?"
"We're doing Mach .85, United 114."
"United 114, heading one-eight-zero, slow to HIISSSS GARBLE HISSSS advise if unable."
"United 114, try that again, Center?"
"HIISSSS GARBLE HISSSS GARBLE GARBLE HIIIIISSSSSSSSSSSSSS"
"Didn't catch that, once more please? United 114."
"United 114, heading one-eight-zero, slow to Mach eight-point-one."
"Heading one-eight-zero, slowing to eight-one, United 114."
"Roger 114, we'll advise when we can get you back on course."
I'm enroute from Los Angeles to Manchester, New Hampshire, by way of Chicago. It's a calm cloudless afternoon in July. Perfect weather for flying. The weather report isn't too good for the analog radio network that controls air traffic, though. Lots of collisions, back-offs, and retries. A stuck microphone. Interruptions from technicians who are testing routes through alternate transmitters to relieve the congestion. Variable quality of service depending on the diction, accent, and vocal timbre of the pilots and controllers who are talking. It's always sobering to consider that the mission-critical air traffic control system works this way.
Today, I'm going to have more time than usual to consider it. Chicago's unexplained sequencing delay costs us twenty-five minutes. By the time I extricate myself from the 747, I've got 15 minutes to get from C19 to B20. I sprint the distance, and arrive with almost 10 minutes to spare, but...the door's shut, I'm out of luck. Here's what the pilot should have said as I was leaving the 747:
"Thank you for flying with us today. We know you have a choice of carriers, and we do appreciate your choosing United. Have a great day in Chicago, or wherever your final destination may be. Except the poor bastard who is connecting to Manchester. Even though we're unloading, and they know you're sprinting to the gate, you won't get there in time to avoid having the door slammed in your face. For the next 10 minutes, you'll stand in line waiting to rebook on the next flight, which leaves four hours later, and while you wait, you'll contemplate your plane, sitting on the tarmac, with your seat empty, preparing for its on-time departure. When you complain, you'll be given an eight dollar meal voucher, which you'll use to buy a six dollar sandwich. They'll keep the change. Have a great day."
Well, at least it's not like the last time this happened, when I missed the final flight and had to stay the night. I'll just settle down and get some work done. Haven't been to O'Hare in a while, but I've heard there are access points everywhere.
Not. I finally ante up five bucks to use a Wayport kiosk to do some Googling, and learn that the access point in the American Airlines Admiral's Club can be reached from the concourse outside. Sounds plausible, and maybe that once was true, but not now.
It's not a red-letter day for air-travel-related networks. Now, I suppose it won't be too long before long Wi-Fi will be ubiquitous at airports. And the mission-critical air traffic communications system has to, at some point, become a digital network. What I'm really curious about, though, is whether the hub-and-spoke (or, we might say, client-server) architecture of air travel itself will ever give way to a decentralized peer-to-peer system.
There is a movement afoot to make this happen. Its literary champion is James Fallows, who has written articles and a book on the subject. Here's the idea. We have five thousand airports in this country. We route most of the traffic through a handful of congested hubs. Most of the airports -- the smaller ones -- move little or no passenger traffic.
An example of one of those small airports is the one in my home town of Keene, NH. It boasts the third longest runway in the state. Air Force One can land there, and does, every four years during the New Hampshire presidential primary. At one time, a string of commercial carriers -- subsidized by the government -- ran flights to La Guardia. I could wake up at 7AM in Keene, NH, be in midtown Manhattan by 9, and back home for dinner.
Fallows invites us to imagine a fleet of air taxis -- essentially winged SUVs, safe, ultra-reliable, highly-automated, super-efficient -- that redistribute some of the traffic away from our failing hubs to a network of smaller airports. If this is going to work, the air taxi itself, which companies like Eclipse Aviation are busily inventing, is only half the technological challenge. The other half lies in the information systems. You'd need to be able to aggregate demand in real-time, so that if two passengers in Keene, NH, two in Rutland, VT, and two in Albany NY are headed to LA (say, to Orange County's Pomona airport) on a given day, you can allocate an air taxi to make those three pickups and then fly direct to the destination. These air taxis would need to be able to steer clear of conventional air traffic, as well as each other.
It's Tom Swift and His Personal Flying Machine, made real with the help GPS, satellite-based networks, and leading-edge navigational computers. Or so Fallows likes to imagine. The cold-water argument is, of course, that it won't be economical even if the technical challenges are solved. What would be the point of creating a decentralized air travel system for a population that is itself not decentralized?
There's a chicken-and-egg situation. Cities always were and always will be the major economic and cultural organs. Some will prefer, as I do, to live in small towns away from the megalopolis, an option that has become much more viable in recent years. I used to pretend that I could work effectively from my small-town home office with a dial-up connection to the Net, but it wasn't really true. With broadband access, it really is true. I'm highly productive, wildly connected, and deeply engaged with many parts of the high-tech industry. Doing that while enjoying small-town quality of life is not something everyone will want. But it's certainly a choice that more would like to make than do, were distance from major airports not a factor.
I honestly don't know whether that obstacle will be torn down in my lifetime, and if so, whether peer-to-peer air travel is the way to do it. But I sure like the attitude of the Eclipse guys. Their catchphrase to describe the airline-industry naysayers' attitude is WCSYC -- that is, "We can't, so you can't." Maybe not, but if things turn out otherwise I know one thing for sure. Nobody is going to miss visiting O'Hare. ... [Jon's Radio]
1:03:01 PM
|
|
EyeTV turns an OS X machine into a PVR. Jim Heid digs into the capabilities of the about-to-be-released EyeTV, a $199 gadget that attaches to any Mac running OS X through a USB port and turns your Mac into a TV viewing and program recording device, much like a TiVo. Interesting...read about it in EyeTV Review on the Macintosh Digital Hub site... [Mac Net Journal]
12:52:38 PM
|
|
Howard Rheingold: "Smart Mobs" This is an interesting twist. I have watched this take off too. It happens in the blog world with regularity (I still wish that people would get off of these automated reputation systems -- a blog is a reputation system, it works via a "I trust you, you trust him/her, etc," and scales extremely well. The difference is that there are real people making decisions vs anonymous voters.).
>>>Smart mobs use mobile media and computer networks to organize collective actions, from swarms of techo-savvy youth in urban Asia and Scandinavia to citizen revolts on the streets of Seattle, Manila, and Caracas. Wireless community networks, webloggers, buyers and sellers on eBay are early indicators of smart mobs that will emerge in the coming decade. Communication and computing technologies capable of amplifying human cooperation already appear to be both beneficial and destructive, used by some to support democracy and by others to coordinate terrorist attacks. Already, governments have fallen, subcultures have blossomed, new industries have been born and older industries have launched counterattacks.<<< [John Robb's Radio Weblog]
11:12:47 AM
|
|
From Hack the Planet, a pointer via cryptome.org to a one of the Hailstorm patents. Great reading, all the usual technology suspects: multiple client devices, loosely federated servers, asynchronous connectionless messaging, schema-driven infrastructure, multiple transports (HTTP, MSMQ, SMTP and multicast all called out explicitly), SOAP/XML/Schema throughout, etc. This whole thing reads like a blueprint for enterprise-grade XML-based distributed computing ("A server federation cooperatively interacts to fulfill service requests by communicating using data structures that follow a schema in which the meaning of the communicated data is implied by the schema"). IANAL but it leaves me wondering how broad the scope of a patent like this really is, ie. what rights does it really confer on the holder? [Peter Drayton's Radio Weblog]
11:08:29 AM
|
|
Liberty Comments.
The Liberty Alliance Project shipped v1.0 of the specifications for single sign-on and "account linking". The website talks about a "Liberty Alliance Developer Forum", but there aren't any links to it yet. The fact that the specs aren't somewhere on the web in HTML format is going to make it hard to have detailed discussions about it. In any event, several things struck me in an initial read of the Liberty specifications:
- Liberty requires & depends on channel security (SSL/TLS/IPsec) for message confidentiality & integrity. This is even the case for server-to-server communications (e.g. a service provider authenticating a user with an identity provider), despite the fact that the messages themselves are SOAP. I wonder why they couldn't make message-level-only security an option?
- Liberty makes heavy use of HTTP redirects and is hardwired to HTTP or WAP/WML transmission of raw application/x-form-encoded XML (no SOAP envelopes) for communications between the user agent and the identity server or service provider. Clients using other protocols appear to be out of luck.
- Communication between the identity server and the service provider is SOAP, but hardwired to HTTP. f I want to use some other protocol, I appear to be out of luck.
- Unlike WS-Security, Liberty is hardwired to SAML (WS-Security allows the use of SAML to convey assertions, but doesn't mandate it).
- Liberty defines it's own separate Reqest (3.2.4) & Response (3.2.5) envelopes that are carried in the SOAP Body to provide addressing info and an extensibility point (only on the return message). I wonder why they didn't use SOAP headers for this?
- Although the messages are defined using XML Schema there don't appear to be any standalone XSDs, which is a pity. There also don't appear to be any WSDL files at all.
- The whole thing feels very browser-centric and RPC-ish to me.
[Peter Drayton's Radio Weblog]
11:07:03 AM
|
|
Microsoft's Palladium is really a global money-flow tax.
Never mind that Microsoft’s Palladium computer security initiative offers digital rights management (DRM) capabilities for the entertainment industry. That someone else will control what you can and cannot run on your computer is just a minor side benefit.
Never mind that even though Mario Juarez, Palladium’s product manager, acknowledges the technology has built-in back doors. Back doors that governments can use to monitor and disrupt the information flows of citizens. Back doors that corporations can use to monitor and control employee behavior. Both are just minor side benefits. Nice, but not crucial.
The real purpose of Microsoft’s Palladium, according to Robert X. Cringely’s latest article, “A Hollywood ending: Does Microsoft really care about protecting the entertainment industry?”, is to control and tax the global flow of money. The DRM bits of Palladium are, according to Cringely, a red herring to draw our attention away from Palladium’s real purpose. After all, Cringely reminds us, “In 2001, the PC business was more than 12 times the size of the movie business and more than eight times the size of the movie and television businesses combined.” Microsoft doesn’t care about the entertainment industry—Microsoft is regularly accused of stealing intellectual property—the entire entertainment industry represents mere chump change in the Microsoft world view:
“Forget the flow of music and movies through your computer, and wake up to the flow that really counts, the flow of money. Put Palladium on every computer, and whoever controls Palladium controls the flow of money in the world. Wasn’t that the real point of .NET? Would Microsoft really pursue .NET without a corollary hardware strategy? I don’t think so.”
Excellent analysis; highly recommended. [ARTS & FARCES internet]
10:40:02 AM
|
|
 Your kid's going to flunk? Sue the teacher.: "One of the students in Elizabeth Joice's senior English class at Sunrise Mountain High School in Peoria, Ariz., was flirting with failure. In fact, it was much more than a dalliance -- she was flunking. The student, whose name Joice wishes to keep private, had plagiarized a test, skipped classes, failed assignments and even missed a make-up session that might have allowed her to raise her grade. Joice had been sending notices to the girl's parents since April, warning them about the failing grade; and both the girl and her parents had met with assorted district administrators, counselors and Joice herself. But it was all to no avail: It was almost graduation, the girl had blown too many tests, and she wasn't going to walk.
Imagine Joice's surprise then, when on May 22, just one day before senior graduation, she received a letter from a lawyer representing the girl's family. The family felt that the teacher had graded unfairly, the letter said; they believed that their daughter hadn't been given enough of a chance, and unless Joice took "whatever action is necessary to correct this situation" they were going to file a lawsuit." [Salon] [GranneWeb]
10:37:56 AM
|
|
© Copyright 2002 Mark Oeltjenbruns.
|
|
|
