|
Nick Gall's Weblog
 |
Monday, September 29, 2003 |
|
Friday, September 26, 2003 |
JBI Deals With the Heteregenous Network.Continuing discussion of Java Business Integration (JBI - JSR 208) and related issues. See my previous comment on JBI. See also the comment thread re the JBI post at Collaxa. I just commented on Carlos' post and I am reposting my comments here:
Carlos, I completely disagree with your analysis. It is only when the network becomes more homogeneous that we make major progress due to the "network effect." Your expectation of the inevitability of network heterogeneity sounded a lot more convincing ten or so years ago -- BEFORE the Internet effectively displaced ALL OTHER internetworking protocols. If network heterogeneity is inevitable, then how did that happen? The same displacement and resulting homogeneity is evident one level up in the Web's universality (HTTP/HTML). HTTP/HTML displaced all other remote-GUI network technologies: Compuserve, AOL, Telnet, Gopher. How did that homogeneity happen. In general, although network homogeneity does take a long time, it does eventually happen. Look at the email network, which settled on SMTP. Or outside of IT, look at the intermodal shipping container, which finally enabled interoperability across the shipping industry.
Web services is simply the attempt to extend the homogeneity of the Internet and the Web even further, and from everything I see, it is succeeding. In my experience, it generally takes five years for a new technology to break though to the 80% use rate. Look at XML. It was ratified in 1998 and it went big this year. That is, it is now in production in 80% of Global 2000 organizations. Also, to say that Web services' interoperability limitations, which are being reduced over time through efforts like WS-I, mean that it has "failed" is like saying HTML has failed due to its interoperability limitations
All that said, there is a role for heterogeneity in the network. Just as the Internet is a set of internetworking protocols (i.e., the Internet standards define no layer one or two protocols) for enabling binary packet interoperation across heterogeneous networks (e.g., Ethernet, ATM, Dialup, Frame Relay), so too Web services is set of internetworking protocols for enabling typed XML Infoset interoperation across heterogeneous transports (e.g., HTTP, MQ Series, SMTP). But if Web services does not succeed, then we are not going to make much progress. I think the major vendors (including Microsoft) all get this, which is why they are so far resisting the urge to fork the core Web services standards.
One thing I know for sure, as hard as homogeneity is in the network, it is impossible (and a very bad thing) at the end points. That is the fatal flaw in the Java vision – that all endpoints should run one homogeneous software model. That is NEVER going to happen, nor should it happen. It is at the endpoint that you want lots of experimentation and innovation.
So while I applaud the JBI JSR, it is akin to ratifying a Winsock or Berkeley Sockets standard API – very useful – but not nearly as essential as the network protocols they drive.
9:16:11 AM 
|
|
|
Wednesday, September 24, 2003 |
Impedance mismatch in Development.John McDowall posts:
The problem is not the programming language but rather the infrastructure we have built. To create an application that takes a string as input through a web form and then stores it in persistently a developer must know a wide range of technologies. This is true no matter which particular approach they use. If I use Java it is HTML/JSP/Servlets/Java/JDBC/SQL at a minimum if I use another approach the stack is similar (HTML/PHP/CGI/Perl/DBI/SQL). The number of transformations the simple string goes through from the web page to the database is significant. The impedance mismatch comes from the need to translate amongst datatypes. If it is a real world problem with complex types the problem quickly becomes hard.
...
While XML has its own set of issues it does reduce the impedence [sic] issue if an XML datastore is used. There are still several technologies in play but one data format. (XHTML/XSLT/XML/REST/XML-Store) - are we better off?
If you're interested in one XML-based framework/language end to end, check out the Water programming language. A SearchWebServices interview discussing Water and its approach to the multilanguage complexity issue and the Water approach is available. I've been following Mike Plusch and Water for awhile. It is definitely interesting technology. Question is, can it have an impact in a language saturated world?
7:25:31 AM
|
|
|
Monday, September 22, 2003 |
Stunting a Framework.Ian Rae has an interesting observation on frameworks in this comment thread on Michael Feather's post, Stunting a Framework:
There are two forces pulling on a framework. From above there is feature creep pulling the framework into a larger and larger scope. From below there are portability issues pulling the framework into creating infrastructure.
Sound familiar. The old spanning layer issue. The key here, which the other comments hint at in their mention of "pattterns," is to base the "framework" on a simple set of identifiers, formats, and protocols that can be used for a wide variety of applications and bound to a wide variety of implementations. This is why the emerging Web Services framework is the right approach.
5:01:22 AM
|
|
|
Saturday, September 20, 2003 |
Why the Future Doesn't Need Us.A colleague just sent me Bill Joy's dystopian vision of technology, The Future Doesn't Need Us, I assume because of his recently announced departure from Sun.
I must say I wasn't that impressed by it. I find dystopian visions as unconvincing as utopian visions. Typically, such visions paint technology as a general good or evil: Completely embrace technology X or absolutely denounce it.
Fortunately, the future is not a utopia or dystopia, which are both nowhere. (Utopia is a Greek neologism meaning nowhere, suggesting that Utopias are nowhere to be found.) The future is to be found elsewhere, in between these two extremes. Let's call it allotopia: allos (other) + topos (place). Allotopia is not better than the present or worse, just different (other). I find that the devil and god are in the details. At any given moment, I believe we should pursue particular uses of technology X into the future and not pursue other uses.
If all Bill is saying is "let's be careful out there," then I am in complete agreement. If he is saying that exploration of a given technology should never be pursued, then I respectfully disagree.
By the way Bill, allotopia doesn't need us; anymore than it needed the dinosaurs. The sooner humanity gives up its ethnocentric view of existence, the better off we will be. But I find this the most difficult Copernican revolution for people to accept.
6:18:39 AM
|
|
|
Friday, September 19, 2003 |
The Lifespan of a Company.From The Living Company by Arie De Geus
The average life expectancy of a multinational corporation-Fortune 500 or its equivalent-is between 40 and 50 years. This figure is based on most surveys of corporate births and deaths. A full one-third of the companies listed in the 1970 Fortune 500, for instance, had vanished by 1983-acquired, merged, or broken to pieces.1 Human beings have learned to survive, on aver-age, for 75 years or more, but there are very few companies that are that old and flourishing.
I'll bet the lifespans of companies forms a power curve, not a normal distribution.
2:15:58 AM
|
|
|
Thursday, September 18, 2003 |
The Evolution of Music.The New York Times reports:
[T]he ability to enjoy music has long puzzled biologists because it does nothing evident to help survival. Why, therefore, should evolution have built into the human brain this soul-stirring source of pleasure? Man's faculties for enjoying and producing music, Darwin wrote, "must be ranked among the most mysterious with which he is endowed."
The courting and social cohesion theories of music's origins assume that there are structures in the human brain that have evolved specifically to handle music. If no such structures exist, then Dr. Pinker's theory or something like it is correct.
11:21:59 PM
|
|
Comfort Foods Switch Off Stress.The New York Times reports:
Until now, no one has known how chronic stress gets turned off. A year ago, researchers in Dr. Mary Dallman's laboratory at U.C. San Francisco removed the adrenal glands from rats and exposed them to chronic stress. When they added stress hormones to rats' brains, the animals remained stressed. But when they fed them sugar, the animals calmed down.
This is the first time it has been shown that the tendency to overeat in the face of chronic stress is biologically driven, said Dr. Norman Pecoraro, a postdoctoral fellow at the University of California at San Francisco who helped carry out the research in rats. What is true for stressed-out rats, he said, is also true for humans.
Now I know why so many analysts (myself included) are overweight.
11:10:29 PM
|
|
Clues to the Evolution of Cooperation: Genetic Basis to Fairness.The New York Times reports:
Two researchers at Emory University, Dr. Sarah F. Brosnan and Dr. Frans B. M. de Waal, report today in the journal Nature that they taught female capuchin monkeys to trade pebbles for pieces of food. The capuchins were caged in pairs, so that each member of a pair could see the other. If one monkey got a grape in return for her pebble but the other only a less desired piece of cucumber, the shortchanged monkey would often refuse to hand over the pebble in exchange or might decline to eat the cucumber — both very unusual behaviors.
These refusals were often accompanied by emphatic body language, like dashing the pebble or the cucumber on the floor, Dr. Brosnan said. The expressions of exasperation were twice as common if the monkey offered a cucumber saw her companion being given a grape without even having to hand over a pebble.
I suspect that such a genetic sense of fairness also includes an aversion to others' freeriding.
10:29:21 PM
|
|
Aoccdrnig to rscheearch...It appears that an interesting hoax/meme is spreading rapidly across the Web:
Aoccdrnig to rscheearch at an Elingsh uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht frist and lsat ltteer is at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit a porbelm. Tihs is bcuseae we do not raed ervey lteter by it slef but the wrod as a wlohe.
Seems that there was indeed a dissertation on the subject back in 1976: The significance of letter position in word recognition. See a discussion of this dissertation and the meme at Uncle Jazzbeau's Gallimaufrey.
12:16:17 AM
|
|
|
Friday, September 12, 2003 |
The Internet Book Race."When it comes to books, Internet selling has not led to uniformly low prices. By Virginia Postrel."
Interesting analysis. When Amazon raises prices by X, it loses a smaller percentage of customers than Barnes and Noble does when it raises prices by X. This implies there is something besides price that users value about Amazon. Let's call it "BRAND."
8:57:23 AM
|
|
Binary XML Standard Considered Harmful --Followup.Omri and I seem to have reached agreement on the appropriate degree of standardization of XML-Infoset serializations. See my prior response to Omri for background.
I repost my comments here:
Omri, thanks for the clarification. As long as you agree that domain-specific XML-Infoset-serializations (wow, thats a mouthfull) are a good idea, then we are in complete agreement.
I completely agree that XML 1.0 (and its descendents, 1.1, etc.) should be the one and only UNIVERSAL XML-Infoset-serialization that spans (interoperates across) all domains. Why? because as your discussion illuminates, an "optimized universal" serialization is an oxymoron because optimization is always relative to a domain-specific set of concerns.
5:31:55 AM
|
|
|
Wednesday, September 10, 2003 |
Layer Six and AOP.I finally found someone making the link between Aspect-Oriented Programming (AOP) and SOAP: Carlos E. Perez!
I repost my comments here:
Hurrah! Someone is at least mentioning the conceptual unity of AOP and SOAP. Just as AOP software architectures enable "separation of concerns" based on the concept of interceptors, so too the SOAP envelope-processing architecture is based on "SOAP intermediaries" processing SOAP headers. Each header (or set of related headers) deals with a distinct set of concerns. Thus the WS-Security header deals with security concerns, while the WS-Coordination header deals with state management concerns.
No question that aspect-oriented architecture (or should the more general term be "separation of concerns -SOC- architecture") is emerging in various domains to deal with the dramatic increase in complexity driven by the increasing number of "first-class" concerns. Concerns such as security, performance, extensibility, availability, deployability, monitorability, which are dealt with by different software architectures in different ways, but which can be unified by SOAP's ability to deal with federated separation of concerns.
Is anyone else discussing how AOP and SOAP are just different "aspects" of architecting for SOC? I mentioned my inability to find such discussion in blog back in July. See http://radio.weblogs.com/0126951/2003/07/11.html .
10:09:48 AM
|
|
Binary XML Standard Considered Harmful --NOT!Omri Gazitt has a thoughful piece on Binary XML, but I disagree with his conclusions. I repost my comments here:
(Thanks for the pointers to the work underway on Binary XML.)
I completely agree with your argument against one universal binary standard, but "not one" does not imply infinite. That is, I think it would be useful to have a handful of "binary XML" standards which more or less cover (80/20 rule) the major scenarios within the three major axes. Of course, the market will eventually discover such "standards" on its own, but standards work could speed things up a bit.
Certainly standardizing a compressed, but simple to deserialize, XML for portable devices makes sense? Another standard would be the gzipped XML 1.0 serialization. All that's needed here is to standardize the nitty gritty details of using the two standards in conjunction; something that WS-I should probably tackle. I also think an XML to ASN.1 standard would be useful, just to cover interoperability between the two standards (a lot of management protocols use ASN.1). A few other serialization standards probably make sense as well, but you get the idea.
In my experience, the hugely successful standards all share the hourglass shape of a "spanning layer" (see http://radio.weblogs.com/0126951/2003/08/12.html#a33 or google the term). That is, they map down onto diverse implementations (in this case serializations) and map up into diverse applications (seemly infinite for XML Infosets). The Internet, Ethernet, SCSI, etc. standards have this shape. Currently, the XML set of standards lack this hourglass shape because they must serialize into the XML 1.0. Instead, they look like an inverted triangle (narrow base and wide top). Such an hourglass shape enables a standard to adapt both to new implementation technologies and new applications.
I believe the heart and soul of the XML family of standards is the Infoset, NOT the XML 1.0 serialization. XML 1.0 was the ladder we all needed to climb to get to a universal agreement on a universally interoperable data structure, the Infoset. Now that we have such a level of agreement, we don't need the ladder anymore--at least not it exclusively.
6:40:55 AM
|
|
|
Monday, September 08, 2003 |
User Innovation.Given my belief that it is the role of the developer-user that is essential to "open development" (which I will discuss later), I was pleased to come across the following MIT web site: userinnovation:
Empirical research is finding that users rather than manufacturers are the actual developers of many or most new products and services - and that they are a major locus of innovative activity in the economy. This finding opens up new questions and avenues for exploration in fields ranging from economics to management of technology to organizational behaviour to marketing research. Examples are patterns in innovation by users, characteristics of innovating users, design of a user-centered innovation process, economics of a distributed innovation process that includes users as innovators, and social welfare implications of innovations by users. [Emphasis added]
I find this view to fit my own experience extremely well. It is often users of a business application that see its limitations and are inspired to build a new application that better fits the business they came from.
I just started the paper "Profiting from Voluntary Information Spillovers: How Users Benefit by Freely Revealing Their Innovations".
3:40:30 PM
|
|
Donald Davidson Obituary.
Dr. Davidson wrote his dissertation on Plato's "Philebus," a dialogue between Socrates and Philebus, one of his many foils, a lengthy defense of the superiority of intellectual activity over physical pleasure. Dr. Davidson told Mr. Lepore that he thought Quine was perplexed over why he chose such an antiquated topic, but kept it to himself.
This is really going to bug me now...So I just went off and looked into it while writing this entry. In tracking this down, I stumbled across the following set of links on Davidson (see #18). Apparently the movie, The Thirteenth Floor, references Davidson! Of course the link is now dead, but Google to the rescue, with a cached page describing why Philebus is relevant.
It seems to be an excerpt from this bio. I guess I'll have to go and read Philebus now...In his honor. (I wonder if Project Gutenberg has it...)
4:31:58 AM
|
|
|
Saturday, September 06, 2003 |
IBM's New Linux Ad: It Moved Me.The Scobleizer says the new IBM freaks him out a bit. Or at least the kid does. But I'm almost embarrassed to admit that I found the ad moving. To see all these experts freely contribute their knowledge, insight and wisdom to "the kid" was moving. What I think Robert may have missed is the line in the ad that says something like: "What he learns, we all learn."
Yes, the ad would be more accurate if it showed the kid interacting more with his "teachers"--asking questions, correcting them, answering their questions. But it still worked for me. It got across the essential nature of open source, which is of course open, shared development.
5:35:06 AM
|
|
|
