Slashdot Mirror
The Economist's Open-Source Quintet
LarsWestergren writes "The latest issue of The Economist has an interesting group of articles on software, including The Beast of Complexity, Open Source, Sun, Microsoft and the Battle of the (next generation) Platforms, XML alphabet soup, and Software Integration. Not technical, in-depth or error free, but still a good overview, and a hint of what the suits of the world find interesting in computer culture." A good thing to point your boss to, if necessary. Economist articles often include some interesting graphs -- in this case, for instance, there's an interesting chart (though from aging data) on Linux developers by their email suffixes attached to the second of these articles.
Fascinating, but I want the next generation platform to evolve, not out of Sun, Microsoft, IBM, HP, Exodus, etc., but out of Mojo Nation, E, Chord, FreeNet, etc.
Open source projects, with ambitious goals for self-healing, self-organizing networks, tolerant of diversity, resistant to any conceivable attack, and free from the manipulations that mega corps inevitably introduce in their unceasing quest to gain monopoly power.
As I recall, in the end they explained why the concerns were unfounded/mitigated. As a pro- free-trade/world-trade publication, they must write this story with this conclusion in mind.
I agree that the economist is a very well writen magazine, but remeber no matter how good the writing you are still being sold thier company line.
--locust
The economist has its own biases. It is a capitalist publication. These views are evident in its editorial policy, the subject matter it persues (the stories that are not writen), and the recomendations of its columnist on given issues. This is not to say its a bad publication... Simply be aware, that just because a you are not being beaten over the head with propaganda your are being given an objective view of a situation.
--locust
The Economist is actually a very intelligent and insightful weekly, and most of their column writers really know what they're talking about.
/. If you dig back a couple of years, you can find a this great article where Bill Gates answers various allegations the Economist made against him. Suffice to say, bad little Bill uses the words 'freedom to innovate' in every other sentence throughout the whole piece.
Being a British paper, they do a good job of getting the news across, neutrally. I never read American weeklies anymore: It's all the same propaganda dog food.
I'm glad that someone had the bright idea of pointing out their computer articles to
"And then there is a raft of fast-growing newcomers, as yet known only to insiders, that are also hoping to get a big piece of the pie. BEA Systems, a Silicon Valley firm founded in 1995, is the most ambitious, with its plan to do for web services and e-commerce "what Microsoft has done for the PC", as chief executive Bill Coleman grandly puts it. "
Uhm....what Microsoft has done to the PC? You mean, fuck it over completely and turn it into a horrid, unusable piece of unstable manure?
OK, the 70s'-80s have been basically moved the boundaries of computing-people relationship from 1-n to 1-1. This decade we are halfway through the shift from 1-1 to n-1 where we clients access multiple servers. This can be very broadly generalised as a shift away from single-user WIMP (Windows-Icon-Menu-Pointer) towards a multiuser LAMP (Linux-Apache-MySQL-Php/Perl) where the traditional model-view-controller paradigm now has multiple views and multiple inputs. Wheras the X event-loop dispatched the user actions, we now have the server serialising access and using the database to maintain a consistent state (baring a power failure or 5) to generate multi-media data-pages on the fly. This is good in the sense that software is now extended through time (think bookmarks) and space (the internet). It is novel to the mainstream because the traditional moving pictures (aka square-eye god of TV) was non-interactive (turn on / tune-out). Hence they can (gasp) click to delve further or (surprise) customise the appearance. This offers intriguing possilbities as we are not limited to keypress/mouse actions as event inputs but higher level functions like complete forms (basically what the fight about XML-schemas is about). Pretty basic stuff for anyone familiar with the guts of X but never before pushed into mainstream. If you model the bandwidth differential between a LAN and the real-world (TM), you can probably predict the type of network applications x years ago that will become somewhat popular next year.
Unfortunately what we are lacking is new courses/theory/tools on how to adapt applications for a n-1 computer-person relationship. Peer-2-peer is just an example of bad marketing of a really difficult concept. How do you get 2 or more machines into a consistent or synchronised state. This is the difference between having a dog on a lease and herding cats. Whatever solves this scalability issue is going to be create a whole new set of products/services (cough*.NET*cough). Why? Because traditionally data processing has been considered a pipeline process (following the manufacturing model) where it gets transformed from low-order to higher more structured order. Instead the network allows multiple pathways, multiple sourcing. This is illustrated by the fact that raw music (MP3s) might actually be more valuable (to someone like a DJ mixing/blending new tracks) than the finished packaged album (which is limited to only listen to). Allowing the consumer to gain access to the intermediate stages of production will be a long-term benefit but for now, it is anathema to vertically integrated businesses which is why they are being very careful with their EULA.
It is going to be an interesting decade ahead.
LL
The Open Source article badly misquote SuSE's claim. This whole article is more like a high school report paper, stealing stuffs from here and there and add flashy graphics.
Nothing written by human hands can ever be truly objective, so the best you can hope for is a magazine that honestly states its editorial standpoint, and gives thoughtful consideration to alternate viewpoints.
And on top of that, there's usually a great deal of wit and humor in each issue. The Economist rocks.
Stop by my site where I write about ERP systems & more
Maybe Microsoft is deliberately creating that situation so that only all-Microsoft systems will work.
Anyone who hasn't already should check out JBoss -- my vote for the next-generation web services platform.
MS built its success on Windows, which benefited from the pervasiveness of cheap IBM compatibles. Sun built its success on SPARC workstations and servers, which were powerful enough to be cost effective network boxes, despite their proprietary technology. Neither platform has a central place in the net-centric future, and both companies know it. Hence various "next generation" initiatives. But no serious observer is impressed by any of these.
Java has always been a solution in search of a problem. Not that Java hasn't had its successes, but it has a longer list of failures: web "applets" (except for a few Yahoo games), thin clients (I don't count terminals that run GUI server apps -- these are "clients" only in marketspeak), platform-independent Office suites, smart appliances... the list goes on and on. Java has been underrated by people who don't understand the strengths of bytecode VM technlogy, but also overrated by true believers. It will always have a role, but that role is limited.
Somewhere a Finnish (no wait, we've been there) or Chinese or Nigerian computing geek is sitting down at his P90 box, cursing his flickering monitor and slow connection, and coding the killer app that he can't afford to buy. He'll upload a copy somewhere, millions of people will discover they can't live without it, and our geek is on his way to being on the cover of Time. That is that future of network computing.
__