Wednesday, February 27, 2002

Here's something that is surprising to me -  I haven't seen anyone put forward an extension to WSDL to support RPC protocols other than SOAP.  I don't get it.   SOAP is designed to be a lowest common denominator, an RPC-of-last-resort.   It's horribly inefficient relative to other RPC mechanisms and it lowers the barrier to entry for vendors outside your market (e.g. Microsoft entering into markets dominated by J2EE).   And the hooks are all there to support specification of multiple RPC protocols.   So, why isn't anyone doing it?    What gives? 8:19:45 PM

So, just for the record, here's my convention -- if it starts with something in brackets, e.g. [rosewater], then I wrote the stuff in italics.   If it doesn't start with something in brackets, I wrote the whole darn thing, whether or not I used brackets.   I arrived at this standard by the principle of "least effort" (for me). 7:55:42 PM

[Scripting News] Essay-in-Progress: Desktop Content Management.  7:52:58 PM

George Scriban is looking for a RSS Aggregator for the Mac.  So far he's had limited luck -- Radio Userland is the best he's found. 7:51:25 PM

[Slashdot] What Makes a Good Web Design? 6:01:30 PM

[Daypop Top 40] "Birth Playmate"   6:00:09 PM

[The Motley Fool] Essential Investing Knowledge. Understanding financial statements is the cornerstone of financial analysis. 5:53:27 PM

[kuro5hin.org] Is software innovation gone?. According to an article by Rob Pike from Bell Labs, software innovation is more or less dead (and Microsoft isn't at fault). For example: high-end PC hardware went from 33Mhz/32MB RAM in 1990 to 600Mhz/512MB RAM in 2000, but in that time software went from Unix, X, emacs, and TCP/IP to Unix, X, emacs, TCP/IP and Netscape. He argues that no one cares about making large changes or something new any more for various reasons.  Hmmm...  interesting, but I suspect we're missing some important dimensions in this analysis.   Probably not fair to lump the entire explosion of the Internet into Netscape.  I think the innovation in the last five years has had more to do with hooking up computers over a network rather than in harnessing the abundant processing power. 5:49:50 PM

I am finding that Porter's five forces is a more useful way to think about competitive strategy (not surprising since he wrote the book) than the more general guidance Drucker gives in The Essential Drucker. Drucker says companies exist to create more and better customers.  Mostly true, but there are opportunities to make fewer, weaker competitors too.   I suppose that's not an end in and of itself.   And you can certainly take that too far (i.e. to the point where you cease to be competitive and focus entirely on taking advantage of your customers).  Still, I don't think your corporate objectives would complete if they didn't address competitive pressures.  A better statement of a company's main objective might be "To create more and better customers while minimizing competitive pressures".    Clearly, that's often a trade-off. 12:23:48 PM

The software industry is tough -  you have to invest lots in getting your design to match the needs of your users.  But once you've gotten it right, it's often easily imitated - followers can take the best of your design and improve on it, avoid your most expensive mistakes, and develop a better product in a fraction of the time.  I guess this isn't news to Apple.  Short of unnoticed disruptive innovation, how will anyone ever compete with Microsoft?  By staying under their radar, I guess, and not competing directly until some foothold is established. 12:07:46 PM

I think IBM gets at least three things out of being an active participant in the open source community.  Cred - They get credibility with the technical decision makers in their target markets.  Their participation shows they are leaders in innovation and that they understand the technology (even if this doesn't always show up right away in their products) Market - By lowering the bar for technology projects (particularly small ones), they increase the overall size of their market.  It might cost them some software sales, but, hey, there's still hardware and services.  Also, small unfunded projects often mature into large, funded projects that buy commercial software. Barrier - Sowing the developer community with free software helps ensure there will be no small vendors to compete with them on price and/or reduced functionality.     It makes it harder for new entrants to get a foothold when a open source developers can take their design and build equivalent functionality on top of free software in a quarter of the time. This doesn't make their participation evil -  just means its good business not altruistism. 11:56:38 AM

[John Robb's Radio Weblog] Interesting.  IM logs.  Microsoft has a plan to introduce this.  Dann has an IM log tool already working for Radio and Trillian.  10:47:51 AM

BEA eWorld attendees: Web services are more than hype 10:36:21 AM

[Slashdot] Building Linux Virtual Private Networks  A favorable book review.  Where was this book when I needed it.  I guess I'm surprised that VPN is not an integral part of Redhat yet. 10:30:44 AM

[BBC World] Greenspan cautious on US recovery. The influential chairman of the Federal Reserve says that the US economy is on the mend, but says the recovery could be slower than normal. 10:25:33 AM

[ZDNet Tech News] Glitch puts Morpheus in the dark. The Morpheus file-swapping network goes dark, shutting out users. Owner StreamCast blames Kazaa for 'technical issues' that are blocking the service.  Shouldn't this be impossible?  So much for the idea that it's a decentralized service with no Single Points of Failure... 10:24:26 AM

[Scripting News] Bill Bumgarner reviews Mac OS X.  7:47:32 AM

Speaking of The Tipping Point, I noticed the other day that Seth Godin (Permission Marketing fame) had written a book with Malcom Gladwell -- Unleashing the Ideavirus.   I will probably pass. 7:34:32 AM

[Daypop Top 40] "Social Network Analysis"  Interesting paper on topics related to The Tipping Point and knowledge management.   Lots of good pointers for further research.   They mention Babble, which is appears to be an IM system with built in social network analysis.    7:27:02 AM

[The Motley Fool] Berkshire Hathaway. "Even with a simple business it is impossible for the investor to know everything."  - A nice little article about indicators of reliability in corporate financials.    I continue to be impressed by the level and quality of postings at TMF. 7:15:22 AM

[Slashdot] RIAA Almost Down To Pre-Napster Revenues a Slashdot article that talks about how cd revenues are finally down. 7:02:52 AM