Slashdot Mirror
Microsoft's Reaction to OSS Adoption
inode_buddha writes "Eric S. Raymond has the eighth "Halloween" memo available here. It looks like Microsoft is really beginning to notice the national and corporate movement towards FS/OSS, and is reacting accordingly."
His main claim to popularity comes from writing "The Cathedral and the Bazaar". The problem is that the seminal understandings that sprang from that lightweight volume are now considered common knowledge. ESR was the first to codify into easily-understood form the innate truths about free software development that people had long since suspected.
He's come out with some more good ones (in particular, I like "Homesteading the Noosphere), but he hasn't written any work with more impact than than "Cathedral". He was also the first to publish the original "Halloween Document", which showed that Microsoft was, at last, taking the GNU/Linux threat seriously.
These days, almost everybody in the free-software/OSS development world understands the difference between the Bazaar and Cathedral development methods. They often consciously choose one or the other, or to develop according to Cathedral methodology, and transition to Bazaar after initial successful release. People understand the success of the development of GNU/Linux now, and despite what some will try to say, most really didn't until 1996 and the CaTB publication.
Lately, he's mostly a critic. Fetchmail is very slow on the development side these days, and his efforts to create a new build system for the Linux kernel were not accepted (killer effort, though, and well thought out, just too politically charged and too sweeping of a change for most people's tastes). However, he's still an exceptionally influential self-appointed Linux advocate. His opinions are read by millions of readers in and out of the free software community.
For the bio on the stuff he's done that has had a massive impact on the free/oss software scene, check out his bio: http://tuxedo.org/~esr/resume.html
Regardless, he has many publications in print and does a lot of speaking conventions. Like Bruce Perens, who is also influential in the community, he chose the role of public advocate for GNU/Linux for himself, and has been very successful in that role.
Matthew P. Barnson
I learn what I think when I read what I write
thus cutting their workstation licensing and support costs dramatically?
A company I was once with looked at Microsoft's Windows NT 4.0 Terminal Server Edition with a similar idea. Among other things, they could standardize on some NT4-specific apps without having to roll out NT4 to the whole organization. But they soon found out that the fine print of the licensing agreement said that since apps running in the terminal server were on NT4, then the user was using NT4, and if the client machine was not running NT4, you get to pay for an NT4 license. The company wound up saying "if we're gonna pay for NT4 on all our desktops, then we're gonna by God run NT4 on all our desktops". An additional downside was that whenever they want to upgrade from NT4 to NT5 (2000), they got to pay for upgrades across the board again. There were some other benfits, like WAN access and centralized administration, but licensing was definately not one of them.
Now Citrix is the company that came up with the idea of making Windows NT "multi-user" over the network. They licensed the NT3.51 source from Microsoft and fixed a lot of the "single-user-isms" and made a product out of it. Then, with NT4, Microsoft said "we won't let you make money from our OS anymore, but we will license the fixes from you so we can make money from it" and Terminal Server was born. Citrix was still making client apps for additional platforms like *NIX and handhelds and such for a while, but I'm not sure what they're up to these days.
Of course, everyone here knows that the MIT X Consortium was running graphical apps on multiuser machines over the network back in the late 1980's.
I am not your blowing wind, I am the lightning.