[DISCUSS] Linux Wins [Was] meeting

Craig M.Buchek craig at buchek.com
Fri Nov 16 01:23:11 CST 2001


Jonathan Drews wrote:
>  Those questuons were quite pertinent in light of this excerpt from
> "Linux Wins":

The presentation was not intended/advertised to be about Microsoft's 
business practices/strategies/marketing or how Microsoft relates to Open 
Source. It was about Microsoft .NET.

While .NET has a marketing/vision portion, the core is about some 
interesting pieces of technology. That's what we wanted to hear about. 
Especially given the fact that we're going to have to interoperate with it 
eventually. And the fact that they're publishing the specifications so 
that Open Source may get a chance to co-opt the whole thing away from 
them.

>  So much for all the clap-trap about .Net cooperating with other  OS'.

Microsoft has no choice but to interoperate to at least some degree. They 
realize that. We need to realize that too. We also need to understand the 
technologies that they are putting forward so we can both compete, and 
interoperate with them.

>  Steve mentioned that .net would allow access to development tools.but
> Linux has them right now. For a measly $80 you too can have ada95
> (gnat), scheme, lisp, forth, Tcl/Tk, FORTRAN, JAVA, C, C++, smalltalk,
> perl and python installed on yor box, With debuggers and profilers no
> less.

But Linux does not yet have a very good integrated development 
environment. At Thursday's meeting, they showed of Visual Studio .NET. We 
are very far away from achieving a level of sophistication like that where 
we can build an interface without touching any code, and then hook the 
code in without touching any layout. And choose from various languages for 
various pieces, and have a consistent API and ABI among them.

This is something Open Source must strive to achieve. One of the key 
points in the first half of the Thursday presentation was that Microsoft 
achieved it's level of dominance by providing a development platform in 
which apps could be built easily and quickly. Once you achieve a critical 
mass, everyone starts building on the platform. While Linux also has a 
high ease-of-use appeal to developers, mainly in APIs, there's something 
missing as far as high-level development tools.

>  Sorry if I offended Christine but I have to challenge BS when I hear
> it.

If you're going to survive as a tech in this industry, you're going to 
have to learn to filter out the marketing and buzzwords. It's not just MS; 
it's Oracle, Cisco, Sun, IBM. In some ways, even the Open Source community 
encourages a certain way of conformist thinking. The key is to look at 
what others are doing, filter out the hype, and identify ideas that we can 
use to our own advantage.

I identified several of these key points from the presentations:
- Create a good development platform -- apps will follow, then users.
- Integration of apps within your platform is key.
- Developers who build upon your framework will develop more than you do.
- The future lies outside the desktop.
- XML is obvious.

On the other hand, there are plenty of ideas that in my mind won't last 
too long:
- Allow your vision to be hard to define.
- Pretend all of your products are part of some unified vision, and name 
them such.
- The desktop is dead.
- XML and SOAP will solve any problem.

And there are problems that are yet to be solved:
- A single entity can't be trusted to authenticate everyone, yet it is 
probably impossbile to authenticate everyone using a distributed web of 
trust.
- XML and SOAP don't provide a complete end-to-end solution.
- Securing web services is hard.

Cheers,
Craig
---
Keep your friends close, but keep your enemies closer.
-------
St. Louis Unix Users Group - http://www.sluug.org/
To unsubscribe from the SLUUG discussion mailing list, send a message to
discuss-request at sluug.org with the word 'unsubscribe' as the body



More information about the Discuss mailing list