Lpc2000

I am getting tired of toolset vendors who offer useful products, but 
then cripple them and make developers lives needlessly unpleasant by 
only offering their products only on MS Windows.

Messages like this one, on one toolset provider's web page bug me:
"We are considering a Mac OSX / Linux port of our compiler tools, 
including ICCV7 for ARM. There is no firm commitment yet and nothing 
likely to be done until toward the end of 2005."

Take note! You will not get my business then. Don't call me, I'll call you.

Those who use both (I have a Linux and Windows system on my desk) will 
know that M$ Windows is an inferior development environment. Take just 
one example: For years now, the Linux desktop has outperformed a Windows 
interface that has remained stagnant more than 10 years. Both KDE or 
GNOME have offered a far more usable desktop, with basic features like 
multiple desktops and window layer control, for years.

Then there is Linux's excellent networking, many choices of graphical 
file browsers, command line access, umpteen other included free tools, 
robustness and fundamentally designed-in defense against viruses, 
adware...  And why use cygwin when you can run native?

If an ARM cross-compiler toolset is developed using one of many basic 
windowing libraries and with portability in mind then porting should 
_not_ be a major issue at all.

Following the recent thread on dongles, It seems that some vendors are 
way more worried about copy-protection than building an excellent 
product that sells itself and that everyone will want to buy.

In my opinion, toolset developers will lose more and more (profitable) 
market to embedded Linux, its variants and gcc because they are 
invisible on the Linux development platform. This is not to say that 
embedded Linux or complex variants are appropriate for many lower-end 
embedded applications. But wake up and smell the coffee! As we witness 
with the LPC series, silicon capability and complexity continues to 
increase, making higher-end environments more and more appropriate for 
many applications. Toolset developers stuck only on MS Windows (and 
sticking their customers with MS Windows) will increasingly be caught 
between Linux for larger apps (anyone see which way cell phone companies 
are heading these days?) and gcc-based tools for both upper and lower 
end applications.

Too bad. There's some useful toolset product development going on, but 
it's wasted. I would like to see some serious competition on the Linux 
platform and have vendors to choose from.

  Brett