>an open development platform for all ISV's and
>Hardware Vendors to add value through services
Yes, ALL. You do know what "ALL" means, no? How about _ALL_ hardware architectures running the same OS? Wouldn't that be COOL? Wouldn't that be NEAT?
"Many observers confine Linux's enterprise opportunity to the market for low-end 'edge' servers such as file, print, Web, and e-mail servers," the study, entitled "Fear The Penguin," notes. "But we are confident that the technical developments and market forces are in place for it also to become the dominant OS on the higher-end servers of the enterprise data center."
Linux on low-end servers. Linux on mid-range. Linux on mainframes. Linux on clusters. A training nightmare if the the OS that runs on desktops is the only one that's unique, isn't it.
It's a simple matter of economics, and thus a simple matter of time.
>Just try to imagine a future where there's only one OS, and
>those skills that you learn on the desktop can be taken with you
>to file servers, mid-range UNIX clusters, and all the way up to
I guess not so cool and neat to any company that goes head-to-head against open source. It's a SUICIDE mission!
Tk is no Mathcad. It's kind of slick, but it's not the general purpose type of engineering tool that Mathcad is.
Mathcad has some limitations. It doesn't have the symbolic manipulations of a Mathematica, but few engineers do derivations.
The ability to pull various programs together into an analysis, is far more important.
Man, you'd better take another look.
Mathcad's an open framework that permits the integration of any MSFT compliant applications.
There's NOTHING like it on the Unix side, because Unix never developed the OLE/COM type of application integration features the MSFT platform offers.
The ability to pull various analysis packages together into a mathcad notebook will make you a believer.
I think there is a version of TK that runs on a server over the web so you can use TK apps on your browser. That would provide multi-platform support. I think Mathcad works over the web too though.
I just ran both for a few minutes here just to refresh my memory. TK is more flexible for those who want to solve problems in different directions. Mathcad is better for generating a nice report. Both seem good for an engineer's workbench. I guess I've always like TK better because it lets me enter equations without ordering them ahead of time and without isolating unknowns.
Yes, TK is quite usable. I only use MathCad when I have to. But there's a lot of good stuff out that's faster and cheaper (or free). I don't know if TK has cross-platform support like most of the professional packages out there.
>Mathcad is overhyped. Most engineers just use Excel, which can do most engineering calculations.
I have been an engineer for over 30 years and have worked for big and small companies. I rarely see Excel used except by the bean counters that torment us and the project managers who need to fake budget numbers. Excel simply cannot do the heavy lifting and only works numerically. My colleagues and I mostly use Mathematica, S (and R), Maple, and our own software. Mathlab is used by a few for DSP and system level modeling. (but note that the answers are not usually good enough for production code). Linux and FreeBSD give us many more options on what code to run.
Ha! Mathsoft went public about 10 years ago and was a complete joke for all shareholders. I shorted it from 12 to 2. Beautiful!
Mathcad is overhyped. Most engineers just use Excel, which can do most engineering calculations. There's even a utility you can buy for under $100 called MathLook which converts Excel formulas into math notation.
>Yeah it hurts you more than it does us right?
When are you going to learn that this isn't about you, or me. And when are you going to grow up and learn that this isn't about "hurt" you pathetic child?