Thanks for the insight, Tom,
This article throws out some interesting statistics. And they
sound reasonable ... Microsoft does have a way to go, IMHO.
I also think they have gone a long way already in terms of the
devices in which Windows CE is appearing, and they are not slowing
down. Tech Ed 2000 was the first time I saw the eMbedded tools
(they were already up to Version 3.0), and the forecast is that
the embedded tools will soon be an integral part of Visual Studio.
I tried the eMbedded tools when I got home and, just with my knowledge
of Visual Basic, I had a VB application up and running on my Pocket
PC in 30 minutes. It was simple to code and still event-driven, with
a timer, and multitasking ... but it was so easy to create that it
made me think of the Ada analogy immediately.
Where VB normally has an Interpreter, the eMbedded VB tools has some
form of source code instrumenter with integrated debugger. When I tried
to run the application in what would normally be an interpretive mode,
the eMbedded tool kit automatically loaded the code into my Pocket PC,
along with some instrumentation and other run-time support, and a task
monitor popped up, showing the tasks running on the device and allowing
me to run the program in a debug mode. Later, when I was finished the
development, I simply ran a Package and Deploy program that bundled the
program up for distribution to the whole family of Pocket PCs and Hand-Held
PCs. I clocked this development experience (from concept to deployment) at
about 30 minutes.
Could I have done the same thing with Ada? Yes, but not the same way.
In 30 minutes? Highly likely that it would not be possible in two hours,
especially if I tried to go through all the steps with an online debugger
and task monitor. And especially not if I started from scratch with a
I had not seen before.
With an enormous installed base of Visual Basic programmers (many, many
larger than the Ada programmer pool), and when the applications are
not safety critical, and they need to handle things like voice recognition
(the Auto PC), handwriting recognition (the Pocket PC), and even web
there are compelling reasons to go to Microsoft for the solution.
Even with our aircraft, Microsoft is becoming very appealing for the
Data System, not only from a technical sense but from the fact that the
is asking for it. Is Ada rapidly becoming a ghost that isn't even on the
screen for this class of embedded applications? Where is the Web Browser
Ada component? The hand writing recognizer?
Note that, on the other hand, Microsoft is not (yet) on the radar screen for
embedded, flight- and safety-critical software. But if you read the reports
Microsoft's new research center in Cambridge, England, you can see that this
change (see past issues of MIT Technology Review).
Richard Conn, Principal Investigator
From: Team Ada: Ada Advocacy Issues (83 & 95)
[mailto:[log in to unmask]]On Behalf Of Thomas A. Panfil
Sent: Wednesday, July 19, 2000 11:59 PM
To: [log in to unmask]
Subject: Re: Ada is very sick and getting worse
Richard Conn wrote: Wednesday, July 19, 2000 11:11 PM
> 2. Microsoft is entering the embedded world
> big-time with its embedded visual tools. Devices
> like Auto PCs (for automobiles), Pocket PCs (including
> ruggedized versions with bar scanners), and the like
> are emerging, and these are being programmed in Visual
> Basic and Visual C++. These are places where Ada
> could be making inroads.
> Is anyone doing something about this situation? Are
> there any Ada compilers for the Windows CE platform?
> Are there any plans for such compilers?
Windows CE might be an area where Ada need not worry,
at least for a few years. Per John C. Dvorak's "Inside
Track" on page 99 of the August 2000 PC Magazine:
"...If China isn't enough of a problem, Micorsoft is
still having trouble penetrating the embedded-systems
business. In last year's Electronic Engineering Times
survey of 5000 readers who were embedded-systems
developers, Windows CE placed 12th on a list of 12 OSs
currently being used. We're talking low single digit
market share. And the company has been working on
this for years. The next iteration is Windows CE 3.0
which according to the company has real-time features.
Seems to me that either such an OS is real-time
or its not. Real-time features sounds like marketing
nonsense. Whatever the case, Win CE may never deliver
on its promise. Microsoft though, must continue
developing it even if it means continuing to lose money.
If Microsoft dropped Win CE, it might loose all future
credibility. Tough situation."
On the other hand, perhaps Mr. Gates is just pulling his
punches in this area for some reason. -- tap
> Richard Conn, Principal Investigator
> Reuse Tapestry
Tom Panfil -- Treasurer -- Baltimore SIGAda
& Registration Chair -- SIGAda 2000