Martin,
I'm all for advanced APIs and making low-level data/control available to
external computers. That's good design practice, and can be done without
releasing source code. Detailed documentation is required, of course. I also
agree that a fast CPU and communications interface would most likely be
needed as well. Perhaps USB 2.0 or (preferably) Ethernet.
I think there's an argument that such an architecture would be advantageous
to the manufacturer. You will recall that the original Microsoft/IBM/Intel
(WinTel) design philosophy was to keep the PC architecture open enough to
allow growth of a strong third-party add-on industry (which Microsoft
manages to keep under their thumb through draconian license terms and not
releasing all details of the API.) In contrast, Apple went with a closed
architecture so that they could supply and control all add-ons,
enhancements, etc. The WinTel approach results in much more flexibility for
the user, at the price of unreliability and chaos caused by
incompatibilities. Apple's approach locks the user into a rigid set of
choices, but has the advantage of superior reliability and seamless
operation. Despite its flaws, WinTel created a huge dynamic and growing
market, while Apple's market share has steadily shrunk. It appears that
flexibility is more important to customers than reliability.
On the other hand, the WinTel approach could allow a ham radio manufacturer
to become dominant in market share initially, but if other manufacturers
follow suit the result would be commoditization of the hardware and lower
margins across the board (some of the money would inevitably flow to the
third party software providers.) The ham radio market is probably too small
to make up for lower margins with volume.
73, Dick, WC1M
> -----Original Message-----
> From: Martin Ewing [mailto:martin@aa6e.net]
> Sent: Wednesday, January 21, 2004 12:27 PM
> To: dick.green@VALLEY.NET; tentec@contesting.com
> Subject: Re: [TenTec] Open Source for TT firmware
>
>
> Dick,
>
> It's good for users to understand the provider's point of
> view. Thanks for
> that. Open sourcing high level (C, C++) code running in
> Linux or Windows is
> easier to think about than specialized firmware. But hams
> want to experiment,
> and some of us want to work with radio control & DSP software.
>
> Maybe the way forward is to look at the "API" (application
> programming
> interface) that TT and other vendors provide us. Right now,
> all we get is a
> serial port and analog audio in/out. The serial port is
> enough to let you
> remote the slow parts of the front panel, but you can't
> remote the actual LCD
> panel contents, like the spectrum analyzer display.
>
> "Friendly" vendors like TT or Elecraft should be looking
> toward a higher speed
> API. This would not infringe on their IP. My short list of
> features -
>
> -Digital signal I/O: IF and audio, so we could do our own
> special modulation,
> signal processing, scanning, etc, in software without flakey
> sound cards. -Higher speed RF control, so you could implement
> your own scanning, AFC, what
> have you. DIY ALC/AGC loops, PSK monitoring, etc.
> -Faster remote of front panel, including display, using X11
> or similar. (The
> IC-7800 apparently gives you a VGA output, but won't let you
> merge your own
> computer's data.)
> -I really don't like to switch between my computer display
> and the radio when
> operating, but I like real knobs and buttons! Solution? Let
> me put my logging /
> qrz.com window on the xcvr's front panel... Maybe I should quit here!
>
> All this could be done with a platform that looks like the
> Orion with a little
> more CPU horsepower and your choice of high-speed interface,
> and the cost should
> be reasonable. It is a logical direction for a company that
> has already done
> Jupiter/Pegasus.
>
> 73- Martin
>
> p.s. I am using a Linksys WiFi Router that has an embedded
> Linux system, where
> a keen user can hack in and insert their own code. It
> doesn't violate Linksys's
> IP as far as I can see, but it is definitely unsupported.
>
> Dick Green wrote:
> > Much as I would love to get a peek at the Orion firmware
> source code,
> > and generally admire the open source movement, I don't think it's
> > feasible or realistic in this case. Unlike older generations of
> > transceivers, the Orion's firmware is really what defines
> the radio.
> > The hardware is relatively simple. It's the firmware that contains
> > virtually all of Ten Tec's Intellectual Property (IP) for this rig,
> > and is the embodiment of what I'm sure was a very large
> investment in
> > research and development.
> >
> ...
>
_______________________________________________
TenTec mailing list
TenTec@contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
|