[Computerworld] - Lineo, Inc. in Lindon, Utah, is now shipping an embedded version
of Linux called Embedix. With an additional software layer, Embedix PDA, due
in the first quarter of next year, the operating system will be able to run
Windows CE applications, the company said.
1.0 will initially run on Intel x86 and PowerPC processors. It requires a minimum
of 8MB of RAM and 3M bytes of read-only memory or flash memory. Bast Inc. in
Sonoma, Calif., will use Embedix in set-top boxes designed for hotels and apartment
based on a modified version of the Linux 2.2 kernel, Embedix contains proprietary
components and won't be free. Manufacturers will pay a royalty per unit shipped.
Embedix PDA, a layer on top of Embedix, will allow applications written for
Microsoft Corp.'s Windows CE operating system to run on Linux-based devices
after a recompile, according to Lineo.
Lineo is a subsidiary of Caldera Inc., which recently won an undisclosed sum
in a settlement of its antitrust complaint against Microsoft.
[Sm@rt Reseller] - Hot on the heels of the unveilings of Transmeta Corp.'s Mobile
Linux and Lineo Inc.'s Embedix, Red Hat Inc. is joining the embedded-system
fray with its Red Hat Tools for Embedded Developers (RHTED).
is an open-source integrated development environment (IDE) for software developers.
It uses any Linux operating system to develop embedded Linux programs for most
computing devices. Such developer tools will enable programmers to port existing
Linux applications to Linux-powered embedded devices. The tools, themselves
open-source, should greatly expand the ability of Linux programmers to easily
and quickly move into embedded-system development.
RHTED combines an IDE with a debugger in a single package. RHTED is built to
work with all mainstream Linux kernels and can be used to deploy embedded applications
for embedded Intel Corp. (Nasdaq: INTC) 32-bit and PowerPC chips.
Linux has already had a big impact in the server market, doubling its new-server
shipments from 1998 to 1999. This signals Linux's attempt to move aggressively
beyond the server market.
the discussion centers on PDAs, we believe the embedded OSes described will
presently be more focused on the desktop and especially Internet/information
appliances. PDAs tend to use CPUs other than Intel and PowerPC chips, preferring
MIPS and Hitachi RISC-based chips. In addition to the CPU issues, the PDA market
is pretty much owned by the Palm OS (approximately 80% share), so boasting of
Windows CE compatibility is of little value for PDAs.
appliance market is fairly new, however, and presently is one of the areas where
Microsoft will try to gain some market share for CE. Providing a "WinCE-like"
Linux solution should help Linux pick up at least part of the ready-made CE
market. In addition, Win CE application compatibility should buy the FOL (Friends
Of Linux) more time while Linux applications get written. It will be up to the
FOL to move quickly on application development, to avoid blowing a potentially
handheld and PDA market is expected to reach 55 million units in two years [Source:
IDC], and so resources should also be expended to focus some embedded
Linux effort on the more prevalent CPU architectures. We also expect Transmeta
to reap some benefit from these announcements.
Although "it ain't here yet", users will potentially find a great deal of value
in these developments. Low-cost appliances have a natural affinity for Linux.
As long as the applications are present, users should consider these products
seriously when they arrive. Low price should not be substituted for true
value - users will need to ensure that the functionality is there before deciding
to save a few bucks. Given the relative meagerness of functionality in appliances,
this hurdle should be easily overcome - if all it takes is an OS, a browser,
or a Java Virtual Machine, the task magnitude decreases significantly.
of the fragmentation of the current embedded-OS market, there is some risk of
incompatibility. At first glance, Red Hat says RHTED will help reduce incompatibility
issues. However, it is often the case that things proposed and things delivered
are significantly different. Thus, users should still exercise caution. Interested
users should check on progress in six to nine months.