I have been looking at the alternatives. While don't know about the others but this is not so nice to see of DDWRT http://xwrt.blogspot.com/
There are a number of Open Source projects in the router space, some of them do a really admirable job. DD-WRT is taking flak at the moment because of going 'commercial' - ignore that, there are probably a lot of users who find DD-WRT really good. There are also a number of other really good Open Source projects in this embedded router space.
There are very few routers available using x86 technology mainly because there are some really good and inexpensive SoC's which are quite man enough to do the job required based on MIPS, ARM, PPC and other architectures. A Router does not require a VGA controller, for example, nor does it require a PCI bus or various other bits of hardware of a traditional x86 platform, all of which results in lower cost silicon.
I have worked a lot with LinuxBIOS, but have had a few projects recently where I have been forced to use ARM architecture. I weighed up the alternatives of using LinuxBIOS against the traditional loaders for ARM processors (I am sure the same would apply with MIPS and PPC processors), and there are compelling reasons not to use LinuxBIOS for these alternatives, mainly because LinuxBIOS is (primarily) designed to boot x86 architecture systems, and in it's design is expecting to have to enumerate PCI busses etc. In ARM/MIPS/PPC environments, the amount of hardware initialisation is often far, far less than what is required for an x86 environment, and hence, LinuxBIOS is a bit of overkill.
I know that there was a lot of work done a few years ago for PPC in LinuxBIOS (for the life of me I cannot remember the name of the contributor - who did contribute a HUGE amount), and as far as I am aware, this development effort has basically ceased, although do not quote me on this please. I am also aware that the targeted processor included PCI bus and a few PC-style peripherals, so in that instance LinuxBIOS may have been a good fit.
I personally have been involved with one GPRS/HSDPA router project based on an Atmel AT91RM9200 SoC, which has a unique and quite tricky boot process, and is certainly not suited to LinuxBIOS. I have also been involved with a number of router devices using the Broadcom MIPS based chips, and purely to retain backward flashing capabilities, the original Broadcom boot loader is used, meaning we have very little use for something like LinuxBIOS.
All this being said, I work a lot with the OpenWRT project (and am now on the core dev team), and this project does have a very active dev environment for x86 platforms. At this stage, the x86 platforms still rely on a traditional BIOS, mainly because this part of the project is aimed at generic x86 platforms, and to port LinuxBIOS to all of the possible supported hardware would be a huge task. I am sure once someone starts developing dedicated router-style platforms, this part of the project would benefit greatly from a LinuxBIOS port, but, it is not there yet in volume.
Hamish