I have been working with HAL for about a week and have some suggestions regarding the spec. I think the ide_host bus should be removed from the spec in lieu of a general ide.host capability. I also think the definition of an IDE host should be an IDE hardware controller which contains two buses (primary and secondary) each of which can contain two devices (master/slave). Currently the definition of an IDE host according to HAL (afaics) is an IDE device which hosts two devices, primary/slave. It seems logical that the reason this was set up like this, is because most computers only contain one IDE controller (builtin to the motherboard on the PCI bus), which has two buses each with two devices. However, there are devices capable of providing additional IDE controllers, such that more drives can be connected and used. What HAL properties would the primary master device on the first additional controller have currently? I believe it would look like this /org/freedesktop/Hal/devices/path_to_ide_dev_3_1   ...   ide.host = 3  (0x3)  (int)   ide.channel = 0  (0x0)  (int) This scheme makes sense logically, but typically the identification for this device on the OS side would be host 1, bus 0, device 0. With the new definition, the IDE device would look like this: /org/freedesktop/Hal/devices/ide_1_0_0   ...   ide.host = 1  (0x3)  (int)   ide.bus = 0  (0x0)  (int)   ide.device = 0  (0x0)  (int) Also, how would I get involved with improving the specification itself, as in solidifying definitions etc