"I do not see.", said /dev/vide0

I find it very odd that a camera that is supported fails to work and a device that isn't supported at all works almost perfectly. That seems to defy reason. I have browsed through some old trouble reports and I can see that like all people they go to what they think is the source of the problem, however, things are not always what they seem. It is clear there is a disjoint in this management of video , which extends to the kernel level. At some point I may make a bid to straighten out the process. It appears that all the needed structure is there, it is simply not managed in a coherent way. Perhaps this is something that I can help with and be effective.

This is the ongoing and somewhat frustrating view of video. I have to mention that this is not a Linux issue and the complaint that they do not support video well is unfounded. The windows environment is a total nightmare and there is no chance that it will resolve as they do not allow anyone to know what horrible tricks they play to get even one device to work, let alone support thousands, and Linux supports devices often, without any help at all from the manufacturers. In windows, you had better just install one device of one kind and limit the number of peripherals that you mix. There is obviously no cooperation between vendors and they would hardly care that another device made theirs unusable after you had paid.

The key here is that Windows forces you to pay up front for the operating system and then delivers trash to your living room and they could care less if it bothers you that they sold you a pile of smelly refuse (The acronym is "System High Information Technology" or more commonly known as SHIT). They have the money and you have been suckered.

So now to Linux, and I am investigating the structure of how video is used and so I am testing all my video cameras. I am now testing one that is on the supported list which shows up by "lsusb" as:

Bus 001 Device 004: ID 046d:0991 Logitech, Inc.

and there is a fault somewhere in the handling and I am trying to isolate that. I have modified the source for my drivers and this could be the origin of the incompatibility now, but I will make a report when I have it running well, to explain how I achieved that and what steps I took.

This follows on from an earlier thread in the blog HERE.I think it is actually quite absurd of the manufacturers to feign support a standard interface like USB and then play tricks to hide how it is used. I often think that it is to hide how ugly their source code is and avoid the ridicule of real programmers. I can guess from past experience that it is riddled with a higher "F" count than the entire source of the Linux Kernel. I am sure that Daily WTF would find some interesting things there and perhaps the number of places where // "Why does this work?" is found in the code would be very informational. I can hardly trust that the firmware in the device itself has any better design quality control.


Automated Intelligence

Automated Intelligence
Auftrag der unendlichen LOL katzen