Hacker News Reader

Comment by mtlynch | Hacker News Reader

I couldn't understand what this article was saying.

There's nothing nefarious about EDIDs. EDIDs are just a way of the monitor to announce its capabilities to the device so that the device and the display can agree on things like resolution, refresh rate, etc. EDIDs are just blobs of data without capability to execute logic, as far as I'm aware.

It sounds like the author is claiming that Windows does some sort of driver update in response to EDID announcements, but OP doesn't explain it at all. If that's true, that would entirely be on the Windows end not on the EDID's end. It sounds like Windows is recognizing LG as the manufacturer declared in the EDID and then downloading a driver for LG. I don't think there's any way for the EDID to declare to the OS that it wants to perform a driver update.

View on HN

This is indeed what is happening behind the scenes. By default Windows Update will automatically download and install support packages for pretty much any device Windows can identify through manufacturer/device IDs, which includes PCIe and USB devices but also monitors. This is most commonly used to deliver drivers but Microsoft allows these packages to silently install any user-facing application as well, presumably due to peripherals such as GPUs or audio chipsets often requiring "control panels" or similar companion apps.

Around a month ago LG took advantage of this feature by publishing a Windows "driver" for all their TVs and monitors that consisted entirely of payola bloatware, resulting in predictable backlash [1] and the obligatory subsequent HN discussion [2]. None of this has anything to do with the TVs themselves being connected to the internet or not.

[1] https://youtube.com/watch?v=Q9uefFYe6bM

[2] https://news.ycombinator.com/item?id=48956688

> automatically

This is the root problem.

The computer should not do something that the user did not specifically command. It should not guess, "Oh, the user plugged in device X. This means I have the user's consent to download and install software."

Every user wants their computer to do things they didn't specific command. You may draw the line differently from random user Bob, but your claim is simply wrong.

The *vast* majority of users do want their machine to silently do whatever it takes for their device to work.

I don't want to go handwrite XF86Config. I want my machine to go negotiate resolutions and aspect ratios and pick a default. I don't want to have to go find a CD drive or go to the OEM's website to get my printer or scanner working. Many users want their audio device to be able to talk over HDMI when plugging in a monitor with speakers. And so on.

In a nutshell, the article is saying that because you can't guarantee OS vendors won't go down the path Microsoft has (i.e. installing unwanted software as a "driver update" when you plug the display into your computer), it's therefore better to block the EDID just in case your Linux distro vendor goes rogue too. I don't agree with that argument, personally, but that's the argument he's making.

I also had to read it twice, if i understand correctly:

If i understand correctly:

1. TV is offline

2. Plug HDMI or DisplayPort into laptop

3. Windows (Update) on laptop recommends to install LG display driver

4. The LG display driver then installs all the bloat

Reply on news.ycombinator.com