Rework documation regarding DT discontinuation

[rcooke-warwick]

---

Please make sure to read the CONTRIBUTING document before opening the PR for relevant information on contributing to the documentation. Thanks!

[rcooke-warwick]

I have come up with the 100 devices number as a rough guide based on the previous project to create the framework. All the device that have poor scores, and/or we intuitively said that they should be removed, happen to have < 100 devices, so it felt like a safe number. Especially after 2 years.

If 100 devices / 2 years is too generous (I think it might be) then please comment

[rahul-thakoor]

i would remove the "Rougly around 100 devices is the current minimum requirement"

[shaunmulligan]

I think its probably too generous, I would say a device that hasn't gained traction in the thousands (>1k) is considered poor adoption and would be grounds for discontinuation. That of course excludes those on Custom Device Support plans of course, where they get discontinued when the CDS plan stops being paid.

[rcooke-warwick]

@rahul-thakoor @shaunmulligan , I think that >1000 is too harsh. It would leave us with about 18 public device types, which is a good number - but a decent number of those are legacy device types that took years to build up those numbers between various users.

I would propose splitting the difference and saying 500

[rcooke-warwick]

As we discussed, here is a draft list of devices to be discontinued "immediately" - i.e at the end of the cycle. This list is the "burn with fire" list, that have very few (<100 , but most of them 0) paid users.

Please comment on any that you disagree on

[shaunmulligan]

Added a few comments

[shaunmulligan]

I think its probably too generous, I would say a device that hasn't gained traction in the thousands (>1k) is considered poor adoption and would be grounds for discontinuation. That of course excludes those on Custom Device Support plans of course, where they get discontinued when the CDS plan stops being paid.

[shaunmulligan]

We should probably chat to Thomas about the Revolution Pi boards, I think we might want to partner more closely with them and for that reason it might be better to keep their portfolio of devices around, but its probably not major.

[rcooke-warwick]

the main reason for killing these 2 revolution pi boards was that they aren't available to buy any more.

[lurch]

Is this 2 separate products accidentally merged into one line?

[rahul-thakoor]

thanks for checking, but it's this device https://docs.balena.io/reference/base-images/devicetypes/#:~:text=Nvidia%20Jetson%20Xavier%20NX%20Devkit%20SD%20Seeed%20ReSpeaker%2D2Mic