On 12/11/18 21:40, John Clark wrote:
We are talking of embedded hardware, stuff that can be replaced with
future-proof
hardware (4+ years) for 20-30$ in most cases.
I don't think it is worth the manpower required to have any real attempt at a
LTS
(i.e. backporting security fixes).
I suppose you have never gone to an electronics manufacturing plant, been told
that the ‘parts placement’ machines were originally sewing machines, repurposed
for electronic placement, and in part run by a pneumatic system that was put in
before WWII. This was at a division of a billion+ dollar international company
at the time…
This is something that new equipment producers face with trying to break into a
market. The existing solution is used till the last gasp.
John Clark.
Apple and oranges comparison, you are comparing industrial automation to
consumer electronics.
Industrial equipment and automation in general is ridiculously
expensive, because it is often one of a kind or heavily customized,
and because it's produced in low quantities so you have horrendously
large manpower costs.
So it makes a lot of sense to reuse all you can, just so you can save
millions (literally).
I'm ready to bet that they also didn't ask volunteers to modify those
sewing machines but paid someone to do it, though.
And they paid for a refurbish of old equipment because it was cheaper
than buying new.
While I'm strongly suspecting that in this case its much cheaper to buy
new low-end hardware instead of paying the actual cost for an LTS release.
-Alberto
_______________________________________________
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
