I'm deeply confused why the person writing this thinks speakers are a technology that even can be "disrupted", let alone all the stuff he does that is for all intents and purposes just reading tea leaves^H^H^H^H^Hspeaker casings.
Sure, the manufacturing is interesting, and there is some truth to the smart aspect of the sonos being kind of "bolted on", but the attempt to read an entire companies outset from the manufacturing processes they use is just stupid.
As much as software likes to go on about "disrupting" this and "reimagining" that, manufacturing, particularly large volume manufacturing is extraordinarily staid, and different companies have in-house experience with different processes. Switching processes just because something is new and shiny is just not done, because the retooling costs are enormous, and you may not have the in-house familiarity with the new processes.
Really, what it sounds like to me is that the engineers at Sonos are far better at designing things for production. They didn't need to use fancy new tools, or design extremely exotic moldings. When it comes to cranking out a product, the LESS fancy new processes you use, the better in almost every case. It means the processes you're using are more predictable, you have more vendors you can use (because they're more broadly available), and you're likely to maybe have a old guy or two in house who's use process XXX for 40 years, and can tell you in excruciating detail every single little thing you have to worry about before you even start production.
Basically, this is a single decent point (sonos is a speaker company adding smart shit, and amazon is a internet-of-shit/retail giant adding speakers), buried in a whole lot of hyperbole and bullshit.
And most of the value added is in the software. The hardware is just an active speaker with the equivalent of a raspberry pi attached. Not new tech.
I owned a set a Sonos speakers for about 5 years now and I am in the market for a better replacement. The software has become unusable, the speakers keep depairing, not managing to read the network drive (and shockingly only support SMB1). But all the alternatives have moved into stasi-style “a mic behind every radiator”. I feel slightly out of choice. I just want the music not the spying.
I have a Chromecast Audio, plugged into the HiFi amplifier and good floor-standing speakers I've owned since I was 21. I use BubbleUPnP to control it, and have MiniDLNA running on a small server to share my music. (Though there are other ways to expose a music collection from a phone, NAS or whatever, or skip both of these and use Spotify or similar.)
I don't know if Google log every track I play — the Chromecast doesn't work if my Internet connection drops — but they haven't got GDPR compliant permission from me, so perhaps not.
Streaming the audio directly from my MacBook is unsatisfactory, it drops out a lot. So the soundtrack to a film playing on a laptop can't be done this way, I would instead plug the laptop directly into the HiFi. The equivalent setup works on Linux with an ethernet connection, but was a hassle to configure and get working so I haven't bothered with my new Linux computer.
I only have the one pair of wired speakers, I haven't tried the multi room feature.
Your comment would be better without the name calling.
My thought was: interesting teardown but where’s the support argument. Scroll, scroll, scroll, ah! Three points at the end, Ben reckons Amazon will win because:
The thing is, all those points are true, but they have zero to do with the hardware design.
If the guy want's to make those points, make those points. Don't pretend to extract them from the hardware with some sort of bizarre hardware phrenology.
Sure, the manufacturing is interesting, and there is some truth to the smart aspect of the sonos being kind of "bolted on", but the attempt to read an entire companies outset from the manufacturing processes they use is just stupid.
As much as software likes to go on about "disrupting" this and "reimagining" that, manufacturing, particularly large volume manufacturing is extraordinarily staid, and different companies have in-house experience with different processes. Switching processes just because something is new and shiny is just not done, because the retooling costs are enormous, and you may not have the in-house familiarity with the new processes.
Really, what it sounds like to me is that the engineers at Sonos are far better at designing things for production. They didn't need to use fancy new tools, or design extremely exotic moldings. When it comes to cranking out a product, the LESS fancy new processes you use, the better in almost every case. It means the processes you're using are more predictable, you have more vendors you can use (because they're more broadly available), and you're likely to maybe have a old guy or two in house who's use process XXX for 40 years, and can tell you in excruciating detail every single little thing you have to worry about before you even start production.
Basically, this is a single decent point (sonos is a speaker company adding smart shit, and amazon is a internet-of-shit/retail giant adding speakers), buried in a whole lot of hyperbole and bullshit.