> AmarOk <[email protected]> hat am 17. Januar 2018 um 20:47 geschrieben: > > > 1. The loss of privacy for ring is you need to use a central point > (explained in the previous mails) which have the authorization to send > push notifications. Push notifications will not be enabled by default.
Hmm is push notification enabled/integrated in the Ring Clients or in the daemon? > > 2. Yeah another way to do that is polling. You can try to get messages > every X minutes. The problem is it consumes battery (and it's another > part and can be used by OpenDHT) and you can also do polling directly on > the DHT. Do you have considered this solution as well? > > 3. The push server contains private informations to authorize to use APN > or GCM or any other solution. So it can't be widely distributed. Any one > can host a node, but you can only use push servers for your application. > The way to self host the push server is to regenerate a certificate for > your application with your device and to host your push server. Again, > like maxim said, push services are centralized. (But again, push > notifications will be disabled by default). Hm so how will Ring manage that for it's users? > > 4. OpenDHT doesn't say this. OpenDHT just tell a new value is stored for > a hash. And OpenDHT on your phone will wake up and decrypt the value later. So there will be no metadata leaked with an enabled push notification? > > > On 01/17/2018 11:23 AM, [email protected] wrote: > > Thanks Sébastien for your explanation again and the shared link to > > Riot*s way to handle that. > > For me there seem to be a few questions left: > > > > 1. The blog entry is talking about "downside is a small loss of > > privacy." -> What would that mean to Ring? > > 2. Matrix is talking about "non-GCM/APNS push mechanisms" -> Are that > > any alternatives Ring could use/develop? > > 3. Will the server-side push server run by you guys? Or is it > > self-hostable? Or is the server-side push server the Open DHT network > > itself? > > 4. And if Open DHT is just telling, that there is a message xyz > > exchanged: Who can someone draw conclusions from who is sending the > > message through the network. > > > > p.s. Could you please "group reply" so i get an E-Mail from you as well? > > > > > Maxim Cournoyer <[email protected]> hat am 16. > > Januar 2018 um 20:21 geschrieben: > > > > > > > > > >> ----- Original Message ----- > > > >> > From: [email protected] > > > >> > To: "Maxim Cournoyer" <[email protected]> > > > >> > Cc: [email protected] > > > >> > Sent: Tuesday, January 16, 2018 1:26:05 PM > > > >> > Subject: Re: [Ring] Sending failes or takes hours even if my > > contacts are online > > > >> > > > >> > Hi Maxim, > > > >> > > > > >> > you mean that my Android contact's should check their power > > settings, right? > > > >> > > > >> Yes, if your contacts are using Ring on Android, they would have > > to make sure > > > >> their Ring client is not "power optimized" by Android. > > > >> This isn't very user friendly; we are working on push > > notifications that should > > > >> make things easier in this regard. > > > > > > > > Btw. do push notifications work in a decentralised network - > > without connecting > > > > to servers running in the background? > > > > > > Short answer: no. The push service frameworks available on > > mainstream mobile platforms (Android, iOS) are centralized by nature: > > a connection is at all times maintained between them and the push > > server, AFAIK. > > > > > > Maxim >
