Google Jibe: Better carrier messaging for everyone(jibe.google.com) |
Google Jibe: Better carrier messaging for everyone(jibe.google.com) |
Basically, they'd like all their SMS revenue back!
Dean Bubley had a great analysis - "Google + GSMA announcement on RCS is no gamechanger" : http://disruptivewireless.blogspot.co.uk/2016/02/google-gsma...
Great! Because I'd like an open platform back such that SMS offers, unlike every messaging platform you mentioned.
the spec even includes the big-brother way of you having to hand over your contacts to pair the phone numbers to other users that whatsapp sadly made the norm.
The spec is available on the GSMA website [0]. The relevant section is 2.13.1.3 of "Rich Communication Suite 6.0 Advanced Communications Services and Client Specification Version 7.0-final draft" available in the specification ZIP.
TL;DR: Encryption seems to be included but for voice and video only. They have _deliberately_ compromised messaging security to be vulnerable to interception.
"SRTP [RFC3711] may be used to provide per message authentication, integrity protection and encryption for both RTP and RTCP streams involved in real-time video and voice sessions."
[...]
"[3GPP TS 33.328] defines two modes of operation for SDES/SRTP: e2ae (end-to-access edge) mode and e2e (end-to-end) mode."
"[...] the RCS client may try e2e [...]"
And:
<Basic description of the messaging protocol>
"When using MSRPoTLS, and with the following two objectives allow compliance with legal interception procedures, the TLS authentication shall be based on self-signed certificates and the MSRP encrypted connection shall be terminated in an element of the Service Provider network providing service to that UE. Mutual authentication shall be applied as defined in [RFC4572]."
I hate being tied up to one device to message people. I change devices on the order of minutes, move over the course of a day between a few locations that all have devices I own, and don't regularly carry or look at my cell phone if I am already staring at something bigger (e.g. tablet/laptop/desktop/TV).
I enjoy e-mail/FB/skype/et al.'s ability to freely switch devices, switch OSes and continue your conversations extremely smoothly without any barriers. I want information to move with me, not with a silly phone.
SMS is wonderful because you can send a message to any mobile phone user regardless of their nationality, operating system, hardware, carrier, etc. However it has a few major flaws: cost, speed, lack of group messaging, limited multimedia support.
The shortcomings of SMS have fueled the rise of IP messenger apps like WhatsApp, Line, WeChat, Hangouts, and iMessage. These all have the features we want, but their major flaw are arguably even more severe: total lock-in and lack of interoperability. Once you have a majority of your communication networks on one of those protocols, you become dependent on a single for-profit company for your communication. Look at what happens to iOS users trying to switch to Android: iMessage blocks their communication for months afterward.
RCS will give us all the benefits of iMessage or WhatsApp but without being locked in to a single company's network or having to get all of your friends/family to adopt one or the other.
However - google and messaging is just a messy, and they're pushing this now? How about sort out your Messenger and Hangouts mess and get all of that consolidated
I'm saying this as someone who is a Android advocate and developer.
And one huge question: will iOS support RCS?
That's an workable idea. What's needed first is to improve the transmission speed of short emails. I was once considering writing a mail forwarder for servers that don't have mailboxes, one that would open an outbound SMTP connection to the destination host before closing the incoming SMTP connection. The message would be forwarded immediately and the status code passed back to the inbound SMTP connection before closing. No mail bounces, ever. This would be the normal case for single-address emails that aren't too big and aren't tagged as spam. It's not essential to do it this way, but a mail server should not delay a short message more than 1 second.
Next, IMAP servers need to implement NOTIFY per RFC 5645. [1] This provides a push notification back to any interested mail client that new mail is available.
Mail clients can then treat emails like message conversations. Maybe using the "colored bubble" display UI on mobile, as with texting. A useful informal standard could be that single-recipient subject-only emails or no-subject emails get that display treatment.
That gives us texting with attachments for images using existing infrastructure. It could kill off a few unnecessary messaging services.
Wouldn't RCS deliver exactly what you're asking for?
Nevertheless, a lot of operators/countries either don't have or don't want to allow these kind of gateways, for whatever reason. A purely IP-based solution would get around these restrictions.
The UI shows several user accounts that you have to keep switching. it is a total nightmare.
then, the worst offense, you have absolutely no idea if you send a message to someone, if it will go via hangouts http IM message or via SMS. you have zero control/feedback from the UI. The ONLY way to know, is to create contacts with just email address and others with just phone numbers. and even then, sometimes it will go via hangouts IM when you send to a phone number-only contact.
the separation between an app for first class SMS (and nothing else) and google voice (for google voice) was the right choice and it still works.
DO NOT pair gvoice to hangouts. you've been warned.
Better it works pretty seamlessly for people not using it because it just appears to come from a regular phone number to them. They just have to have that number also entered in their phone. That's one of the major hurdles to having a good replacement for SMS is that if it doesn't integrate into normal SMS there's a friction if Alice doesn't have a phone that can use the fancy new SMS replacement that Bob and Carol are hooked on.
edit: I see someone beat me to this suggestion.
It would be killer if Telegram could interoperate with at least 1 or 2 of the other services, which would allow it to ramp up adoption. Unfortunately Wechat is very insistent about not allowing 3rd-party applications to use their messaging protocol, and Facebook seems to have closed their XMPP interface as well. I would guess Whatsapp probably has the same attitude.
In this respect I really, really miss the days of Pidgin and those other similar applications which used to be able to put MSN/ICQ/AIM/QQ/Yahoo/Zephyr all on one interface. That seems impossible with the state of mobile apps now.
I fear it could eventually become another closed garden.
You have these problems with carriers as well. They're technically interoperable but they hit you with prohibitively expensive rates if you try to do anything international. Carriers are also effectively government-controlled monopolies or biopolies in a lot of places so users end up locked-in anyway.
> lack of interoperability
Again, you have these problems. Yes, phones are interoperable with phones, but they're not interoperable with anything else. IP-based messaging apps have the potential for phones to interoperate with computers, tablets, VR googles, smartwatches, and whatever else you want (although not all do at the moment -- Wechat and Whatsapp suck in particular -- but at least they have the potential, and some actually do interoperate with non-phone devices -- such as Facebook and Hangouts).
Phones may not necessarily be the center of everything 10 years down the road. I'd rather the freedom to innovate be with the software companies, not the operators of one particular infrastructure.
And very poor transparency on those short codes... I once got some extra charges, because of a shortcode (it was actually Yahoo!) I happened use was international. (No way to tell that by the number alone.) It was surprising to hear from the carrier I didn't have any way to block international text, too.
> Look at what happens to iOS users trying to switch to Android: iMessage blocks their communication for months afterward.
It used to, but Apple built a "deregister my number" page that'll be able to unlock that within a matter of minutes.
[1]: https://en.wikipedia.org/wiki/Short_Message_Service#Initial_...
2 - the iMessage to Android experience is still broken in many other ways. For instance if you were in any long-running group messages that were iMessages, your messages will be silently dropped until the thread is re-created explicitly.
The airtime is basically free (although, if send/receive volume is high, this ends up with more of the control messages sent than would have been with an idle phone), but routing and storing the messages in transit isn't. Also, a lot of carriers have contracted out their SMS systems and are paying their contractors per message. Everybody charges everyone else for SMS between carriers as well.
With "forward while connection is open" mail forwarders, that code would be passed back to the originating SMTP sender, which can then display an indicator in the bubble that the message has made it all the way to the recipient's device.
Now it looks just like messaging, but works over email infrastructure.
I respect your opinion about separate apps for sms/messages, but I personally don't believe that was the right choice. I strongly prefer having one app to use over switching all the time.
no dropdown whatsoever. Also to dial via gvoice you need a hangouts dialer addon, and everytime you last selected sms on hagouts and select to dial with the dialer addon, it complains that the sms account can't dial.
it's a train wreak.
Given that -AFAIK- Signal documents its protocols, I expect that there will be Pidgin support for it not long after the Signal desktop client moves out of population-limited beta and is officially released.
But yeah. It's a goddamn crying shame that this new crop of devs have decided to not only reinvent Instant Messaging, but to do it in such a way that leaves us back in the same situation we were in in the 1990's. [0]
That wheel keeps turning and periodically crushes us all, I guess.
[0] Of course, one could do the very same thing that was done back in the 1990's and reverse engineer the new wave of chat protocols. We still possess general purpose computers that can be used to snatch the plaintext of conversations that they're a party to that are sent over encrypted channels.
Also a lot of apps in various other countries (notably China, because I'm familiar with it) do registrations via SMS, in which you need a +86 number in order to register, period. Smaller companies don't have the infrastructure to send international SMS verification numbers. Larger Chinese companies (e.g. Weibo, Wechat) permit registration with a US number, but they're larger companies who have servers in the US and infrastructure with US SMS gateways. So running around the world with a US number isn't always practical. If SMS is dethroned as a "de facto" communication method that everyone is expected to have, this might change. Other places besides China may also have a similar situation.