Google Tries Publicly Shaming Apple Into Adopting RCS (theverge.com) 187
Google is kicking off a new publicity campaign today to pressure Apple into adopting RCS, the cross-platform messaging protocol that's meant to be a successor to the aging SMS and MMS standards. From a report: The search giant has a new "Get The Message" website that lays out a familiar set of arguments for why Apple should support the standard, revolving around smoother messaging between iPhone and Android devices. Naturally, there's also a #GetTheMessage hashtag to really get those viral juices flowing. For most people, the problems Google describes are most familiar in the form of the green bubbles that signify messages to Android users in Apple's Messages app. While the iPhone app uses Apple's own iMessage service to send texts between iPhones (complete with modern features like encryption, support for group chats, and high-quality image and video transfers), they revert to old-fashioned SMS and MMS when texting an Android user. Not only are these messages shown in a color-clashing green bubble but also they break many of the modern messaging features people have come to rely on.
Just use Signal (Score:4, Insightful)
Re: (Score:3)
Re: (Score:2)
Okay.
Google.
Re:Just use Signal (Score:5, Insightful)
Re: (Score:2)
Yep, it's frustrating. SMS is what pretty much everyone has and uses even if it is not secured. :(
Re: (Score:3)
You mean the guys who count on Intel SGX [slashdot.org] to maintain security?
Yeah no
Re: (Score:2)
Re: (Score:2)
Just use email?
Re: (Score:2)
Installed it today. Tried to message my mother. Didn't work. Went back to Facebook Messenger.
People "just use" whatever communication system gets the message to those who need to hear it. Nothing more, nothing less. Using signal doesn't help me if I'm talking to myself.
Desktop client (Score:3)
One benefit of Signal is that there's a desktop client (Win/Mac/Linux). It's paired with the mobile app, so you can send and receive messages from the keyboard as well as from your fondleslab of choice. Voice and video also works from the desktop.
Re:Desktop client (Score:5, Insightful)
Signal desktop is total crap, because it doesn't even work if it's not paired to a phone.
Nobody should be using a messaging service/protocol which forces you to have a mobile phone and/or a SIM card/mobile plan with a carrier.
Any PC/tablet/whatever with an Internet connection should be enough.
Re: (Score:3)
Are there any popular desktop messaging apps that don't tie you to having an account with a specific gatekeeper company? You know, like how phone numbers are compatible across all phones and service providers?
Seems like everything from Facebook messenger to ICQ requires an account with a specific company to be able to participate. I seem to remember decades ago some groups pushing a decentralized ICQ alternative, but it never seemed to attract any users.
Of course we have long had e-mail that does *exactly
Re: Desktop client (Score:2)
Wire used to allow you to sign up with an email account only. No one I knew used it so I stopped as well.
Re:Desktop client (Score:4, Informative)
iMessage works on my desktop too.
Re: (Score:2)
In Windows and Linux? macOS, sure.
I wouldn't consult Google (Score:3, Insightful)
Re: (Score:2, Funny)
when it comes to messaging.
Why not? They've tried and failed about 100,000 times. If nothing else, they know what doesn't work more than Apple.
Re: (Score:2)
SMS IS OLD (Score:2)
Re:SMS IS OLD (Score:5, Insightful)
Sorry. But "Because it's old" is not an argument for a protocol that's meant to send text messages, and that still works perfectly fine for exactly that.
Re: (Score:3)
Re: (Score:2)
The argument isn't whether SMS can still send a text. It's the fact that it can't do any of the many other things people expect from messaging these days. I'm not one of them, but I at least acknowledge the reality of the situation.
Then use Whatsapp or Viber or FB Messenger or Instagram DMs or Signal or Telegram or Threema, or Line or Wire......or spin up your own server for Rocketchat or Mattermost or Matrix or Zulip...
Users who want cross platform '...is typing' notifications and read receipts and high-resolution video transfers have the options. Yeah, yeah, relevant XKCD [xkcd.com], but this is already a solved problem. Those who want blue bubbles can get an iPhone, those who want lowest-common-denominator have SMS, and those who want to go b
Re: (Score:2)
Then use Whatsapp or Viber or FB Messenger or Instagram DMs or Signal or Telegram or Threema, or Line or Wire......or spin up your own server for Rocketchat or Mattermost or Matrix or Zulip...
Those aren't cross platform. You're just shifting the type of platform they're locked to.
Re: (Score:2)
They are all cross platform.
Re: (Score:2)
Re: (Score:3)
Re: (Score:2)
When Google Talk/Hangouts/Chat/whatever the fuck it's called today uses an open interoperable standard....
And this was a point that I completely agree with and forgot to make. It's rich of Google to complain about the absence of an interoperable standard when they themselves can't manage to pick a standard for Android in order to compete with iMessage. They tried it with Hangouts and got bored, then Allo, but Messages support RCS except many OEMs put their own SMS clients on phones, so Google is one to talk. Hell, AOL figured this out back in 1997.
My point about the third party messaging services is that the us
Re: (Score:2)
What things?
I expect (in priority order): 1) my messages to get through and receive all messages sent in a timely manner 2) ...
3) other unimportant glitzy crap like video or whatever is a nice option sometimes.
People expect read receipts (showing whether the other person has seen your message), response notifications ("so and so is typing...") and emoji reactions (e.g. giving your message a thumbs up). There may be some other ways in which SMS is considered to be deficient, but those are the main ones.
Re: (Score:2)
None of those "features" would be of any benefit to me. I don't want the person who sent me a message to get any return message I didn't send.
Besides, my plan includes unlimited text and doesn't include unlimited data. I don't see any reason to replace SMS.
Re: (Score:2)
Re: SMS IS OLD (Score:2)
I don't expect any of those things.
I also think "emoji" are as retarded as it gets.
Re: (Score:3)
Re: SMS IS OLD (Score:2)
Then they should use an app that provides them. Go pick one and leave the rest of us alone.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
and that still works perfectly fine for exactly that.
That's great if you're sending text messages. In other news Gopher was a great protocol for sending text over the internet too, but these days we expect a bit more than that.
Re: SMS IS OLD (Score:2)
The wonderful thing about standards is that there are so many to choose from.
Re:SMS IS OLD (Google isn't sexy anymore) (Score:5, Insightful)
An "industry standard" is what the industry has chosen as their standard/default messaging format. Messaging client vendors and formats come and go (see: BlackBerry) but very few actually connect/respect carriers' industry practices.
Bullying Apple when Google has repeatedly failed to properly and reliably deploy RCS within carriers (much less focused across carriers worldwide) is just desperation marketing tactics. It's not a fault against the concept of a new industry standard; it's a fault by the entity trying to bulldoze the global messaging landscape.
Google has proven themselves untrustworthy across a plethora of technological and social endeavors, and RCS Messaging is just another feather in that cap. They even went to far as to re-brand the parent company to Alphabet to illustrate how wide a gamut their ambitions entail. But that doesn't mean they're right, in-the-right, or successful at anything beyond their core competency (search advertising).
Re: SMS IS OLD (Google isn't sexy anymore) (Score:2)
Re: SMS IS OLD (Google isn't sexy anymore) (Score:4, Interesting)
https://www.wired.com/story/go... [wired.com]
Google's original plan was to get carriers onboard with its RCS implementation, which it created when it acquired Jibe Mobile in 2015
https://jibe.google.com/ [google.com]
That's Google's RCS thingy. The concept behind rich messaging has been around for decades, but RCS (especially Google's RCS) is distinct and is what's being pushed. Even the Jibe site lists out how Google's RCS is both spec and tech infra.
Combine with the recent article (shared by other commenter below) https://techcrunch.com/2022/06... [techcrunch.com] and you being to better understand that Google's hands are all over and up the RCS game. Literally and figuratively.
Re: SMS IS OLD (Android Police) (Score:5, Insightful)
Here's another explainer, this one with the fabulous quote:
The history of RCS may not have started with Google, but it ends with it. Google delivered the version of RCS messaging most customers will experience through Chat and the Messages app. As a matter of context, though, this is far from the first time Google’s worked on a messaging service, and basically every single one of its earlier efforts failed — usually because of Google itself getting distracted or making bad and dumb decisions, like randomly starting new overlapping services or abandoning projects that just needed a little love. In many ways, RCS messaging is Google’s last hope for an iMessage competitor.
Re: (Score:3)
Bullying Apple when Google has repeatedly failed to properly and reliably deploy RCS within carriers
Why should carriers have anything to say in the matters? Carriers should be dumb pipes. They just process my data, and should not even be aware if I am sending a message or any other type of data.
Re: (Score:3)
Because RCS is supposed to replace SMS/MMS.
Which is a carrierevel transport protocol and has nothing to do with Internet TCP/IP.
Re: (Score:2)
Bullying Apple when Google has repeatedly failed to properly and reliably deploy RCS within carriers
Why should carriers have anything to say in the matters? Carriers should be dumb pipes. They just process my data, and should not even be aware if I am sending a message or any other type of data.
In instances where Google manages the entire end-to-end experience (across national carriers), they've had reliability issues. This indicates Google's competence within the space is suspect to say the least; Google's management/control of the RCS infrastructure regarding advertising indicates their tentacles go deeper than leasing space within a data center but truly controlling all data flow within the RCS ecosystem.
And Google is also a "carrier" (dumb pipe) yet they seems to know everything going on wi
Re: (Score:3)
What's wrong with SMS and MMS? If a protocol works, why change it? SMTP has been around for ages, and still is usable.
SMS/MMS have their issues, but RCS does not give much more, and is often used as vehicle [techradar.com] for spam [theverge.com]. Overall, why open oneself to another spam vector?
A SMS/MMS replacement needs reliability. Stuff is held server side until it is sent, and until the device confirms it is received. It needs end to end encryption and authentication. This is a must, so the protocol can be used as a way to be
Re:SMS IS OLD (Score:5, Informative)
What's wrong with SMS and MMS?
Doesn't work over wifi and no idea if the message was received or evaporated into the ether.
Re: (Score:2)
Re: (Score:2)
Then wh did you not call in the first place?
Re: (Score:2)
>"and no idea if the message was received or evaporated into the ether."
Are you sure? I use Textra and after I send a message, a short time after, I have a "Delivered" tag with the time attached to that message. And this is texting to people on different carriers and with different phones and SMS applications. Rarely, there is a long delay before "Delivered" appears. Extremely rarely I will get a delivery failure and can click on it to retry.
https://www.technipages.com/an... [technipages.com]
https://www.reddit.com/r/ [reddit.com]
Re: (Score:2)
Re: (Score:3)
What's wrong with SMS and MMS?
It's tied to a phone number / SIM card (this one alone is a big fail)
The phone number is tied to a country/region.
Phone carriers have the ability to bill per message. (Worse, some even do!)
It doesn't work on most Internet-connected devices on earth (including PCs, non-cellular tablets, etc. with no cellular plan)
It has low, artificial limitations (message length, pictures and video quality)
It's not reliable (if your phone is not on, with signal, there is no warranty that you will ever receive the message)
Sh
Nay the Internet, Yay the Metaverse (Score:2)
All Internet publishers should be ashamed of themselves for supporting a standard that's 30 years old. HOW DARE THEY?
br
Re: (Score:2)
Re: (Score:2)
Does iMessage have features not yet in RCS? If so that is reason why Apple will not adopt it yet.
I think the bigger issues is whether RCS has features iMessage lacks. Apple doesn't have much motivation to move to a stanard that provides no benefits over what they currently use.
This is the same thing that happened with Lightning and USB-C. Apple adopted Lightning to have a port that could be plugged in from either direction because at the time the standard, Micro USB, was a pain in the ass to plug in. After they've been using Lightning for quite sometime USB-C comes out and has feature parity. Great. Bu
Re: SMS IS OLD (Score:2)
As an Apple user, it doesn't bother me. I have no interest in any of the features touted by RCS and would prefer Apple spend its resources on shut that would benefit me, the Apple user.
Re: (Score:3)
The only carrier supported messaging apple supports is SMS and MMS. These are 30 year old standards.
While that is true, it's also true RCS is not reliable [reddit.com]
Neither is iMessage. I recall times when messages on my iPhone have gotten delayed by an hour or more on either AT&T or Sprint (can't remember which I was on at the time). If your signal is bad, messages take forever. Part of this probably caused by the complexity of trying to do SMS fallback when iMessage fails.
Re: (Score:3)
Re: (Score:2)
Part of this probably caused by the complexity of trying to do SMS fallback when iMessage fails.
Complexity... check! Every time my iPhone reboots it throws a "Failed to send message" dialog for a 4,000+ day old message, but you can't actually do anything with that dialog to get to the problem message to delete it or somesuch. Short of deleting the entire conversation history for that contact.
Re: (Score:2)
If a 20-minute delivery delay is not reliable, then I've got to say even SMS is not reliable. Messages *usually* go through within seconds, but I've seen them take hours, or get lost entirely. MMS is even worse. Which really isn't surprising when you consider how they're implemented on the network.
Re: (Score:2)
I receive about 50% of text messages (SMS/MMS) sent from AT&T to my phone (on T-Mobile). It's possible that I receive zero SMS messages and the messages that I do receive are MMS. I don't really know.
It's only messages from AT&T that are problematic. I suspect that my phone number is listed in a database somewhere in AT&T as being actually an AT&T number because my number was originally with a compa
Re: (Score:3)
That's the problem. We have this standard called "the Internet." Any new messaging protocol should use it.
We definitely do not need to replace hacked together protocols that depend on individual carriers not to fuck up, with new hacked together protocols that depend on individual carriers not to fuck up.
There are lots of open IM protocols. Pick one.
Re: Yes, but RCS sucks (Score:2)
Re: (Score:3)
Works flawlessly for me and most of my friends.
The few that use android, are just mostly hamstrung in that they cannot send high quality videos to anyone....but I can send to them.
So ... "works flawlessly, as long as you and everyone you communicate with stay in Apple's walled garden, when consumers are all supposed to be"?
Re: (Score:2)
Mostly just saying most people I deal with, use iPhones and iMessage...friends, family, co-workers....
Re: (Score:2)
So ... "works flawlessly, as long as you and everyone you communicate with stay in Apple's walled garden, when consumers are all supposed to be"?
You have reading comprehension problems.
If I sent you a message with iMessage you receive it es SMS. Likewise in the opposite direction.
Revision Control System (Score:4, Funny)
Revision Control System, didn't that become obsolete in the mid 90s ?
Re: (Score:2)
Re: (Score:2)
Revision Control System, didn't that become obsolete in the mid 90s ?
Came here to say that. No need for Google to shame Apple publicly -- Apple did it themselves if they're still using that RCS.
Hey Apple, time to switch to Visual Source Safe. [*ducks*]
Re: (Score:2)
"Revision Control System" can also be a generic term for a class of programs.
I had first interpreted the caption as Apple not using any revision/version control system for a project.
Use a modern version control system (Score:2)
Clearly Google is trying to sabotage Apple's development process. I mean, RCS [wikipedia.org] had its uses back in the day for small teams, but it's entirely inadequate for a large development organization geographically spread out over multiple locations and including remote devs.
Re: (Score:2)
RCS is nicer for config files than git. You can have a bunch of ,v files side-by-side with your configs. and use checkout/checkin to control the read-only access to configs to slightly improve their resistance to accidental edits. Good for labs where you have a lot of unskilled people mucking around in the system. A more professional deployment would use centrally managed configuration, probably rolled out with Ansible. Way more work up front to do it the "right" way.
Once you want branching, merging, or dis
Re: (Score:2)
Branching and merging in RCS works just fine.
Under the hood CVS uses RCS.
The other systems you mention are strong for distributed development and/or large teams.
Re: (Score:2)
I never liked branching and merging in RCS, or rather it felt rather incomplete for anything significant. I feel that CVS was a big quality-of-life improvement and used it for a very long time until Subversion reached feature parity with CVS.
And CVS hasn't used actual RCS under the hood for a very long time. It was rewritten and while structurally similar it departs from RCS in some key places (that's my warning just in case you attempt to use rcs or rcsdiff on CVSROOT files)
Re: (Score:2)
Rewritten, nevertheless the file formats are still the same.
If you dislike branch/merge then it can only be a tools/UI problem.
Under the hood it is exactly the same.
Re: (Score:2)
Well they're not the same. You will break your CVS repo if you use RCS commands on it too freely.
Original CVS is the tool/ui that solves some of the hardships of RCS. Eventually, after a decade and a half of development you can end up with data in your ,v that RCS doesn't really parse properly.
But of course everything is just bits on a disk, and I could use a magnetized needle and claim that everything else is simply a tools/UI problem. reductio ad absurdum
For use cases where you aren't sharing a file-sytem
I remember something like this in history class (Score:2)
Something about one being right, the other wrong. Arguments like "my tech is better than yours," and "yours will kill people." Eventually it led to an old circus elephant getting electrocuted or some such.
Don't kill elephants Apple, just adopt RCS.
Just a new way to push ads (Score:5, Informative)
No doubt Google's RCS team is looking for new revenue after giving up their juiciest market.
Gee, I can't think of any reason why a protocol vulnerable to spam might be an unwanted addition to a "luxury" phone brand. And certainly, certainly none of those Indian companies will start spamming US targets over RCS. Why, India doesn't spam or robocall the US at all! Once the people with spending power are on it, RCS spam will skyrocket in the US.
Next time somebody invents a communication protocol, make the conversations require permission from both parties, limit the frequency that users can try to add new permitted contacts, and require ID. If people really want to get randomly spammed by anonymous people then make receiving those messages opt-in.
Re: (Score:2)
Next time somebody invents a communication protocol, make the conversations require permission from both parties, limit the frequency that users can try to add new permitted contacts, and require ID. If people really want to get randomly spammed by anonymous people then make receiving those messages opt-in.
Google Workspace's shifting and resetting ad-tracking permissions feels seen.
... and they will accidentally not migrate your opt-out/opt-in permissions between builds and versions updates. Oops, sorry not sorry.
Re: (Score:2)
Exactly. I expect Apple to retaliate by showing how RCS is simply being used to deliver you ads.
Followed by big ads showing "Why Apple is not adopting RCS" and examples from India showing phones flooded with spam.
Heck, they'll probably tie it in with the current robocall scourge as well - RCS - text-form robocalls!
Re: (Score:2)
The simplest opt-in method is the Contacts list. If you are in my contacts list, I want your messages (and maybe calls). If you are NOT in my contacts list, I really do not want your messages or calls. Then make something like NFC standard on all devices to make it easy to exchange contact information between two humans. Bluetooth is on every phone and also has contact sharing as one of its protocols. Also make it easy to attach optional, user-selected phonebooks into the contacts list. For example, a
Re: (Score:2)
> If you are NOT in my contacts list, I really do not want your messages or calls.
cultural differences: In Europe we don't have such a big robocall problem, and delivery people use regular phone calls to alert their arrival, instead of the doorbell. This way they save minutes if you're home, and more if you're not and they can skip your address.
No, we don't do "leave it on the porch" regularly. Its the exception.
State administrations contact us primarily by phone for little issues that don't have
What incentive? (Score:2)
I don't see any incentive for Apple to tear down the wall around their garden. One of the iPhone's strengths is to simplify complex interaction; the "it just works" mentality. As such, most iPhone users (at least those that I've encountered) aren't particularly tech-savvy, although there are plenty of IT pros who do use iPhones*. So the non-techies don't really care about iMessage vs RCS.
The only way for the shaming to really work, is to really push the message that 87% of the mobile market is Android. Then
the garden is as popular as lightning cables. (Score:3)
I don't see any incentive for Apple to tear down the wall around their garden. One of the iPhone's strengths is to simplify complex interaction; the "it just works" mentality. As such, most iPhone users (at least those that I've encountered) aren't particularly tech-savvy, although there are plenty of IT pros who do use iPhones*. So the non-techies don't really care about iMessage vs RCS.
The only way for the shaming to really work, is to really push the message that 87% of the mobile market is Android. Then again, many iPhone users see it as a status symbol, although I don't really understand that considering the phone is competitively equipped but it's nothing special in today's market. Maybe Google needs to compare iPhone to some of the more reviled proprietary formats of companies in the past. Sony ATRAC?
*I'm not a fan of iPhone due to the vendor lock-in behaviors, but many are equally not a fan of Google's spying which is probably equally valid and a thing that I struggle with as well.
For starters, some incentives are anti-trust and customer happiness. When your mom on Android sees your kid videos as garbage, it really sucks. Now you either let the android user suffer or you need to share a link and add extra complexity. Apple users don't like Apple's shitty behavior. They just put up with it because they like Apple or feel socially pressured to keep up appearances and have fancy apple gadgets. There are a lot of shitty ethical elements of the apple ecosystem, like the working condi
Re: (Score:2)
Lightning sucks and charges much more slowly and means you have to carry an extra cable everywhere for your phone/headphones/AppleBatteryPack. It's really pitiful.
That's not the cable, it's the box you plug the cable into. Plug your Lightning cable into an iPad (or other higher amperage device) box and you'll see it charge up super fast. Apple does this because slowly charging the battery makes it last longer.
Lightning is limited to 18w. USB-C is now 90+W (Score:2)
That's not the cable, it's the box you plug the cable into. Plug your Lightning cable into an iPad (or other higher amperage device) box and you'll see it charge up super fast. Apple does this because slowly charging the battery makes it last longer.
Nope....Lightning is officially limited to 18w on USB-C. USB-A is limited to 12. I have a dedicated amp/volt meter for USB as well as a bunch of cables with them built in (I love those). I also use high-end lightning to USB-C cables on expensive chargers and have measured actual output. TBH, it's not a huge issue. My phone is pretty new and holds a charge well, so I don't charge it often.
I have my old pixel plugged in now and it routinely exceeds 20w. My USB-C iPad tends to top out at around 30w. M
You're saying "From where I sit," but... (Score:2)
Sigh... If you don't like Apple products, don't buy them. FULL STOP.
Now from where I sit, Google's spying is much more intrusive and harmful than Apple Lightning cables. Listening to Google defenders talk about 'their open standards' in the face of Google's own problems with search, with anti-trust regulation, with AMP and other "we own the web so we set the standards" strikes me as hypocritical.
You're saying "from where I sit," but not providing a tangible reason as to why my life is worse. Can you give a specific tangible example? What event in my life would have been noticeably better if they had never collected my data? Do I like Google spying on me? Not really, seems pretty creepy.
That said, can you provide a non-theoretical instance where my life was made worse by Google's spying? Google is an expensive operation. They need to fund it somehow and the only 2 realistic options are adve
Try sharing anything from reminders or notes (Score:2)
like how I can't share anything with a non-Apple user. Wow, we found him! The greatest idiot on the Internet.
...or high quality video. It's challenging to share with android users and most won't bother. But hey, don't let the facts get in the way of your chosen narrative about your favorite brand. Anyone who has anything but praise to say about Apple must be "The greatest idiot on the Internet"...in your words. You're one classy dude!
Re: (Score:2)
Apple customers... (Score:2)
But yeah, what's wrong with SMS? It works.
"Get those viral juices flowing" (Score:2)
A phrase that should never be uttered in any context that I can think of.
Problem: Carrier support (Score:3)
The problem with RCS is the carriers are involved. Change your number and you're gone. iMessage = carrier independence.
Re: (Score:2)
iMessage also = Apple lock-in. I'd much rather have a choice of phone brands and price-point options.
Re: (Score:2)
Hopefully apple will never adopt RCS (Score:3)
Re: (Score:2)
RCS isn't a Google thing, it's a cross-platform standard.
LOL (Score:2)
When SMS is gone, (Score:2)
I'll be doing a LOT less texting. I'm willing to turn on mobile data so I can send and receive photos via text, and then immediately turn it off; but I'll be damned if I leave data turned on all the time just so I can send and receive plain text messages.
I run LineageOS and I don't have many apps installed so my phone is probably reasonably secure anyway. But severely limiting the amount of time I'm connected to the internet vastly increases my phone's security, and I'm not giving that up just so I can rece
Most Android users outside the US use WhatsApp and (Score:2)
Subject says it all. Not that WhatsApp is a better or worse choice, most people just do not care. Whatever works.
Re: (Score:2)
Yep, Apple has a powerful lock-in mechanism here and they're not going to give it up, even if it means adapting RCS in the background while leaving a visual differentiator visible to users.
We all know which will win in a fight between a popular proprietary system and an open system that is technically superior...
Re: (Score:2)
and an open system that is technically superior...
Not even a hint of obfuscating your clear bias on the matters. NICE!