Hacker News new | past | comments | ask | show | jobs | submit login
Google changes its messaging strategy again: Goodbye to Allo, double down on RCS (techcrunch.com)
274 points by Garbage on Apr 20, 2018 | hide | past | web | favorite | 336 comments

Didn't Google have messaging figured out with Google Talk? Everyone was on Gmail all day (as many still are) and people actually used it.

Then they let it wither, came out with Hangouts, and it's been a downhill train wreck from there.

Google Talk had the potential of being one of the biggest messengers in the world...they own Gmail and Android (!!!!!). Are you kidding me? There's really no good reason for them NOT to be in a hugely dominant position here.

But instead they screwed it up & left a gaping hole in the marketplace.

Google dominated non-SMS chat in the late 2000s/early 2010s. I don't think they realized how insanely valuable that was - they were basically sitting on the future of communication. All they had to do was sit back and let network effects build their monopoly (with some more tech build-out and straightforward marketing in international markets).

Instead, they myopically saw Facebook as a threat and started one of the most user-hostile rampages I've been victim to. They hobbled existing products and over-complicated new products to force people onto a platform no one asked for. Facebook (and others) used this opportunity to grab communication away from Google.

It's all an interesting business case study of missing the forest for the trees.

Sometimes doing nothing is the best strategy but it's hard to get promoted that way.

Totally agree. This is something a lot of people miss.

IT culture in general and Google especially puts too much pressure on engineers to re-invent the wheel to get better visibility and promotions.

When the primary thing you're evaluated for is "impact", it's a wonder that Googlers bother to maintain any existing products at all.

(Plenty do out of love for the product/customers/etc, but man it's soul crushing to not be recognized for it)

it's a wonder that Googlers bother to maintain any existing products at all.

My understanding is that not everybody there knows where the corporate ladders are. If you don't (yet) know the terms of promotion at the GOOG, you'll be happy to knock out bugs and tweaks day in and day out. You're a Googler!

The ladders and breakdown of what is required at each position are well known and talked about frequently, especially around performance review time. If you're a googler and don't know what's on your ladder you aren't paying attention

I don't have any Google experience, but from what I've gathered over the years, there are always unwritten rules. Not everyone follows the spec. Often times, the spec even seems intentionally incomplete in retrospect.

There are almost always unwritten rules; everywhere.

For example everyone says value "work life balance is balanced here and encouraged". But irrespective of whether they actually do it or not - if the two people output around the same ballpark but one is visible as more serious/responsible and seems more invested usually by virtue of more hours at work that becomes more "visible" than the "casual" 7-8 hr person with the same output.

So everybody onboards with the knowledge that launching new stuff is the only way to get promoted? Because that's Google's engineering reputation out here on the street.

Google dominated non-SMS chat in the late 2000s/early 2010s

My social group was dominated by BBM & Skype until iMessage & Facebook Messenger took over. I've always assumed Google's efforts around messaging were because they had a niche product inside Gmail that they wanted to make more consumer friendly, hence the constant attempts to rebrand it.

In India WhatsApp, now, is not a gold standard of online communication but it is pretty much the only mode of communication largely. I don't about other places, but at least here - people are not going to Google new "Chat" or Hangout anytime soon. Even Facebook chat is slowly losing out to WhatsApp and I don't think Facebook minds it. In fact they saw it coming and bought the app for that record price.

The reason I jumped out of Google ecosystem is, as you have mentioned, they forced me into closed ecosystems and I remained there for sometime but then the tools offered to me in the form of app, UI, and interfaces were beyond pathetic so I was out. So were my friends, family, relatives, and other acquaintances.

Orkut > Buzz > Wave > Chat > etc they did it one after another and they are still doing it. I mean you would think they would learn at some point.

Dont worry, its the Americans who live in a bubble.

I live in one of the wealthiest and advanced countries in the world. Ask a 12 year old girl what SMS is and she will not know. She will know WhatsApp.

Interesting! I never looked at it that way...

Google messaging must have some of the worst project managers in the industry. Not only did they let their lead disappear, when they decided to start competing with Facebook and WhatsApp they released Allo! A crappy MVP in a crowded market hugely dependent on the network effect.

And now they've decided to involve the carriers, using a technology that's not made to be cross platform but instead be dependent on phones. I can see RCS being a good solution for companies to send out rich boarding passes and similarly, but nobody around here uses SMS or MMS for person to person contact. Chat shouldn't be locked to or even routed through phones, and using a telephone number as an ID is impractical.

Using a telephone number is impractical? That's our identity in the third world. I think they are focused on the developing world in this.

Some of the people here are in a bubble. There are billions of people in the 3rd world countries who are now coming online and they will never have a GMail account in their life. But they will have a Facebook account and they will have Whatsapp. So many of my relatives started using internet after Jio made it so cheap and none of them use Hangouts, Allo, GMail to communicate. But everyone is on Whatsapp and everyone has a Facebook profile.

Disc: Googler, but this knowledge is my own as I grew up in India.

None of that disproves his point. Google Talk could have been WhatsApp, very easily. They were in a position to make it happen before WhatsApp had even started development - Google Talk had enjoyed a meteoric rise and dominance over the messaging market for 4 years before WhatsApp came out, and it took WhatsApp quite a bit of time past 2009 to gain the steam it did.

The fact that they failed to capitalize on it is what has allowed WhatsApp and others to become as big as they have. If Google hadn't messed it up, everyone would be using Google Talk, even in the 3rd world.

Exactly. GMail did not have to be the only way to access Talk. It could have allowed parallel identity schemes easily for phone-number based login.

You live in a bubble, but your bubble is different from their bubble. Thank you for being kind enough to share what you leaned growing up. It helps us all see past our own bubbles.

I am from a rural part of India and everybody has a Jio phone and everybody is on WhatsApp and Facebook. Most of them don't even know what is email and I don't know anyone who uses email or any other chat service other maybe Hike in rare cases. I was surprised since they already have Facebook and it would have required an email.

I often get disapproving looks when they realise I am not on Facebook :)

Users can sign up on Facebook with phone number only. Most users in rural parts are signing up on Facebook through phone no. only.

What if the user is on a WiFi-only tablet, or laptop, or desktop PC? What if the user is at an Internet cafe? What if the phone user has no cellular signal but has WiFi access?

Clearly it would be a mistake to limit this to just phone users.

I'm from China where WeChat dominates everything and SMS are nothing but to receive verification codes. Almost everyone here use WeChat as their contact and Contacts.app is merely opened no more than once a month. This phenomenon is common from giant cities, like Beijing and Shanghai, which are at least as developed as NYC or SF, to small cities in the rest part of the vast country, which are absolutely 3rd or even 4th world.

I agree with your point that in some countries people don't use Email and that's the face in China as well. But people use phone numbers just for register other services, as the online identity. We don't use SMS, and we prefer WeChat Audio over phone calls when it's available.

I think you really need to check how people use Internet in India, Indonesia and other Asian/African countries. I doubt anyone uses laptop or desktop PC unless it's needed for work. And Internet cafe is a thing of the past.

Having wifi but no cellular signal is such a rare usecase.

What I'm saying is that telephone number as ID is fine, but non-telephone number as ID should be supported, too.

They should be trying to include everyone, telephone and non-telephone users alike.

I am from the same place and what what_ever is trying to say, I think, is that in these places mobile number is pretty much the identity and username and emails are largely irrelevant for the majority and when I say that majority that's like overwhelming majority and they are coming online for the first and they really don't have time or any interest to learn about email and messaging protocols.

For example, we recently launched UPI payments - that's like a banking address sort of thing (you can send from it and request too) and you usually get your UPI address like this "9876543210@upi". The number part is simply the 10 digit mobile number and that's it.

But those things don't sell here. Maybe businesses and professionals buy it but not regular folk. All of them are buying smartphones in millions and going online everyday. Cellular signal is the norm all over the country and we have around 80% of it as villages. So I'm pretty sure all of those use cases come at a distant second. Please read about the impact of Jio in India.

> Please read about the impact of Jio in India.

I read this piece [1] which centers on Jios impact on the telecom industry. It states that the rate of digitalization has been boosted, but doesn't really detail the societal impact. Do you recommend any particular article?

[1]: https://www.ft.com/content/3f1fe4d6-e4e0-11e7-a685-5634466a6...

I don't live in the third world, and I don't think any new messaging service should focus on that limited use case. I'm pretty sure Google focused on India with Allo, but what's the point of focusing on one market when you're a global internet giant? I mainly use Hangouts for messaging, because I use the "desktop client" (Chrome extension) more than the mobile app.

It would be possible to let people use both phone number and e-mail, that way I can log in on computers or my phone with a temporary SIM card while I'm traveling, and people in the developing world doesn't need to get an e-mail account.

Please understand that the size of the population which uses this limited use case in my country is greater than multiple western countries' populations combined. I've been a developer for nearly 7 years and make a good amount of money in my country. Yet I've never touched an airplane. I'm telling this to point out the difference in living styles between the two zones of earth. All tech companies are rushing to capture market in the developing countries because we are poor now but in 20 years we may well be up there along with you all.

Wow. That's surprising. I am from north India - based in Bangalore - and I fly almost every other month.

Are you from a remote/small town or like tier 2-3 city? Freelancing maybe? As for the questions - it just got me curious.

250kms from Bangalore. Road is like 4 hrs journey and I live in HSR. So considering the journey to airport and everything road is cheaper and takes almost the same time.

Also flights are cheaper only nowadays. It used to be unreasonable.

Aah.. makes sense. And indeed flights were ridiculously unreasonable even 1.5 yrs ago or so.

Because GSM is everywhere whereas IP connectivity is spotty?

To be fair, Allo only came after Google had killed several other chat implementations because for some reason they couldn’t tolerate the idea that GChat was successful.

> but nobody around here uses SMS or MMS for person to person contact

Where are you located?

Mac/iOS Messages and/or SMS is my primary way of talking with friends and family. That seems pretty normal for my age range 25-45 and location (Seattle).

I think it depends a lot on the country you are in. In many European countries, the SMS is practically dead: in Germany or Switzerland it's all WhatsApp these days, with a bit of Telegram or Signal mixed in. Given this, I was surprised to discover how many people are still sending SMS here in Canada.

I think it has a lot do with mobile carrier pricing: here in Canada, mobile fees are enormously expensive, but SMS are almost always included for free. In Germany/Switzerland, many people use prepaid cards where SMS are (still) charged per message.

In Portugal all carriers have had free sms between all carriera for a few years now on almost all plans. I believe people still reasonably use it. Mobile data cannot be assumed to be always on as it is too expensive most of the time and never forget that having mobile data on drains battery too fast. If anyone wants to be sure to that someone replies to something in urgent matter better use standard sms/phone calls. I havr no idea in what kind of bubbles people are mostly nowadays. Anyway, almost everyone has whatsup or telgram too and most people also use those as a main way to communicate and mostly to be in chat groups. I'm the weird one using only sms and signal and very little of facebook messenger.

Head on over to Europe, and:

> Ik steur je wel even een appje

(I'll send you an message on WhatsApp)

is the norm in The Netherlands. My entire family, extended family, and everyone around them is all on WhatsApp. While I communicate with iMessage/SMS with friends in the US, that's not the case outside of the US. Everyone seems to have WhatsApp, most people don't use Google hangouts or other services alongside that, let alone SMS.

i am curious, where does it say "whatsapp" in dutch? you don't use international names?

"appje" is the diminuitive form of "app", which is a shortened form of "whatsapp" (referring to a whatsapp message).

In English, you would never shorten the phrase "I'll send you a whatsapp" to "I'll send you an app" because the word "app" would be confused with a mobile app.

The Dutch still use the word "app" to mean a mobile app; "appje" means a whatsapp. So the diminuitive form allows them to disambiguate.

The -je suffix in Dutch is commonly used and means 'small'. 'app' is used here as an alias for WhatsApp.

I'm also Dutch but I think 'app' is more like an alias for any chat service. WhatsApp is indeed huge though. Texting is basically dead in Europe afaict.

WhatsApp is huge in Brazil, even businesses now offer a WhatsApp number for contact. Maybe that's a by-product of Apple devices being too expensive here.

SMS is mainly used for bank alerts, confirmation codes, advertisement and scammers. I personally haven't used SMS for conversation in years.

Yep - what's weird is gtalk used to work pretty well even in business context. Then they went the crazy forced google+ route (I actually liked google+, but they just kept jamming it everywhere - even youtube seemed to want to force a linked google+ profile - it was unusable) and started blowing up how chat worked over and over. Where these things linked? Google talk / hangouts seems to have started going down with the whole crazy google+ mess.

Over to slack we go (which works fine).

I like imessage etc - little to no spam. I'll be curious how this carrier controlled approach will handle spammers.

I wonder how often they get internal suggestions to kill google+ which get vetoed out of brand pride.

You can’t kill that which is already dead.

" Thou canst not kill that which doth not live. But you can blast it into chunky kibbles. " -- Quake Manual, describing Zombies.

One of the best features of using Google Fi as my cell provider is that I can still use Google Talk. Still works amazingly well and I can send text messages directly from Gmail or from my android phone. I love it.

Also on Fi - I use hangouts for everything. On desktop hangouts has my SMS and Fi phone number available for voice calls. On the run, hangouts is on my phone for SMS, and I can use my phone as normal for calling people. It's far and away the best experience I've had with a Google product outside of search/gmail. They really need to look at what the Fi folks are doing, because it is truly excellent.

I can do this and I don’t have project fi. I’ve got a google voice number from a long time ago, and I can text from that number via Hangouts in gmail or on mobile.

Yeah, same. Hell, as a kid, I'd use GVoice really often just to send texts when I had wifi but no cell signal, or had a device without cell service (like a tablet). Saved me a couple times when I was younger and, say, ran out of toilet paper.

I was a Fi user for about a year, but had major issues getting the phone to switch back to the cell network when I left my house. We live in a little valley with poor reception, so the phone seemed to "give up" on cell service, and then I'd have to reboot it after I left my house to get it to reconnect (wifi calling worked fine while I was home).

pushbullet (plus a few other apps) also recreate this functionality. Buggy and it's $5/month but I really use the hell out of it.

> Everyone was on Gmail all day

Not everyone is on Gmail. There a untold millions of users who aren't interested in creating a Gmail account. Those users will never be reached with any "just use your Gmail account" strategy.

It's a pretty simple explanation. Google talk was XMPP which is just not a good mobile protocol.

Could not they just replace the XMPP protocol with whatever better protocol, while keeping the GTalk alive.

They kinda did exactly that. GTalk became Hangouts which is no longer quite XMPP.

Well, not only did they change the protocol, they ruined the UI too.

It's kinda sad how they ruined a superb product and replaced with something some par!

X in XMPP stands for eXtensible. They could simply augment XMPP to match their needs better - and guess what, it's exactly what happened in last years. Now XMPP is a good, modern, battery- and UX-friendly IM protocol, with some clients even E2E encrypted by default.

I had Google Talk on my phone for ages, I'm not getting what you mean by this. I had no issues. It was fast, to the point, and worked.

What makes you say it's not a good mobile protocol? It had some problems in the early 2000s on mobile, but these days it has stream management similar to TCPs and XML compresses very well. As I mentioned in another comment, there are people running it over very low bandwidth HF radio, if it works well there it works well over phones, but I'd be curious to see why people still think otherwise. (Google Talk, naturally, never did implement any of the parts that make it good on mobile, just lilke they refused to do TLS on server to server connections, so note that I'm not arguing that Google Talk was good on mobile, just that if they'd worked on it a bit it could have been)

What service is using it at scale these days? I can't really think of a popular user. After a certain point it just becomes a headache.

When I was working with ejabberd clusters servicing mobile clients it worked but I've had a lot less pain just using a custom protocol over websockets. My personal experience also seems to be mirrored throughout the industry as all the big players eventually move away from xmpp.

Most ironically, Google is using it for their Cloud Messaging solution (https://developers.google.com/cloud-messaging/ccs and https://firebase.google.com/docs/cloud-messaging/server#choo... for the replacement) to allow applications to talk with a backend.. The advantages over HTTP are multiple: better battery life, bidirectional communication, asynchronicity.

Riot Games is also using it for the messaging part of League of Legends (https://engineering.riotgames.com/news/chat-service-architec...)

XMPP is also used by military organisations such as NATO (https://www.isode.com/solutions/military-xmpp.html)

The sad truth about XMPP is that it just works, but it has a very bad publicity because there's just no marketing around it, and it's never used directly by users. So it suffers from that wrong image that is just completely outdated.

You'll notice that the google links mention that the xmpp api is deprecated. I'm not sure what Riot uses these days but the xmpp connection is no longer available to third parties.

That sad truth is that people who actually use it end up dropping it. Its not some conspiracy. It just wasn't cutting it.

It's not XMPP that's deprecated, it's the whole CCS (first link) that's being replaced with Firebase (second link). The XMPP api is still present in the replacement.

While not fully consistent the XMPP api is referred to as legacy on much of FCM documentation. https://firebase.google.com/docs/cloud-messaging/server

You also misunderstand what this api is referring to. These are APIs for your server to connect to Google's. This is not a direct http and xmpp connection to mobile devices. That's all abstracted away from this.

All big players who have an interest in building a walled garden - yes, that sums up "all big players".

GTalk was awesome while it was federated.

I don't think lack of technical expertise to fix whatever needs to be fixed is what caused Google's demise in the messaging domain.

By "not a good" you mean an open protocol not controlled by Google?

No, by "not a good" he probably means that it's not well suited for low battery consumption expected from a mobile device.

Don't push your words on other people :(

The battery argument is just misinformed. The real reason Google dropped XMPP was not a technical one. Yes, Google claimed it was not "cloud ready" at that GoogleIO. The same one where they announced Google Cloud Messaging, based on, wait for it, XMPP.

IIRC WeChat started from the ground up with a custom protocol designed for compatibility/efficiency with low bandwidth mobile networks for maximum reach, IMHO a great example of thinking outside the box of "let's use XMPP since it works".

Just curious, who else is using that protocol?

Don't think anyone is, I thought there was a hacker news story on it but I could not find it with search, sorry.

Maybe it wasn't in 2006, but it wasn't that hard to make it a good mobile protocol and now it certainly is.

i'd take gtalk back right away - the best message solution there ever was.

Google talk was brilliant and it was a sad day when they shuttered it.

>downhill train wreck

The downhill train wreck is not isolated to their messaging services. Google, as a whole, is a fragmented mess that has manifested with Android OS. Not only is Android OS crippled by Google's ADD-fueled incompetence, it's crippled by their inability to remove carrier-influence from their devices. Now, Google wants to double down on their boneheaded decision by doing the same with messaging. Google Messaging needs to be carrier independant, have SMS fallback, integrate with all Google services, and be bundled with Google Services on Android. I have lost /ALL/ faith in Google.

RCS is an improvement to SMS but it does not address the core limitations: it is owned and controlled by carriers, so your ability to contact someone is entirely dependent on that user's carrier. It is tied to a single, physical device. WhatsApp shot to popularity because it was the opposite of SMS: an entirely over-the-top service that everyone could use, on any platform, without the consent, extra billing or buggy implementation of their carrier. As an app, it could be installed on the widest imaginable array of devices - including the obsolete, crappy ones that most of the people on Earth use - without OS updates.

I don't need stickers or Google Assistant. Those things are nice, but they're not what I need. What I need is the ability to contact anyone, anywhere without having to wonder whether it's going to work when I hit 'send'. Hi Bob, did you get my photo? Sorry, it didn't go through, Bob's carrier in Bangladesh has set limits on photo size and mine was too big. I tried to send Raj a message, but he recently got an iPhone and my RCS message sent as SMS and his carrier blocks international SMS messages.

WhatsApp & FB Messenger (and to a lesser extent iMessage) are successful because they reduce the thousands of carriers and hundreds of countries in the world to a single, seamless service. RCS is taking us back to the days where a single obstructionist carrier (remember lots of carriers are owned or controlled by governments) can screw things up for all of their users. WhatsApp et al showcased the ability of the Internet to make the world a smaller place.

I don't Google understands how big or complex the world really is.

Frankly, I'd rather have a messaging system controlled by carriers but distributed among all of them (like SMS and RCS) than a messaging system fully controlled by a single company like Whatsapp, Telegram, Allo, iMessages and so on.

I can still switch carriers and be able to communicate with my friends.

The only issue I have with SMS is that it costs money when I'm not in my home country.

> Frankly, I'd rather have a messaging system controlled by carriers but distributed among all of them (like SMS and RCS) than a messaging system fully controlled by a single company like Whatsapp, Telegram, Allo, iMessages and so on.

I think that's true in general, but I don't mind Signal because it's a nonprofit and pushing the technology forward.

Re. being able to switch carriers and still communicate, what happens when you travel and (temporarily) get a local SIM? Are you suddenly messaging all your existing contacts from an unrecognized foreign number "Hi, it's John, this is my temporary Thai number..." and then relying on that message not getting binned by some spam algorithm? What prevents carriers from charging extra for RCS messages originating abroad?

> what happens when you travel and (temporarily) get a local SIM?

The real fix would be sensible roaming charges so you don't need to.

That would seem to be less likely than getting every carrier to implement RCS.

It happened in the US long ago for the most part. I often use a tower that is run by someone other than my carrier, but everything just works. This has been the case so long that I can't recall when the last time anyone I know had to think about roaming.

More recently (since 2015) I get free text and data (but not voice) in more than 100 countries around the world. Some countries include voice as well. In short it is coming, if you don't have it ask your carrier why. (odds are it is regulations so get after your representatives and tell them to stop it or you will vote for someone who will)

In France we get to keep our phone number when we switch operators.

Travel is an issue yes, but I think it's worth it to have a truly decentralized system.

At a conceptual level, it does have many advantages. For one thing, it's just part of your phone, you never have to install anything. You don't have to market it to anyone, it just automagically rolls out through the world as people buy new devices and carriers adopt it.

But I'm concerned that the intransigence of carriers, especially those that operate in non-competitive environments, will get in the way of rolling it out properly and effectively.

RCS should be backwards compatible, so if the carrier is blocking it, the users simply see a normal SMS message.

Isn't it the same with Whatsapp, Telegram, Signal, etc.? Those are all tied to your phone number.

With Telegram you can change the phone number and keep all the messages and contacts. It's also possible with Whatsapp, but with some limitations.

Why are you changing your number just because you switch carriers?

Because in my example you're traveling to a different country.

The problem is that carriers are terrible at this kind of thing. For example, here in Canada, Rogers, Fido and Bell support RCS.

Rogers/Fido use Google's Jibe infrastructure, but Bell uses Samsung infrastructure.

The result is that if you're on Fido, you can't send text messages using RCS to your contacts on Bell because there's no interoperability between their infrastructure and Jibe. How insane is that?

Whatsapp relies on a phone number for account activation though, doesn't it?

And if you lose that phone number and phone, you need to get a new whatsapp number. I have to keep my SIM from New Zealand with me so that when I get a new phone I can activate Whatsapp again.

WhatsApp uses your phone number as your "ID" but your message history and contacts can be migrated to a new device or number. So there is a concept of an account beyond just your current phone number but it's not exposed to the user directly.

You can migrate your Whatsapp account between different phone numbers. I've done this with SIMs from different countries.

WhatsApp, nothing more than another internet messaging app that requires data, shot to popularity because it fooled users into thinking it was somehow an SMS client because it used phone number as account.

No, it shot to popularity because it functioned exactly like SMS messaging, except that messages were free. "Fooled users" is an incredibly mean-spirited way of looking at it - iMessage certainly "fools users" into thinking it is SMS far more. Users care about the result, not the method used.


So basically it works, it was $1/year, it allowed to text, send images and send videos and one only needed to know someone's existing identifier (phone number) to connect to them.

Who would have thought that making something that works would eat messaging?!

Here in Italy, it shot to popularity because it was "free unlimited SMS for everybody".

Also "can send images & shit for the same price".

No, because it was (and probably still is) the only messaging app that works with feature phones.

> No, because it was (and probably still is) the only messaging app that works with feature phones.

Sadly, Facebook has since dropped feature phone support to support their flanking maneuver on Snapchat (Instagram on the other flank). Instead of keeping WhatsApp simple and admitting it was feature-complete, the grand plan seems to be "copy features from Snap, and drop feature phones that can't keep up".

> without having to wonder whether it's going to work when I hit 'send'. Hi Bob, did you get my photo? Sorry, it didn't go through, Bob's carrier in Bangladesh has set limits on photo size and mine was too big.

Pretty sure you'll know prior to sending whether Bob is able to receive the message by just seeing the message is being sent by RCS.

The carriers are idiots. It's where innovation goes to die.

We had to pull them kicking and screaming into the smart phone and data era. (Remember When AT&T sent a paper copy of every image you downloaded on your iPhone in a bill?)

Quite frankly the core problem with Google's strategy is involving then at all. Instead, go the manufacturers and campaign against supporting sms at all. It's insecure, slow, unreliable, monitored, among many other things. Push the manufacturers to an open Android messaging api that runs over TLS (or double ratchet) and bypasses the carriers completely, but the manufacturers can still rebrand.

The other issue is Apple. For as much as they claim "privacy is a human right", they have been jerks about making sure iMessage drags it's feet using secure technologies to communicate outside of their ecosystem. Apple wants to ride a white horse on encryption, but in reality they're having an opposite, negative, effect on the vast majority of the world's population.

> unreliable

Really? It's the only messaging format that works without a data connection and is an absolute godsend when I'm out camping, rural biking, or at large sporting events and the towers are absolutely swamped.

I'm in a situation at least once a month where the only reliable communication method is SMS.

Unreliable in the sense, that it is on the discretion of the network, whether it will drop your message somewhere in transit, without letting you know. Just like UDP.

Is it a practical issue where you are? In France I can't remember the last time I've had issues with undelivered text messages and I use them quite a lot (I generally use the cheapest plan I can find which usually doesn't contain data, so I default to texts).

For important messages I'll expect an explicit confirmation anyway, it's not enough that the message reaches the recipient's phone, they also have to read it!

My main problem with texts is lack of privacy which is why I'm slowly pivoting towards Signal instead.

If you've had a long running group chat over MMS, ask to see everyone else's phones. Messages will be in a different order on every phone and there will be many missing messages.

My wife and I do this regularly because we won't understand something said and we'll see one of us didn't get a message from the group.

It's 2018 and I learn that there's MMS group chat o_O

-- a guy who used SMS and then had 90% of friends being reachable via IRC and email, and SMS for the rest - before entering the world of smartphones.

SMS supports delivery notifications, I think I've been using them for almost, if not 20 years

I get a little display under the message saying 'delivered' in the standard android SMS app

Sometimes carriers will send a "delivered" even though it only made it as far as a cell tower (and not an actual phone)

That's another message that may or may not be lost, if it is checked at all.

I've got my lesson about SMS reliability few years ago: I've bought a bus ticket via SMS, it was never delivered, but I got charged for it and the operator claimed that it was sent, when disputed. Except I've never received it, and couldn't ride the bus without it.

> That's another message that may or may not be lost, if it is checked at all.

that's a general problem with acknowledgements and is not specific to SMS

TCP is exactly the same

But tcp detects this and recovers or explicitly kills the connection.

Dropped SMS messages, particularly between Android and and iPhones, have caused some catastrophes at my work. I no longer trust them enough to use them to communicate with staff.

Isn't that usually because iMessage doesn't actually send SMS if it thinks it can send something else?

Wow - infrastructure varies. Here in Eastern Oregon the pattern is pretty reliably that data and signal drop out at the same time, at least on Verizon which is the carrier with the best coverage.

I really thought the pattern that you describe, which I've seen plenty of in the past, had just died out over time and tower upgrades.

I routinely see SMSs fail to deliver, deliver multiple times, deliver weirdly out of order, or deliver at random times, hours or days after they were sent. And this is in a relatively well-served area fifty miles out of Boston.

I still have the issue where MMS seems to just sometimes not work for days or even weeks on end, then suddenly start working again. No OS updates and no APN settings changes in the meantime. Normally it's not an issue, since I don't really use SMS all that much anymore, but my mother and I tend to send pictures back and forth a lot and she still uses SMS.

SMS itself usually works fine, but MMS is particularly bad.

Same with Eastern WA and CA, you basically just drop behind a big hill/mountain or into a valley and lose all service. I still sometimes use texts since I can turn off data to save power and still send SMS.

It all depends if there's a 2G network to fall back on if 3G+ data goes down. Most places have shut down their 2G.

>Most places have shut down their 2G.

Citation? I haven't seen this to be the case in California - at least on Verizon.

There was a big story about it when ATT shut down their 2G network in SF since MUNI used that to report their locations.


Only AT&T shut down 2G. T-Mobile reiterated recently that their 2G network will probably be shut down after their 3G one, since it hardly uses any spectrum.

With places that have shutdown 2G, there is no "voice only" network anymore since 3G is required for the voice traffic.

T-Mobile and AT&T have very different planning. AT&T shut down their 2G because it mostly ran on 850mhz and this was needed for LTE and their legacy band 5 WCDMA (hspa) network.

T-Mobile mostly ran its 2G on 1900mhz, but had beefy midband spectrum holdings there and in AWS (and AWS-3) for most markets. Since you can run GSM down in like 200khz or so it's not a huge deal to have a small slice allocated to legacy voice fallback. You can't do that with 3G... you can thin it down but it's still going to need 3.8mhz or so. (T-Mobile has experience thinning HSPA down prior to the MetroPCS CDMA shutdown... which they had legacy CDMA running in guard bands)

Verizon and Sprint used to mostly be on CDMA-based technology. So their 3G 'EVDO' can't do calls.. it needs cdma2k layer to go back to, or full voLTE support. VZW has its shit together on voLTE, sprint will probably never launch volte at this rate.

Sorry was thinking up in Canada only Rogers maintains 2G still.

SMS is a best-effort protocol; that is inherently unreliable.

Business class SMS is quite reliable.

> Quite frankly the core problem with Google's strategy is involving then at all. Instead, go the manufacturers and campaign against supporting sms at all. It's insecure, slow, unreliable, monitored, among many other things. Push the manufacturers to an open Android messaging api that runs over TLS (or double ratchet) and bypasses the carriers completely, but the manufacturers can still rebrand.

But it's universal. I would never buy a phone that doesn't support sms. How do I contact people with flip phones? People with iPhones?

Also, Gchat is "monitored" as much as sms. Imessage claims it's not, but has that/can that be verified?

Its both universal _and_ relies on the voice network rather than the data network (if the device even has one). SMS is a fundamental part of the GSM spec, has been for over 30 years, and is one of those super lightweight battle-tested technologies that there are certainly some pretty great IP-based alternatives to, but which replacing wholesale for some yet-another-IP-based-rebranded-messanger seems crazy. The beauty of SMS riding on the voice network is that it gives you redundancy. Phone doesn't have a data signal, but can get voice? SMS still works.

Until we live in a world where GSM and friends are dead and everything everywhere is VoLTE, SMS still has a place as a fallback.

I'm always a little surprised to come into HN comment threads about SMS and see so many people rallying for its replacement, yet we go into a thread about Slack and see many of the same arguments being made to defend IRC, etc.

IRC is more modern in a sense. It usually works over an IP network.

iMessage is end-to-end encryption.

How do you know that? Have you seen sources and verified that those sources are actually compiled and shipped with the final product?

Unless you write your own compiler, look at every line of the source by hand well enough to ensure there no "underhanded C code contest" backdoors, compile the application yourself, load it onto your phone, and do the same for your friends and share keys with them in person, you don't have this kind of assurance for any communication app.

You have to draw a line somewhere and accept some level of trust proportionate to your threat model. iMessage is good enough for most people who aren't Edward Snowden.

That’s a good point. Apple attests that it’s end to end and people have tried to track and fail and leaks don’t reveal vulnerabilities.

By comparison, no other majors attest that (eg, Google explicitly says it’s not end to end) and leaks have shown data accessed in clear text by government actors.

Look. Its simple.

Signal work on Android just like iMessage - you set it as SMS app, and it uses Signal messages with people who have Signal (even on iPhone) and SMS otherwise.

We already have iMessage-style app on Android. It's just that pretty much nobody is using it.

Because no carrier installs Signal as the default messaging app.

> The carriers are idiots.

For decades, ISPs/cell networks' biggest fear is becoming "dumb pipes". But they all seem to land there eventually.

Yes.... lack of vision is what seems to define an ISP. Creating content, while 'sexy', is fleeting and incredibly expensive. Delivering the content reliably and quickly is a predictable busy model. ISPs should be clamoring to advertise how they deliver Netflix the fastest, or offer equivalent to Cloudflare's service but for their customers. Instead, we get an XFINITY VOICE REMOTE O.M.G. and a platform thats update once every 5 years.

This is caricature, but based in reality: https://xkcd.com/1810/ - SMS, like email, is ubiquitous. Unlike email, it doesn't even require a data connection, just a cellphone signal. (Yes, this is a thing, even in supposedly civilized countries) Unlike email, it doesn't even require a feature phone - even the dumbest of the featurephones can read SMS, from the venerable Nokia 3210 up (yes, even e-mail can be too complex for Grandma).

And the other thing - asking Google to stick to a messaging API? Google, which has launched-and-killed more mutually incompatible messaging platforms than I've even managed to use? Not gonna happen. "Use Talk! Use Hangouts! Use Allo! Use TwoCansAndAString!"

The point is not to kill SMS, just to make the default an IP-based protocol, that's not tied to the carrier/operator.

Currently people are dependent on phone numbers for too many things. (Too many "2FA" uses SMS.)

Agreed, SMS needs to stop being supported by handset vendors so people stops relying on it as the universal text platform.

When a communication system silently drops messages, this is unacceptable. But people are ok with it because they're accustomed to it being shit. The sender wonders why you never responded, and you are unaware they contacted you. Or the sender asks if you want to meet for lunch, and you get it 8 hours later. Group messages? I've compared phones on a table and we all have a different context to the conversation because each phone has different set of missing messages.

The messaging platforms are in silos and it sucks, everyone is annoyed about having 3-4 chat apps. But at least they work reliably, and that's the better alternative to garbage SMS.

> When a communication system silently drops messages, this is unacceptable. But people are ok with it because they're accustomed to it being shit.

Or they are OK with it because they have never had their SMS messages dropped, like me (that I know of).

At least here (Finland) SMS is used for a lot of stuff like bank transfer verification and postal package pickup notifications, having many dropped messages would seemingly make it unsuitable for those purposes.

Of course, MMS is a completely different story and I've seen issues left and right on various phones and carriers. I don't think anybody really even uses MMS group chats here.

SMS is important. It's universal, and it can be used as a way to communicate with anyone. I have friends that I chat with exclusively over SMS. They use Facebook Messenger, and I won't use Facebook anymore. They could join me on Signal or something, but I'm not going to ask them to download a whole app just to chat with me.

It's the one messaging platform that's not in a silo, and that's just as important (maybe more so?) than 100% reliability.

> Remember When AT&T sent a paper copy of every image you downloaded on your iPhone in a bill?

No, no I don't - that has to be apocryphal.

>Remember When AT&T sent a paper copy of every image you downloaded on your iPhone in a bill?

I'd never heard of that, fascinating. https://en.wikipedia.org/wiki/300-page_iPhone_bill

Its shocking how Google is embarrassing itself with these incompatible chat apps released one after another and abandoned right after the release. GTalk Gmail Chat Google Voice Hangouts Android Messages Allo Duo for fucks sake Google, just copy Whatsapp but make it stored on cloud, not on the device, like FB Messenger. Enable account recovery using email in addition to sms and enable usernames or email in addition phone number to add contacts. Thats pretty much all you need to do to beat the competition thanks to your android influence

That's what you get when you apply reinforcement learning company-wide and set rewards +500k +promotion for completely new, "revolutionary" products ;-)

"and set rewards +500k +promotion for completely new, "revolutionary" products ;-)"

Do you have a source for this? Not disputing you, but would like to reference.

Thank you. That's a clear accounting of perverse incentives in action at Google.

"I adopted a new strategy. Before starting any task, I asked myself whether it would help my case for promotion. If the answer was no, I didn’t do it."

I think Google gave it a pretty good shot at competing with WhatsApp with Allo. The problem is that at the end of the day, the network effect is what matters with messaging. No matter how good the product, it doesn't matter if everyone I know doesn't use it as well. It's not an engineering question, it's a social one; a lot has to change to get people to switch from iMessage, WhatsApp and Facebook Messenger today.

The theory says that you can get people to switch if you deliver a product that is x10 better than the competition, but really, how can you possibly improve messaging as it is today? It's not like people are going to go "Wow, this app sends my messages so much better than <current established app>, I'll go ahead and switch and tell everyone I know".

Networking effects matter, but only when after you abandon a common protocol in favor of walled gardens.

Is it that hard?, you force every android phone maker to install it as the default SMS client, and if someone else is using it you use allo network instead, and if the user's paying for sms say "sent with allo for free!"

If only allo actually supported SMS, which is the weirdest thing of it all...

GTalk, Gmail Chat, and Hangouts are all the same thing, and Google Voice works seamlessly through Hangouts. Android Messages is just an SMS app, it's no different from iMessage with the "iMessage" feature turned off. They were almost there until they released Allo and Duo instead of just refining their existing product.

Google has a culture where rewards and power go to the people who release products, not managers who maintain business lines. It means that they create great products but often don't support them well.

that + a desktop client/web app is all anyone wishes for, how can they be do blind and stupid.

It never ceases to amaze me how blatantly incompetent Google is when it comes to software, blind and stupid is not hyperbole, the proper thing to do here is absolutely crystal clear, users are begging for it (I know, I've googled trying to find a desktop solution that works and doesn't cost $5/month ffs), and they just keep releasing pieces of crap over and over. It's hard to believe this is real life sometimes.

You're talking about company, you expect coherent product vision/strategy of a single entity.

Reality is that all these products are result of autonomy given to teams - it's not really pushed top-down, so new 'innovative' ideas are spearheaded and pushed up for strategy's adoption.

Are you actually calling out arguably the most successful web software company of all time (primarily through search and adwords) for being incompetent? A company that profits billions a...quarter?

Maybe chat just doesn't make them any money so they don't really give a shit and all of these things are pet projects that get shut down on a whim?

The key to Google is that their real Users are advertisers. That's it. The "users" you're talking about are the product they're selling to their real Users(Advertisers).

Yes, yes I am. Because most of their software absolutely sucks.

Boom. This is why Slack and Discord expand so much. Usability plus availability. Sad people at Google get paid a lot of money to not figure these simple things out. Know your enemy and then reimplement their work in your own better way. Allo needed a desktop client and the video stuff should of been built in stop trying to copy Apple in that regard.

Hangouts has all that. What's wrong with hangouts? Why should it be replaced with something "newer"? Why does Google HO keep getting worse?

Hangouts doesn't have a desktop app.

PS. Please don't call that chrome window a desktop app.

You forgot Google Meet.

Don’t forget Google Wave.

Everybody forgot Google Wave.

They recently added a chat function to the Youtube mobile app too.

I like the Whatsapp solution. Why does it have to be stored on cloud? Web interface with Whatsapp Web (via phone) works well.

you must have never tried migrating whatsapp from iphone to android, major pain in the axe

So the solution could be to make backups and migration easier

I have and lost all messages (got them exported so can still search for them). Not optimal but I prefer it to FB messenger where it's much harder for me to access my data.

I also think the switch Android<->iOS is solvable, probably just an edge case and they don't think it's profitable to do.

I personally hate that aspect of Whatsapp. Having to scan a QR code to use chat on my desktop/laptop is a huge pain and is a major reason why I stick to Facebook Messenger and Google Hangouts.

Google Voice and Duo are not messaging apps. GTalk, Hangouts, and GMail Chat are the same thing at different iterations and were therefore fully compatible with one another.

If you're going to criticize Google for this, at least be correct when doing it.

You make a good point but also make a broader, negative point against Google. OP's incorrectness is evidence that Google is obfuscating the purpose of their products. There’s a bewildering amount of names and products. At some point fatigue sets in for casual users/observers and everything starts to get lumped in together. I’m not going to sort it out and so in my mind these are all messaging apps.

Apparently Google is trying to do something in the communication space involving voice and text. It doesnt seem to be working and it’s ended up confusing a lot of people.

No, GTalk was XMPP-compatible and federated, Hangouts was siloed.

So non-google XMPP users could speak to GTalk users, but not Hangouts users.

This was frustrating to say the least, if they'd just done a clean break it would have been annoying, but at least we all (I was using a third party client at the time) would have known what was going on, but they more or less silently broke it, but ony some of the time, and then strung it out for ages. I suspect it was a deliberate strategy to try and make the blame fall on the shoulders of third party clients instead of them for cutting it off, but who knows. I wish they'd just shut down the interop cleanly when they moved to hangouts, it would have saved me a lot of pain and debugging early on.

Google Talk was federated and could chat with people who had accounts other Google, so it feels a bit unfair to lump that in.

It's like saying "Gmail is compatible with hypothetical Gmail where you can only send emails to other Gmail users, it's just a different iteration." Maybe true, but suddenly a very different service.

Nope. Google broke GTalk (on purpose?) when Hangouts came out, that's not "fully compatible" in my universe.

Messaging is a core feature of Google Voice.

Long time ago, Google chat supported XMPP. Later on that support has been cut off.

How do they expect people to trust this statement: "We don’t believe in taking the approach that Apple does. We are fundamentally an open ecosystem. We believe in working with partners. We believe in working with our OEMs to be able to deliver a great experience,” ?

It’s funny because they squashed open standards when they were the big player and then push for open standards when they’ve lost market share. I wish companies believed in open standards for the value they bring to society, rather than selectively supporting them based on competitive interests.

I think there is a pattern with Google:

Social network: Orkut was their one and only working social network. Google - it is not great, so we raze it to the ground and make a better product... and they were trying to do so for last 12 years.

Messenger: Talk was their one and only working messenger. Google - it is not great, so we raze it to the ground and make a better product... and they were trying to do so for last 12 years.

Photo blog: Picasa was their one and only working photo blog. Google - it is not great, so we raze it to the ground and make a better product... and they were trying to do so for last 12 years.

New site: Reader was their one and only working news portal. Google - it is not great, so we raze it to the ground and make a better product... and they were trying to do so for last 12 years.

Casual blog: Blogger was their one and only working casual blogging site. Google - it is not great, so we raze it to the ground and make a better product... and they were trying to do so for last 12 years.


Google Photos is such a faceless product to the extend you get surprised time after time again to the fact of its existence when you see it popping up under different names after firmware updates.

Picasa was a photoblog, something with essence, community, and vaguely stated purpose. Photos, on the other hand... well, just a file dumpster with UI.

Orkut was great ... but not exploitable in terms of commerce. It lost almost all its market share in the industrially developed West, but was huge in BRIC countries. Not the target demographic Google was looking to exploit at the time. And thus it got killed.

Google is primarily a marketing company. They need to make money off of their products. In that context, all the products you reference, which I agree were pretty great, were not useful to Google. So they got killed, or allowed to languish, in the hope that they would at least drive new users into the Google ecosystem.

I do have to disagree with part of your assessment of Google Photos. You are 100% accurate ... it's a file dumpster with a UI that I routinely forget about. And that's perfect. I do not need to have a relationship with my software. But it quietly functions in the background, and periodically rewards me with an animation or AI-generated image? Hey, that's pretty cool. I'll share that with my family and friends. And that's what Google wants to see.

Well, to be precise, orkut was big in the B of BRIC...

Some of your examples aren't as convincing as others. Picasa was mostly renamed and migrated to Google Photos - which is a fairly good product. (The desktop app was a casualty in the process however)

sometimes question are ask is web code old, insecurity? does it scale? power profile, can we support it?

look at flash was massive but it was insecure and had to be replaced.

from my understanding talk had to be change due to mobile network usage and battery drain.

sometime it isn't black and white what kills a app and most of these products are free so it isn't like google owes you anything.

i think they could communicate this better with people be more open on why a product is put to rest or replaced

That's the old MS playstyle, which I expect MS themselves got from even older companies.

It's almost as if they don't understand that much of the value of iOS as a platform comes from Apple's consistent willingness to tell carriers to go pound sand.

I would guess that they do know that, but they think for whatever reason that they can't get away with doing the same thing. (Or maybe they'd have to sacrifice something else that they care about - if they started setting higher standards and got 90% of carriers on board, but the other 10% stopped selling Android devices, would that be worth it for them?)

I'm not sure if they're right, but I am pretty confident there are tradeoffs here.

>We don’t believe in taking the approach that Apple does.

I bet the person who wrote that is naive enough to believe it due to lack of historical knowledge. Probably 95% of the current Google employees were hired after XMPP support was dead.

Probably the person who wrote it is in marketing, gets handed down a list of talking points, and has no idea about the technical background of the decision.

These are all weasel words: "open ecosystem", "partners", "great experience"

They are signals that mean "closed" and "invasive"

Because people want party parrots. And there isn't a standard to send party parrots through XMPP. /s

Okay, a little snarky, but it makes a point. Chat programs have a lot of proprietary features that won't translate through XMPP.

Most of the features these programs have translate fine; you won't have a good experience unless you're using their client or another client with a similar feature set (though things to tend to degrade gracefully), but that's fine for the most part. It still gives you a nice realtime API for developing other stuff on top of it, most people will use your clients anyways, and if you're on a feature phone or something and can't use the first-party clients, you can still connect. It's a win-win.

> Chat programs have a lot of proprietary features that won't translate through XMPP.

Most chat networks are still built around xmpp - they’re just not federated and likely using private extensions.

Pretty sure XMPP included extensions for sending arbitrary HTML and binary attachments. Now, support for janky mobile connections, that's another matter.

XMPPs actually been really good at this for quite some time; despite XML being a bit verbose, it compresses very well and it's had TCP like stream management at the application protocol layer for ages, among other things. It was pretty bad at low-bandwidth connections in the early 2000s, but now there are people running it over extremely low bandwidth high frequency radio, so a janky phone connection is no problem.

XMPP had extensions just for that crap. The trick was getting clients that all supported them.

That's exactly my point. Thank you for making it for me.

> How do they expect people to trust this statement

Because they're using an open, industry-created standard. It certainly isn't in their interests to do so (unless Apple also adopts it, which... we'll see)

Right. Embracing the open standards - like they did previously (e.g. with GTalk), and then suddenly..."oh look, we've improved the standard, and we're not publishing our extended version. It's so much greater, but only works with this one client, made by us - what a coincidence."

Gee, I wonder where I saw this pattern before...nope, it's not like they're going to exterminate their Official Blessed Client once they tire of it, 18 months from now.

Surely with that attitude they're damned if they do and damned if they don't? If they make their own standard it's bad, but if they adopt an industry standard they're probably going to ruin it, so that's also bad?

If they've embraced an industry standard previously, extended it and exterminated, multiple times before, surely it will be different this time.

No end to end encryption = no interest.

If I use this to communicate with a business (or person!), the carrier can read every word. No thanks.

Hell. You can't even download the standards documents without giving GSMA a name and address and consenting to receive promotional material...

Oh. Thank you for pointing that out. I thought that Allo supports E2E so will chat, but that was my sloppy reading of The Verge article [1]:

> But, like SMS, Chat won’t be end-to-end encrypted, and it will follow the same legal intercept standards. In other words: it won’t be as secure as iMessage or Signal.


[1] https://www.theverge.com/2018/4/19/17252486/google-android-m...

I am starting to believe that these "new chat projects from Google" are actually all the same one with a different name to keep, whatever work they are doing, alive.

I heard some time ago from the colleague who got an internship at Google that they have to release a new project almost every 3 months and then fight against other internal teams to do a public release. This way they — mostly interns — can justify a full-time contract. At the same time, because these projects also need a project manager, some high profile engineers help the interns in order to get bonuses and also to justify a change of responsibilities to, maybe, become full time engineering managers.

Whatever the story behind the chat re-brands is, many will agree with me that this is becoming irritating.

> Unlike Apple, RCS is designed to work with carriers who can develop their own messaging apps that work with the protocol

No no no please. We should be doing everything we can to relegate carriers to “dumb” data providers. They don’t have he skills or expertise to provide anything beyond this, which they’ve demonstrated time and again.

I don't see a single reason to invest myself into Google messaging product once again. I was adamant gtalk/hangout user but you can only fuck me over so many times.

You can take “messaging” out of that first sentence.

Bing and Apple Maps here we come?

I don't have an iPhone, but my understanding was that users had strongly adopted Apple Maps after the initial stumbles.


I've had occasional issues with Apple Maps, sure.

I have no clue if Google maps have the same problems, because what I would have to give up to use it doesn't make it worth it, no matter how much better it may hypothetically be, and it wouldn't be worth it, even if Apple Maps were as bad as the Google maps users all claim it to be.

Also, not bing. DDG, of course.

more like Maps.me or Osmand, i find first superior to anything on android

I've actually just quietly gone on using hangouts and ignored whatever strategy statements they make. It's not perfect but it still works and I see no reason to stop.

...which is exactly why the service will get shut down. "We made something, but people wanted to use that! Kill it, kill it with fire!!!"

Yeah, me too. It is definitely better than all of their other choices.

To be fair, the migration from Talk to Hangouts was pretty smooth, and Hangouts still works.

Google gave up pushing Hangouts for consumers but I don't think it's going anywhere, as it's a big part of G-suite for corporate customers. Though with Slack and Discord I'm sure it's usage isn't as strong as it once was.

> the migration from Talk to Hangouts was pretty smooth

I take it you never used google talk via xmpp.

I did actually. They kept the XMPP compat running for quite a while after, so for me Pidgin and my old blackberry GTalk app kept working.

Eventually they shut it down though, but in an uncharacteristic move they gave quite a bit of lead time.

They gimped the hangouts extension with the great chats that would slide up on the screen and could be individually rearranged though and replaced it with the new useless "app". Now I either have to load inbox/gmail or hangouts.google.com, which loads slowly for something such as messaging

> the migration from Talk to Hangouts was pretty smooth

So smooth that I lost access to hundreds of contacts that used a different, non-Google XMPP server.

I was furious when they dropped desktop application support for Talk.

Hated hangouts, finally switched to Telegram and will never look back.

For me, nothing demonstrates how self defeating Google's internal structure is, than chat. They had a lot of adoption of their system in the early 2000's (and granted it became unmaintainable but that was an engineering issue not a technical issue) and they killed it. They had a great system they could have copied (iMessage) and they ignored it. They had a standards friendly system and they spurned it.

The combination of killing gchat and reader really struck me as "Wow, it feels this company has no idea what they are doing strategically."

They don't get it. People don't want new apps, they want improved versions of the app they already have. It takes so much time to adopt and adapt to an app... nobody I know uses Allo. I have a google phone and I don't use it, because no one else uses. Allo has so many interesting features but I can't use them if no one else uses the damn app.

I don't have an iphone nor want one, but they got it right when it comes to building upon their core software. Same can be said to Windows/Microsoft. Yes, they are not perfect, yes sometimes behind everyone else when it comes to "the latest and greatest", but they have USERS! They know their audience and slowly improve their apps to keep them happy enough to keep using it.

I can't believe a company this big with so many resources doesn't have an attention spam greater than 6 months.

By that definition, they do get it. Because RCS is an improved version of the Messages inbox already in the phone. No need to take any time to adopt and adapt to any new program.

Great example of how expensive it becomes over time to name (almost) every product you create "Brand" + x.

Google is eroding trust in its product offering & regularly adding an adoption handicap to their whole product portfolio.

P&G, Unilever et al know this. I'd expect the software companies to learn from the challenges of the "supermarket"-era brands.

The tradeoff:

- Add "Google" before the name any product announcement, and it's sure to be front page news[1].

- Shutdown any product with "Google" before the name, and it's sure to be front page news.

[1]You can especially see this effect when a random Googler publishes a random project, but it's under github.com/google/ because it was done at work. Even if it's not "a Google product", it'll shoot up to the top of HN.

RCS is shit. It’s a good idea in theory, but in practice it’s a crappy, over-engineered clone of iMessage or WhatsApp without any of the good parts like end to end encryption or spam protections.

It relies on the carriers to keep it functional and we all know this won’t end well... given the robocall and SMS spam situation do you really think RCS will be any different? But at least robocalls and SMS spam cant carry anything besides plain text, while RCS could contain HTML and a lot more “active” stuff which no doubt will exploit some bugs in the phone’s implementation to install malware.

This is absolutely true.

And as has been said in other threads, Google Talk was the only sane strategy for Google and it was lost.

Sometimes it's hard to believe that Google's course of action and decision making to messaging isn't satire. I can't imagine what a dumpster fire must be going on somewhere within, eventually I hope we get to hear the stories behind their calamity of multifarious messaging products.

I'm not even being facetious when I say that they should include an estimated date of when they expect to cease supporting said product (2 years seems to be about right)

If I was to summarize the problem of Google is this: Google does not hire entrepreneurs.

They're great at hiring engineers. They're great at hiring MBAs from top business schools. But their hiring process does not leave room for unusual rouge entrepreneurial people.

I mentioned this to a person whose company was bought by FB and later became an investor. He agreed and he said, he was interviewed by google but they didn't hire him. So he started his company and a few years later sold it to FB. I know this in anecdotal. But I think Google just lacks the entrepreneurial blood. Someone to hustle a product with a cohesive strategy and vision and finding the way to build it up and turn it into a massive success.

I think the issue is they want too hard to be seen as entrepreneurs. They are always reinventing the same products rather than just improving what they have.

I'd go so far as to say, Google would benefit from a two-year ban on any new product releases. Engineers should spend that time focused on adding value to their current products.

Not only do they keep cancelling chat projects but this one isn’t even end to end encrypted. What’s the point. Either release a superior product or just let it go.

Can anyone explain why I should be interested in RCS, though? It seems like it's more designed to let carriers try and stem the flow of customers switching to other options, and I don't really see what benefits it offers me.

Years ago 1998ish(?), there was a fight to open up chat applications AIM, ICQ, YIM, etc. We had to have 20 apps to chat with each other.

Fast forward 20 years, and here we are again, FB Messenger, Kik, WhatsApp, Allo, Signal, Telegram, SMS, Slack, etc.

At least this appears to be a "standard" that everyone can use again.

The thing that actually worked was a multi-protocol messaging client, like Pidgin.

Which I still use all the time, because most of the chat clients I have to develop against have introduced asinine restrictions that force their official clients to run in single-instance mode, so your other alternative is spinning up entire VMs in order to have two accounts signed in to talk back and forth on the same machine.

Why would any of those products get onboard, unless they’re already failing?

They wouldn't. And so I have to have 20 apps to chat with all my friends. And that makes me a sad panda.

I think I'm going to start a 21st app that implements a new protocol that unifies the other protocols...


But it doesn't really seem to be a standard that everyone can use. Like others have pointed out, FB and Google both originally built their products on XMPP, and only later decided to lock out competing options.

With SMS I know that I can send a text to anyone. With RCS, I don't know what would be supported by each carrier.

> With SMS I know that I can send a text to anyone.

Unless their phone is turned off, or in a different room, or they’re in a different country, or they’re outside cell coverage and have wifi only.

I’m not arguing googles shit is good, but sms is a fucking woeful alternative.

I send most “messages” from my laptop, and about 1/3 are “international”.

I suppose if their phone is turned off outside cell coverage long enough then yes, they won't get it, but the rest of it isn't really an issue I've had to deal with. And with carriers determining RCS functionality, I'm not sure I see how having your laptop but not your phone is likely to make a real difference.

My point was to compare to something not controlled by people who for their own financial gain, push the idea that a phone is the only way to communicate.

Xmpp, iMessage, etc are just data over an ip network - no inherit tie to a mobile phone and its network.

So if I'm out of cell coverage, or my phone is flat, or whatever - I can get iMessage/XMPP/whatever messages on my laptop or iPad or whatever other device I have.

That is how having a laptop but not my phone makes a real difference.

I've been using jmp.chat recently for this (SMS sent over XMPP and phone calls via SIP). I actually liked it so much I ported my old number and got a data only plan, now I can launch Dino or something and get SMS on my desktop. I really wish one of the big carriers would embrace something like this.

no, you didn't, there was Miranda and later Trillian and Imo.im combining all these messengers in one app

I remember around 2012 it was also great time, I could chat with my Skype, Facebook and Gtalk contacts from within Outlook.com website thanks to XMPP, the best part was it acted like proxy so i could reach the people even though these services were blocked in China, then Goggle and Facebook decided to kill it and I stopped using messaging in outlook.com

https://xkcd.com/927/ Relevant xkcd.

I'm a bit tired of people pulling out this comic each time a new standard appears.

Whatsapp, Telegram, FB, iMessages are NOT standards. They're closed systems controlled by a single company.

XMPP is a standard from the days of ICQ, AIM, MSN, etc. Unfortunately it failed.

SMS is a standard for mobile phone text communications. It's too limited now.

RCS is the evolution of SMS.

I don't think RCS is a redundant standard, it's a necessary evolution of SMS. It's a standard for a decentralized system. I know Americans hate their carriers to the point they'd rather have everyone's data in a single SV company, but the solution is to fix your carriers. Not to give all your data to a company (Facebook) that had been proved to be less than trustable about it.

Xmpp didn’t fail - the major companies who all built their messaging platforms on it just turned off federation.

RCS is still focussed on phones only, so it’s a non starter IMO. But it’s driven by the same companies that charge the equivalent of $1000+/mb for sms delivery, so it’s unsurprising that customer needs are not really their first priority.

RCS is a standard on which development started in 2008 and was built upon the core of another standard called IMS, which started development in 1999. The same year the jabber.org project was announced.

> Whatsapp, Telegram, FB, iMessages are NOT standards. They're closed systems controlled by a single company.

That's what the term "Defacto Standard" is for.

Not really. A de-facto standard is when multiple implementations from multiple parties agree on some behavior without it being described in a formal specification. Whatsapp does not have multiple implementations from multiple parties.

Not really; the proliferation has been due to proprietary protocols, not due to efforts to create standard unifying ones.

To be honest, in the fine print in of that XKCD it says "See: ..., instant messaging".

The closest thing to a standards-based, non-walled-garden option that has the possibility of widespread adoption?

Like XMPP that Google abandoned? The solution is hardly to create a new standard, but improve in existing, perhaps rewrite them to address fundamental issues. But if it's meant to be used collaboratively, it should be developed collaboratively.

XMPP is already that, and Google Talk supported it until they found it inconvenient/got enough userbase and removed the gateway & support.

Only being deprecated 1.5 years later?

Applications are open for YC Winter 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact