Hacker News new | comments | show | ask | jobs | submit login
How did Google Talk change from a dream to a nightmare? (tnhh.net)
803 points by calcifer 42 days ago | hide | past | web | favorite | 376 comments

Google Talk was only ok when they had a jabber service. It wasn't much better than the other options besides that it tied delayed messaging (email) to instant messaging (jabber) through the same contact info.

Then they killed off the jabber client to force users further into their software ecosystem. Then Hangouts came along and has to some degree replaced Talk with further ties into that ecosystem.

Then they removed browser support for anything but Chrome in Hangouts, initially promising to re-extend support, yet today the docs have changed to just say you must install Chrome.

The most recent eco-system tie-in that this article doesn't mention is Google Hangouts Meet. The corporate conference room version of Hangouts that has a different interface, custom Google hardware, and less memorable links (/xcf-fges-sce vs. /organization/meeting-name).

Pixel has similar limitations when it comes to screen mirroring, only a Chromecast will do for the Pixel! Yet other android phones freely connect to Chromecast/Roku/Firestick/etc all because of a hidden menu setting that Google has disabled and set to 0 by default. You must root your phone to get around it. Great flagship right?

All this points to a company that's so fucking worried that their tech is going to be outpaced by the little guy that they have to resort to handcuffing users to their wares. I've almost completely switched to Slack in the mean time.

Even if you didn't use Jabber, the #1 reason I recommended Google Talk was that it was very, very lightweight at a time when MSN Messenger and friends were adding stickers and other teenager/young-adult friendly* features.

Today, ironically your choices for relatively straightforward messaging on the PC/Mac are: iMessage (Mac only), Skype for Business (the consumer client is too distracting for words), Whatsapp Web, or go with a "heavyweight" website/app like Slack (which is painful if all you want is IM and none of Slack's extra features).

I have to wonder what the product managers were thinking.


* this is not used pejoratively, I recognize people use IM tools in different contexts

There are others of course. You can go Telegram or Signal as a great cross platform chat. Not geared directly towards business, but still a great alternative.

Chat apps are the posterchild of what happens when open standards (irc, jabber, xmpp,...) get replaced with walled gardens.

On my phone I run: Telegram, Messenger, Viber, whatsapp, hangouts, signal, slack and SMS.

I should have installed skype as well, but its the worst chat app ever that kills your battery instantly. There are others as well but enough is enough.

How did we get here? I regularly need to think where should I message someone or where is a specific chat group....

Imagine you had to do this for every email service provider.

> Chat apps are the posterchild of what happens when open standards (irc, jabber, xmpp,...) get replaced with walled gardens.

To me it feels like the open standards never really kept up with the times, which in a way paved the way to walled gardens by the giants (and, in case of Slack, not so giants) to fill the gap which the open standards refused to fill.

I mean, sure IRC is awesome, text-only, channels, but emojis are limited to ASCII, inline GIFs are non-existent, and file-sharing (and/or storage) is rather clunky from what I remember from the old times.

And then we have XMPP which doesn't support things like video or voice calling (or at least it didn't last time I checked) and it's not very surprising that companies just capitalize on that and make their own platforms.

Don't take me wrong though... it creeps me out how the company where I work trusts Slack with reams and reams of confidential documents, but I guess it's convenient to an extent which no other platform is.

> And then we have XMPP which doesn't support things like video or voice calling

The Jingle extension is about a decade old and it's what Google based its support on. They even played ball for a while and released a good quality open libjingle, then decided that fighting spam from the other open XMPP networks is not worth their time and killed the whole openness concept - despite being an orders of magnitude simpler problem than email spam.

I think now the only escape from the walled gardens is regulatory pressure forcing interconnect of the larger players (Facebook, Google, Yahoo etc.) at least at text chat level.

XMPP is a bit complex, and suffers from some problems derived from too much flexibility. The mandatory part of the protocol is too small. This could be fixed by creating a meta XEP that lists all XEPs needed by modern clients.

However, it's a very capable protocol. Just see how nice conversations.im is. It doesn't even use GCM, and both latency and energy usage are fantastic.

Such a meta XEP exists already: https://xmpp.org/extensions/xep-0387.html#im

If you use a client that conforms to the Advanced Client requirements of the IM and Mobile Compliance Suites (with a similarly up-to-par server), you will have a very good experience.

The problem is that selection is still limited if you factor in OMEMO support (end-to-end encryption):


Recommendations for good clients for desktop systems? Maybe my google-fu is particularly weak today, but I couldn't find a clear statements which clients outside conversations.im fullfill those (obviously only the IM suite, since mobile wouldn't apply)

https://gajim.org/ tends to be best for desktop (binaries exist for most desktop OSes).

For macOS, you may want to try Adium or Psi. Or you can use https://movim.eu/ which, though a web client, has fairly good feature support as well.

I second Gajim especially 1.0.0 version that looks good. There is also https://dino.im but that's still in alpha.

thanks you two, will give Gajim a try!

Remember to install some essential plug-ins (OMEMO, HTTP Upload, URL image preview, etc.) for the best experience :)

> I think now the only escape from the walled gardens is regulatory pressure forcing interconnect of the larger players

Sounds good on paper, sure, but ultimately it will end up being a predictable bureaucratic mess, causing more harm than good:

a) The mandatory 'open' standard that gets produced will end up being designed by committees of management teams from 6-7 major companies, each with their own list of feature requests and no central 'vision'.

b) A lack of a cohesive strategy and (critically) a lack of real incentives by the members to partake will result in endless delays, slow moving technological progress, layers of old cruft that never gets removed, and toxic political infighting causing confusion among vendors.

c) The standard ends up being so complex and involved that it isolates other small/medium sized players (or large foreign players) from joining in, eliminating the 'openness' the original regulation envisioned and crippling competition.

d) Ultimately reduces the ability for developers to get paid via monetization and grow via capital investments in the US, as non-regulated open-source projects (or foreign private apps) gain a major advantage of not having to be forced to use the standard. Cannibalizing the market the big players spent so much time/money building.

It's not just about good intentions and spotting a tough problem, it's about whether they can realistically and effectively achieve the end goals...

TLDR: open-source and the global nature of technology, lack of incentives, design by committee, regulatory agencies staffed by the very same companies it's regulating, etc, etc will result in the crippling of innovation and harm the quality of chat apps in the US.

I don't expect US to move a finger on the issue - all major players are american. However, something like USB charging was imposed by the EU with great success, ending a massive source of e-waste and proprietary cruft manufacturers imposed onto consumers. For EU, the economic motivations to force american companies open to European competition would be very tempting and the consumer benefits significant.

The takeaway from the USB success is not to design a new protocol by committee, rather pick a mature open standard - for textual chats there are several mature ones.

You mean like what happened to the web ? I'm a bit tired that every time someone suggests a new regulation/standard on HN, someone feel smandated to explain why regulations are bad and why neo liberalism is awesome.

Could we assume that we all understand the downside of regulations ? but that we still suggest some when the market fall into a bad optimum (bad for the consumer). If you don't agree that the market is stuck in a bad place, fine you can argue that.

cornholio brings up a really good point. The fact that Whatsapp and other messaging apps have significant more European marketshare and growth rates than in other parts of the world could make an EU regulation's effect felt around the world, much like the USB one.

And this is what happened back in the day. AOL was forced to add interoperability with competitors. But our government today is unwilling to step in front of corporations on behalf of the people.

HN readers don't all have the same government. I assume you mean the US one. The EU has in some cases proved to be willing to confront corporations and/or to defend some consumers right. Not always or in the most efficient way, but here's hoping that it will continue and improve on that path.


I'm well aware, and really happy about the work the EU's competition commission is doing. Unfortunately, for US-based corporations, global change is unlikely to happen until the US steps in.

What would be the advantage of adding tighter voice/video support to XMPP, vs just using links to sip: URIs? A major point of open standards is that you can use use them together with other open standards.

> Chat apps are the posterchild of what happens when open standards (irc, jabber, xmpp,...) get replaced with walled gardens

Slack is showing that the usefulness and popularity of a chat app has exactly 0% to do with whether it uses an open or closed protocol.

It's all about product design and user interface.

And the failing of the walled gardens is that you can't choose the product design and user interface you want and use it to talk to everyone.

Slack provides IRC relay, which initially was almost feature complete, but got out of date with time.

Anyway, in some time Slack will probably fade away as quickly as it appeared, further contributing to the problem grandparent noticed and you seem to misunderstand.

And the more i read about Slack the less i want to touch it.

Emoji, gifs, fuck off.

Signal and WhatsApp require a smartphone though, so not fully cross platform, and tied to their (closed) client software. Not sure if Telegram requires a smartphone too, but it does seem to require a phone number at least.

I had signed up for Telegram on a phone, so can't say whether you can sign up w/o a phone (I think it would require SMS verification).

It doesn't need you to be logged in on any phone however. You can use just the desktop app. In fact I am on some really crowded and hyperactive Telegram groups so can't even imagine keeping it on my phone. Besides not a single one of my IRL friends use Telegram. I see 3-4 names, maybe they had signed up once.

A smartphone isnt required, only a telephone number for authentification.

> A smartphone isnt required

It is for Whatsapp. Not only for sign-up, but actually all communication on the web version goes through the app running on your phone. The web "session" times out constantly, so I'd have to re-pair it with my phone all the time. If I still decided to use whatsapp.

To be really pedantic; a smartphone is not required for WhatsApp. You can activate a Google voice number with a landline, then use your Google voice number to activate WhatsApp running in BlueStacks or your emulator of choice.

Source: I did this for a few months. I'm not quite sure why...

A smartphone OS is required for whatsapp

> [A]ctually all communication on the web version goes through the app running on your phone.

How on earth does that work? Say my phone's on data, my laptop's on my home wifi with NAT. How does Whatsapp on the web reach my phone?

Your phone connects to WhatsApp's servers. So does your laptop. The bridging is likely done server-side. I assume this is done because WhatsApp is (probably) using end-to-end encryption, so everything must be ran by your phone (which is the only place where your private key is stored) in order to encrypt the messages.

It works impressively well, though -- I use it far more than I do the mobile client directly.

The problem is that you need to own that same phone number permanently. Phone numbers are tied to exactly one SIM card, which is tied to exactly one telco in one country. Too many ties.

Threema doesn't require a phone number but it's not free.

Depends on a country, in Sweden you can move your mobile number to any telco, and anyone can look up your number and your home address using something like [0]. But yes, it is still tied to a country.

- [0] https://hitta.se

Yes, what I meant to say is that the SIM card is tied to exactly one telco at any particular time. It's also tied to one device at a time.

Perhaps one of the virtual SIM card offerings would work with Signal though.

I'm juggling too many SIM cards and phones already. I don't want any new dependencies that complicate matters further.

You only need the sim card active to activate your signal with that number. After that you don't have to be connected to a cell network at all.

But I still need to keep the SIM card active or I could never switch phones (or reinstall the phone OS) without losing my Signal identity.

I cannot ever let go of the phone number I used to activate Signal. In practice, that means I need to keep paying for a phone tariff I may no longer want.

True. I was assuming most people have at least one phone number they want to keep long term (even if it's not always active with their cell). Note though that signal will also work fine with non-traditional services that provide free or much cheaper phone numbers like google voice (US), skype, or twilio.

It is true though that you do need to maintain a phone number. That is one of the trade-offs they made to allow it to actually be usable. I have not seen a better solution for encrypted communication yet.

How does the phone number tie-in make it usable? Phone numbers and SIM cards are about the most user hostile thing I can possibly think of.

What's wrong with the way Threema does it for instance, which is essentially to bind the identity to a key pair that you can back up?

Threema appears to optionally do the exact same thing. Signal eschews the flexibility to ensure more universal usability by being able to assume that anyone who uses signal can be associated with their phone number. An assumption that holds for the vast vast majority of potential users.

I don't see how making phone numbers mandatory makes Signal easier to use at all. There must be another reason for this restriction. Some say it's for spam protection.

Signal will work with Google Voice, but as far as I can tell you still need a smartphone to use it. Without that, you can't even set up an account because you need to scan an OCR code first.

No, you can change the number associated with your Telegram account.

I was talking about Signal. Sorry I didn't make that clear. The phone number is your identity. Changing it means reregistering with a different number.

[Edit] And I think it's essentially the same with WhatsApp, only they now make it a less manual process.

Ah, I should have mentioned that I was talking about Telegram in my original comment, too. Can't edit it anymore :/

In these cases you could use a https://jmp.chat/ phone number.

No smartphone is required for JMP, only an XMPP client (see https://news.ycombinator.com/item?id=16352711 for some client options).

A smartphone isn't required for Telegram either.

Incorrect. Whatsapp refuses to open on my iPad and isn't even available in the App Store. The only reason it got on there was a glitch in Apple's app sync which must've auto-downloaded it since I have it on my iPhone.

Sorry, I was talking about Telegram. Should have made that more clear.

You do need one to set it up, right? I had to scan a QR code displayed on my desktop with my phone. (I could continue to use the desktop app while my phone was being repaired, though, which was great.)

No, you can do the complete setup with the Telegram desktop app, but you need a device which can receive SMS or a phone call.

Oh hmm, looks like I got tangled in the threads - I thought we were discussing Signal. My bad!

Ah I see ;) Yeah the QR code scanning sounded like Signal / WhatsApp, never heard of that method in a Telegram app.

Wire doesn't. It uses the same protocol as Signal, has mobile, desktop and web apps and the source is on Github.

I'm using Signal on my linux desktop right now. It appears to be a packaged webpage but it works fine and is linked to my phone and laptop.

All it require is a phone number, besides it's Snowden approved :)

You're using signal in a browser on your linux desktop. Poke me when there's any way to connect via bitlbee or libpurple, like there is for ICQ, jabber, etc.

They lose all credibility when they force you to reveal your phone number. Doesn't matter who endorses it.

They are entirely honest about their trade-offs. While those trade-offs may be deal-breakers for you, it should bolster their credibility.

I'd just like to add that one of the key things I love about the current crypto craze is the fact that it gives us something that never existed before - the capability to monetize protocols. I know that most ICOs will fail, but through this process we might get something that solves this shitty situation with chat applications.

Perhaps try Skype lite? It has almost zero bloat, and has been working flawlessly for all my needs.

"Introducing Skype Lite Built for India - Chat and share with friends."

it appears to exist only for Android phones.

It's nothing but a joke that you'd have to switch to a "lite" version that is built explicitly for India [because of bad infrastructure] just to make it work in a country with much better infrastructure.

It's like buying a low-end car built for Indian market, because your regular car doesn't fit on the narrow roads in the US, the fuel consumption is too high, it leaks oil constantly and there is no service because nobody cares.

This is a worthless analogy, because some software is paired down and is "built for india" it sucks?

A paired down debloated skype sounds like a godsend to me.

...that's not what they were saying at all. The insults were being heaped on the 'normal' skype.

And skype lite is definitely lacking in configuration, like the ability to make emoticons less annoying.

(I've also switched to skype lite for the remaining two chats I'm in that use skype, now that the slightly-older android version stopped connecting and the newest interface is amazingly painful and laggy.)

It that like the Facebook Lite app that's geofenced to emerging countries? How bloated an app has to be to make a company with the size of M$ to create an alternative client?

I (relatively) happily use the Facebook Lite app in Ireland where I got it from Play Store. I'm not sure we're emerging from anything.

There're thousands of companies in Russia using Telegram as main communication tool.

> I have to wonder what the product managers were thinking.

That billions of people were using WeChat and WhatsApp, and they wanted a piece of that? I'm guessing that's the motivation.

As for the enterprise use case, I'm not really sure why nobody cares. Every company I've worked for has had their own internal system for IM-like functionality, some better than others. Microsoft has one included with Outlook whose name escapes me. At Google we used Hangouts, and despite everyone complaining about it, it mostly worked well enough. You typed a message in it and at some high percentage probability, the other person got it. It was fine.

My complaint has always been intrinsic to the medium, it lets people bug you Right Not for very low cost. "Hi, I see you're currently triple-booked with meetings, but I'm bored and I want you to chat with me and I'm much too lazy to think about my problem long enough to type a one-paragraph email and then wait for you to reply when you have free time." No.

For personal stuff, my group of friends uses Discord these days. It doesn't alleviate any complaints that you might have about other services, though. It is IRC-like and has voice/video chat. It has a native app, but it's whatever that framework is that calls bundling 1 kilobyte of HTML with three hundred gigabytes of a Chrome fork a "native app".

I also think you'd get better customer service from your local DMV than Discord:


>Microsoft has one included with Outlook whose name escapes me.

Mircrosoft's current stab at this is called Teams. I think it actually has potential: it's pretty much a clone of Slack, except that MS is willing to declare it compliant with, e.g., HIPAA.

Teams is a nightmare. Buggy, slow UI (I suspect that they use their browser internally), annoying. They started to worsen Skype too, but it is stil heaven in comparison. I suspect that they make bad UX on purpose. No way to release such a bad IM client by accident.

Teams is a cross between Slack and Facebook. It has channels, but also a threaded-messages structure that is just painful. Everybody in the company I work for wants to like it, but they struggle. Slack would be so much better, but then we’d lose Sharepoint and AD integration and we’d have Yet Another Silo We Have To Pay For.

I found it horrible, clients are Electron monsters (but still bo Linux version..) that they shove in your face whenever you try to use the web version.

So far I can stick with Lync/Skype for Business at my day job, which is bad too but.. slimmer, and does the job. Most of the time.

I'm positive that they run the whole damn thing in a VM and electron together :). Its the very epitome of slow.

> My complaint has always been intrinsic to the medium, it lets people bug you Right Not for very low cost. "Hi, I see you're currently triple-booked with meetings, but I'm bored and I want you to chat with me and I'm much too lazy to think about my problem long enough to type a one-paragraph email and then wait for you to reply when you have free time." No.

Still easier to ignore on a chat than if it entered the inbox or if they call you in the middle of a meeting?

I'm working on a lightweight native client for all major messengers:


It's only 90 KB (!)

Right now there's an old barely usable alpha out, but a new release is coming up next week.

Not native, not 90KB. The first thing it does is downloading an embedded web browser. 90KB the installer. Not native either. libpurple is native.

v0.26 used to do that. Not v0.27. And yes, v0.27 is a 90KB native app. It will be released on Feb 14.

Pretty impressive. Any chance of tying it to IRSSI or making an IRC gateway like bitlbee? I connect to almost all those services through bitlbee already and it works pretty well. Add Naver Line support as well and I'm sold..

IRC is supported, so you can use bitlbee with eul.

Line messenger is not supported, but eul is modular, and everyone will be able to add support of custom messengers in the future as long as they have an open API (which Line seems to have).

Sounds impressive. Unfortunate timing that the new version is not out yet, but if it needs final touches then it does ;)

Did you consider open-sourcing it? More people would probably be willing to give it a try and support you on patreon if it was.

Sounds too good to be true! I'll check that.

Looks promising, keep working on it!


Oh man, Slack really is a beast isn't it? I do like the multiple network nature of it (even though everything is saved to Slack's servers) but I find it hard to believe it does so little for it's footprint.

They too though made a Google-like transition and went from an open protocol (IRC/XMPP) to a closed/custom API. At least here, the server admin can enable those protocols if they want and the API seems to offer enough features that someone could probably integrate Slack into another IRC-like client.

You may explore Mattermost [1] if hosting on your own servers is a requirement. It is very easy to deploy and I found it very satisfactory, albeit testing with a relatively small team. I imagine scaling up would work. It ties in nicely with GitLab, which is another neat product for those who must not or wish not outsource information.

[1] https://about.mattermost.com/

Matrix [1], mentioned elsewhere in these comments, looks interesting too and the Riot [2] app is similarly styled to Slack.

[1] https://matrix.org/

[2] https://riot.im/app/#/room/#matrix:matrix.org

Matrix is basically open source Slack. It fixes a lot of the problems that IRC had, enabling continuous presence, central authentication, easy file transfer, bot integrations, and more. I really hope it catches on more.

Disclaimer: I help run a small Matrix network.

Yeah, the GitLab omnibus package comes with MatterMost bundled. It's a really nice solution for self-hosted chat.

> They too though made a Google-like transition and went from an open protocol (IRC/XMPP) to a closed/custom API. At least here, the server admin can enable those protocols if they want and the API seems to offer enough features that someone could probably integrate Slack into another IRC-like client.

As for the XMPP gateway it's just barely usable. From my personal experience it even tended to drop messages from time to time. It seems Slack added it just to claim XMPP compatibility. Just look at where is Slack placed in this ranking https://conversations.im/compliance/

Apparently they were built to mimic IRC but don't actually use it under the hood. But like you mentioned, they DO have some integration with IRC.

WhatsApp web (or the desktop) is really frustrating at times (though that's not completely WhatsApp's fault). Half the time I want to use it, it is not connected and I have to unlock my phone and open the app and wait for some time when the WhatsApp is connected again and the desktop app is connected too. Because WhatsApp web/desktop uses the phone app as its server.

I have the same issue with the web app (using an iPhone). What I figured out is, that you don't need to unlock the screen to make the web app reconnect. All you need to do is switch on the phone (without unlocking), so just give the home button a little tap and 2-3s later the web app is reconnected. Not a great solution, but well...

You are right. That sometimes works. But what I've noticed, maybe it's some issue with my set up - not sure, that often the WhatsApp desktop/web app doesn't connect for 1-2 minutes even after I have unlocked (or woken) my phone - so I just click on WhatsApp app on phone anyway and let it reconnect. Android and iOS both are pretty strict on background tasks these days. I hope battery tech catches up with the speed at which apps' resource appetite is growing.

Rest assured, Googlers mostly saw the same thing, and asked the same question "what are product managers thinking?".

They were probably thinking "How many features do I need to claim responsibility for in order to get that next salary / stock option boost?"

By default feature creep tends to win out over less visible improvements: supporting the latter requires active effort from management.

Oh, I was mostly thinking about bystanding Googlers not working on hangouts.

Your description is reasonable for people working on a project. (But even inside hangouts there was dissent. I don't know too much about the specifics, and probably wouldn't be allowed to tell, if I did..)

Discord, is great, works well with groups up to 40,000 and can work as a simple IM messenger, though you still have everything from the group message features as in Slack.

Try to go back and delete old conversations.... Discord really falls down there.

They have decent search at least. Lots of filters you can customize. It feels a little janky to use though.

> Skype for Business

Which has a horrible UX.

Personally, I use trillian to connect to facebook, jabber and others. But FB is always working hard to make that integration as bad as possible.

the relevant XKCD : https://xkcd.com/1810/

I really want Jabber to succeed, but it looks like all of the social providers don't want this and want to keep your communication on their servers.

Also, I really don't understand the appeal of pushing 100s of these chat apps out. I.e. Slack, Discord, allo, whats app, etc. They're not compatable with each other. You have to have a 100 different clients installed. Many which their desktop version requires 2 cpus and 128gb of ram to run each of them.

XMPP is the prime example of too little too late. XMPP missed both the shift to mobile and the shift to more engaging and complex messaging.

There is a nice overview in "The State of Mobile XMPP in 2016"[1] where you can see how far behind XMPP is compared to nearly everything else. By 2016 nearly everyone could work over mobile networks, offline messaging, push notifications, syncing between multiple devices, file uploads, end-to-end encryption etc. XMPP had a plethora of experimental XEPs with unknown support across clients and servers. And not much has changed in the past year.

So let's say you have the only client in existence that supports all these XEPs, Conversations for Android [2]. What can you do with it? Oh, you'd have to find a server that supports all your features and talk to people on the same client. And that's about it. For everybody else you're stuck with plain text messages.

It will only get worse for XMPP. AliChat and WeChat has long been the way to pay for anything in China, with over a trillion dollars flowing through their systems annually. Apple, Facebook, Google, Telegram are busy adding payment capabilities to their platforms. It will be years before a relevant XEP is drafted, and another few years before maybe one client and maybe one server will start supporting it.

[1] https://gultsch.de/xmpp_2016.html [2] https://conversations.im

Yeah and email doesn't have hardly of those features either. It's basically just text or perhaps HTML. Sending files is super inefficient and awkward. Encryption is hardly supported on clients.

None of that changes the fact that email is a useful and reliable way to send a text message to someone. In the same way XMPP is a useful and reliable way to send a text message to someone. If all you have is a lowest common denominator there is no point in dismissing it for something that doesn't exist.

> Yeah and email doesn't have hardly of those features either

That's why:

- there have been countless attempts to improve email (anything from threading, to automatic contact info to inline images to file previews, to adding interactivity to emails, to... to...)

- the younger generation prefers anything but email

Slack has the saving grace of optionally supporting an XMPP gateway, for those still committed to that dream. For now.

(I sincerely hope this comment isn't too prescient.)

They don’t do federation. And it doesn’t work very well.

Indeed, Federation was only a part of their Enterprise offering. :rimshot: It is only Slack-to-Slack, so it is not interoperable in that way. Shared channels are a thing now too, and danged if I can keep up with all the new bells and whistles.

Slack is very much an organizational groupware product, with chat done well. Not the utopia federated XMPP promises, but frankly, XML just needs to die. (Which, not coincidentally, is part of what attracted me to work on the telehash specification.)

And indeed, an IRC gateway.

When they started pushing Duo and whatever the other piece of the associated current Google chat/conference... duopoly, is. Supposedly with some move away from and/or deprecation of Hangouts, not to mention the explicit language and dates WRT killing Talk.

Well, fuck me. I can't be bothered to keep up with their changing product lineup, naming, marketing, whatever TF this is.

I just use non-Google stuff, now, for chat/video.

FB Messenger may be busy selling my soul, but at least it's still "Messenger" and actually fucking works. Which is why pretty much everyone has it installed and knows at least basically how to use it.

If I could just get my associates and friends to start using Signal or the like... (Of course, Signal could improve their UI a bit. NOT more fucking emojis, but instead making it very clear and easy to opt out of making it the default texting app (on Android). That is, not a big bar to opt in, with a little X within that bar to opt out. Tired of explaning that each time, to "normal" people, before they get back to me complaining that their texting doesn't look/work the same, any more.

P.S. As I recall now, that other piece is Allo.

If I even understand things correctly. I couldn't bring myself to read up on this thoroughly; I kept feeling the desire to throttle somebody.

P.P.S. And I remember when, a few years ago, they REALLY wanted everybody to move to Hangouts. Including integrating their SMS activity. Only, this had significant bugs, including aspects of data loss and the irreversibility of changing to their set up. That they seemed to have little real momentum in fixing -- or even communicating clearly upon.

I guess I should actually read the OP. But I'm not sure my blood pressure can take it, now.

Facebook Messenger Web is becoming unusable. Back in 2015 it was great, but slowly, they made it so that the website lags in all browsers if more than 100 messages are on screen, and the scrolling is now wonky.

Ah. I haven't used it on desktop in a couple of years. And even then, I wasn't actually "signing in" to it. It still worked just fine, via Facebook's page(s), so why bother, and this also seemed to avoid their announcing your online status.

On mobile, if you don't know how else to reach out to someone, Messenger's your best chance. Closest thing to a modern-day AIM that I've seen, in terms of ubiquity.

Duo is a mobile only video/voice chat app that uses a experimental transport protocol.

Allo is a text chat app with lots of options for stickers and end-to-end encryption disabled by default.

I don't have a strong need for either of these apps.

Was so excited when they finally added a video call option that wasn't a conceptual travesty so that we could finally have a cross-platform facetime alternative, unfortunately video always cut out and audio was inaudible, and still, despite being phone number based, won't let me call a number if I haven't added them as a contact. Oh well, back to, "let me go grab my iPad so we can facetime."

I tried to like it, I really did.

Well, my Motorola G5+, purchased in August and on 7.0, came with an icon for Duo prominently displayed. Hangouts not installed by default. (Nor Allo, I now see.)

So, Hangouts is not facing some declared or expected EOL, at this point?

It's already marked as "classic" in the help section for Meet.

> If you use hangouts.google.com for video calls, or need help on chat, visit the classic Hangouts Help Center.

I can confirm. New Android phones ship with Duo and no Hangouts.

The reason they "killed off the jabber client to force users further into their software ecosystem" is because all the other messaging products had decided to move away from an open standard (XMPP/jabber) to proprietary protocols, so there wasn't much point in restricting Google's own ability to innovate around their chat product by keeping it tied to XMPP. We can argue whether that was a good idea or not, but it wasn't purely Google being a bad citizen.

Sure there were other and earlier offenders, but that doesn't mean Google had to follow suit. They could have contributed something open and federated - an upgrade to XMPP. Branding would have been easy and the darling of the web would have held that image a little longer.

But they didn't, the closing and interconnecting of Google's product ecosystem was the beginning of a trend which is now seen across most of their offerings:

The Pixel has silly restrictions as I mentioned. Android apps, Drive, Docs and the office suite apps have gone stale since they came out of beta and lack the commitment to open source that Google once championed.

Even their new HTML replacement, AMP, is heavily tied to Google resources, requiring entirely different implementations of the same experiences between the HTML and AMP versions further handcuffing the buyer to their ecosystem and the buyer's customers to Chrome.

Sadly that's also my thoughts about Google. When previously they were open and leading the innovation now they are just blindly copying competition in hope of somehow making it big. Twitter? Google Buzz... Didn't work. Facebook? Google+... Didn't work. Whatsapp? Allo... Didn't work.

Google+... Didn't work.

Actually, I just went over there and Google+ looks like it's working just fine. Many parts of it are thriving and there seems to be a lot of activity in communities. Google+ is niche. Just because it doesn't have the ubiquitous adoption level of Facebook doesn't mean it failed.

I really hate this attitude in our circles that unless an app devours everything else and becomes a unicorn, it's a total failure.

> I really hate this attitude in our circles

Maybe I wasn't clear. My point is that it is Google's attitude right now. If a product is not "a unicorn" they instead rush to something else (Talk->Hangouts->Allo) or kill it (Reader, Wave). Being niche product in Google is just dangerous for that product.

I use hangouts exclusively on Safari. It's actually frequently unusable with Chrome (nobody can hear me).

Meet though is exclusively Chrome. They have now cut Safari out of the picture as well... https://support.google.com/meet/answer/7317473

Wow, not even Firefox support? Hasn't Google heard of something called the open web?

Firefox was first to go. I'm sure they were ecstatic to have an excuse to end support when Firefox changed how their add-ons work.

Didn't they change it so it's much more similar to how Chrome extensions work? Shouldn't it in fact be easier now to have an extension work in both Chrome and Firefox? The Mozilla wiki page on WebExtensions even says "Much of the specifics of the new API are similar to the Blink extension API."

The Chrome extension used should work just fine with Firefox, but Google would rather try and force vendor lock-in. At this point I just use Jitsi, it works every time without any browser extensions to do a quick video conference or screen share.

When it launched in 2013, Hangouts used an NPAPI plugin that worked in most browsers. In 2014, Google ported Hangouts to WebRTC for Chrome only because it relied on non-standard WebRTC features only implemented in Chrome. Note that Firefox did support WebRTC at this time and Mozilla in fact developed its own WebRTC video chat service in 2014 called "Hello" that worked in Firefox and Chrome.

In 2015, Mozilla announced that it would drop support for NPAPI plugins in 2017, giving websites using NPAPI plugins ~18 months to switch to a non-plugin solution. When the NPAPI deadline came, Google announced the Hangouts would stop working in Firefox, they were "actively working to develop a solution" for Firefox, and until then Hangouts users should use another browser:


It's 2018 and Google is still working to support WebRTC in Firefox, but they're definitely working on it. In fact, WebRTC Hangouts does mostly work in Firefox now, but only for users with a G Suites enterprise account whose admin has manually enabled Hangouts. I don't know why.

"Hangouts Meet" is a different product using WebRTC whose support for Firefox is still in development.

I've been documenting this saga on this Mozilla wiki as part of the NPAPI deprecation and Win64 Firefox rollout:


Thanks for the info. I totally forgot Hangouts was still using an NPAPI plugin, and I didn't realize Firefox dropped support for those too. Doesn't explain why Safari wouldn't be supported though, as Safari still supports NPAPI plugins.

I haven't tested Safari, but Hangouts' system requirements say it supports Safari and IE using the NPAPI plugin:


They have, but they don't seem to be in favour of it.

Since October 30 2016, Hangouts is unusable on me on Safari.

I always use the pop-up feature in order to get individual chat windows that are real operating system windows, rather that the javascript windows in Gmail. Since that date, it doesn't work in Safari. It works for a few minutes, then the windows close themselves and disappear.

The only solution was to use Chrome... which worked until very recently. Now I get the same problem with Chrome too, albeit the timeout is much longer, perhaps an hour or so or sometimes longer.

Everything is fucked.

By the way, this is Apple bug 29018740. It might help if someone makes another bug report and references this bug, though I doubt it because when the latest Safari version was released (with the bug still present, of course), I mentioned this to Apple and they responded:

> Thank you for contacting us. If this is still an issue for you on current releases, please file a new bug report.

Thanks Apple.

By the way, this is just for text messaging. For voice/video Hangouts became unusable earlier than that. The video quality is approximately 120p and the audio is 1kbps or less, and I can't talk because noise cancellation doesn't work and I hear myself back with a 2 second delay and 80dB gain. Don't even get me started on CPU utilization yet...

Jabber still kind of works. It's what I use at least; all my Hangouts conversations come through on my Mac through the Messages app.

I am a Firefox user and when I "need" to use hangout I just fire up Vivaldi.

Hangouts runs fine in the most recent version of Firefox.

Slack and Zoom, for my teams.

Instant messaging is an example of something that used to work, and now doesn't. It used to be that everyone was basically on AIM, which functioned as an open enough network. Status information worked pretty well - you could fairly easily tell if someone was online, and so people could respond.

And now - we have dozens of clients, such that there's no way of knowing which subset your contacts use. With the prevalence of phones, it's hard to tell if someone is available or not, so you just send messages in the ether, and get surprised if anyone responds.

And the worst part is, there's no obvious fix, since any new system will just add to the fragmentation.

I don't think that's accurate. There were a bunch of silos in the past, too. People used to be on AIM, MSN Messenger, ICQ, Jabber and IRC, and there were others. That's the reason chat apps like Adium and Trillian supporter multiple accounts. I had friends on all those networks, and nobody was on all of them. It was a mess back then, just as it is now, except now we have even more choices.

The difference was you'd fire up Pidgin and connect to all of those accounts simultaneously, at which point it really didn't matter which individual network any given contact was on.

Now, in the smartphone era, there's nothing like that. There's a bunch of different completely siloed apps, and I can't talk to people all from one place.

Yes for over a decade I was able used Gaim/Pidgin as my one stop instant messenger. In the early days, I used it for MSN,AIM, and ICQ. And when Google Talk took off that worked too. Could even talk to my coworkers in China over the QQ network. And when I joined a company that used Skype, there was a plugin for that too. Life was good.

The consumer really has been the loser now that Apple, Google, and Facebook have walled off their messengers.

I miss those Pidgin days so much... Even Facebook was rocking XMPP for a while. Things were so simple back then.

I remember setting up Trillian (later Pidgin) and setting up a bunch of accounts just in case I had a use for them - not sure I ever used AIM/ICQ, but I _could_.

Now it's just more trouble than it's worth.

Franz is the modern day equivalent of Trillian/Pidgin.

Not really. That's just yet another instance of web browser, this time with chat-specific tab switching and - as I learned recently - a need to create yet another cloud account...

In Pidgin, you actually had it all in one place, with the UX defined by Pidgin.

That's another bad thing about current fragmentation. You have no way to unify the UX.

That's by design of course. The client side software is where the interfacing with the user happens, and if your business model is 'first become huge and indispensable, than figure out a way to monetize', you need absolute control over what the user sees and how they interact with your service.

Let's not beat around the bush - if you "need absolute control over what the user sees" so that you can "figure out a way to monetize", then you have already decided that you are monetizing with ads. That is the only possible reason to maintain iron control over the user's eyeballs.

This business practice of pushing the product unsustainably in order to drive out the competition was rightly condemned when WalMart did it. I don't see a similar public backlash in the tech sphere these days.

I was a big fan of Meebo, personally. Since it was a web app, you could sign in once on any computer and all of your chat clients were there.

...Guess who bought it and killed it? https://en.wikipedia.org/wiki/Meebo

For a while, imo was meebo's successor, but they saw the writing on the wall for third party interoperability after Skype cut them off for the nth time and decided to focus on their own walled garden chat and Twitter service instead :(

the answer is... Google

> In June 2012, Google acquired Meebo to merge the company's staff with the Google+ developers team.

Yes, it was really good. It pretty much Just Worked(tm), had a good UI for a web app, and was relatively lightweight (I remember using it even at crappy airport computers).

It used libpurple to connect to the chat networks -- same as Pidgin :)

Yeah I really liked Meebo too

You might be forgetting that AIM would change their settings randomly for a few years to prevent unofficial clients from working.

Yeah but I doubt most Windows users use Pidgin. What happened back then ICQ, MSN and AIM dominated at different time and geographically, even though they co-existed since the 90s. Now, not so much.

I used Trillian back then on Windows.

Same. Messengers used to just work, now we have a bunch of walked off shit with no intercommunication. "Are you on Slack / Discord / Skype / Telegram?" question is getting pretty annoying.

I used Pidgin on Windows for years. Ah, the good old days...

This is the sad part about today. We lost user centricity with the dawn of mobile apps.

The web (http) and internet were conduits for information and communication. Now the pipe is a dumb channel each companies territory.

I miss pidgin where you had AIM, ICQ, IRC, Gtalk and whatever others. I also miss how companies are pushing their app instead of improving their mobile presence on the web.

It also had a tiny footprint.

I'm amused that IRC is the only one of those you listed that is still alive and doing well.

It's alive, but I don't know about 'doing well'. It survives among some special interest communities, but the usage is tiny compared to any of the major platforms now. You won't find your uncle on it.

To be fair, I wouldn't have found him on IRC in 1998 either. IRC has really always been a relatively niche thing and I think it's fair to say it's doing well if you define that as remaining popular within the niches it's always been associated with. Though I think this is starting to change with the popularity of Discord.

Yes Discord is starting to really starting to dominate.

More and more often these days you'll find at the bottom of a community info page a discord rather than an IRC channel. Even big pirating communities are making the switch.

How does that work? I'm a regular in the C# Discord community and one of the rules we enforce and have to enforce regularly is no illegal or shifty activity (pirating, cracking, hacking, aimbots, cheating, etc.) because the discord admins don't take kindly to that sort of thing.

Well there's obviously far too much conversation for the admins to actually snoop in on what everybody is saying so unless anybody in the group reports it then it would just fly under the radar.

If you've been to an open source conference that lasts a week the pattern is always the same: the bulk of messaging ends up on IRC.

It doesn't start out that way. First you have a lot of people pushing their favourite barrows. Then they set up IRC gateways. Then they bleed users to IRC. I've been to a few conferences that keep actual statistics of messages published, and it's always the same.

But it's well hidden. No one gives up pushing their barrow, so if you just listen to the conference gossip you would never know it's happening.

>You won't find your uncle on it.

Bug or feature?

Openness and standards matter in the end.

I sure hope so!

Afaik ICQ is still somewhat popular in eastern Europe.

ICQ’s find a friend feature got me into contact with someone in south africa (I’m from europe), who I ended up visiting for a month.

It’s the sort of random encounter / friendship that you could have in the pre-2005 internet and that’s much harder today.

I don't actually remember why ICQ died, it was the best then there was MSN Messenger for some reason...?

I think the main reason is because Windows XP included a basic MSN Messenger client enabled by default in 2001.

Indeed, ICQ was vastly superior. In fact, in my opinion, many of its features (such as the discovery features) have not been equalled since them. It was quite traumatic to me when it was replaced with MSN Messenger.

AOL bought them. Slow burn.

ICQ is now owned by the Russian company behind Mail.RU. AOL sold it for $187.5 million, less than half what they paid for it:


in my circle of friends it was because aim let you pick a username instead of a number

I remember that they bloated up their client so much with menu bars, icons and later ads that it was not snappy anymore and required a lot of memory. So people where looking for alternatives.

> It used to be that everyone was basically on AIM,

Given that the A in AIM stands for America, I'd call that a pretty damn narrow definition of "everyone".

Yeah, pretty much no one here (Germany) used AIM, it was 80-90% ICQ and some using MSN.

As a french I spent countless hours chatting with my classmates on AIM when I was in junior high.

Pedantic nitpicking:

The A in AIM stands for AOL. The A in AOL stands for America.

The A in AOL used to stand for America, but now AOL doesn't stand for anything, much like the K in KFC used to stand for Kentucky.

> much like the K in KFC used to stand for Kentucky.

It always did (they just weren't going to pay Kentucky for their fraudulent trademark) and still does - https://www.snopes.com/lost/kfc.asp

> In November 2006, KFC and the State of Kentucky finally reached an undisclosed settlement over the former’s use of the trademarked word “Kentucky,” and the restaurant chain announced it would be resuming its former name of “Kentucky Fried Chicken.”

From the "More information about this page" link at the bottom of that Snopes page:

> You’ve just had an encounter with False Authority Syndrome.

> Everything in this section is a spoof.... As for Mississippi’s doing away with teaching fractions and decimals in its school systems because kids find them too hard to master, that’s no more true than Kentucky’s imposing a licensing fee on uses of its name

Well, damn. I (correctly) feel a right old fool now. Thanks!

(Sadly I can no longer edit the original otherwise I would add a Mea Culpa.)

It’s really hard to build a billion dollar platform business from just staying out of the way, so apps layer on the complexity.

Plus there’s always the VC around the corner hoping to fund the next Whatsapp, so we get a proliferation of me too-apps.

Very US-centric view. In the UK everyone used MSN (well, people that actually used IM did) and now literally everyone uses WhatsApp. Even my technologically illiterate mum uses WhatsApp.

Did presence ever really work though? I’ve been online since the late 90s and I don’t remember ever considering it reliable except perhaps for people on IRC who I knew had a transient internet connection (namely dialup).

I’ve never really paid much attention to or trusted presence status, with the exception of statuses that are manually set (e.g. do not disturb in Slack and similar).

It’s kind of funny, considering that presence is probably the most challenging technical requirement for large-scale messaging platforms. I will say that my two most-used messaging platforms, iMessage and Slack, both usually do a great job of delivering push notifications to only the ideal device (my phone, tablet, or computer). My intuition is that this logic is roughly the same as presence.

AIM handled presence quite well- custom "away" messages, easily editable "profiles" which were like a .plan file to include a note about where you were logged in, idle timers, icon to tell if an IM was forwarded to a cell phone. Finally while it lacked anything like Slack has to store your messages centrally and persistently, AIM would notify a user when you logged on to more than one location and allow remote logoff of your other sessions to ensure subsequent messages only went to your current session.

I'm glad you've been blessed with whatever Apple Magic lets iMessage work across devices for you but between my iPad iPhone and MacBook all I can say is I follow up important messages with a phone call because it's one of the more unreliable and indecipherable systems I've never been able to figure out.

The best is when as an android user that has an iPad, texts sent to you from an iPhone will go to the iPad that gets opened maybe once a week and not to your phone.

No going half-way into that ecosystem.

It’s interesting that your iMessage experience is much worse than mine. I’ve seen some weird behavior, but generally only on very spotty connections.

Facebook Messenger definitely does the best job. I believe it shows the delivery status of every single message.

>It used to be that everyone was basically on AIM

That was never true. AIM was very popular in some places and had no foothold in others. In my area it was first ICQ and later MSN. Nobody, except one buddy from Seattle, used AIM.

IIRC, AIM required you to have AOL as your ISP. Even where applicable (i.e. in USA), this was still inconvenient until they dropped this requirement sometime in 2000. Other, ISP-independent messengers existed before, but the pastoral idyll of "everyone at the same chat network" never existed in reality.

> Instant messaging is an example of something that used to work, and now doesn't.

It's a miracle that the web works the way it does. So perhaps we need a consortium (like W3C) for messaging too?

Ugh. No. You need vendors having common interest banding together to standardize on some parts for everyone's profit (what WHATWG essentially was).

Otherwise, no matter what the committee does, it's doomed to fail because there is no incentive to support them.

And given that no one wants interop with their systems (except for some users, but who cares about them?), an interest group is unlikely to appear.

In the business space, there was a consortium of companies/contributors that got together to try to solve enterprise messaging (not IM/person-to-person messaging, but messaging computers/message buses/queues/etc.)

That consortium created AMQP, which is a fucking abomination of a protocol. You can read through the spec and see the design-by-committee chimeric parts.

I'm not convinced that it's a good approach, basically.

So don't make new systems.. help dismantle old ones?

I also haven't seen any substantial innovation in chat for like 30 years. The main value prop hasn't changed since pre-internet days.

It is remarkable how thoroughly Google has munged their communications platforms.

Especially as an iPhone user, I am constantly befuddled trying to get a semblance of a well organized contact list between my google apps, never mind a correspondence between the iPhone native contact list.

I have a google voice number, and generally I assume that messages get to me one way or another.

The most recent baffled was finally understanding that there is no way to manage contacts from within the iOS Hangouts app. I’m actually not sure how that contact list is generated.

It may be a byproduct of how Google incentivizes their engineers.

As I've heard from chatting with engineers and PMs at Google, the biggest rewards go to those that take a product to its first major release and the initial marketing fanfare.

After hitting 1.0 you typically see teams break up, the star engineers moving on to the next big thing. Once that happens the pace of the products development slows significantly, leading to ignored user feedback and a struggle make improvements.

The product mothballs for some years until - in the case of something like Google Talk - momentum is generated around replacing it with the next 1.0 product.

And so on...

Writing really good software takes years of user feedback, and a team of knowledgeable engineers digesting that feedback to write new versions. A company can afford to do that when its users are its customers. When they are its product instead, you get a steady stream of half-baked and quickly-abandoned software, often installed via forced update.

Most of the industry has settled on the CADT model instead. The GP of this post suggests Google has done a particularly good job formalizing it these days.


Did you try that link when HN is the referrer? ;)

In theory the goal is "launch and iterate", but yeah, the incentives add up to "launch and leave".

It isn't exactly clear how it's managed from the desktop either. Over the years they've really screwed the whole thing up. It's no longer clear whether someone is online, offline, online but hidden, or just lost in a contact list change reorganization. it also isn't clear whether someone is available to chat, or if the IM I send will be inconveniently received on their mobile phone, something I try to avoid doing when possible.

yeah, yahoo did all this same stuff 20 years ago.

It’s not just Google. Have you tried Skype for Business lately? Meetings are incomprehensible.

Teams is supposed to fix everything... but my O365 tenant doesn’t have it yet.

What the hell is going on with groupware? IBM, MS, Google, are they all just befuddled by whatever web fad is current to forget making their stuff actually useful? I feel like there has been nothing but regression since 10 years, ever since Skype brought free and stable group calls.

I suspect that nobody has figured out how to make money on it and that is what is 'going on'. It does make one ponder how to do this stuff in a sustainable way.

Seems like Slack figured out how to make a lot of money out of interoperable groupware. If they'd been slower at it, I'd still be saying they're an acquisition target for those companies.

edit: "All" they did was come along at the right time and make the interoperability easy. The prerequisites to have done it at all? Smartphones, WebRTC, and ElasticSearch.

Calling Slack "interoperable" is a bit much to me, though I admit it's better than everyone else in the "new crowd".

I wonder if we need a new word for "interoperability in the time of SaaS" - the kind of where SaaSes talk to other SaaSes via locked down APIs, under absolute control of the vendor.

I mean, you can call the APIs locked down if you insist, but SaaS is SaaS. Would you count S3 as a locked down API, for example? OpenStack's Swift (and plenty of other products) will emulate it. I chose the term "interoperable" specifically instead of "API compatible" but I agree this gets very confusing when talking about SaaS.

In order to do what Slack is doing, they have to be somewhat committed to open standards. They are just doing webhooks, like GitHub does. We tried specifying microformats, and we tried specifying webhooks, but the "loose RPC" model... Seems to work way better than XMPP server interoperability ever has.

I recommend this recent nested Twitter ("new crowd") thread amongst Stewart Butterfield and many other early web folk, about whether Slack is a web app: https://twitter.com/stewart/status/961704310613491712

I don’t think so. Cisco maybe, but Microsoft is printing money with this tech.

Capturing email, files and love communication drives subscription revenue for E5 SKUs of 365, plus most companies will need Azure AD to meet their security and compliance requirements.

We ran the numbers at work... taking the blue pill and going all on Microsoft will 3x our lifetime value to them vs a standard E3 shop.

It’s funny you said that.

One of my colleagues pointed out that the IM client from Exchange 2003 era (I think it was Live Communications Server) and LiveMeeting were a better collaboration platform than SfB, even given the severe limitations on bandwidth and cpu.

It's really not a high bar is it? Frickin MSN Messenger and Adium was a better platform than what we have now. While never having used it I suspect the best was probably blackberry's platform in its heyday, even though AFAIK complicated to integrate into a business, but at least then it would work solidly.

Teams is an afterthought from Microsoft in its continuing lineage of Collaboration Applications.

We recently switched to it for a world wide support team and it's a step down.

- No direct quoting, only threaded replies in group chats.

- No linking to Threads, even though you can personally bookmark it

- No message indications except in 1 on 1 chats ad hoc chats, so if you're in a large group thread, you have no idea if someone is paying attention or not

- Poor UI decisions (the menu which appears over messages for "options/bookmark/like" will obscure messages if the same user sent two messages in a row; because it appears over the right-most part and the second of two messages from the same user doesn't include the user info, the menu will hide whatever's at the end of the message).

- Scrolling back in a message thread (group or personal) is pointless, as messages take so long to load, and it seems to load sequentially. (i.e., you can't just scroll to approximately where you think the message is; if you scroll for 10 minutes, it will still just load the most recent messages before continuing)

- Search has no proper way of search only one conversation, it's all or nothing.

- The mobile app is incredibly slow to sync regardless of whether it's on data or wireless.

I could go on for awhile on this. I honestly wonder if anyone at Microsoft even tried to use Teams before they released it, since everything about it feels like it was just tossed in there and the devs weren't allowed to look at what other collaborative chat apps did to make them good. Even their "me too" implementation of things like giphy integration or inserting photos in to the chat is very poor (dragging a photo from desktop will insert it into the thread, but dragging it from the web will upload it to sharepoint, though both take very long to upload). Accessing Sharepoint content takes a long time because Microsoft's login to the Office 365 space takes a long time. All text is actually rich text, but for some reason they included a pseudo markdown syntax which just toggles the rich text functions (bold, italics, etc) whether you mean to or not. You can't just escape these characters either since it's not that well thought out of a function.

When we briefly had Skype For Business, I thought that Microsoft couldn't do any worse, but they really did their homework for Teams and made a real horrible product that works for basically no one.

> Have you tried Skype for Business lately?

On Linux it does not even show the screen shared by others (Windows) users. In the old Skype (not for Business) client Linux users could even share. For corporations progress means removing working functionality.

That's been the same situation with webex and Cisco. (Spark is even worse)

Teams is a joke. It’s an alpha level product at best. At worst it’s black hole on the user voice forums for features that were in IRC and a constant memory leak. The vscode team needs to teach the teams team how to properly write an electron app.

If I move the teams window from one monitor the my other slightly different sized one the close and maximize buttons don't work until the window gets minimized and then re-shown. smh. (win8)

Skype for Business isn't really Skype, it's the new name for Lync, which has always been garbage.

> The most recent baffled was finally understanding that there is no way to manage contacts from within the iOS Hangouts app. I’m actually not sure how that contact list is generated.

You also cannot manage contacts in any app on android besides the contacts app, or, probably better for you: http://contacts.google.com

The rationale for Google's behaviour has always been that it will help them achieve user lock-in. But disregarding both the technical and ethical merits of this strategy... is it actually working for them? Six years ago I spent very significant proportion of my time using Talk and Reader. In their absence, I have shifted to using FB Messenger (/WhatsApp/Telegram/etc.), with Facebook and Hacker News kinda substituting for Reader. I still use Gmail and Google Apps, but my daily time inside the Google ecosystem has probably decreased by about 2/3rds.

If I'm their target audience, then this strategy is an unqualified failure. Obviously I'm not their target audience -- I'm an outlier in almost every respect -- but I've yet to see an analysis which suggested that Google has actually succeeded in capturing more eyeballs or generating more revenue via this strategy. In fact I haven't really seen any business analysis of this at all, other than blaming the business-types for making these decisions. Which is probably correct, but I'd still be very curious to see whether those decisions paid off, purely on their own terms.

I can relate to this. Google's services became less and less useful over time. I even replaced Hangouts with self-hosted XMPP server while previously I thought custom would mean a lot of trouble. After a while test driving it I migrated entire family (Conversations.im on Android works great and we've got encrypted E2E chats).

Now I'm thinking about hosting my own calendars (kind of easy) and email (scary).

I've been finding Google Calendar incredibly buggy as of late as well as other services. It's like they don't care anymore.

Google can't organize the world's data if the world won't trust Google with the data in the first place.

I transferred my American number to Google Voice before I left the country and have been able to hold onto that number for over a decade.

Unfortunately, everything about Google Voice/Hangouts/Talk/Whatever-it-is now sucks. Trying to find someone is damn near impossible. I totally can attest to that. The removal of federated and regular XMPP was the wrong direction.

I'm in the wonderfully weird situation of not having a Gmail account either. I deleted it and ran my own mail server back in 2012. I use DavDroid/Radicle for contacts, so in the web interface, any contacts after 2012 that are SMS are phone number only. I have to use my phone to see the names.

I don't think it's even possible to find someone via gmail address anymore. Occasionally I'd find someone on Google Plus .. and have no freakin idea how the hell to send them a personal message (either in G+ or Hangouts).

The whole Hangouts/G+/Gmail ecosystem is awful. If you don't have a gmail account, it's beyond unusable. I pretty much just use Hangouts for legacy chats.

AIM/Yahoo/MSN all worked .. and now they're all gone (I think Yahoo is still there, but web only). Facebook was unreliable as shit and god awful until around 2013/2014. It took them that long to create someone their competitors had done better a decade ago.

Voice/Hangouts has felt like an abandoned application for many years, even though it's potentially one of the best products Google has ever made. They really seem to stop at v1 for almost all of their products and have not the slightest idea how to monetize them.

> Voice/Hangouts has felt like an abandoned application for many years, even though it's potentially one of the best products Google has ever made.

That's probably WHY Google Voice was one of the best products Google put out.

it was originally a company called grandcentral and eventually acquired by google. they basically left it as it for YEARS


There's a reason users/clients have come to fear the Google acquisition of a product they've come to depend on and in which they've vested hopes of continued improvements and solutions.

At least Google hasn't outright put a bullet in Voice, yet.

Have any recommendations for commercially similar solutions, that also support a proper SIP interface?

I'm preparing for that day, when GVoice dies.

https://jmp.chat/ is a good alternative. Text and picture messaging over XMPP, and voice over real SIP (with voicemail transcribed to text).

There are lots of great XMPP clients, such as Conversations on Android, and Gajim or Adium on desktop. Or use https://movim.eu/ if you like a web client. iOS has a few options as well, including Tigase Messenger and IM+.

I use Google voice # + OBI220 device for home phone service, and hangouts on mobile/tablets. The home phone service using obi device in nice bc it's portable, the hangouts integration is fragmented, frustrating and unreliable

My understanding is that Google Voice has some partial cover in that it is the backbone of Google Fi.

And just try keeping a distinct Google Voice set up, if/when you sign up for Fi.

Unless you're very careful, they mush together your Voice and Fi accounts, irreversibly.

When Fi was in its early days, a lot of Fi users found this out the hard way and were pissed about it. People who'd been using Voice, maintaining a separate Voice #, and who wanted to keep it that way. Who DIDN'T want that number suddenly tied to a/their cell phone.

(Hint: Create a separate browser profile (not just a Google account, but an actual separate profile under its own sub-directory). Sign up for a different Gmail/Google account. Use one account for Google Voice, and the other for Google Fi. Don't accidentally cross-pollinate.)

So basically an anti-competition move from search? (Don't ask your friends, they're unreachable anyway. Use search instead!) Suitably dystopian, I like it.

Unfortunately, they recently remembered they have it and have begun removing features and options.

This sounds like a cynical joke, but for those people who don't use Voice, it's sadly what's really happening.

I want to get off Google products for many reasons, but the main reason is becoming that I can't trust them to listen to users.

Rereading it I can see how it might sound like a joke... it's not. Truly very sad, they let you go back to using the old interface but sometimes you get jolted back into the new one and there's no way to access some of the old options... maybe if you have a really old browser user-agent you can trick it into only giving you the old interface?

For people that use Google Voice, that's often the hardest Google service to leave. You can get similar features with https://jmp.chat/ - it supports text and picture messaging (and voicemail transcription) and you can port in your Google Voice number to make the transition seamless for your contacts.

Thanks. Am I correct in assuming this is your project?

How long have you been working on it? What's the exit strategy look like? How big is the team currently?

I noticed that your blog hasn't updated in over two years, while I recognize that making blog posts is time-consuming, when it comes to software that just gives the feeling that the author has abandoned the project. Maybe consider linking to an area you still frequent?

https://jmp.chat/ is a project I started, yes. There are about a dozen other people who work on it to varying extents (you can get a sense of this from the commit logs, some of which are linked below).

Based on https://gitlab.com/ossguy/sgx-catapult/commits/master and https://gitlab.com/ossguy/jmp-fwdcalls/commits/master it looks like I've been working on JMP's codebase for a little over 13 months now. Prior to that I worked on JMP's precursor (see https://github.com/ossguy/sopranica - also called "Phase 0"), which started about 4 years ago, and was more part-time.

Since I and a lot of my friends depend on JMP for their everyday communication, I'm not really interested in "exiting". If the project were to be sold, it would have to be to someone who cared about keeping the software free and open source, and who wanted to keep it on a federated network. I suspect going public would be a bad idea, because most of the public (i.e. investors) doesn't(/don't) care about JMP's values.

We publish an update on JMP and related projects every 1-2 months. We recently started using Mailman so you can find the past couple updates at https://soprani.ca/cgi-bin/mailman/listinfo/jmp-news and signup if you like. For day-to-day updates, you can join our group chat at https://anonymous.cheogram.com/discuss@conference.soprani.ca (more options for joining are at https://jmp.chat/#support , including directly from your XMPP client). That's where the majority of project-related communication happens.

But how do I know they'll be around in a year or let me save my number?

As with all carriers that offer US (and Canadian) numbers, they are required by law to let you port your number away. So you can always bring your number elsewhere if they decide to close up shop.

And if that happens, likely someone will setup another instance and keep the same service running. That's another great advantage of JMP over Google Voice: all the code is free and open source, so anyone is free to run their own instance if they like. In fact, you could run your own instance right from the start and then be in complete control of whether JMP keeps working for you.

Instant messaging used to be so simple. You connected via adium or pidgin, and it just worked. You could hook it up to gtalk and msn. Even ICQ and AIM. Then msn wouldn't play nice any more, then gtalk.

Now we've had a dozen renames and replacements, XMPP is long gone, and what's left is an almost unusable mess. Add to that the dozens of people I once had connected on Adium (from memory most on Google talk) are now strewn across loads of different chat apps.

These days I'm texting more again! This really isn't the future I expected.

That's been my experience as well. Facebook/Hangouts/WhatsApp/Allo/whatever might have amazing features but I truly don't care. SMS/iMessage (since it's seamlessly integrated) has completely taken over with my group of friends because of one defining feature - everyone has it. If you have a phone number, you can receive a text from me.

A while back one member of our group got tired of the 8 person running group thread we have and suggested moving to something "better". The group unanimously shot it down because we don't want to download new apps, create new accounts, then inevitably have to migrate to the brand new flashy service in 3 months.

For context, we're all late 20s/early 30s.

Edit: Actually, I believe SMS is still superior because it works even without a data connection. We spend enough time camping, skiing, attending huge sporting events, etc. any of those other services would be a significant downgrade.

I suppose you live in a country where the telcos offer free SMS. In Spain they cost money, so almost no one uses SMS for personal messaging. All the SMSs I get are from companies.

In Germany there's a bunch of different options when signing up with a provider, eg I have one where I pay 2€ extra per month for 100 SMS which is plenty for me. For 4€ I'd get like 300 or so. Its the same pricing for free minutes for domestic calls (normal landline or mobile, no service numbers obviously).

I think it's similar here, but 300 SMS/month (i.e., 10/day) wouldn't be even close to enough for the average Spanish user. Whatsapp usage is huge here. Many people do all their routine personal communication using it, and then they have a group for the family, a group for coworkers, a group for friends, a school parents group where they all complain about their kids' homework, etc. Of course, they also send photos and videos (which I'm not sure if are covered by SMS plans or you need that exotic thing called MMS).

And even if it were enough, a Spanish user won't pay 2 or 4 € for something that they can get for free. In fact when Whatsapp had the idea of charging a symbolic amount and suspended access to some accounts due to not paying, people started switching to Telegram en masse (Telegram even went down due to that), then they gave back access and abandoned the idea of charging.

Certainly has a big plus - SMS will often get through when even voice can't. Data on the other hand! That's without getting into how badly some apps handle intermittent or no signal.

Have you tried Telegram? It's the best.

See, there's the devil, in the details as usual: "Have you [and everybody you wanted to talk to] tried the X [, all having tried it at the same time]? It's the best [when it works, unless it's subtly and invisibly broken, void where prohibited]."

Not forgetting the rapidly escalating permutations. They say Telegram, work wants Wire, someone else Whatsapp, someone else says Messenger and on and on.

Either install 27 apps or decide enough is enough. I've had just about enough until one can aggregate like Adium used to.

is this sarcasm? (honestly it's unclear to me)

I enjoy telegram, but if I need to talk to more than the two people I know that use it... well..

When I talk to the 20 tech people who I respect most, what I notice is that everyone respects Google less now than 5 years ago. Is it a successful branding strategy that generates so much dislike?

I'm especially curious because Google is famous for basing its decisions on "data". I have no idea how things work in Google, but I can say that every company I've worked at that supposedly valued "data" in meetings actually valued something darker. The use of "data" in meetings tends to be a passive aggressive negotiating tactic for a group of people who for cultural or emotional reasons don't think it is reasonable to express strong disagreement or actual anger. Instead of expressing strong emotion, people are taught to quote data -- they then cherry pick whatever statistics back up their beliefs.

As far as I know, there has never been a company that said "We want the worst informed people to make the decisions" so in a sense all companies have always valued data. But they didn't make a fetish out of it. They simply expected people to be well informed, and to make intelligent arguments, based on what they know. That would have been true at General Motors in 1950. That much has probably been true at most companies for centuries. When management says that the company is going to be "data driven" they are implicitly asking for a particular type of interaction to happen in meetings, an elaborate dance where people hide their emotions and quote statistics.

I can't cover all the nuances of company meetings in a comment on Hacker News, and of course I am not advocating that meetings should be abusive, but I do think it can be healthy to tell meeting participants that it is culturally acceptable to advocate strongly, and with emotion, for what they think the right policy is.

Google seems like an example of how a "data driven" company can go off the rails. I'm not sure what their meetings are like, but I know that in interviews the management at Google talks about their focus on data, yet their brand image continues to decline.

And of course, I'm on the record in believing there should be less group meetings and more one on one meetings, at every company that I've ever worked:


Wow. That is a really apt critique of the culture you are describing.

It sounds like you're basically saying that as with programming styles, wellness, activism, and everything else in the world: there is a basis of truth that underpins data-centric management, but you can easily make it into a cult.

You put that idea together so succinctly - thank you.

I've migrated off Google services as much as possible over the last six months; I use Duck Duck Go for searches, and fastmail for email with my own domain. I also use the fastmail client's built in calendar. So far I've only had to fall back to a Google search a few times in six months, and the fastmail client is light years behind gmail inbox and gcal, but it's fine for my usage patterns. A year ago I was spending half of every day on a google service, now I barely touch them.

This seems to be a common and popular trend these days, especially on HN. Is anyone doing the opposite and becoming more ingrained into the Google ecosystem?

I'd love to dive more into the Android/Google side of things, but these sentiments are (unfortunately) making me think twice about doing so. I'd like to read different perspectives if they are out there.

Personally, kind of.

I’ve moved further and further from google for my personal email and communications, for all the reasons elucidated above. However, for my commercial presence, I opened a gsuite account because it seamlessly brought together - and kept together - disparate components that I wanted to be able to use in a device-agnostic manner for myself and my team. I could have brought things together from a half dozen different, possibly better, services but gsuite was basically one-stop-shopping. I didn’t want to spend time getting the tools together, I wanted to spend it on the job at hand. The communications tools are good enough.

That said, I may change my mind about that. For getting work done I still find Outlook kicks the shit out of gmail, and Microsoft Live or 365 or whatever they call it has a very very solid online office suite (and their offline tools remain best in class, IMO). Todoist integrates with both. I’m just lazy about non-mission-critical upgrades. I don’t have a good replacement for google hangouts, though. Skype is a festering tire-fire, and I don’t ever want to use a webex unless I’m being held at gunpoint by the Cisco CTO.

Yes, kinda. I tried the iPhone 8 (this is Brazil..) after 8 years on Android and feel like Android is much more modern, integrated, free, and even has better designs on so many places (yes, its odd). Particularly the Play Store is beautiful compared to the App Store.

Every app on iOS has a paywall, integrations are limited, apps can't process your SMS, stuff like this enrages me. Actually I couldn't even find a good way to transfer images/codes both ways, like I do with PushBullet on Chrome + Android all the time.

For me iOS also misses taps and lags on some operations. If using Swiftkey, sometimes the wrong keyboard will pop-up. I feel like the image I had as an outsider that iOS was perfect and smooth kinda broke.

So I guess I cannot compare to other people because they use their phones differently: I have to test everything...

What data are you basing the statement that Google's brand is in decline on?

Granted I'm by no means saying Google's messaging strategy is optimal, but it's not totally bonkers.

And the adage "you aren't the user" seems apt here. This applies both to the statement that Google's messaging tools are in decline, and to the statement that Google's brand is.

HN has been decrying google for years now, but he isn't your average consumer. Much as my extended family can all use allo and duo with ease, but I doubt my grandparents know what irc is.

I kind of think that it's difficult if not impossible to appeal to the broader set of people whole also appealing to the hn crowd. And, well, we're a lot smaller.

To loop back to your post, thats part of why data is so important. If you don't base your decisions, at least in part, on data, you end up designing a tool that's useful for you and not for your users. That's not to say that emotion has no place, but I think you're being disingenuous towards data here.

(Note:I work at Google, if that matters)

There is a running joke within Google that the best way to get promoted is to create another chat or video conferencing app.

Is it really a joke at this point, though?

Googler engineering leaders and teams engaging in the replacement of good chat tech with worse happens every year or two. All to demonstrate "cross-org impact" and climb the promo ladder. This is in fact the driving force behind many of the complaints mentioned in the article.

A sad and systemic failure, indeed.

It really makes you wonder why there is no one in the organisation saying "This is a fantastic new product, with truly innovative work. Why did you feel the need to create an internal competitor to it rather than help make it better"

One of the most destructive issues I see from my time in management has been engineers who are more interested in re-writing what other people have already done. It seems amazing to me an organisation the size of Google hasn't identified the waste here.

Yep, "demonstrating impact."

There are apparently some moves afoot reforming the promo process to make this less common though. Here's hoping.

We used to have a relatively open internet based on developing, using an adhering to protocols and open standards. This is probably because everything was new and there was a lot of money on the table for everyone.

We would joke about peoples' theories of the "Balkanization of the Internet" and mostly we would look at it from the wrong angle (nation states).

Now we have an internet of walled gardens. Your data is precious and no longer yours.

> Now we have an internet of walled gardens. Your data is precious and no longer yours.

And you can't do shit without agreeing to some ToS or EULA. At this point I'm actually waiting for someone to "innovate" a way to dynamically sign EULAs with the same API calls you'd use to get access tokens, just to add insult to injury...

> And you can't do shit without agreeing to some ToS or EULA.

And it's guaranteed to have some damned binding arbitration clause[0], if it comes from a company based in the United States, so when[1] everything that company knows about you is "unintentionally" transmitted to every hacker and script kiddy with a modicum of PHP knowledge, you can't even band together and financially ruin the company. Not that you'd be able to get anything out of them, anyway, since 90% of them are so knee-deep in venture capital money that you'd be 9,392nd in line behind the likes Sequoia, Andreessen, Y Combinator, and Honest Achmend's Certificate Authority and Investing Company.

I guess I'm getting old but, damn, I'm really starting to hate the corporate Internet.

0 - I loathe every single one of you who has ever posted on a Hacker News thread about how horrible binding arbitration clauses are and who are also founders/CEOs/operators of companies with binding arbitration clauses in your terms of service. This goes quadruple for companies that are B2C. I do not care if "everyone else is doing it" or your "lawyers recommended it." Grow a spine and delete the clause if it's so awful. Go first. Be bold. Half points for burying an opt-out that requires faxing a notarized statement to a phone number in Saint Pierre and Miquelon. That's better than nothing.

1 - It's not "if" anymore, it's when.

Unfortunately, spines were delaying the release and that feature was cut.

A typical process at a big corp:

Initial small enthusiastic team has a cool idea, works day and night, having a lot of fun on the project. They release it, it's cool, useful and everybody loves it. Then after a while original team members get bored or focus on other interests, move out of the team and new team members come in. Those team members aren't necessarily worse than the original ones, but have to maintain the project they didn't write. They need to show their usefulness, so they try to conjure new ideas that might not be useful at all, but hope they would be visible enough to grant them promotion within the company. They focus on some mistake/wart/non-refactored portion of code, make an essay or two to convince management that changes are needed, then slap on another feature they could be known for and proceed changing experience. Once they are done, management applauds them and customers either are lucky and have a blast with the new changes, or those changes were complete misses and started to ruin experience. This process repeats often until product is EOLed or just maintained for whatever reason.

Agree that it happens, and think this case is different.

See: https://news.ycombinator.com/item?id=16351137

This case is no different, right?

I can only agree with that post. Sometimes I'm wondering though if this is just us techsavy people or the fact that I'm getting old and everything was better in the old days. Maybe the average person not working in tech prefers how this works today? Many times I'm already lost at the basic presentation of modern websites or apps. Big colorful landing pages that try to aggregate different data and events in an intelligent way but just seem like a jumbled mess to me, everything has to be dynamic and load additional data via ajax so that often times the back button doesn't work or at least makes sure the scroll position won't be restored properly... I could go on for hours. I guess I'm too old for that sort of nonsense any more.

What happened at Google? I remember when gmail was first released - such a great product.

Hangouts destroyed the google talk experience and community. Someone should have gotten fired over it. Everyone seemed to use google talk before the plus/hangouts conversion. Such a massive lost opportunity.

There are many other examples of half-baked and unmaintained software from G. For example, play music's web UI has many irritating bugs that never get fixed. I'm close to canceling my subscription.

I'd be pretty hesitant about committing all in to any of Google's enterprise offerings, like GCE, based on my experience with their other products.

Google makes money off of data. Essentially, they are a big ad company that does tech. Whatever that means to you, that's generally how they make money. The more of your data that they can have going through their services, the better.

All of their "products"/tools (at least after a certain point) have this end in mind. If a product does not get them enough data[0], they either a) bastardize it until it does, or b) kill it.

I'm sure this is not news to a lot of people, but it surprises me how often people forget that Google has no reason to care about the quality of their tools as long as enough people are using it. I'm not saying they don't make good tools, but that is why we see good tools get killed (Reader, Wave) or ruined (Hangouts, Voice).

[0]: or they can't figure out a good way to get good ad data from it.

EDIT: formatting.

Google just can't do UI's, I can't think of a UI put out by google that I don't dislike, if not completely hate. It's almost like there's someone super senior in the company who forces bad decisions on designers.

The crowning achievement of absurdity is probably the YouTube app, it's not as bad as Skype, but given time they may get there.

Their initial search UI was great, though they ruined that too.

I switched away years ago, and am increasingly baffled whenever I accidentally run a google search.

Switched to what though?

DuckDuckGo is really good - maybe where Google was a year ago but without the user tracking. Also, bangs are awesome.

Search results are as good? If so maybe I'll try, I've basically gotten to the point that I could sider google to be mostly evil.

For me, one of the worst parts of this transition is that as more people use hangouts, the experience becomes worse. The utter lack of control or signaling on it leads to tons of missed messages, threads that fall out of visibility, and general "did they see my message? do they even use this any more?" questions.

All of which Google Talk managed just fine. And if you wanted more control, install a third party app, because it worked with them all.

What did we gain for this?

Not to mention you can't even delete messages on hangouts. One of the reasons I switched to telegram.

you guys want to talk about frustration and pain from hangouts? pffff, try this:

an important new client calls you on your google voice number, which routs to hangouts. You don't yet have their number as it isn't in your Google Contacts or your Phone Contacts.

You answer, but are busy, or need to look up something for them, and you tell them you'll call them right back.

You get the info you need and you look to your recent calls for the call THAT JUST CAME IN LITERALLY A MINUTE AGO. It's not there. Hangouts does not save incoming calls. (or at least it didn't a year ago)

Why? Who the fuck knows??!

Google Hangouts: We Don't Give A Shit.

Exactly my experience. I was always wondering, if I'm just too stupid to figure out how to see recent calls. Glad I'm not alone.

This is why I have my Google voice number forward to another platform.

There is at least one major attempt at addressing this fragmentation with a modern protocol, that being the Matrix protocol (matrix.org). Yeah, if it takes off then it will just be yet another bit of fragmentation, but at least it is designed in such a way that chats dont get siloed into a single echosystem.

I have to say that, although I began to use Matrix for the open source / open protocol aspects, their flagship client Riot is actually much more fully featured than most other propriety products. For sure it is much more useful than Google Hangouts and I'd say about equal to Slack in terms of user experience and features.

Riot is impressive, especially for something so new. Equally impressive is how trivial it is to create plugins to bridge to other protocols. For example: https://github.com/ImmanuelBaskaran/Hangouts-Bridge/blob/mas...

>So you have to use the Google Hangouts on a separate tab in the browser, or it gets juxtaposed to your Gmail, which will slow your lastest i5 laptop down and makes your fan whirr like crazy just to have one conversation.

Fucks sake, this annoys me to no end. Here is a product that's clean, lean, and fast. 5 years of development later and you have a bloated, energy-chugging, slow and irritating piece of shit, with absolutely zero added functionally to show for it.

God damn it.

The story of basically every native Windows app that got rewritten as a web app. It'll run in Chrome on Linux, but good luck writing a GTK+ or Qt app that will interface with the underlying protocol.

Oh hey Michael, how's it going?

Yeah, Microsoft has a lot of amazing initiatives to virtualize, emulate, and wrap interfaces around everything in the universe, which is a smart idea long-term, but they never seem to get to the phase of actually optimizing it. Someday, when we convert our entire race into pure energy, we will be able to seamlessly shift between between different levels of indirection in our matrix-like universe, but for now I'm having a lot of trouble just trying to open a single application in the first 4 minutes of booting up a Windows machine. It never used to be so bad, but every year it just keeps getting worse.

My only gripe about Matrix is the lack of a good desktop client, but at the same time trying to wire WebRTC into a desktop app feels almost as bad as C++03 did. Gstreamer just got support for it, but I have no idea if thats enough to implement voice / video conferencing in a Matrix implementation.

It's not exactly what you're describing, but to see an example of a video conference running within a Matrix client...


Yep, web apps are just plan bad, but i can understand why a startup might go that way to shorten time to market, but replacing a native app with a web app should be a shootable offense

This is a good rule of thumb, but web apps have the serious advantage of being (nearly, excluding differences between browsers) platform-agnostic. This saves development time as you pointed out but also allows easier cross-platform support. For example, Skype which wasn't officially supported on Linux (it was left frozen at some ancient version) until it was ported to a web app. After this change, Linux gets the same latest version as all other operating systems.

I don't know about you, but I'd trade any reasonable performance degradation for broader support of a product I use.

Reasonable, mmkay? Every other webapp-in-a-wrapper seems to go the Win95 way, which was "mine is the only app on the whole computer, let's just hog all the resources!" Nowadays they must be mining bitcoin or whatever, because I don't see what else a messaging app needs a whole core for its entire runtime (looking at you, Slack Desktop).

The problem is that it is exponential, to the point of having a couple of applications running at the same time, requires a CRAY class computer.

But Google+. Do you not realize the value of Google+??

I just realized that Google+ isn't dead yet. Guess someone at Google is still hopeful about its future?

P.S. Also the amount of shit Google+ has fucked wrt to Google services in amazing. It also killed the "+" sign in Google's search results too.

They have to keep Linus happy.

Google+ is very far from being "dead", it's quite lively, though niche compared to Facebook.

Yeah, generating lots more page views and clicks, because I can no longer effectively search for what I want and instead have to manually sort through a ton of poorly targeted and differentiated results.

Nope. What is it? Honestly. I've never seen it. All they seemed to have was slightly cleaner sharing permissions UX.

I suspect you wouldn't been downvoted so far if there was a sarcasm marker on this, of some sort.

Wait, is using more than one question mark not an internationally recognised sarcasm marker???

Google Talk was like the “cool” version of AIM to me when I was using it between 2010 and 2013. I could access Talk from my browser or from a desktop client (Pidgin, and I believe there used to be an official Talk desktop client).

For some like me, I didn’t mind to have my messages/conversations stored like an email (I am aware of some people’s privacy cocnerns). Even today I can go back to my Gmail and look up my conversations with friends and family. I own the conversation history, and I can export them at any given time! I can search them and sort them directly in Gmail.

Now I use WhatsApp, Messenger, text messeage, and Wechat (for different groups of people), I don’t know any mechanism to export my conversations. For WhatsApp and Messenger, I know search is only available on their desktop client... and IRRC only text search (can’t do date/time search alone) is supported and is on per contact basis. No global search...

Communication used to be, hmm, simple communication: no snap, no video, no crazy emojis and stickers :)

Google, what happened? :<


Edit: The only thing still usable is the free phone call from and to U.S. I use it all the time since carrier signal is so unreliable and uneven (and subject to surrounding environment) so I just hop on Wifi to make calls, esp at work. PLEASE DON’T KILL IT .

> For WhatsApp and Messenger, I know search is only available on their desktop client... and IRRC only text search (can’t do date/time search alone) is supported and is on per contact basis. No global search...

WhatsApp has search across all conversations in the mobile app (on iOS just pull down on the main screen) and in individual chats (tap on the chat title and select search from the menu).

My personal experience with Hangouts is that it would often hang and disconnect so that I would have to forcefully close the app. Then it wouldn't update the read receipts, or possibly worse, lie about who had seen what. I put all of this down to my having an old phone.

Recently, however, I got a new phone. I was excited to have a usable Hangouts once again, as that's where many of my close friends are. Unfortunately, now I know that the problem is Hangouts, not my phone.

That's alright, the competitors are even worse. I used to use Skype on Linux. They stopped development on the Linux client a decade ago, but it kept working just fine. And then they released a new client and forced everyone to upgrade. The new client is so bad in terms of sound quality that I've been forced to switch to Hangouts.

You might be able to use Slack or Discord, though I don't believe either offer video options and the only real mention of voice that I saw for Slack was for calls.

Still, Discord might actually be worth looking at - it's targeted at gamers, but it does at least small-group voice chat reasonably well in my experience.

Discord has 2+ person direct audio and video calls.

Our company switched from Skype to Discord after the new Skype client came out and have been happy with it.

If you haven't already, consider having at least some of your folks sign up for Nitro so Discord can stay viable.

As to the contact list being some random list of names, I've noticed companies seem to go through this sequence:

1) We get more engagement if people add more friends to the product! Repeatedly nudge them to add ALL their friends! 2) Wow, people have a ton of friends they aren't interested in. People don't seem to know how to manage their friends list. Let's automatically pick what to show them.

I used Pidgin once and it worked fine on Windows and Linux.

So instead of one universal client, one requires Skype, Facebook Messenger, Google Hangouts/Allo/Duo. (I don't use iOS but Apple have their own, I guess)

Thus if anyone wants to contact me, stay old school and use SMS or email! :)

100% this article. Google Talk was awesome, especially as MSN got more and more bloated and then once they forced into Skype and everyone's contact list got screwed up or people fled, it was like, yes, a clean client for use with friends. Then it turned to garbage. I'll have someone call and 4 different devices start ringing except the one I'm on, and then the others will continue to ring forever. I called my friend back when it didn't work and we had like 2 different voice calls going at the same time together. He works at Google and has to use it, so that's the only reason I keep it, I don't even know how to use it it's so confusing now.

Shower thoughts had me thinking just the other day how I wished we could have a new Pidgin like system. I have a hard time figuring out which friends use which platform and where I can get replies. I want to keep in touch with friends now that we all live in different places, but it's hard when everyone lives in a different chat app that you only have 2 contacts in. It was awesome when everyone just used one or two things.

The fallback to shitty text messaging is the only guaranteed medium for almost all people on one platform.

Just want to say thank you for writing this and bringing it to HN attention.

Gtalk was absolutely amazing, and you could even send files to your friends just so, just drag it into the window! It's 2018 and this functionality is hardly available anywhere anymore.

Telegram's Desktop client has this functionality.

I gave up on Talk / Hangouts a few years ago, when I kept getting into a state with it where someone would send me messages fine but my replies would be delayed for several hours. It was possibly the most frustrating computer-mediated interaction I've ever had - and I say that as a very cynical programmer! After a few weeks of it happening intermittently, I said to my handful of friends - "If you ever want to speak to me again, we have to find a different platform". Now we're on a combination of Telegram and Slack, and I'm happier. It's not as good as the world was in the AIM days, but, it gets the job done.

I don't use Google Hangouts, Allo or whatever else they produced since GTalk.

WhatsApp, Signal and Viber are the only ones I use and can reach most anyone worldwide.

Google screwed the pooch on their IM strategy long time ago and there's no coming back. Kind of like screwing the Google Pixel 2 by removing the 3.5 mm aux jack.

I don't even know someone who used to use those products mentioned here.

However, I'm still on IRC...so what do I know of this fancy new stuff.

Our entire company ran on GTalk for a while, then Slack landed like the Spanish Flu and the IRC++ like features pushed every other medium out, except for some, email. Hangouts was used in meeting rooms for a while but it was so horrendously unreliable that we switched to GotoMeetings then Zoom.

Zoom is less shit than the others but still fairly awful. Slack, imo, solved instant messaging in businesses completely, but conferencing is still a wild west of garbage solutions.

Try https://appear.in it has worked wonders for us wrt to conferencing.

Shameless plug, our startup makes a video-calling-in-Chrome application. Our goals are being super-reliable, simple, and free: https://daily.co

This happened right about the time that Google abandoned the desktop as its primary target and started focusing entirely on mobile. Things just haven't been the same.

Hangouts may be a disaster but even so it is still in my eyes the best messaging solution in that it is truly cross platform, most people have accounts (even if they don't know it, via google accounts), it supports at least some open protocols (remnants of XMPP still works) and does all the essential things (voice, video, text chat, presence, etc). I honestly think that Google's best move still would be to drop everything else and go back to Hangouts / Talk as the single solution.

The problem is that the whole space is a nightmare where none of the players actually has the best interests of the users at heart.

There are several contributing factors to Google's disastrous history with messaging services, but a significant cause is Google being spooked by Facebook.

In an earlier post of mine [2] (which includes an older revision of the timeline linked in [1]), I speculate that it was Facebook Chat that killed the mid-2000s chat networks of old like AIM, Yahoo Messenger, and WLM, rather than Google Talk or any particular missteps of those incumbent chat networks. For example, I was surprised to learn that AIM was present in the iOS App Store at launch on 2008-07-11 -- of course, there were no push notifications at the time -- not until 2009-06.

Facebook Chat went live 2008-04-06. Google didn't view them as a threat in messaging but a threat in social networking; they thought they had messaging in the bag with their 2008-09-23 release of Android 1.0, and their 2008-11-11 update to Google Talk which brought voice and video calling, and their 2009-03-11 acquisition of GrandCentral, which was soon rebranded an invite-only Google Voice. But Facebook kept growing and growing and it had an integrated chat on a website where people went to spend their time, instead of Gmail, where they went to manage email.

To combat Facebook on social networking, Google launched Buzz with aggressive auto-opt-in on 2010-02-09. Unfortunately a few months later on 2010-05-20, Android finally got push notifications, which enabled third-party messaging apps to be viable on Android. This led to the rise of cross-platform messaging network apps like WhatsApp, whose success came back to hurt Google later.

Buzz fizzled and attracted controversy for its aggressive piggybacking on Gmail, so Google tried again with Google Plus on 2011-06-28. That was a better effort, and it included the features "+Messenger", a text chat, and the video chat "+Hangouts". By this point Facebook had more than 700 million active users, and on 2011-08-09 they introduced a standalone app for just Facebook Chat, called 'Messenger'.

On 2011-10-12, Apple released iOS 5, which came with iMessages. This significantly improved the iOS platform's story on messaging, putting further pressure on Google. Having put many of its eggs into the Google Plus basket, it needed a win, which it didn't get.

On 2013-05-15, Google liberated Hangouts from Google Plus so that it could address messaging and social networking separately. Hangouts was being repositioned as the Android-default messaging app. This helped onboard new users, but on 2014-02-19 Facebook announced it was acquiring WhatsApp. At this point Google lost, and it began to flail about, sometimes reviving long-forgotten apps like Google Voice, sometimes deprecating integrations it had encouraged previously, and then making two more messaging apps.

Sources (in detail at):

[1] https://news.ycombinator.com/item?id=13465483

[2] https://news.ycombinator.com/item?id=11114518

Thanks for the timeline!

Fascinating. Within my social circles, we use Hangouts extensively. It still feels a lot like the original Talk for us. The Android app works, on desktop the Chrome extension is solid, I've never lost an important message, and the features he laments the loss of (busy status, XMPP support, off the record, contact blocking) just don't matter to us. I haven't had his experience with adding people (if I want to chat with someone, I input their @gmail.com address, done), message delivery is reliable, multi-media (photos, video) works fine, voice and video calling works, search is in the Gmail we already use, and group chats are easy and straightforward.

My social circle is almost exclusively Android, which may help, although the few iPhone users do work, and a lot of chatting happens through the Chrome extension/gmail window, so there's that.

I don't particularly want to chat through Facebook (Messenger is a busy, ugly mess, and the app is pretty heavyweight) and many of my contacts aren't on Whatsapp or WeChat. SMS often delays message sending and delivery in areas with spotty mobile coverage. Hangouts uses wifi first, which is ubiquitous in our homes and at work.

It's really interesting to see the different perspective. I do worry about the future of Hangouts, it feels like it's just coasting now. Duo/Allo/Meet may be an okay replacement, if they migrate all of the data over and swap it in Gmail. I've played with their slack-competitor (Chat), and it's interesting. Still doesn't feel as good as slack, but it's been okay.

One of the things I love about Google Fi is that I'm still allowed to effectively use Google Talk and SMS still works in Hangouts on my phone. I can do all messaging from the hangouts app with people over SMS or if they use hangouts, the message can go over that. I can text and call with my real phone number through the Gmail interface. Unfortunately, those with Google Fi are the only ones still allowed to use it that way.

Sure but the web interface for hangouts ruins the whole experience. How search functionality can be ignored in a Google product is beyond me.

You can search for label:chats in Gmail (extremely poor workaround) but this doesn't work in Inbox, forcing you to navigate back to Gmail to search chat history.

It's funny how this seems to happen to chat systems over and over. Google Talk, Skype for Business, ICQ... They start out pretty good, but then the simplicity gets abandoned by what appears to be really awful committee driven product design decisions until they basically become unusable.

Maybe their inherent simplicity makes them the ultimate bike shed.

I constantly have obvious spam accounts trying to send me Hangouts messages. That there isn't simply an option to never allow anyone not already in my contacts to message me is crazy.

Agreed with most of the points in the article. Google Talk was minimalistic, but that was the whole point of it. And it was in a point of time where Google actually tried to use an open protocol which was interoperable and extendable.

I could use Google Talk in Pidgin along will all my other IM accounts within the same client, and I also could use some plugins like OTR to encrypt my conversation if I needed to.

Now we're forced to use the Hangouts client which frankly isn't very good, and everyone else is also trying to lock everyone in their platform by putting interoperability on ice.

XMPP is still a good protocol, but no one seems to care :(

I have never agreed more to any single article ever than this in my life! How accurate! So true and the pain to leave GTalk still lingers in me to this day. I experienced my first love in that thing ffs!

I use Google Hangouts from time to time, not my favourite IM app but some of the criticism in the article is inaccurate. I can definitely tell when someone is online (article says you can't), and I have been using the desktop app for a while (article says there is no desktop app).

Note: the desktop app doesn't seem to be native (I think I got it from the Chrome store) however it's definitely a desktop app in the sense that you can run it without having a browser open.

Are you kidding me, you can't even click on hyperlinks in that "desktop" app. You have to copy-paste them manually in your browser.


It seems the links do open, but they open silently in one of my minimized browser Windows. And of course it opens in chrome even though my default browser is Safari. Wonderful quality.

I haven't stated anything that contradicts what you said

Came hoping to read some insights about how the Gtalk debacle happened and why. Left disappointed.

Great piece. I used to use GT a lot, then it changed to Hangouts and I used it less, then people were saying 'it's better to use Ello for encrypted chat' and I thought 'why bother'?

I've had this experience with quite a few Google products - they just keep getting larger and cheesier with no discernible user benefit. Everything just keeps getting bigger and worse, even search.

Hangouts also ha the worst video call quality of the bunch. It eats the battery extremely fast (vp9). There was a two year period where there was a feedback echo because the microphone wasn’t silenced, and then thrown to the side for a newer and less complete/just as buggy duo and allo. Let’s not forget the sms integration and then later deintegration.

Jabber days were good.

Google has the talent and the flexibility to make the very best instant message client that would be useful for everyone... but they fucked it up somehow.

It means they really don't care, which shows in their product. My chats often get lost or delayed. I can't control my "contact" list, and there is no reason why I shouldn't be able to except they don't want to allow it. I see my friends who have died 3 years ago still on it, and I can't remove them unless I block them. If my contact is "Off the Record" they don't even show up in the list, which is infuriating.

I would love to know the reason why they let Gmail and Gchat go. My killer app for the Mac is iMessage, mainly because it integrates my text messages and Gchat into a single client. It's so convenient throughout the day for me.

Wtf are you talking about? I use it with almost everyone, inside and out of the org. Issues are usually local internet connection issues. And it's free, which is nice. There are pay for services I suppose are better, once others are set up with it. But I love Hangouts.

I haven't had much problems with Google Hangouts but on the other hand I'm a light user: I just wanted a real-time text chat and didn't suffer from most of the problems decribed in the article. Yes, the video chat was choppy but so was any other video call at the time; I still managed to talk to my wife over video on business trips so it wasn't unusable. I still like the Gmail chat that you can pop up into a separate window. It's fast and reliable.

However, the issue with Hangouts is that I don't have people there anymore. I talk to 1-2 friends on Hangouts who refuse to use Facebook or Whatsapp. That's all nice but ten years ago I had a significant portion of my friends in Google Talk. They still have Gmail accounts but they chat elsewhere.

Wow, I never once in my life just pinged a stranger on Hangouts and it seems to me like a incredibly obnoxious and rude thing to do. Nor did my i5 laptop flip out because of hangouts in Gmail.

I'm not a big fan of Hangouts, but if it stopped someone from pestering strangers online? Good on it.

It died when they killed XMPP federation (by refusing to support server to server encryption).

Google+ has been quite a bit of a disaster. They wanted to tie everything to it, and it ended dragging down everything else. Nowadays, for a quick chat, Jit.si seems to be the easiest thing - it just works, you send them a link and get online.

You know what's even worse? Google Voice.

Terrible web-app, even worse iOS app. So many times where messages were duplicated, or frozen, or stale. As in requiring a reset of the app to fix.

At least hangout was stable despite having equally poor UX.

Continuing the train of thought of the comments bemoaning the state of the instant messaging: can anyone recommend a simple instant messaging / voice call combo like the old-timey Skype?

I use Wire[0] now. It offers IM, voice messages, voice & video calls, & a decent set of unobtrusive "silly" features (voice filters, ability to send drawings, etc).

Finding people is generally easy but for some reason, it will sometimes not find someone when you type the entire username but will find him/her before you finish typing the username out.

The best part is it works well on Android without GApps & seems to work much better than Skype in video calls on a slow connection.

0 - https://wire.com/en/download/

I've been using "telegram" with several friends, it seems to be okay, it's both desktop and mobile

I worked at Google. An observation:

The internal codebase should be structured like the yellow-pages of a phonebook[1] but instead it's structured like a flea market.

Now, consider that in light of Conway's Law[2]...

[1] https://en.wikipedia.org/wiki/Phonebook

[2] http://melconway.com/Home/Conways_Law.html

People on here bemoan the loss of XMPP integration and federated access to these platforms, but do Facebook/Google/Microsoft really give that much of a shit about that? I suspect the % of their userbase who wants, let alone understands those features is very small

Of course they don't give a shit. Because they understand the game - they're the ones putting things on the market, and users have zero choice (network effect means good marketing beats any feature set here).

Yes, the clients suck, and it's a pain to deal with so many different messaging apps.

This is one more bit of evidence that Google is on the way down. When basic product management is too boring to take seriously, the company is just coasting along on past glories and hype. Yahoo was in that situation for years before it imploded.

That's funny how a huge company may bring a technology to the mainstream and then kill it just like that. Jabber, RSS. Open protocols that treated their monopoly. What's next on the kill list? HTTP, email?

The following are postulations that I have not managed to refute since 2013.

Google clearly knows what they have made. The theory that nobody cares to fix this is false. They have good engineers.

Google was always tracking our location and scanning our emails. Few things happened simultaneously in 2014. Google tried to replace the messaging app with hangouts, blatantly stepping on privacy. Inadvertent signing ups for Google+ could occur. I switched to Cyanogenmod.

Navigating new privacy settings would took so long. Then Google maps started suggesting where I intend to go.

Not very creepy. It seems to me that Google intend to do the right thing, which is to declare publicly that our privacy is compromised, (e.g. Snowden) when we use Online centralized products. That is a good thing. I stopped trusting my ill-founded sense of privacy.

I lived in Canada, perhaps the timing was different in the states.

Google needs to streamline their communication product in to one that works across all platforms. And also need better product manager that understands that choices are bad. And not working choice doubly so.

I have a folder on my iPhone for all the various google chat apps, allo, duo, hangouts, meet and the list goes on and on.

Not an application, but a whole folder filled with chat applications!

All my friend have moved onto slack.

And, as if to further illustrate the problem with dysfunctional WWW stuff, the ony way to comment on that article is via Disqus...

I miss google talk but it was just chat. I use hangouts this days for chat, voice and screen sharing and it works well enough.

The things that sucks most is the fact that everything is recorded and it can't be erased. Goodbye privacy.

Totally abandoned all chat applications after the demise of Gtalk and cross network communication breakdowns.

I thought talk and hangouts do the same things, google needs to do a better job in marketing

Google Voice was similarly “updated” over time and is very good anymore

I don't use Google Talk, so I don't know how accurate the article is.

But if it's at all true, it should become an HBS case study in how to destroy a community and how not to write software.

Same way they managed to screw up Google voice.

Shhhh, don't give them the idea to cancel my voicemail transcribing app.

Or does voice actually have other features?

How many product managers do they have? 50?

All these people bemoaning the loss of jabber. Xmpp is awful for mobile connections. There's a reason no one else is using it. Google tried the longest and they get the most flack for it.

> There's a reason no one else is using it.

because they cannot own the client, and therefore, cannot completely collect the valuable part of the service - user data and activity data etc.

Applications are open for YC Summer 2018

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