Slashdot Mirror
Ask Slashdot: How Would You Solve the Instant Messaging Problem?
Artem Tashkinov writes: The XKCD comics has posted a wonderful and exceptionally relevant post in regard to the today's situation with various instant messaging solutions. E-mail has served us well in the past, however, it's not suitable for any real-time communications involving video and audio. XMPP was a nice idea, however, it has largely failed except for a low number of geeks who stick to it. Nowadays, some people install up to seven instant messengers to be able to keep up with various circles of people. How do you see this situation being resolved?
People desperately need a universal solution which is secure, decentralized, fault tolerant, not attached to your phone number, protects your privacy, supports video and audio chats and sending of files, works behind NATs and other firewalls and has the ability to send offline messages. I believe we need a modern version of SMTP. [How would you solve the instant messaging problem?]
People desperately need a universal solution which is secure, decentralized, fault tolerant, not attached to your phone number, protects your privacy, supports video and audio chats and sending of files, works behind NATs and other firewalls and has the ability to send offline messages. I believe we need a modern version of SMTP. [How would you solve the instant messaging problem?]
If everyone stopped messaging on insecure lines, the problem would solve itself.
"People desperately need a universal solution which is secure, decentralized, fault tolerant, not attached to your phone number, protects your privacy, supports video and audio chats and sending of files, works behind NATs and other firewalls and has the ability to send offline messages."
I don't see the sense in that. There's so much evidence to the contrary.
May as well say people desperately need a universal language. May I interest you in Esperanto?
I keep it simple and don't add to the problem by using more chat software. I use e-mail for work and longer messages, I use SMS/text for shorter messages with friends. That's it. I'm not going to get bogged down using a dozen different messaging services. If I really need to get in touch with someone I can always call them.
Obligatory XKCD
Do we really need to improve how our electronic devices constantly interrupt our lives? I actually kind of like the fact that there's no universal IM platform because it serves as a throttling mechanism. I imagine a lot of slashdot'ers put time into filter/sorting their email because of the sheer volume you get. Now imagine that with instant messaging. No. F*cking. Thanks.
https://www.caminodesantiago.m...
It's simple really, everyone should be able to accurately identify everyone else without being identified themselves. Surely someone could do ... something with public key encryption ... ?
This is pretty low on my list of wants. Lots of other shit way more important. How about a universal translator? That would be cool. Maybe if everyone could understand each other there would be less war, maybe? Eh
What problem? I'm glad to not have to have these damn kids on my lawn spamming me with celebrity gossip through Snapchat and Instagram!
there are clients that support a lot of services the rest is up to the service itself mostly.
THere's already a solution for that: XMPP
The reason we don't see it is that the people that _are_ capable of supporting the necessary services behind it (like... for people that don't run their own servers) is that it's difficult to monetize. AIM dropped open support because too many folk use Adium or Pidgin with it, rather than the AIM client, and thus AOL couldn't push ads down it.
Google chat uses XMPP essentially... so if you want a well supported platform, that's it right there.
But more important, the requirements listed are simply out in left field. Video and audio are not instant messaging requirements. Video and audio are both, by their very nature, dealing in linear time. They cannot be "instant".
Cut back to a more realistic list of "instant messaging" and you have some hope of finding a solution. Perhaps accept that "secure" isn't as necessary, too. If you're dealing with top secret things, or assuming that a message that claims to be from your boss telling you to do something expensive or stupid RIGHT NOW, then maybe you shouldn't be "instant messaging" in the first place. Or at least not trying to shoehorn your critical security issues onto an application that most people don't need anywhere close to that level of security for.
When we have the technology to communicate instantly, all we talk about is the ability to communicate instantly. The technology only solves problems that can be solved instantly, and we have a world full of giant social and economic problems that can never be fixed in 140 characters or with an email or with a Facebook Like.
To quote Radiohead: Idiot, slow down.
This has been going on since I used VMS to send chats over BITNET. No end in sight.
What we need is interconnecting routers that manage the protocol translation invisibly behind the scenes. Then it wouldn't matter what chat program you used, it would be routed to the correct destination. Of course the entire point of proprietary is to lock in your customers so you don't need to spend resources competing with others.
SIP is ideal solution, if only people bothered to use it.
Stumbled on this. Seems interesting. Can anyone tell me more?
https://briarproject.org/
That's all you need.
with its circle drawn with a breach in it?
Every programming language implementation eventually includes a half-baked lisp implementation, every non-instant messaging implementation eventually becomes a half-baked version of SMTP, and every instant messaging implementation becomes a half-baked version of IRC.
Rather than starting over, it would make more sense to extend what we have. For instance, the medical field has a goal of eliminating spam and insecure transmission of medical information by creating the Direct Project, a network of curated SMTP servers with a pool of trusted certificates, who are contractually required to only create accounts for people whose identities are known to be in an appropriate medical field. Delivery confirmation is mandatory, as is end-to-end S/MIME encryption. To support the latter, there are number of different methods for locating a destination provider's address and certificate. (Then they applied a ridiculous amount of violence, sorry... XML and SOAP to the damn thing, but at least their intentions were good)
Likewise, Matrix is basically IRC (the canonical pubsub implementation) with a number of added features like federation, decentralized message storage, encryption, and so on.
Just that everyone wanted their own damned proprietary protocol, and the users allowed it.
The future is that the various walled gardens will become ever more powerful. So that you can use attachments, search messages, do everything with them that you can do with email. Why would kids want to bother with clunky uncool Email?
If you want to communicate with my daughters friends, you just by an iPhone so that you can use iMessage...
People desperately need a universal solution which is secure, decentralized, fault tolerant, not attached to your phone number, protects your privacy, supports video and audio chats and sending of files, works behind NATs and other firewalls and has the ability to send offline messages.
So why don't they flock to XMPP then? Anything that there isn't a at least an extension for it?
bickerdyke
Seriously, it is just like SMTP in that it is a protocol.
Want it to be a killer protocol? Then extend it for not just IM, but using it for other forms of messaging.
More importantly, develop the apps to run on all the major platforms (linux, androi, osx, ios, wind, bsd, perhaps mainframe).
I prefer the "u" in honour as it seems to be missing these days.
Pick up the phone?
" People desperately need a universal solution which is secure, decentralized, fault tolerant, not attached to your phone number, protects your privacy, supports video and audio chats and sending of files, works behind NATs and other firewalls and has the ability to send offline messages. "
Well, this just sounds like the web...
Something like google docs allows multiplayer notepad, so, many people using a simple AJAX page with some widgets could do just fine.
Google wave was a valid attempt at solving this problem yet google didn't allow it to settle.
There is XMPP but there are too many vested interests.
Matrix is a federated chat protocol, similar to XMPP/email, with bridges that connect to various chat services (similar to XMPP) but uses modern web based protocols and practices and doesn't have the negative stigma that XMPP has. Other options of course are to use multiprotocol messengers such as Pidgin or bitlbee, which support most of the protocols in the cartoon.
Check out matrix.org. It is not only a rich IM solution with all the bells and whistles, including multi-devices end-to-end encryption, but Matrix also provides for bridges and proxies to other networks, so that it can be used to unify communication.
It's only 2.5 years old but has already come quite a way!
https://matrix.org/
echo mailto: !#^."<*>"|tr "<*> mailto:" net@madduck
It supports many chat systems by just running custom web access layouts to all with web support. It makes all the chat systems in to one app through its tabbing system.
That's a long list of stuff that's mostly been solved. What we need is interoperability and proliferation of a standard, like how HTML is handled.
We don't need everything in that list for every usecase either:
Study the US and other brands the CIA was able to read in plain text and don't do what they did in hardware and software.
Learn form how their staff designed their hardware and OS. Think about how their very best internal security reviews found nothing.
Think of how the CIA got in, stayed in, was not detected and why staff could not, would not or did not go looking for such issues.
A lack of imagination? Skills? Budget? Staff understanding the product or just the parts they worked on? Other issues that distracted the staff from security issues? No ability or skills in house to imagine, ponder, consider or observe or find something like the automated implant branch (AIB)?
Can crypto be trusted given the CIA, NSA and GCHQ interesting in communications over the decades?
The past weak, junk US cryptography been offered as a standard is now really fixed? A few academics at some international conference published a book chapter on how perfect all crypto now is? Just like academics did in the 1970's?
Or do the security services have a few ways in? Junk hardware designs, the weak tame OS, the software and crypto that offers up plain text every time?
The world is yet to have a walk out with the modern crypto side details so the big brand crypto is still trusted?
How to get your staff to understand how the US and UK think about crypto?
Buy books that detail the past efforts of the US and UK crypto efforts from say 1910 to 1990.
Embassy codes, telephone, fax, global communications, cell phones, the home PC and the internet in the 1980-90's.
Over time the USA and UK weakened or provided junk crypto to all products sold in the West.
To get privacy back include a comic or cartoon about creating a one time pad message on paper and entering that as the message.
That returns privacy.
Given the GCHQ and NSA have global reach, anonymity is not really an option on most networks.
Domestic spying is now "Benign Information Gathering"
solves all the scenarios posed.
Seriously, we have the protocol. We also have 12 competing standards from 12 different walled gardens, the only one that added anything of value being Signal.
WebRTC is essentially this just need a decentralized account system
It's not hard, just don't install them or just the couple you want. Or, just use SMS that works with everyone everywhere...
...you just invented iMessage!
fake foresight
One product that has been around for a while is Trillian but this has definitely not caught on nor does it integrate well into the idea of a unified desktop/environment
- https://www.trillian.im/
Until there is a more universal method that supports instant messaging to an address (similar to email) but I see this a highly unlikely as the providers do not want to release control of their user/supporters/etc as they are a market that can be advertised to directly without a very low bar to exposure.
Maybe IEEE should take a crack at defining a protocol?
https://xkcd.com/927/
If they are they will be willing to pay a dollar or two to buy and install a client and refuse to communicate with ad supported spyware. But privacy is over valued by a small section of techies, who shout very loudly. Most people sell their most private and intimate info about themselves for 25cents off a loaf of bread. For another 25cents they will sell info about you too.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Email and voicemail got replaced by text. Long docs via email are fine (I play online games via email), but most of my email nowdays are retailers who think they have a relationship with me but I never look at them unless I need to buy underwear and want a discount. Voicemail? Fuck off. Yeah, I can have android do email to text with hilariously bad results. Fact is, you leave me a voice mail and I pretty much ignore you unless I recognize the sender. Voice mail for me. Long, drawn out voice taking a good minute telling my phone tree options.. I enter my phone number. Long, drawn out voice. I enter my code. Long, drawn out message, I enter my code again. Long, drawn out message, I hit 1 for play. Long, drawn out rambling message from some dipshit asking a stupid question. I tell them they're a dipshit. Long drawn out message saying my mailbox has 2 more messages. Repeat 2 more times. Long drawn out message, my mailbox is mt.
Text version? Get a text. It's a stupid question. Might also say why in 142 characters.Delete, or answer, whatever. Problem solved in seconds instead of minutes.
Voicemail used to be great, now it's the refuge of people who can't condense their thought in less than 30 seconds.
I had been doing fine for years with multi-network clients like Pidgin/Trillium/Adium until semi-recently when big players started kicking 3rd party clients off their networks.
You could be that guy. You know the one: the one who tells all his friends "This is what I use. Use that to contact me, or e-mail me instead."
For the most part, I'm that guy. I use one IM program for personal use, and another for professional use (due to corporate mandate), and that's it. The only exception to this is as I do have a Facebook account, if someone wants to message me there I'll accept these messages as well -- when I'm at my computer and logged into the web interface. I have no intention of installing their Messenger client on my mobile devices.
Then again, I don't feel the need to have people messaging me all day. My messaging contacts list consists of about four immediate family members, and that's it. Guess I'm just not social enough for "social media" and IM (for that matter, I don't own a cell phone either. I go out not to be disturbed by IM and phone calls -- why would I take the annoyance with me?)
Yaz
There *is* a standard, problem is vendors would rather lock you into something proprietary. That won't change no matter what standards exist.
Use XMPP because creating yet another protocol just fragments things further.
Get over it.
A little bit of chat is pretty easy.
Inevitably people clammer for more.
A robust extendable framework requires a boring standards like approach.
It's a similar dilemma with graphics.
To get decent funk, xml is required to deliver robust svg.
Don't forget, the tools implementers requires robustness of spec to deliver reliability.
Look at the astonishing tool that is Inkscape.
XML has a place.
I lived in Gotham where there was no XML.
The summary gives two good answers to its own question:
1. Use non-proprietary, open, universal protocols. There's a reason why SMTP works so well -- nobody owns it, everybody supports it. Unfortunately this provides no path for some entrepreneur to take over the Internet and become the next trillionaire, so nobody's going to put much work into making it into an easy one-click app. You may have to do some work yourself, both deploying and promoting your chosen solution.
2. Install seven apps. This seems to be the solution that most people prefer. If you need to be babysat by corporate nannies, then eat what you're served and enjoy it.
The problem isnt a technical one, it is a business one. More specifically, every business giving you a "FUCK YOU" attitude when it comes to interoperability with different platforms. Facebook Chat? That was XMPP. Google Chat? (that thing before Hangouts and Voice), also XMPP. Countless other systems out there are XMPP too. It works. It works GREAT. There pretty much wasn't anything wrong with it. Then businesses were like "FUCK YOU", and decided they didn't want to cooperate anymore, and so it died.
First World Problems :-P Can't we fix the million people a year starving to death first? Just a thought.
If you go back to the RFC, you'll find SAML and SOML as smtp keywords: they mean deliver as mail or immediate message (unix write(1)) or as both mail and IM.
davecb@spamcop.net
We have no instant messaging problem. We just have a robust constellation of competing systems, serving different communities. Why is that as problem?
- My teen has a Snapchat community, an Instagram community, as Facebook community, a Pinterest community she hides from me, a Twitter community she denies, an SMS constellation, and a variety of less visible communities gathered around video, music, photo, and mixed media paradigms. Some of the members overlap and are pasrt of several communities, some of these communities serve specific purposes, some are flash mobs instantiating and disappearing quickly. She manages her various communities by platform sometimes. These are 'where she is' at any given moment, sometimes in more than one place at a time. Oh, and she has email too. Several of them.
- I don't want a messaging platform mixing my Facebook and G+ communities. Leave them separate. Some overlap occurs, but I can manage that.
- SMS is not very useful on my desktop PC You want to do some Universal Inbox of 'Follow Me' concept for 'messaging'? Please don't.
- I get messages from entities also. When Amazon delivers an order to me, I get SMS, an email, An Amazon app notification. I got one when it was scheduled for delivery. And when it was 'shipped'. And when it was ordered. I get 12 messages for that one order. If I ordered multiple items from different fulfillers, add 9 messages for each different fulfillment channel. It pollutes my life. I turn some of them off, and they creep back in. Multiple apps send me notifications. They are 'messaging' me. Some let me turn off notificaiotns, abnd they keep right on sending them. Some 'apologize', they blame their own app, most ignore me. The cost of 'free' is real.
- I rarely use or send videos. They are horribly inefficient for simple, spoken or written communication that does not require visuals, and I loathe how-to videos that waste 70% of their duration on establishing shots, personal anecdotes, uncomfortable drivel, wasted time and noise. Give me a step-by-step please. A list.
- Email is highly underrated, still. I carry on conversations in email very well if the correspondents keep up. At work I get IMs from the loathsome Skype For Business client I'm given, and despite the 'instant' intention people regularly turn away and let a chat linger for minutes. Instant is the behavior, not the app. Email is better than you think.
- I'm guessing the real complaint is having to manage the address books, friend lists, etc. that these platforms use. I refuse to use my Facebook/Linkdin/Google Contacts to log into multiple platforms. I don't want to share my contact info in Facebook with my Linkedin community. Or with Google, G+, Pinterest, etc. I have good reasons to keep separate communities separate.
We do NOT have a problem with proliferation of messaging platforms. If you think you do, leave some of them. Everyone you deal with online is either a member of more than one of your communities, or they are as member of one you will keep.
No problem.
deleting the extra space after periods so i can stay relevant, yeah.
Signal is trying to be the answer. They certainly cover instant messaging, and I think they're adding video and audio. I'm not sure about email, but like any encryption system, it only works if you get both ends using the same system, and to use it, you have to go with their interface. Maybe with plugins for Outlook, GMail, Thunderbird, and whatnot, it might work, but it still requires everyone switching to it.
And none of that fixes the spam problem.
"Perhaps accept that "secure" isn't as necessary, too."
This message brought to you by your friends at the NSA, CIA, and other organizations that are eager to learn more about you.
I think there's nothing wrong with considering what security means here. I'd certainly prefer non-technical people conversed electronically using a protocol and free software programs which used encrypted message transfer by default. I don't think it's wise to continue in the older way of doing unencrypted message transfer for everything and then being unpleasantly surprised how many parties spy on users. How much trouble one wants to go to in order to ensure a desired level of security should be a concern and part of the needs requirements. I'm also unconvinced that multimedia can't properly be a part of IM. Every message in any medium takes time to compose and read, watch, or somehow deal with. I take the "instant" in instant messaging to be a (somewhat inarticulate) description of the ease with which one can send a message of any kind to someone else. In other words, minimal setup with a simple UI, not the medium of the message. My understanding is that younger users expect to be able to make and deliver short audiovisual messages, so I'd give such messages more consideration than the parent suggests.
Digital Citizen
I use Slack (for work), then Facebook Messenger and SMS for personal stuff. Don't feel the need to use anything else.
Torchat - Anonymous encrypted communications
Until enough users are tech savy enough to demand that systems work cross platform, each company will continue to carve out their own feifdom full of surfs who live and die by their rules. As soon as a couple of companies (WhatsApp, IM and skype, for example) start to offer cross platform functionality, those who don't offer cross platform functionality will be progressively marginalized. Then it will be just a preference of what UI you like better.
If you disagree, please post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like
It's like saying it makes no sense to have a house way across town when you could rent a living space at your work. People don't want to live at work. They want natural barriers between the different aspects of their life and the groups of people associated with them.
For those who keep active Snapchat, Facebook, Instagram, Twitter accounts, etc., one of those will be how they casually chat with close friends, one will be how they keep in touch with family and work friends, one will be where they flirt, one will be where they post about their political concerns, etc. Having distinct non-interoperable services suits them just fine.
When things get complex, multiply by the complex conjugate.
Tox is fine so far.
I'm using it with close family members.
I solve this problem by never having a total of more than about five contacts.
The only reason XMPP failed was cause everyone (ie: Google, Facebook, etc) cared more about user capture than interoperability. This is all the more obvious considering that they HAD support for it, but then killed it off for no reason.
XMPP supported almost everything except possibly real-time video. Well, apparently the protocol itself *does* support it, but because no one actually cares, it's never seen the light of day. At least I haven't.
Everyone conveniently ignores the fact that because HTTP was a universal standard, it allowed a *ridiculous* variety of tools and systems to be developed on top of it. The internet as it exists today, wouldn't, if not for that ubiquitous standard.
But as usual, lessons in history pale to short term profits.
Of course, both were made obsolete in RFC2821, sixteen years ago. While there were such commands, there was nothing that meant that email was "instant", even if it could cause "messages" to appear on someone's terminal.
Given that "email" in the time of RFC821 was often transported via UUCP, claiming that "email" was designed to be "instant messaging" is just silly. To claim that it is not a good medium for "instant messaging via video" is a very incomplete and misleading statement.
What problem are you saying exists? Other than what is in your lame mind.
More than half of humanity has a cell phone, and you can call them instantly if you have their phone number. No additonal protocols or servers are required.
Now get off my lawn.
If the problem persists despite wide availability of your proposed solution, it's not actually a solution.
That's what I felt when the boss at a previous job wanted us to use instant messaging when there was already a phone for urgent things and email for non-urgent matters.
Was one of your co-workers deaf or hard of hearing? If so, the use of instant messaging for urgent things may have been an accommodation for your co-worker's disability, pursuant to the Americans with Disabilities Act or foreign counterparts.
I am signed into Skype using the third-party SkypeWeb plug-in for Pidgin, which supports the JSON protocol that Skype for Web uses.
Back when there was AIM, MSN, etc., a chat program called Trillian came along and let people use the major chat platforms with one client. Then there was an open source one called Gaim, which was later renamed Pidgin. Then they tried to standardize it with XMPP and such, but that died down for some reason. Pidgin is still around, and supports Facebook messenger. We're back where we were 15 years ago. Now get off my lawn.
Email has different design priorities than instant messaging, which is why all of these instant mesaaging protocols were created after email was already popular. Possibly the biggest difference is that email is designed to be reliable rather than instant - when a hop is down, it'll keep trying for hours or days. Your email client checks for new messages every ten minutes or so - that's much more efficient, and obviously very much not instant.
You mention a field something like "preference: instant" which would presumably cause all of the servers involved to use some different protocol. At that point it's no longer email.
(already mentioned above by some anonymous cowards)
https://webrtc.org/
https://www.youtube.com/watch?v=p2HzZkd2A40
Isn't this why the government and Skype collaborated so hard? You mean it didn't work out when people found a way to track other users IP's? gasp who would have thought. Centralized communications only really helps companies and governments to track users, it's not in their own best interest to use one solution.
Point to one of the current solutions that are proven to work and to scale. The server side is done, and they call it XMPP/jabber. ejabberd can run on a thousand nodes without a hitch, and you can have multiple networks operating together as a federation. It's distributed and webscale, but perhaps that is the problem. If it were centralized and proprietary you could create a business around providing IM as a cloud service.
The client side, needs a lot of work. But aren't there a million web and app developers just sitting around trying to implement the next big thing in user experience? What is fundamentally wrong with XMPP that prevents them from doing this? Because I don't see it, and I think it's too soon to say that XMPP is dead. It's a fine standard that can be extended in numerous ways. Perhaps a new standard for the sake of giving the impression of newness is all that is needed.
It's a shame BlackBerry dragged their feet on porting their BlackBerry Messenger to other platforms. They were the gold standard for simple secure messaging in their prime. BBM today is an excellent messaging system that nobody uses because they don't know how good it is. I had high hopes for Jabber er...XMPP but as others have pointed out companies couldn't figure out how to monetize it so they ultimately opted for a walled garden solution.
"A person is smart. People are dumb, panicky dangerous animals and you know it." - K
Quick Ideas I will roll with. What to use instead?
1. Fidonet FTS-0001 - http://ftsc.org/docs/fts-0001.016 (bluewave/golded + pgp) (pgp gorilla? with the LARGE keys..)
2. cw on am(fm, usb, lsb) CB radio (it's dead right?) using a couple resistors and a 4PDt -- ya can isolate the RF vs Audio just like the ham pros (sic) + dm-780 and now you have most of the protocols. CQ DX de yourfunkandjunkhere
2a. On the little FM broadcast transmitters for your car... SAME THING. hook it to the DM-780
2b. VHF radios hack the microphone port. Same thing. Get tat signal down to below 1v PtP
an ol IC-701 it don't have data, but NOW SHE DOES, want me to send you a BINARY FILE?!
3. Run a LOG grepper. Run a web server. Have your firenads hit some 404 errors which have nice long sentences. Testing 1,2,3 (text only but if you run a lamp / namp you know what I mean about LOGS. You could type in http://slashdot.org/-=[ I hate ya ]=- and that will show up in the logs "if slashdot's sysop looks." the weird charac-tator-s won't look prretty but they'll all be there OR you could type in http://slashdot.org/-=[ the article about the 3D printers has a 404 in the story--fix it fast!!]=- and the sysad would be head's up'd he'd know who did it by the -=[ that would be me doing that to your logs you see now you understand. ]=-
You could look at it like this.
browser gets a 404 to confirm the message has been sent. you could script your proxy so that a 404 says, THANKS FOR YOUR FEEDBACK, YOUR MESSAGE WAS SENT. ( i know you hackers gotta love this one ) Router people rejoice. Probably got this junk built in. OTP anyone?
4. IRC (lots of nice things goin for this, use as many tricks as you can imagine)
5. KDX (dusty ol mac users know what I mean) a bit aged, I wonder what's wrong with it's CRYPTO?? maybe we can layer something on it, or it IN something.
6. dead drop + that new usb condom thingy to guard against firmware attack..
7. PirateBox - old but effective.
8. Direct Line of Sight a mile or less? Use a Wooden DISH (Yes same shape as LARGE OLD satellite dish, Face them at each other and then talk into the bowl standing in front of it.) Of course you would need to coordinate the TIME, or have someone standing there all day and night.
Okay what do YOU THINK?
But additionally, that was the POINT of that factoid in regards to the Babel fish: If people CAN fully understand each other, that makes them MORE likely to kill each other, not less, because the culture and values they support are likely an affront to the other culture and its values. This is especially true when one culture would be considered open and accepting and the other is considered closed and xenophobic. The latter group will often attack the former under threat of their culture being absorbed and obliterated, even at the same time as the latter is trying to convert members of the former into willing vassals of the themselves.
This is exactly what Google Wave intended to be - a federated extensible decentralized instant messaging service with enough features to replace the features of every other existing messaging service. Their server/protocol implementation was deeply genius (built on top of XMPP) but unfortunately they were cut off way before the client was particularly practical and before federation could be realized. Had Google given Wave a decade to evolve I think it could have solved this problem - but there was no profit in solving this problem and early adopters struggled to figure out how the very rough (but innovative) client fit into their lives.
No company will ever build something as decentralized as SMTP or IRC because the business model of the web involves owning the eyes of a set of users. Allowing those users to jump ship to another service provider using the same protocol doesn't help the bottom line of any startup. On the other hand - open source efforts don't seem to be able to get enough developer attention, and I assume it's because developers know that not enough people will use these alternatives if the big players like Facebook/Twitter/Google with the massive user bases aren't going to cooperate. Wave and diaspora* were both innovative attempts to do decentralized versions of currently centralized communication tools but neither ever had sufficient traction among developers and early adopters.
Who the hell wants to be in INSTANT communication. I don't. I have many irons in the fire, all of which are more important than just about anything that might pop up in an IM window.
Send me an email, and I'll answer it when it is appropriate.
Instead of having a new messaging app, just have a contact app that remembers what app you use to get in touch with everyone. When you want to message someone, the contact app will open the relevant messaging app for you. Receiving messages can happen through your phone's/PC's notification system, so it's not so much of a problem.
People I care about know how to contact me. People I don't care about may not. I'm fine with that.
Case in point.. I have a POTS line at home that keep around for the very occasional fax I need to send. When my "home" phone number rings, most of the time I don't bother answering it. If it's anyone I care about, they'll ring my cell.
Universal anything is a bad idea
I knew ya could.
Most people who have to exist on differing networks do exactly this.... Especially when we have to be on walled networks for work.
uses email as the handle and sending your chat recipient an email of the chat that he or she can responsd to directly.
Supports email reply processing, direct chat in some email clients, or click to a websocket capable Angular chat client.
Go ahead, try it!
... like a fucking phone???
Idiots ....
I rarely read replies, it's my opinion and if you thought about your opinion a little more, I'm OK with that.
http://www.cisco.com/c/en/us/p...
That's the solution. It's well documented, it's a federated protocol, it's already here.,
Of course, you need to convince Google to restore it and people to give it a try (again).
With the Google part being very important!
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
I solve the "problem" by not being connected 24/7. This thus removes the "problem" of constantly having my 2-bit attention span interrupted.
Or did you refer to any other so-called "problem" with IM?
Want to contact me? There's always e-mail. Which I can use on my computer, no telephone number needed (or on my mobile phone if I want to). Which I can mark "unread" when I want to come back to it later for a reply or action on it. And I don't need any proprietary app - anything that is nice to use goes.
By the way, if I have my phone's data switched on, I get notified as an e-mail comes in; same with my computer - like any other IM. I would reply using the same little keyboard as I would for an IM.
Not "instant" enough for you? Well, you probably also save hours and hours of time by typing "ur" instead of "your" and "k" instead of "OK".
Free, as in your money being freed from the confines of your account.
So you want an updated version of SMTP, meaning that you can use the server software implementation you like, on the server you like, and with the clients you like, and presumable across domains, etc. - but for chat, and presumably for security.
Well there you go - that's basically what XMPP is.
But here's the thing... It was created too late. People know the name LINE or Google Hangouts or whatever, they don't know the name "XMPP". And no large popular messaging system other than google has shown an interest in actually using something based on it. (IBM also uses it for their SameTime Chat software supposedly, but that's mainly company internal chat).
I agree it's a good idea, but the average end user is more interested in the community than the technology, so they will just go where their friends are. Companies have to want to do the right thing, and they don't everyone wants to keep their users for themselves. It's like AOL mail, all over again.
Is it an instant messaging problem or a contact management problem?
Never trust a man in a blue trench coat, Never drive a car when you're dead
https://xkcd.com/927/
is the logical conclusion
Too bad they both kinda suck, and only support a subset of these proprietary protocols.
All these companies want to keep their users inside their walled gardens. They will do everything to prevent open access to their valued users.
Might makes right irrelevant.
The biggest concern with XMPP is of course that you need to access a public service, which may be shut down at any time - unless you run your own XMPP server (I do). Another concern is privacy. Even if messages can be encrypted, meta-data about who is communicating with whom may leak to those who have the powers to listen in on the IP packages to the server(s).
For whistle-blowers, journalists, and anyone living under repressive regimes, there are not too many options today. I am working on my own solution, allowing IM over the Tor network, using the legacy Tor Chat protocol (https://github.com/jgaa/darkspeak). This takes care of the privacy concerns - but it adds another protocol to the mix.
Having played with RCS on IMS it goes a long way towards a standardised general solution.
Having played with RCS over IMS it goes a long way towards a standardised solution.
Before we can move everybody to one platform, one needs to be developed. At the office, we are talking about setting up our own internal IM server, but so far haven't been able to find one. We've looked at:
- XMPP and SIMPLE, but we are a Windows shop, and those assume a Linux server, or have a Windows version that's basically a recompile of the Linux version, which ends up as a program to be started manually, rather than a Windows service.
- Commercial servers with a free version. These always end up requiring some obscure or insecure software such as Java. The last one we looked at (which looked great from their own presentation) wanted to install in C:\Windows and have its data files in C:\WhatEverTheProgramWasCalled. That directory layout hasn't been used since Windows 3.11, yet the software was presented in a modern, responsive, five words before you need to scroll down web page.
- Some software exists only in a cloud version. Sorry, we used to all use MSN Messenger, which we lost when Microsoft decided to close that down to force everyone onto Skype. Not going to make that mistake again, this time we want to be in control of our own server.
actually, i think xkcd missed one:
"messaging on random /. threads"
That's just delivery. There's no talk of it appearing in a client and facilitating a live discussion. The reality is email is a digital version of snail-mail. It's been with us for decades, and it's going nowhere.
The problem is IM has no standards, and those that control the de facto protocols lock away the specs. This is why most people IM through crap like FB, twatter, snapcrap et al. The control freaks like MS do not want to work with developers in an open world, so people drift through services they happen to like that provide IM functionality that's just about good enough.
It's that simple.
The *problem" with other channels is: their signal-to-noise ratio keeps decreasing to the point where they become unusable. I don't care about social interaction, photos of your cat, children, lunch, dog, grandchildren, and so on. I moved from ICQ to Skype to Fring to BBM to Hangouts, because all the people I care about can (and do) use them. Now I am trying out Allo, waiting for Google to pull the plug on Hangouts. I am not using whatever you think is "better". There is no "better", if all you want to do is send and receive information. I am not choosing a channel based on its entertainment potential, we have TV for that.
If I have something important to say, I will find a way. If not, I don't care about your funny pictures.
I don't mind paying a couple of pennies if necessary, if I want my message delivered. I don't care if your messaging system is cool, is new, is free... If you think your message is not worth 15 cents, then don't send it.
Nowadays, some people install up to seven instant messengers to be able to keep up with various circles of people. How do you see this situation being resolved?
I see them dropping some of those circles of people or more likely those people will over time converge to towards one platform. Also remember Zawinski's Law.
Instead of having a new messaging app, just have a contact app that remembers what app you use to get in touch with everyone.
Oh hells no. If you want to use some off the wall messaging app knock yourself out. I'm not going to be bothered to try to keep track of everyone's preferred app on top of the rest of the mountain of data I deal with in my life. If you need an app to track everyone's communication preferences you are making the problem worse, not better. If you have special needs (hard of hearing, etc) I'll make a special accommodation but otherwise you're just being a pain in the ass.
I use email and phone and text messaging because I can reach everyone I need to reach through those. If I can't reach you through those, we don't have a relationship worth worrying about.
The fact that there are many networks is not, by itself, the problem. It's a fun thing to joke about, but not truly bad.
The truly bad thing is that as of March 2017 there is not one network that is secure, reliable, and popular.
Non-free-software networks WhatsApp, Kik, Allo, Facebook, Line, KakaoTalk, WeChat, etc., are not secure by definition. I don't trust a word of what WhatsApp is saying about Signal-protocol security. It's not just about my safety from government surveillance—I am even more concerned about processing of my data by marketing companies who want to learn who I am to improve their sales (or political campaigns—it's quite the same thing).
Signal is Free software, and it's widely recognized as secure, but it's not reliable because many people complain that messages are delivered slowly or not at all, and it's not popular. Also, it doesn't have much of a bot API, and it's becoming important these days. Finally, its partnership with WhatsApp is truly puzzling.
Telegram is partly, but not completely, Free software, and it's more reliable and popular than Signal, but not very much so. Personally, I love it for the features, like channels, groups, bot API, and cloud storage, but I acknowledge it has problems. (Most of my friends are on WhatsApp, and those of them who tried Telegram all said that Telegram's features are better.)
Wire is kind-of curious, but very unpopular.
What I really wish is that Mozilla stopped all its activities except developing Firefox on desktop and Android (I'm talking about pointless stuff like the recent rebranding, Webmaker, overblown international conferences, etc.), and then acquired Telegram or Signal and focused all of its non-Firefox efforts on one of them. If it acquires Telegram, it should make it fully Free software. If it acquires Signal, it should invest in its reliability and popularity.
I maintain contact with everyone I need to maintain contact with via email, phone, and SMS. Pretty much everyone can receive at least one of those three options and it's easy to keep track of them. I route my SMS messages through a voicemail service so I can respond to them on my desktop computer if I care to. Same service transcribes my voicemails when I get them and gets them to me via email and SMS. If you can't be reached effectively with email, phone or text messaging then we don't have a relationship that is deep enough for me to worry about you. Sorry.
I cannot be bothered with Facebook or Twitter but those are common enough that I can see some people adding those to the list of options but I consider them optional and to a large degree superfluous. I don't see any IM apps that provide any meaningful benefit in my work or personal life so I don't really use any.
Just use multiple messengers.
Nerds do not have a problem choosing some messengers, so they can reach everyone.
Noobs will install a new app, if it has colorful smileys. So they do not have a problem either.
If you want a standard, use XMPP (and OMEMO). And read the XKCD comic about standards ... until then we just use multiple programs with a fallback to e-mail, phone and sms.
The reason why there are so many mobile applications for room chats is that they want to harvest your data. If you are not paying, then you are not their customer. In a country where 98% of the smartphones use Facebook's Whatsapp, the expectation is that you need to give Facebook your personal data. I do not want that, so I do not have a Whatsapp account.
Email was a direct outgrowth of ftp, and in that era (not later in the uucp era) was endpoint-to-endpoint. I implemented SAML and SOML in gcos smtp and it happily interworked with unix (and I think tops-10, but that was a while ago (:-)).
It fell out of use when people started setting up mail-hubs, and was long dead by the time MX records came along.
davecb@spamcop.net
Seriously.
These companies keep coming out and closing down their IM services or replacing them with non-open services.
And having to run multiple different nasty, intrusive IM clients just to stay connected is straight-up BULLSHIT.
If their services aren't integrated into universal IM clients like Trillian, Adium or Pidgin, DON'T FUCKING USE THEM. Simple.
Because if you do, you're basically committing yourself to a technological ghetto scene.
Chas - The one, the only.
THANK GOD!!!
This post sounds like the lead-up to a sales pitch.
The summary suggests that the desperately-needed answer to the issue of people being split between multiple communication platforms... is another platform. I probably don't need to go over why that doesn't make much sense but, to summarise: not everyone would go over to it, and then you've just introduced one more circle to the diagram.
Personally, my approach would be an umbrella app, linked to whatever existing platforms you use but abstracting out the particulars, and configurable based on the user's priorities (security, functionality, speed, cost etc). When you want to talk, you add the people (potentially setting some other parameters as well), and it intelligently decides what the best platform is to send your message/host your discussion. "Oh, you want to have a personal discussion with Roy and July? I'll use WhatsApp, they both use it and respond quick.Oh, you need a confidential business discussion with Paula, Derek and Sam? They all use email, but that's a low security channel: do you want to use it anyway, set up a Slack channel with P & D and invite S to join, or Slack with P & D and send a separate message to S by Yammer?"
While universal IM clients go some way towards this, the next steps are to group contacts across services for individuals, start including none-IM contact methods (email, SMS etc), and to pull the decision of how to contact people from the user to the client. That said, I'm out of date of UIMs, it may be that some already do some/all of this?
Twitter works fine with SMS, this should be fine within your scope of accessibility.
No argument but I regard twitter as an optional extra. No use to me personally but I get how others find it useful and I'm aware of it's compatibility with SMS. Other services work well with SMS too but I think tying yourself to them without very good reasons to be problematic.
Use Internet to send messages with some specified degree of 'richness', such as unlimited length and large attachments, falling back to SMS if either end cannot support the rich protocol. Note that I mean the iMessage approach, not Apple's specific implementation of it. I doubt that a standard enriched protocol would include balloons backed up by cheering, for one.
What ever happened to writing people letters? I just sent my Mom a Birthday card with a lovely note inside. I get like one text every 6 months on my phone. Then again, I am not in a relationship.
Trillian already solved this, so did pidgin, and nimbuzz, and whatever else. And they all suck, for various reasons. See the thing is, you think you have a really good idea for an app, but you don't. You have a tired old idea that will never work.
Apple "iMessage" which cheats to hijack text messaging is really the only successful multimessenger client, and that is because it exploits market position, and impossible access to devices and services.
A new standard would be a great idea if any of the messenger platforms had any motivation whatsoever to follow it. It is 100% downside for them to do that.
"Oops, hold on, I have to switch back to yahoo messenger because Trillian doesn't support that yet."
"Sorry I am on pidgin, what is that emoji supposed to be, I can't tell if you wanted to get dinner or not"
"Hey, sorry, I got disconnected and my client doesn't always save messages, could you copy/paste the last half-hour for me please"
These are the joys of multi-messenger.
The answer is the one in my head.
You can have it for a cool million.
Good try trying to get it out of me slash dot.
https://tox.chat/clients.html Tox is open source, cross platform, free, and uses encryption and out of everything that's happened in the last few days, it hasn't been mentioned in Vault 7, as far as I know. There are apps for both desktops and mobile devices. You create a profile much the same way you create an OpenVPN config file. In other words, it can be shared across devices and isn't stored on a server. That profile, which stays on your computer/phone, contains information that's password protected regarding user info (user name and a profile pic if you want) when connecting to a Tox server. The Tox server only acts as a relay until it finds the person you want to talk to. Then, it's encrypted p2p. In most cases, it only works if both sides are running clients, much the same way as other instant messaging apps. Tox supports video or audio calling, text messaging, and file sharing. You can change your "availability." There is no phone number or "signup." The code used to act as a "phone number" is incredibly long and randomized. If someone contacts you via Tox, you either know them or you posted your Tox string online somewhere. Most clients support QR Code reading to make sharing contact info much easier. Be very careful though, Tox is also the same name given to a randsomewear designer on the dark web. They have nothing to do with each other and call me paranoid, but I suspect it was on purpose to scare people from something that works really well. I only say this because if you DuckDuckGo "Tox" by itself, bad stuff may or may not show up instead. Only people that care enough to use a Tox client are also probably the same kind of people that know what "randsomewear" is. Use the URL I gave at the beginning and you'll be fine.
This author brings up a great point. We should create a new standard way for people to chat because that will solve all the problems.
People don't "desperately need a universal solution". It's not that big of a deal. Feel free to cut down to whatever clients you want to use and people will either meet you there or use the lowest common denominator (e-mail). Eventually this problem sorts itself out (remember AIM vs ICQ?).
Not because RCS is superior.
But because everyone is against XMPP!
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
It's called email. It'll never go away, ever.
there cannot be a universal solution because communication is about taking sides. ...
the more groups, teams or inoperable programs exist, the more (secret) identities, secret-keeping, information partitioning and scheming is possible AND THAT is what it's all about
the hilarious part is that the communication facilitators SEE everything: it must seem strange that grown up people with the assurance of "privacy" act like pre-school children?
> Email was a direct outgrowth of ftp
Nope. mail was present in 1st edition UNIX, FTP was not.
I'm pre-unix: ftping messages and then submitting them into a local-only mail system was how we first implemented inter-machine mail. In fact, you can still do that on an IBM by ftping a file and specifying it is jcl. That's how I used to submit jobs a few years back, from S390 Linux to MVS.
davecb@spamcop.net
In the comic, there are two all encompassing subsets, SMS and email.
The universal solution seems painfully obvious to me. Use email. If you have a contact that is SMS only without the ability to do email, then use a email-SMS gateway. Every carrier offers one for free.You can also use paid-for universal gateways.
OK, how about: https://en.wikipedia.org/wiki/ARPANET#Network_applications
"In 1971, Ray Tomlinson, of BBN sent the first network e-mail (RFC 524, RFC 561).[49] By 1973, e-mail constituted 75 percent of ARPANET traffic.
"By 1973, the File Transfer Protocol (FTP) specification had been defined (RFC 354) and implemented, enabling file transfers over the ARPANET."
Identity is everything. Would you use a product that could (maybe blockchain?) guarantee your identity, and provide an audit trail of all access to your persona? Who would be trusted to keep the original data? Would we all have a chip that could safely store and retain our personality? I could see plugging in to make encrypted backups taking on a whole new meaning.
We live in a society that grants corporate/legal entities rights, who would be the "chip keeper" for that? Maybe a more meaningful board of directors would occur when a quorum of directors would plug-in to share and update the corporate persona (and yes, their individual persona would need to sign the chain).
The biggest benefit/difficulty would be anonymity. Would the "agent for service" become the fall guy for the shielded corporations that are the conduit used to shield the owners of corporations from scrutiny. Money, credit, identity theft, tax fraud, etc.. would all be radically altered. Things that weren't part of an audit-able chain would get the scrutiny that they deserved. I could envision even "CASH" transactions requiring a two party receipt.
Bring on the dancing pigs and autocratic cows.
You will be saying people still use it next.
I have excellent Karma and I am not afraid to Troll it.
We had the programs before the RFCs
davecb@spamcop.net
Whatever. You may have had file transfer tools, but you didn't have FTP.
First, give me a universal ID that I can use on ANY system, with SSO and the ability to keep all my data safe, secure, and unhackable.
Then we'll talk about chat systems.
Oh, and XKCD forgot LINE in their comic.
Is taking care of it. People want choice. 1 million billion choices. Freedom baby. Suck it.
Tox is secure easy to setup and supports text, audio, and video. Also has a client for many different platforms.
https://tox.chat/
The answer is right in the comic. The wall (bathroom) guy has it solved. If you can't contact him any other way, then he can't possibly have this problem. Who needs to quibble about the communication form when there are billions of people out their to communicate with?
It worked quite well. Google, Facebook, Whatsapp, and undoubtedly more built their entire chat platforms on top of it. The issue was a lack of standardization and federation between "competitors." That won't change until we start getting our IM services from a third party the same way we do with email. RCS is a great step in the right direction, but I hate that it is still tied to a phone number. The comic also leaves out the mess with iChat and all the fruit-lovers who don't even understand that it's yet another IM platform.
The awful thing is that we went through this problem once before, with AIM, Yahoo, MSN, ICQ, etc. and they actually *were* starting to interoperate. A wealth of multi-protocol IM applications emerged (RIP Pidgin), and for a while, everything was actually pretty good. Then suddenly mobile apps changed all that. I don't really understand it. I curse everyone who uses WhatsApp, kik, snapchat, every other stupid, new chat platform for contributing to this problem. These applications perform one very simple task. They aren't innovating--not really. In this case, increased competition isn't accomplishing anything positive. It's time to dump these products. Sadly, the millennial generation just doesn't give a fuck, they will flock to whatever is new and shiny.
As a programmer, i have had to read a many white papers and technical specifications. For a work project I started writing a perl script that used XMPP protocol to send messages to co-workers on a OPENFIRE server. For this I had to read the XMPP specifications, and OH MY GOD are they the most beautifully written and clearly explained technical documents I have ever had the privilege of reading: https://xmpp.org/rfcs/rfc6120....
And developing apps using the XMPP protocol was super easy and fun. Sadly, the protocol was mostly abandoned due to lack of features we have seen in proprietary implementations. E.g. XMPP clients don't have universal support for in-lined pictures and video (showing someone embedded youtube video or the picture instead of http://imgur.com/aabbbaa or http://youtube.com?watch=blah) It has no support for video messaging (it was build for chat and IM) nor does it have support for screen-sharing. Also if you close your XMPP client and someone sends you a message, you don't get that message when you log in again. Some servers implement an extension that kinda does this, but because its not an official part of the XMPP protocol its spotty and unreliable. Various XMPP clients tack one or multiple of these features onto the XMPP spec, but this fragmented support tends to drive people away instead of too it.
But if ALL you want is a protocol for real-time chat rooms and instant messaging of text-only, man is XMPP fantastic, cheap, easy to use, and reliable.
We are not going to solve this problem at once. The first step is to stop using the crappiest instant messaging solutions and move on until we get only open standards, and hopefully there will only be one.
1. Stop using proprietary, single-vendor solutions (iMessage)
2. Stop using proprietary solutions not working on common internet-connected devices (Whatsapp doesn't work on PCs)
3. Stop using open solutions unable to work on all internet-connected devices by design (SMS)
4. Stop using proprietary solutions not accepting third party clients (can't work inside pidgin)
5. Stop using proprietary solutions not using an open standard as a back-end (many protocols use XMPP but are still locked-down)
6. Use only open standards
iMessage is a huge step backward as it's not only proprietary, it's single vendor. Let say every phone vendor cloned iMessage. We would still be stuck with hundred of incompatible messaging protocols. Sure, they would all fall back to SMS, but that would also mean that you couldn't change phone number (especially internationally), and that we would all be stuck on our tiny phones even if we have a real keyboard on our much faster PCs.
The sole purpose of iMessage is to form a closed community of Apple device owners.
Develop TORchat and make it available on iOS and Android as well as windows, linux and mac
http://secushare.org/
I'm surprised that nobody has brought up bitmessage yet
https://prism-break.org/en/projects/bitmessage/
It seems to satisfy (almost) all the conditions above and is supported by multiple clients.
I guess I don't view this as a technology problem - more of an 'economic demand' issue. Looking at the evolution of email (and I know this is a simplification)... There were lots and lots of proprietary walled-garden email solutions pre-Internet. They worked fine for what they were - electronic memo systems for corporate users. The Internet changed the game, of course, and all those corporate users became interested in inter-operability - i.e. Company A want to exchange emails with Company B. My perspective is that email standards only gained real traction because of this economic imperative. People were willing to pay for email inter-operability and vendors were only too happy to adopted technology standards to make some money off their products. Where instant messaging is concerned, I don't think we have a economic imperative to drive a market that can pick winners and losers. Interestingly, there are plenty of walled-garden IM platforms in corporate use - e.g. Lync, Groove/Spark, etc. There is some amount of demand for inter-operability, but it's minimal in my experience. So without an economic impetus to standardize - here we are.
CrazyLegs
"Pork!!" said the Fish, and we all laughed.
A good summary how XMPP is still a good choice today for the mobile world:
https://gultsch.de/xmpp_2016.h...
Here in Brazil we can say that Whatsapp solved this problem, virtually everyone use it. Really. I usually order pizzas through Whatsapp. The only problem is that sometimes a bad-mood federal judge blocks it in the entire country.