Slashdot Mirror
AT&T Re-ignites Instant Messaging War
travisd writes "Looks like AT&T Worldnet is trying to play Microsoft - no, not in being a monopoly, but in trying to horn in on AOL's IM turf. The Washington Post has the story.
" Yes, it's the return of the IM Wars - we had covered this this past summer. Microsoft was in the right on this one - a standard does need to be made, by an industry-wide group.
I guess the battle over access to cable modem networks has spilled out onto other fronts.
There is a lot of animosity between AT&T/TCI/Exite@Home and AOL right now, and I wouldn't doubt AT&T folks did this simply to show AOL who's "boss."
if you couldn't call your friend's house because they were using Microsoft Phone and you had AOL's You've Got A Phone. Sometimes standards are a good thing.
Apple was supposedly working with AOL to integrate AIM into the system at some level. I wonder whatever came of that...
I use Macs for work, Linux for education, and Windows for cardplaying.
Am I taking AOL's side here? No. I think that if AOL wants to keep the their protocol non-standardized then Yahoo!, MS, and AT+T need to get together and make a standard protocol. Soon people will drop AOL's IM because they can communicate with more people through the new standardized system and AOL will suffer for its wrongdoings.
MS nor AT+T is accomplishing nothing by keeping this war going on, except maybe for once MS is in the role of the good guy and they don't want to lose that
It strikes me that instant messaging is one of those areas where _everyone_ needs to either be on the same service, or the services all need to be open to each other.
Since there is no financial incentive to allow your competitors to access your servers (and you customers), it seems to me that the only way around this is a global service run on a not for profit basis, that allows _anyone_ to use it with any client they like (ie, open the message protocols to the public).
No company could make money from this (after all, if you can write your own client, you'd leave out the bit that displayed the advertising, wouldn't you?), so it seems to be a prime target for a government funded initiative.
Now, I don't like the idea of "The Government" being involved in my messaging any more than you do, but it does seem that some sort of international task force, funded by a variety of governments (or the UN) would solve a lot of these kind of problems.
All critisicms are, of course, gratefully recieved.
My Journal
Speaking as the sysadmin in a cyber cafe, I'm really beginning to hate instant messaging. Every time we crank up a computer, it loads yahoo, MS, icq, aim...
Having seen what people use (and why), I can be fairly confident that AT&T's effort is doomed, if not to failure, then to obscurity.
MS's version took off like a rat out of an aqueduct, simply because of that big-ass banner on the top of the hotmail page. To people who only use the computer for hotmail, this was an added feature, and one they embraced[1]. Aim, of course, already has godzillions of users.
So what about worldnet? Its new messenger will appeal only to those who haven't already got an account with one of the biggies. Anyone else'll ignore it.
[1]hotmail used to have a banner ad for internet telephony. At lest five times a day I had to explain that it was an ad, and not something they could use in hotmail.
ICQ has > 55,000,000 people using it
AOL has about the same
Yahoo had 20,000,000 I think the last time I heard
How many more people can want to keep in touch with people? Or is it going to be a case where we need once client to talk to Dave, and one for Sally and Fred (being the guy who has to have the newest software) we need a 3rd.
The article refers to an IETF effort to pruce an instant messageing standard, if anyone is interested here is the url to it's chater page.
For anyone who hasn't been following this issue you have several different messageing programs all backed by one company or another which do not 'talk' to each other. An equivelenet would be if it was imposible to phone somone from your phone because there phone came from a different telco. The situation is completely stupid, all logic dictates that communications systems designed for the same purpose should be able to talk to each other.
AT&T are doing the right thing in the wrong way, inter operability should be a priority but not by some strange kludge that only works one way. Does anyone else agree that if this carries on for much longer it may be a job for legislation to bring the IM providers into line
One thing that really impressed me was the success of icq, I never really use it myself but a totaly centralized propriatry system being that big a success is not exactly the norm on the on interent.
As I understand it, there are two issues here, and a lot of knee-jerking going on as well. First off, the protocol issue. I think Microsoft / AT&T have every right to try and clone the protocol used by AIM. I think there may even have been legal rulings over in the US about the legality of reverse-engineering protocols. AOL, of course, are entitled to change there protocol, but leaving a gaping security hole (allegedly) in the clients to do so is unacceptable. Secondly, there's the question of access to servers. AIUI, Microsoft's client was connecting to AOL's servers in order to communicate with AIM users. Regardless of what you feel AOL ought to do, those servers are their property. They have the right to decide who should and should not have access. If they decide the servers are for use only by AOL/AIM users, it's their choice. If I understand correctly, Microsoft or its software's users might have been committing an offence under the UK's Computer Misuse Act.
That's like saying "We had a gate, but no fences, and no one came through the gate"
Security, by definition, can't be bypassed. Maybe AOL did have the right to stop MS using it's servers, but that should have been defined in the licence, and enforced using real security measures. "Security by obscurity" is not security.
You expect the occasional bug in a product, but just not telling people how it operates is not a viable security method.
I've found this wonderful new technology. Its called email. It allows you to send a message to a person, and they can reply to it too. Unfortunately it's completely open and every program works pretty much the same.
Why on earth do people need these things? What is wrong with an email. If you need to contact someone really quickly, why not use that other piece of incredibly useful technology. We call it a phone.
Unfortunately, as AOL own both AIM and ICQ (the two real biggies), they pretty much control the majority.
As a result, any standard is unlikely to have such a high take-up unless AOL can be persuaded to be less arsey about it all.
Now, if Microsoft and Apple both integrated compatible messaging products into their next versions, *that* would be a different matter (and most likely, a whole new lawsuit)
The fundemental principle underlying instant messaging schemes is identical to that underlying office productivity software. The software is *inherently* worthless and becomes worthwhile if and only if there is a critical mass of people who use it. Once the population using a particular incompatible with anything else software reaches this level, new users will tend to use this service as well, establishing a de facto standard that other providers must comply with if they want to reach the majority of users.
Treat this as flamebait if you will, but I do not believe that establishing Yet Another Standard for Instant Messaging is the solution to these problems. As long as users have the freedom to switch from service to service, as they do now, the relative popularity of each service will fluctuate with respect to it's utility to the user. The real issue is not whether standards are neccesary (they aren't at this point and won't be beneficial until the technology matures) but whether companies have the right to use the networks of other companies. As I see it, creating artificial boundaries splinters the community and benefits nobody, neither the company that does the restriction, nor anybody else benefits. With bandwidth availability skyrocketing and new communications technologies being developed almost daily, does it truly make any sense to lock in an IM standard that will be obselete practically as it is approved? I would much rather have a de facto standard that is flexible than a rigid actual standard.
--
Flames? Think I'm a karma whore?
Well the way I see it MS has no business asking for open standards in instant messeging protocals if they have closed protocals in other forms of networking. In particular SMB, thats the protocal network neighborhood runs on. The Samba team has reverse engineered most of it, but the last time I checked Samba still couldn't be a Primary Domain Controller you have to buy NT if you want that. I haven't followed all this HTML stuff, but those MS extensions are they an open standard, I don't think so, someone correct me if I'm wrong. And their file formats for Office. MS thinks its of deathly importance that all IM clients should be able to talk to each other, but they don't think I should be able to read spreadsheets, databases, word-processing documents, etc on anything except their platforms. Really which is more important? Fscking hypocrits!!!
From the recently posted C|Net article about AT&T being blocked by AOL:
Despite warning signs, AT&T executives had been hopeful AOL would cooperate, especially given the online leader's recent push for so-called open access to cable networks.
It took nearly everything I had to not break out in a fit of laughter upon reading this. AOL is desperate to get direct access to *some* form of high speed gateway, to the point where they dumped a couple hundred million into *satellite* technology(DirectPC, to be specific. Great for rural areas, but rather horrific to scale--and one way, too).
So here you have AOL, banging on the doors, trying to get the law (very very justifiably!) twisted so that, gee whiz, there can actually *be* ISPs that don't own the direct lines of communication.
Meanwhile, at the very same time, Microsoft is banging on their door, trying to be an IMSP(Internet Messaging Service Provider) that doesn't own all of the direct communication servers.
The communications architectures are publically supported(guess whose land all that fiber's going over!), while the Instant Messanging systems *aren't*, but those are just *details*. On one hand, AOL wants open access to AT&T's property, and on the other, AOL wants closed access when it's their own damn network!
Oh, what a brilliant gambit on the part of AT&T, calling them on an apparent contradiction! I honestly would not be surprised if the only reason AT&T ever partnered with Tribal Voice in the first place was because they realized AOL Just Couldn't Win Both Battles--and either way, their "enemy" would be weakened by either loss.
AOL's counter should be interesting. I know some of the guys at AOL--hell, one of 'em is probably the single smartest networking guy I've ever met. We're talking about a company that, in response to Microsoft's attempts at circumventing their network security, consistently and repeatedly exploited security holes in Microsoft's AIM client, and likely threatened to announce exploits for their client unless Microsoft caved in. (They did.)
AOL is bound to have a truly fascinating response to all this. I, for one, am going to be watching, popcorn in hand.
Yours Truly,
Dan Kaminsky
DoxPara Research
http://www.doxpara.com
First off, it is very thoughtful of Slashdot editors to tell us who is right and who is wrong, without their expertise we might actually have to form our own opinions on the matter.
This is quite simple in my opinion, AOL provides the IM message service which it AFAIK created. In return for this free service AOL recieves money from the advertising banner it flashes across the screen. The key issue here is this. This is _AOL'S_ software. Other companies want to walk in, grab a hold of their large userbase, and walk away with it. Now if these companies were half as inventive as they were greedy, they'd create their own chat network. But of course not. The scream of "Open standards" quickly rallies slashdotters into a frenzy before they look at the issue here. AOL _created and maintain AIM_ it is _theirs_ nobody has the right to tell them what standards they should and shouldn't use. Now if these companies wish to make an open standard for their _own_ chat network, heck, more power to them. Personally I must say that AIM is one of the best pieces of software I've ever seen, it's fast, it's not filled with useless options, the advertisemens aren't annoying popups, etc.
Microsoft was in the right on this one - a standard does need to be made, by an industry-wide group.
... they should have found a better way to do it.
... email, but instead of setting up my own mail server, I hacked into yours and pointed all my users to it, then when you fixed your server so I couldn't get in, I hacked into it again to give my users free use of a resource I don't own, maintain, or hold accountability for?
When did hacking into a network and leeching it's resources for your own gain become a valid action when striving for industry wide standards?
I'd like to see standardized IM as much as anyone, but saying Microsoft is right is a bit far fetched. Microsoft is wrong for hacking into AOL's IM infrastructure after access was denied and AOL is wrong for exploiting their own client software to keep MS out
What if I wanted to get in on this exploding new technology called
And don't respond back with "email is an open standard, while IM is not" because that's not my point, my point is MSN HAS their own IM infrastructure and could have used it, but they decided to leech on the resources and investment of another company and when they were asked/told/forced to stop, they hacked it again. They are FAR from right in this case.
That isn't a valid comparison.
Why not?
You are running a clone of the AOL computer on your computer, and using the servers in the way they were designed. Saying that just because AOL owns the servers they shouldn't need to specify they can only be used with AOL clients is like saying Netscape should expect only Netscape browsers to hit www.netscape.com just because they made the first commercially successful web browser.
As for your property thing - sure, you are right there, but if you owned a large patch of countyside in the middle of a national park, and you didn't put fences around it you shouldn't be surprised if people trespass - and infact they would probably have a good case in court.
Here in Australia, the is an item of common law (inherited from England) that if an area of private property is used for public purposes continuously for a long period of time, a "right of way" is formed, and the owners can no longer stop people from using that area.
That is why places like Adelaide University, which has a large "public" throughfare through the middle close the gates once a year - to stop the public use. Apparently they didn't once, and the final year law students tried (in court) to get a right-of-way.
To get back to the point - perhaps something like this might (or should) apply on the 'Net - and with one netyear = 3realyears... do the sums for yourself.
I'm not really serious about the right-of-way on the Internet, but it does make you think, hey!
I certainly agree that there needs to be a standard. And I applaud Microsoft for saying so considering all of their media coverage, no matter what Microsoft's real perverted intentions were. And, lo-and-behold, there is an industry group workign on the standard as we speak. Let's just hope they adopt it. But, I do have a list of complaints I've seen with current IM-like products that I think should be addressed so they don't get worked into a standard. Stock/News/Weather tickers: There is a time and a place for these. I simply don't want special features cluttering my screen, memroy and disk when I'd rather use a third-party stock-ticker from elsewhere. Just don't put it in the protocol. Let the indvidual IM client software add what they want, btu keep it out of the protocol. Advertisements: Again, keep them out of the protocol. Names: OK, this really has to do with client-sofwtare only, but I'd sure like to be able to put aliases on the names in my buddy list. I can never remember who GoldenFry77 is. I'd rather see an alias as "John Smith from Work" and have the client software use GoldenFry77 in the background.
IETF's IMPP Working group (charter) Maybe this was mentioned deeper in someones thread, but all the same i figured I would point out that there is a standard being developed. Last I heard the group was getting fairly close to having a spec.
You want to legislate a computer standard?
Oh man... read your history books!
I totally, absolutly disagree with you. I understand the comparison with telephones, but the Internet is a new game.
Let the market work it out - if one company loses a couple of $mil on the wrong standard, that's no big deal. It's not like they were rolling out incompatible telephone wiring.
It is a little annoying for consumers, but it doesn't cost them anything except convenience, and it allows new products and protocols to develop in ways a legislated standard wouldn't.
Anyway.. what country would you legislate in? What's to stop the non-compliant messageing servers moving to somewhere that has a concept of intellectual freedom?
If there were an open standard, then Microsoft would "innovate" and build the client into Windows as part of the operating system. In the fullness of time, just as with browsers, competing clients would be driven out. At that point Microsoft would be free to "enhance" the product making it incompatible with non-Windows systems. AOL is certainly aware of this possibility and that is why they (rightly) will not go to an open standard.
...and it's called "IRC". Internet Relay Chat already provides all of the features of the various IM clients and web-based chat pages. You can share files; you can be notified when certain people leave/join channel(s); etc. Too bad there's no cutesy flower icon, or "bing-bong" every time someone /messages you, otherwise it'd be the Next Killer App(tm).
And there's even an experimental RFC (1459) that describes the protocol! You can't get more standard than the IETF!!
Rev. Dr. Xenophon Fenderson, the Carbon(d)ated, KSC, DEATH, SubGenius, mhm21x16
I'm proud of my Northern Tibetian Heritage
I just recieved the newest beta version of Millenium here at work, (4.90.2419 for those out there that are keeping count) and something very interesting installed itself by default. You got it: MSN instant messanger.
Now, let me pose this question: If it comes with the computer, why download another application? Seems to me that, standards or no standards (or anti-trust lawsuit or otherwise), MS is back to it's old, dirty tricks. Think: Internet Explorer.
--
rJames.org - illustration
Jabber.org (http://www.jabber.org/) is working on just such a thing. We're one of the *ONLY* IM systems out there that already has plugin support for ANY IM system, including the IMPP standard once it's released..
And the best part is, it's non centralized, and Open Source.
-- I'm the root of all that's evil, but you can call me cookie..
Eventually means a few monmths more.. ;-P 1.0 Very soon, with luck..
-- I'm the root of all that's evil, but you can call me cookie..
They are still working on incorperation of IRC and Jabber, which is an Open Source instant messaging system which can support nearly ANY other IM system out there..
http://www.jabber.org/
-- I'm the root of all that's evil, but you can call me cookie..
Actually, a nit-picky point of clarification... Instant Messenger was created by MCIWorldCom (not sure if it was pre-merger or not) and is exclusively licensed to AOL.
I think this would be a great surprise to all the AIM developers down the hall from me. AIM was conceived, designed and developed at AOL.
That works for the IM part, but I suspect there is a scaling problem when you have a few million simultaneous users, and every server has to send login/logout data to every other server for buddy lists. That gets into the tens or even hundreds of thousands of messages per second per server pair. That's tough with the latencies on a worldwide network.
Er, It wouldn't work that way. Ever server doesn't send login messages to every other server. The only server a given user logs into is his own. At this point, his home server sends an announcement that he is online only to the servers that have subscribed to his/her/its presence. Servers are decentralized in that they do not know of eachothers existence untill they attempt to send a message to it.
As far as I can tell from the FAQ, Jabber doesn't really decentralize the back end. It seems to be a "middleware" server that sits between the Jabber client and all the proprietary back-end servers. To your client, it must seem to be the server, and to the real IM servers, it appears to be just another client. I'm not sure how it gets all the buddy data from the servers, but it's possible that at high volumes it runs into the same problem.
Nope. Jabber is an IM system in and of itself, serving as the entire backend for Jabber users. It supports another tier when using gateways to other networks.at which point it is serving as a middle tier, but only for the gateways, and these are run as a seperate service entirely. This is made possible by it's modular nature, and transports can serve whatever purpose they need, be it a gateway to an external system, a news ticker, a chat server, or a news ticker.
-- I'm the root of all that's evil, but you can call me cookie..
It isn't a standards problem. AOL published the protocol, which is why open-source groups, Yahoo, and Microsoft were able to write AIM-compatible applications in the first place.
There is a difference between protocol standardization and free access. Just because Telnet is a standard does not give you the right to log into any Telnet-speaking computer and use it however you like -- you must conform to the administrative policies of the computer's owner.
Similarly, being able to speak the AIM protocol does not give you the right to log on to AOL's server network and use it without conforming to the administrative policies of the server owner. The server owner is AOL, and its policy is you cannot use the MS client on its network.
Microsoft wants people using its client and AIM to be able to talk to each other? Then Microsoft can set up a network of AIM-compatible computers and publish the address. AIM, after all, allows you to specify the address of the server you wish to log into.
This has plenty of precedent. The Realtime Blackhole List is a prime example of standards-compliant messages being blocked from servers because the servers' owners do not want to accept the messgaes. Password-restricted FTP is another. Just because something is connected to the Internet does not mean you have the right to access it.
Talk is very nice, but doesn't do quite what instant messaging does. Talkd runs on a machine and allows logged in users to receive 'instant' messages in text. By contrast, an instant messaging service is a server 'out there' that you contact, by means of which you 'advertise' your presence. The difference is important, because in this day and age, most people connect their machine to the 'net directly, they don't log into a mainframe. Not to mention that anonymity is possible in a proper implementation of an instant-message server but not so much in a traditional talkd. This is not to say that a talk-compatible server would be a bad thing. ;) Just modify the talk-daemon so that instead of using 'logged in users' it uses 'connected users' and instead of writing to terminals it writes to sockets...
;)
Of course, this doesn't allow for file-exchange possibilities. It might be better to run an 'instant communication' service that essentially ip-forwards talk, ftp, and speak-freely protocols along with its own management-information connection.
Or, the short version: instant messaging does more than talk, but it sure would be nice if they'd build on what we already have.
--Parity
--Parity
'Card carrying' member of the EFF.
The structure that AT&T is trying to setup, of allowing connections to other servers, is very much like the free, open alternative that the Jabber team is creating. Although we are not creating a client only solution. We have a server based around a XML stream protocol that allows pluggable transports for different protocols, with clients for almost every platform. This allows for an extremely extensible system. Currently we have betas for AIM (libfaim has not been blocked throughout this entire ordeal of blocked clients), ICQ, and MSN is in the works. I have just completed a group chat (similar to ICQ) module as well.
Although the side of having an IM client that can connect to a lot of others is appealing, Jabber is much much more. We have plans to jabberify many programs (CVS, Abiword, Bugzilla, and more). By doing so the power of these programs increases in many orders of magnitude. Just imagine multiple working on a document in Abiword at the exact same time, or CVS automagically pushing you updates, or Bugzilla yelling at you when you get a new urgent bug? It all sounds really appealing to me.
Being free and open source, Jabber has the potential to be so much to the internet, and help settle some of these annoying arguments that a lot of the corporate players are having. Come visit us on IRC in #jabber on openprojects network or just visit the web site.
AOL has control of two major "standards" today. AIM (I still don't know why you'd use it), and ICQ. AOL, wanting its own monopoly, has gone and tweaked both protocols all they could. However, the ICQ protocol is still open because of the older v3 and v4 connection clients. However, I've noticed recently that Licq and ICQ 98 (which don't have the latest weird byte-swapping techniques that ICQ 99b uses) seem to have trouble connecting to the network.
Would it be so hard to take the most flexible protocl (ICQ v5), change it around a bit (ie: make it easier to impliment), and produce that as a standard? I'd certainly like to see that RFC. Then we can be free to choose our own clients, ala choice for browsers (well, none for MS users, see "Millenium" installing MSN Messenger by default).
Anyone willing to write an RFC about this?
---
--
Internet Explorer (n): Another bug -- that is, a feature that can't be turned off -- in Windows.
As far as spoofing messages.. do you even _use_ irc? its called /nick . ever have nick wars on irc? channel wars? people stealing channels via server splits
How often do *you* use IRC? As far as I know, EFnet is the only network that *doesn't* let you register channels. I haven't seen a channel takeover in about 4 years now. Every network but EFnet and Undernet also let you register nicknames. That gets "message spoofing" out of the way, and even the dimmest of people would recognize that a person might not be who they say they are and use "/whois" to make sure.
AFAIK no IRC server in existance can support more than about 1500 clients
Most servers on major networks *routinely* handle over 1000 clients, and there are dozens that handle thousands.
*** Connecting to port 6667 of server irc.concentric.net
*** There are 4842 users and 47781 invisible on 37 servers
*** 163 : operator(s) online
*** 21825 : channels formed
*** I have 10340 clients and 1 servers
*** Highest connection count: 12849 (12848 clients)
Yes, that's twelve thousand clients.
I want to talk to people, I don't want 14 year old script kiddies portscanning my IP, ping flooding me, etc.
You'd rather they spoofed UINs, changed your ICQ passwords, hacked peoples' Personal Webservers, etc.
-A.P.
--
"One World, one Web, one Program" - Microsoft promotional ad
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
What I meant, was, when I hook up to the net,
I'm assigned a dynamic IP by my provider.
As far as the internet is concerned, I have
no hostname, only an IP address.
We'd need a proxy-daemon on talk.provider.com
so people could talk to parity@talk.provider.com
instead of parity@012.345.678.901
--Parity
--Parity
'Card carrying' member of the EFF.