Slashdot Mirror
Microsoft Messenger Architect On The Future Of IM
CowboyRobot writes "ACM Queue has an interview with Peter Ford, chief architect for MSN Messenger, by Eric Allman, CTO of Sendmail. They discuss the present and future states of IM, the current big players as industry shuffles toward standardization, some of the social implications of IM versus email or telephone, and technical issues such as using SIP as opposed to XMPP (Microsoft is pushing for SIP, everyone else seems to favor XMPP). They don't bring up Wallop, Microsoft's community application that will be built into Longhorn, but that's surely part of the long-term discussion."
Trillian works great for all my needs. IRC! Man that's where it's at. What bothers me, greatly I might add, is that while the majors like Nokia, Ericsson and Motorola are busy selling IM at whatever cost their last meeting brainfarted, it is highly probable that most non-nerd people think this is the way to go. They are 0wn3d by the marketspae'k, and it's trendy so hey, cool, they love it. And there's money in it for companies to gain money per character of text, or per 32byte-max transfer. (or is it 255? tee hee)
The pundits of chargeable IM services socialize the use of the service, as a Freudian brainwash, by forming IM parties with other-sexy-trendy-phone-pundits, and I sit back wondering what the fuck is happening to the world; it should be all free, or at least the cost of hardware. It's obviously a ploy to put a price on a few bytes of data, and slap a carriage charge on top of it. Which is why I'm not at all surprised this Microsoft guy, PETER FORD (from the interview) is talking about IM. It seems that the fancier the names of the new protocols are, the more money it's going to cost. But it's mumbo-jumbo to the end user, who would gladly fork over the cash just to make it go away (and just work). That's what these pundits are counting on.
One part of the article I found interesting was the design of voice mail. I agree. It would be better to build the message at the sender's location and *then* send it.
"Does that mean 10 years, five years, two years? I couldn't predict. Quite frankly, the thing that fights against it being quickly this time around is that the communities operating with these mutually incompatible protocols are quite large. If you look at AOL's cloud or the MSN cloud or the Yahoo cloud, you're talking about fairly large, significant systems. To have them migrate and interoperate with standard protocols will happen, but it is going to take time." This also assumes that AOL or MSN or Yahoo will cry uncle first. Who serious believes that any one of them will be the first to abandon their standard for an open standard when it could mean the end of their software? Remember, we are dealing with some *SERIOUS* egos here...
Microsoft is pushing for SIP.
IBM, which sells the #1 selling business IM solution (Lotus Instant Messaging), is using SIP.
Apple is using SIP.
So who are the "everyone else" who want XMPP?
GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
The interviewer says:
Microsoft, Lotus, Sun, and Novell seem to have settled on SIP. Intel, H-P, Hitachi, Sony, and more or less the entire open source world is going toward XMPP, sometimes better known as Jabber.
and the poster says:
Microsoft is pushing for SIP, everyone else seems to favor XMPP.
Yeah, it's fun to paint the world in black and white but this is just a blatant lie.
When men used to be men
Microsoft's community application that will be built into Longhorn
...
So IM will be build into Windows, and Netmeeting, and this and that and whatnot. Isn't this getting slightly ridiculous to bundle everything in an OS ? I'm sure nobody wants *all* of that installed on their hard-drive, just as I wouldn't want to install all the packages that come with my Linux distro CD, but instead I want to choose what I install and nothing else, and save disk space.
What's beyond me is why don't we hear a great number of people (regular users) complaining about this waste of disk space, and also why so few OS experts voice their concern about the fact that the OS/application boundary in Windows is so blurry it's frightening in terms of security and stability
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
IMs are fine and dandy but when are they going to work on improving video confrencing. Typing is tedious but strides haven't been made in free video confrencing software. Perhaps that should be part of their implementation of the next "IM" software. Afterall even the old Netmeeting has a chat window you can bring up.
I used ICQ for a while, then uninstalled it, multiple times had to uninstall YIM that got installed with Netscape before Mozilla really came into play, fought kids installing GG (polish IM) on classroom computers, generally did a lot to get rid of instant messengers from my life. Am I weird or what?
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
There are two doors. The door to your right leads to SIP, and the salvation of Redmond. The door to the left leads back to the matrix, to XMPP, and to the end of your species. As you adequately put, the problem is choice. But we already know what you're going to do, don't we? Already I can see the chain reaction, the chemical precursors that signal the onset of emotion, designed specifically to overwhelm logic, and reason. An emotion that is already blinding you from the simple, and obvious truth: XMPP is going to die, and there is nothing that you can do to stop it.
I do agree that you will see consolidation over time. You could probably argue that the e-mail experience that you and I and a lot of others lived through over the past 20 to 25 years is probably going to be repeated--perhaps more quickly than last time because the Internet makes that kind of evolution easier.
We're gonna go through this spam thing again, aren't we? Man it's like living in Groundhog Day. On the other hand, this does give us a use for Bunker Buster bombs - instant localized retaliation against any spammer. And their families. And friends. And neighborhood.
Which is as it should be.
To celebrate the occasion of my 1000th post, I will post no more forever on Slashdot. Goodbye.
Protocols will become more proprietary, telco companies will continue to *squeeze* money out of consumers for sending text messages over networks which would otherwise be utilizing much more bandwidth for a normal voice call, and proprietary IM providers such as AOL, Microsoft and Yahoo will not collectively work toward a standard, because they have their hands too deep in consumers' pockets to see that it would benifit more people than just them to work together for a common good.
No, I don't think the major IM players will settle on a standard. The best thing we can hope for is that the Jabber protocol catches on and we all have an open IM standard.
That's most likely not going to happen, though, until the rest of the world catches on to the whole OSS movement. And at that point, there are going to be so much better things out there than text IM that people are working on together that it won't matter anyway.
It is pitch black. You are likely to be eaten by a grue.
Meanwhile, I plan to wash my hands of the whole mess and use Jabber. Remember back when we had standards, and the internet was decentralized? It actually worked - there wasn't a single point of failure. When was the last time the entire email system went down? Jabber can offer the same reliability, and you don't aren't locked into a single server or client.
Besides being decentralized, Jabber tries to offer gateways, and many Jabber clients (such as GAIM) also play the "keep up with the proprietary protocol" game. So have the best of both worlds - get a Jabber account somewhere, and whenever your friends's servers lock out their clients of choice, convince them to get a Jabber account also.
Litigious bastards
That's a great story... I don't use IM or chat very often, so I haven't thought much about them. So a lot of what was said was fairly relevatory for me.
The thing that interests me is the way that Ford talked about differences in accessibility (can people you don't know communicate with you?), and verifiability (do I know who you are?) in various systems, and how one system (say chat) might be used to allow rough and tumble anonymous communications with strangers, while another (IMing) might be limited to friends on a whitelist.
Another characteristic that's particularly important to me is real time vs. instant response. I *hate* systems that interrupt me in real time, which is why I use email instead of IMs. I've pretty much stopped answering my phone, too, because I can, and now I depend on my machine to queue up calls, so I can deal with them when it makes sense to do so.
The question that all of this raises, for me, is whether or not it's practical to have a comprehensive messaging service that will allow people to tweak all of these different parameters in combinations that they like. Is there any need for email and IMs to be distinct?
Maybe we need a messaging "account" to be open, and another to be whitelisted, or one to be real time, and another to be queued -- but can't they be the same general sort of accounts, configured differently?
(I'm not talking about trying to twist email itself into this shape... but about a new system that would cover much of the same ground.)
SIP (Session Initiation Protocol) has been around for a long time and AFAIK is a binary protocol. SIMPLE is built on top of SIP and provides the instant messaging functionality.
XMPP is relatively new and is based on XML (hence why it's so extensible.) There are two parts, the core (which might as well be equivalent to SIP's core) and the IM extensions.
The glaring practical difference is that there seem to be about zero open-source SIP servers, and about a dozen open-source XMPP servers (going off the list at JabberStudio which might not represent all of them.)
Karma: It's all a bunch of tree-huggin' hippy crap!
Basically, apt-get is a kick-ass system for making sure your Debian system is up to date, has the latest packages installed, and manages conflicts. At the core, what is an IM system about? Making sure your message 'packages' are up to date, has the latest messages 'installed', and manages conflicts, that is, a reply had been requested, yet hasn't been sent! All the key infrastructure was already in place, including an interface (dselect), which could easily be ported to all the required platforms to allow easy reading and sending of instant messages.
The first step was to use apt-get itself to distribute a modified apt.sources file, which contained the IP addresses of all of the IM clients on the network. Some people had suggested DNS as a solution to this, but my feeling was that DNS wouldn't scale so well (this was a large LAN, with over 10,000 clients...I'd like to see DNS cope with that!!). Once each client had it's apt.sources file updated, you could basically send a 'message' (your ASCII message encapsulated into a .deb file by a custom packager I created that runs as a background process) to any host specified in the apt.sources file. To do this, I had to create a daemon-ized version of apt-get, listening on a predefined port. The daemon would be contacted by the apt-get client, would receive the .deb package containing the message, and then 'install' it to the dselect based client on the receiving system.
Without trying to sound like I'm blowing my own trumpet, the system was a huge success, and the many features of apt-get for package management really came in handy for managing IM flows. For instance, just say you've just sent a message to a colleague via apt-get saying "Let's meet for lunch at 1pm":
apt-get install host=fred-pc "Let's meet for lunch at 1pm"
But then...you're called into an emergency meeting and you can't make lunch until 2pm. You need to 'upgrade' your message to the latest version:
apt-get upgrade host=fred-pc "Make that 2pm!"
Easy! The whole project was essentially wrapped up in 6 months, and because of the open-source nature of apt-get, we'd managed to port to all of the platforms in our specification. If Microsoft can swallow their pride a little, I think they could really learn something from the power of apt-get!
There was, in my opinion, a glaring omission in this article: no mention of IRC. I find this interesting because there is no reason why IRC shouldn't be adopted as the protocol of choice for text instant messaging. It's more stable than all the others. It interoperates nicely. There are IRC servers running on all kinds of operating systems. Endless clients.
How many millions of people use IRC? Why not adopt it as a mainstream system? I was surprised that the interviewer, being from Sendmail, so glaringly ignored throwing this into the mix. IRC can do everything instant messaging can, and then some.
Both the Mr. Ford and the interviewer failed in their mission: the former may not be much of an architect if he's willing to overlook this, and the latter should've asked more incisive questions.
Cheers,
Eugene
http://eugeneciurana.com | http://ciurana.eu
He's a first-rate architect. He's one of those people who understands more than just the protocols he's dealing with at the time -- he gets the reason those protocols came into existence, what drives them, who wants to use them, how they fit with othe protocols, etc.
Peter has been pushing for SIP inside Microsoft for a long time. I was part of the design process for a couple of years, and it was a real pleasure to work with so many excellent engineers and thinkers. There is a real desire to make interoperable, public network products at Microsoft -- don't laugh, it's true. We spent YEARS making H.323 work (which is a public protocol -- anyone can implement it), but it didn't matter because, in the end, H.323 sucked. Even the Windows Messenger guys want to move to SIP, because it solves a lot of headaches for them.
The best thing about SIP is that it is fairly decentralized. It's exactly as decentralized as DNS+SMTP. If you have a domain, you can publish your SIP service records, and you can handle your own communications any way you want to (similar to SMTP). This is in contrast to the way that all of the current IM protocols work -- extremely centralized, where all of your messages go to a server, that just re-sends them to the other person.
I don't know anything about XMPP. If it's a good protocol -- awesome. But whether it's XMPP or SIP, or whatever -- it's gotta happen. Instant messaging (and other similar services) need to be decentralized, standard, and open. And for once, the people inside Microsoft agree, and are actively working on it.
I just hope they can convince the upper management layer.
As long as the underlying protocols stay free and open (be it soap, irc, jabber or whatever) then if someone wants to write a closed source interface to it, that's their perogative, and of course they do so at their own risks. As great as it is to work as an (open) team, there is still something to be said for going it on your (closed) own.
I'm a writer, a poet, a genius, I know it. I don't buy software, I grow it.
There is no technical comparison of SIP vs XMPP. From indepth one I would expect to see why features of SIP are better or worse than features of XMPP. They don't even list features to compare.
Also, they talk about XMPP and ignore Jabber user community, which has recently overgrown ICQ community by amount of users.
They talk about interoperability IM-gateway in a future tense, whil most of Jabber users already use interoperability today. For example, my Jabber client doesn't communicate directly to my ICQ or AIM buddies - it does it through the Jabber server instead.
I don't wonder they don't talk about personal/SOHO Jabber servers, which some percentage of Jabber users connect to, instead of direct connecting to public server, in a process to communicating with the rest of the world. Of course, Microsoft prefers everyone will connect directly to MSN - they don't like people building communities out of their control.
And, yes, IRC is missed. I don't like some features of IRC protocol personally too, but the fact is that IRC is here for many years, has a community, applications, and still good concepts.
Well, what do you expect from the guy, who works for Microsoft (the company responsible for so many viruses due to poor architectural design of their products) and Sendmail (the company responsible for so many spam due to poor architectural design of their main product)?
I am so disappointed that I wasted my time on RTFA.
Less is more !