I'm not going to mince words: this video is something Apple should take a very serious look at if they want to make any sort of headway in the enterprise desktop space during the next couple of years.
For those of you reluctant to use RealNetworks' junk software or lacking a plugin for Windows Media, the video depicts Vista's built-in basic application sharing and shared workspace features. They call them "collaboration sessions", and it seems extremely easy to use.
I've been involved in CSCW ever since we started pushing H.320 down an ISDN pipe using a computer instead of dedicated hardware, and despite the dismal success that stuff like T.120 has had ever since, I've kept an eye out for this kind of stuff, and it looks like Microsoft might finally be doing something right.
They did, of course, do a lot of things wrong. For instance, in the early days of T.120 they pretty much ran away with the ball (and killed off all the competition) by giving NetMeeting away for free.
And for the last couple of years or so they've been reeling Groove in as if it were a halibut. It had an auspicious start as a moderately open collaboration tool, but then "went Visio", i.e., it became progressively more tied to Windows, eventually culminating in its acquisition and complete dismemberment.
The clever bits are now being bolted on top of Office (mixing and matching with their twin enterprise collaboration behemoths, SharePoint and Exchange), and my guess is that the basic features (or something that they might have provided inspiration for) are what's being shown on that video.
Mind you, I'm well aware that Microsoft Research knows more than a fair bit about collaboration, P2P protocols and IP service discovery, so it might just as well be completely in-house stuff. In the end, the technology's origin is pointless.
NetMeeting was, by and large, a failure. People never really took to it, both due to the crummy interface and the difficulty in connecting to your peers and actually using the damn thing. The relative scarcity of Internet connections and the fact that NetMeeting was completely NAT-unfriendly also helped.
Mixing NetMeeting and MSN was their second attempt, and it pretty much tanked as well - you only have to look once at the current MSN eyecandy-centric interface to figure out that IM users don't really care about the collaboration options.
But having basic P2P file and application sharing on the base OS with auto-discovery and painless workspace creation, now that's another thing altogether.
If I am to take that video at face value, setting up such a workspace and sharing a desktop takes all of 20 seconds.
Before the Linux groupies start hitting the comments link in droves to say that this can all be done with VNC, my main point is that it is all seamlessly integrated into the OS and required zero fiddling about with configuration settings.
I can't even begin to count the number of meetings where that sort of thing would have been useful (even before Wi-Fi became a standard accessory on all our laptops), and it is sure to be a feature that I'll use every working day when my company rolls out Vista on our corporate machines (which is sure to take a while for all sorts of practical reasons).
Every Apple Is An Island
But, more to the point, this is the sort of thing that Apple really ought to be doing to make their operating system and their machines more attractive to corporate buyers (as opposed, to, say, Dashboard widgets).
Besides the obvious points of leveraging Bonjour/Rendezvous to handle peer detection and getting iChat to do a bit more than fancy animations and video calls, most people aren't aware that the Remote Desktop service in Mac OS X can also share the screen using VNC's RFB protocol.
It's one of the poorest implemented VNC servers in existence, but it sort of works, and the only thing holding it back is their insistence on selling it as a classroom management product. Yes, even as Microsoft gave away a "good enough" version of their Remote Desktop service as part of Windows XP, Apple insisted on selling their Remote Desktop solution as a (rather steeply priced) classroom tool.
My take? Well, I don't think Apple is really into the collaboration thing. Their repeated shunning of traditional groupware (even down to the brain-damaged Exchange support in Mail.app), their focus on keeping iChat "cool" instead of functional and their dismal positioning of Apple Remote Desktop are more than enough evidence that Apple just doesn't get it.
Sure, Apple is more focused on the home segment, but even so, they can't ignore corporate IT trends (or undercurrents) for ever. The thing is, corporations are more and more focused on actually using computers to increase productivity (it's a novelty trend, you might not have heard of it yet), and, whether we like it or not, they're pushing for more and more CSCW tools.
Most of these are the usual centralized (and increasingly web-based) document libraries and time reporting tools (more or less loosely coupled with ERP stuff), but those are already behind the times.
The Ad-Hoc Workspace
And the reason those mammoth CSCW tools are behind the times is that the very nature of work has changed - people now slice their working hours by hopping in and out of small, tightly focused groups (not all of them physically present on the same place), and what I like to call "ad-hoc workspaces" are the tool to cope with that.
Groove did those very well indeed, and I could only find fault with their single-platform approach and a touch of creeping featuritis.
Microsoft has apparently pared all of that down to basics (see my desktop, take these files), and dropped that in to Vista (whether or not you'll have to use MSN or any of their newfangled web services to have more functionality is yet unknown, but my magic 8-ball says "pretty damn likely").
Should Someone Fill The Gap?
Now, there's all sorts of hypothesis as to why Apple would shun desktop sharing. It's either complete and utter blindness to how people actually use computers in groups (which seems a reasonable flaw for a company that focuses on delivering the best individual experience possible on its computers), or (and here's where the Watson/Sherlock conspiracy theorists can jump in at any time) It's almost as if they were unsure of what to do and expected third-party developers to come up with a winning formula.
But the trouble with third-party developers is that they won't do something unless a) it's wicked cool, b) they can charge for it, or c) both.
And collaboration, despite its former coolness, isn't that cool, and not something a lone developer in a basement somewhere finds very appealing.
Maybe I'm being cynical, but these seem like pretty good reasons why, if you discount SubEthaEdit (which is utterly, utterly brilliant), the number of interesting real-time collaboration applications on Mac OS X is embarrassingly small. Maybe even zero.
Plus the real collaboration market is now equated with big, lumbering server-side stuff, and it's getting crowded. The applications are getting very complex indeed, and surely something as simple as screen sharing can be done with VNC, no? Who would pay for that?
And the answer is: not many people, in fact. Maybe some brilliant folk working out of a coffee shop feel like picking up on this, but I don't think that's likely to happen.
In The End, It's Too Damn Useful To Ignore
But that doesn't make basic ad-hoc screen sharing any less useful, and I hope someone, somewhere out there is reading this and wondering why the hell Apple hasn't done this first, especially when they already have all the technology at hand...
It would be a hell of a lot more useful than Dashboard or Front Row for you to gain mind share at corporations, and you can't honestly expect to sell ARD if every Mac user can't experience even a little bit of it...