FUD or dud? An overview of Longhorn for Mac users

A couple of days ago I promised to respond to some of the comments on my (brief) post about Longhorn, and why it’s important for Mac users. Sadly, work prevented me from doing anything before today – well actually, it’s not that sad as it means I was busy and therefore getting paid!
But before I go into why it’s important for Mac users, it’s useful to have a look at what Longhorn actually is, because the comments in my last post showed a little bit of a lack of knowledge. For example, there was a lot of talk about how major features for Longhorn (and particularly WinFS) had already been dropped, which really isn’t the case. There was also a lot of “Longhorn is just playing catch up with OS X”, which isn’t completely true either. Parts of Longhorn are just replicating in Windows what Apple was able to do when moving from Mac OS 9 to OS X. But parts of it go beyond anything that Apple has now.
So what is Longhorn? On the most simple level, it’s Microsoft’s next-generation of Windows. It’s as much of a jump for Microsoft as Windows NT was, although thankfully it doesn’t require a complete rebuild of all of the Windows code base as far as I know. That alone, of course, makes it worth keeping an eye on.
There are four core parts to Longhorn, which I’ll look at in turn. These are Fundamentals, Avalon, Indigo, and WinFS.

Fundamentals
This is a catch-all category that includes some very important core elements of Longhorn, aimed largely at administrators. One such is Manageable Apps: tools allowing developers to build applications that can more easily be managed by system administrators. Another core fundamental is the much mooted Next Generation Secure Computing Base, designed to make it much more difficult to write viruses and much more. And, of course, there’s enhancements to the identity system, and to Tablet PC.

Avalon
Avalon is the codename for the presentation system in Longhorn – the code that allows developers to draw on screen. It includes quite a few bits of “catch-up” with OS X, for example by providing native hardware acceleration, just as Quartz Extreme does. Like Quartz Extreme, thought, there’s a lot more to Avalon than simply a toolkit for drawing pretty windows. One key part is XAML – Extensible Applicaton Markup – which is a markup scheme for .NET objects, and something that aims in part to bring designers closer to the code when creating interfaces. Conceptually, what XAML does is put code and interface on the same level in Windows – something that’s not been true in the past. Avalon also streamlines using media in applications,

Indigo
Indigo is a set of .NET technologies for building connected systems, including a new communications infrastructure built around Web services. It brings together features like .NET remoting, ASMX and others into one consisten and unifed whole. That means that, in part, it’s a tidying up excercise. but it also makes it a lot easier for developers to build applications that communicate over the Internet, including peer-to-peer.

WinFS
The most misunderstood part of Longhorn is probably WinFS. WinFS is the storage subsystem of Longhorn, allowing users to search, organise, and share data. Essentially, it’s a database of what’s on your machine, allowing you to find files by author, title, properties, or whatever. For end users, this model makes a lot of sense in a world where users work with thousands of files, from music to images to Word documents, and need to be able to find and categorise them in a way that makes sense to them. For developers, it gives them the chance to build applications which share data from common stores far more easily.
What WinFS isn’t is a replacement for NTFS, the file system that’s currently the standard on Windows. WinFS sits on top of NTFS, but it’s file-system independent at least in theory: the documents that WinFS accesses could be sitting anywhere on a network, accessed via WebDAV or some other method. However, the “feature cut” that got so much coverage recently wasn’t, really a feature cut at all: from what I can gather, WinFS will not be a part of Windows Longhorn Server. But, and this is the important bit, WInFS is still part of Longhorn. It’s not going away.

One more thing…
The last thing about Longhorn, and the one that’s probably stirred up more ire from Mac users than anything else, is its new user interface: Aero. Note that Aero isn’t part of the core of Longhorn, and it’s the part that’s in flux more than any other. It wasn’t part of the developer release of Longhorn given out at this year’s developer conference. And yes, the screens that have been released are remarkably Mac-like. There are, however, some nice little touches – RSS integration, for example – but so far the best thing to do with Aero is wait and see.

So… the point is…?
If you’re with me so far, you’re likely to be asking a simple question: so what? All this stuff sounds interesting, but it’s not the kind of thing that will shake the world of any end user. It doesn’t sound as cool as Rendezvous, or iSync, both of which are great technologies that it’s easy to get to grips with.
And that’s kind of the point. At the moment, Microsoft isn’t interested in end users with Longhorn: everything it’s announced and released is aimed towards developers. In particular, it’s aimed at exciting developers and allowing them to build fantastic net-enabled applications. And this answers one of the key points that many people made in the comments to my last post, which was that Longhorn is at least two years away so why care about it? The answer is that, for developers, Longhorn is real, now: you can already begin working with it to plan new applications for release after its final launch. Of course, as a developer you’re not going to pitch totally into Longhorn development alone now: but you can begin to plan and work out what it’s capable of.

Developers, developers, developers!
And that’s where Longhorn is a challenge already to Apple, because the mind share that Mac OS X has with developers is vital to the platform’s future. Apple has to regain the high-ground in terms of developer excitement, something that it did very effectively when OS X was launched. Can it do it? Of course. But in order to do so, it must treat Longhorn as a real threat, and act now to counter it.

  • Dave Dombrowski

    So Ian…. let me understand you. Since I’m a developer of Windows and web apps for a brick’n'mortar company of around 2000 users, Longhorn matters because I can develop something TODAY that the clients I support can use?

    No. Like Scoble (and I argued this with him through comments) this is nothing but a sales pitch. Of the worst kind too – since it distracts many from what is the most important: what SOLUTIONS can I develop that can be used TODAY.

    LMAO. You see, Scoble even tried to tell me that if I “desired to remain relevant (IE not lose my job to someone in India)… ” I _have_ to attend last October’s PDC. No… I only lose my job if I can’t develop solutions my clients can use TODAY. Oddly enough, if I spend time learning something that as yet has no release date and the hardware requirements mean they won’t want me developing for Longhorn in ways that will use the features you mention here for AT LEAST two years AFTER the release…. get my picture?

    My biggest gripe is how MS handled certain things internally. The .NET Framework is fantastic. The combination of Windows, Office and MSSQL is extremely powerful. Yet, they put so damn much emphasis on Yukon and now Longhorn that it will be AT LEAST 5 years between releases.

    Yeah yeah, Apple had the same exact thing with OS X. But consider this: MS also stopped development of things in use today. At one point they were not releasing any new version of MSIE. They still are not planning on a new release of Outlook Express. Combined, these two archaic apps are responsible for a multitude of virii.

    Why is the XP firewall defaulted to OFF on installs? Why is this not available as a separate download for older version of their OS? They finally relented (months late BTW) with XPSP2. But my MSIE still isn’t tabbed.

    This lost/misplaced focus – of which YOU so very eloquently exhibit here – is resulting in a void in the marketplace. Even when Longhorn, with all of these great features (I mean that sincerely), won’t “matter” one bit to my 82 year old mom if it means spending a few thousand on a new PC. Care to explain the compelling reasons a small business won’t think the same?

    This void in the marketplace…. the NON-enterprise user of tomorrow… the user of TODAY… will be filled. The question is by what? At this point, MS dropped that ball big time in 2003. BIG time. So the question for them is not “does Longhorn matter” but “can we support our customers today”.

  • Tom Barta

    The trouble with Microsoft is that no matter how crappy their implementations, millions will buy their products regardless. Hopefully, OS X will pick up some momentum outside the Mac community in the next couple of years. It has to some extent already, among the UNIX crowd, but not in nearly the numbers we need. I haven’t seen the kind of increases in marketshare I had expected at this point; if anything, the numbers are looking stagnant. Yes, I know these kinds of numbers are grossly inaccurate, but if Apple were roping in LOTS of switchers, it would show up in the stats.

  • http://technovia.typepad.com Ian Betteridge

    The short answer, Dave, is no: you don’t have to worry about Longhorn at all. You develop applications for a company, and so you serve an installed base. You don’t have to even think about it, if you don’t want to.

    Slightly longer answer: it’s up to you. If you find features in Longhorn that will help you to create an application which will better serve your clients, then it’s *you* who’ll start pressuring them to upgrade (or pressuring The Powers That Be to upgrade them). But again, that’s up to you and realities of your work. Personally, if I were you I’d certainly start learning about it, more out of curiousity than anything else.

    But beyond that, I think some of your claims are wrong. Has Microsoft “stopped development of things in use today”? Given that I’m currently using XP SP2 on two machines (and the improvements on Tablet PC alone are worth waiting for) I suggest they aren’t. Would you rather them release a buggy SP2 that you would end up having to pull off half the machines in your company, or would you rather they delayed it a couple of months and got it right?

    Will Longhorn mean buying a new PC? I don’t know that – and neither do you. Will those same small businesses be buying any new PCs between now and the release of Longhorn? I’d be very surprising if they didn’t – and I’d be very surprised if a machine that’s bought today won’t run it.

  • http://www.snappingturtle.net/jmc/tmblog TM Lutas

    Longhorn matters for Apple but not necessarily for the reasons mentioned so far. The transition to Longhorn is going to be long and painful. There’s going to be plenty of low hanging fruit for Apple to pick off in the business space and they’re aiming for that space.

    The announcement of Xsan is a perfect example. You get a compatible solution for a lot of existing SAN implementations so it’s not a completely unknown quantity and you’ve got extremely significant hardware savings.

    The adoption of Samba is also significant. When Active Directory support becomes seamless, there’s no reason to pay client access licenses for Windows anymore. This is a significant savings for a great many businesses who need file and print even more than server side applications.

    Remember, Apple is a hardware company first and foremost. Xcode, GCC, all the eyecandy that is aqua extreme, all that is the candy coated shell on the real sales product, Macintoshes.

  • http://technovia.typepad.com Ian Betteridge

    TM, you make a very good point: every time that there’s a major transition in Microsoft’s OS strategy there’s a chance for Apple to pick up business. But only if it has a compelling case, which of course means ensuring that OS X doesn’t stand still (not something it will, I hope!)