Longhorn rescope redux…

So here’s what it comes down to…. Longhorn is going to ship without WinFS and Avalon is getting scaled back (I think I remember reading that they’ve chopped vector-based graphics, but I can’t find a reference), and the remainder of WinFX will be back-ported to Windows XP and Server 2003. There’s a bunch of other changes in all of the pillars, but these seem to be the ones that everyone is talking about.

I think this is a good strategic move for Microsoft.

The shareholders should be happy, because this decision makes good business sense. The original roadmap made all of us geeks happy. Things like database file systems and scalable 3-d screen elements make us giddy. While we are the early adopters and we generally drive the platform forward, we don’t make-up the majority of the user base. The new roadmap should be more in-line with the needs and wants of the majority of customers.

As an example, try to imagine how you would explain to your parents or grandparents, or to a non-geek business manager why they need WinFS… Yes WinFS is a good idea, and Microsoft needs to keep working on it, but implementing it within 2 years not only would be technically challenging, but the user-base isn’t ready for it, and they won’t be ready for it until it is seamless, easy-to-use, and easy-to-understand. To those of us that utilize databases every day, WinFS just makes sense, but getting everyday users on-board is going to be tough since we’ve spent the last couple decades teaching them to use files. And files were somewhat easy because we related them to documents & file-folders. How are we going to explain WinFS?

Me: “Well grandma, it’s like this… when you save this picture, you’re actually going to be putting in 4 different file cabinets… in this one the tabs are labeled by date, in this one, the tabs are the name of the person who took the picture, in this one they’re organized by size, and in this one the tabs are the names of the people and objects in the picture, and you might put it in a bunch of different folders in that last cabinet…”

Grandma: “That sounds like a lot of work… I’ll just stick to my 35mm..”

Many home-users and businesses are already lagging behind on the upgrade path. There is a combined problem of a lack of upgrade dollars, and a “what I have is good-enough” mentality. With that said, Microsoft needs to focus on making the Longhorn platform(s) a complelling and worthwhile upgrade for business and home users. For home users it’s about “What features will make my life easier and more fun?”, and for business users it’s “What features will make my business more efficient and more profitable?”

Robert Scoble has some good perspective here and links to lots of blog feedback here.

I’m psyched that we’ll see Longhorn in 2006. I’m disappointed that WinFS will lag behind. I’m hoping that Microsoft can keep this on-schedule.

Blog Comment Spam, and Thoughts on Rss-based Messaging

Raymond Chen complains about comment spammers: Spammers look stupid when they don’t read the blog they spam on

I deleted my first comment spam a couple of days back. I don’t even remember what it was about. I like having comments enabled, but I’d hate to see blogs get as cluttered and obnoxious as email has become.

I think email is going to die a slow death, mostly because the open-ness of it makes abuse too easy. Ideas like SPF may work to some degree, but if you trust too many entities, then spam gets through, and if you trust too few, the usefulness of email diminishes.

I’ve been toying with the idea of building an rss-based person-2-person messaging system. One of many ideas I don’t have time to develop right now… I hope someone beats me to it, for the sake of the Internet. Basically, individuals would “publish” multiple feeds. Some would be public (like most blogs today), some might be accessible by small groups (family news, small organizations, companies, teams and workgroups, etc), and others would be unique to individuals. You would subscribe to the feeds based on what you want to read. One of the big benefits is that your messages would be pre-sorted based on where you retrieve them from. Going on vacation? You could turn off all your advertisement feeds. Your personal feeds could pile-up, and you could check your business feeds from your PocketPC in the hotel. Prioritize information-load based on your priorities! This would make auto-filtering email obsolete, and it would give users _total_ control of the information they receive!

For an rss-based messaging technology to take hold, the process of building social networks has to become seamless. If I meet a guy at a conference, I need to be able to build the link with him in the last 5 seconds of our interaction. Something has to replace the email-on-my-business-card method. I’ve been kicking around ideas like temporary access tokens encoded with a 2-d barcode, or having an “connect with me feed” that people could subscribe to, and the client would automatically set up individual feeds on both ends.

Well, I don’t have time to pontificate any more on this now, and I think it would be best hashed out in front of a very large whiteboard.

Ian Griffiths blogs about the removal of Raw sockets in XP SP2

Ian Griffiths’ Weblog: “Raw Sockets Gone in XP SP2 ”

I haven’t decided if I agree with Ian on this yet. He argues that hackers will find ways around the limitation, and that it limits the platform’s usefulness as a security tool.

I think that making things harder for hackers, even if it’s just a little bit, is a good thing… But I also would rather not have to fire up a Linux box to run NMAP. One more good reason to keep a Knoppix CD handy.