Infoseek tried it, Lycos tried it, AltaVista tried it [...]

> issues of not have good network all the time

Gears

> people what to use their devices for more then just internet access

Put an http server on the devices.

Instead of starting with Live Mesh, Microsoft should first build Windows on top of an embedded SharePoint server.

ugg. Young jedi. They aleady did that, it was the Green screen and a mainframe. Today you could connect to your citrix server with a browser with mixed results. IBM tried it, WordPerfect tried it, Corel tried it, Oracle tried it, others have tried it. They all failed and some died because of it. Besides the issues of not have good network all the time and html/java sucking as a UX, people what to use their devices for more then just internet access. There are many folks still left that just want to get stuff done.

Who is going to be porting the library to all the non-Windows devices and providing support for those products? Novell?

>> How about when you combine it with XMPP? Meshy enough for you?
> Not sure what your suggesting.

Use XMPP for status updates of sync meta data. Establish a community CDN on top of BitTorrent for large files shared by a lot of users. Automate the downloading of the rest from personal cloud over HTTP or http://FTP.

> the idea is to keep everything as a feed.

I do like modeling lists of lists as feeds of feeds and see value in SSE managing these structures. I think calendar data fits nicely in this model. I even believe that UIs can be effectively represented as recursive feeds. However, I believe that there are limits to the effectiveness of modeling everything in this fashion. That’s a very large ocean to boil. Just convincing others to buy into that vision will be a difficult task.

> With +97% market share on their own platform [...]

97.46 percent of the global desktop operating system market does not mean anything to someone that believes that the desktop is dead. One day in the near future most devices will boot directly into a web browser. No local file system exposed. Widgets replace applications.

> Apple needs good apps and platforms more then MS needs iPhone users.

Who is going to convince Steve Jobs of this? Not Microsoft.

> Who cares about Android

I do. Everyone at work has iPhones and I still have this crappy Palm piece of shit. I have been holding out for an Android phone because I have developed apps for both platforms.

> it needs a lot of work

Really?

http://www.engadget.com/2008/05/28/google-demos-the-htc-dream-at-i-o-conference/

> why switch topics?

I am not. If Mesh succeeds at all, it will run smack into Android and fall over.

So what? .Net library is free and callable from many languages. So your saying MS can’t create their own libraries and apis or you just in the camp if it comes from MS you can’t use it?

> How about when you combine it with XMPP? Meshy enough for you?

Not sure what your suggesting. Encoding torrent stream in base64 and including in xmpp or including the torrent link in the xmpp and downloading out-of-band? If first, torrent not needed. If latter, going out-of-band with another api is not the idea. The idea is to keep everything as a feed. We could already do out-of-band sockets or ftp or ws-* today, but that is another mental context switch and more complication for the dev. The user could care less as long as it works, the devs want a single unified model which is why they are using feed model.

>Dude, you need to stop drinking the kool-aid before its too late.
>Mesh is vaporware as far as I am concerned until it supports other >desktop and mobile operating systems.

Dude. I drank a lot of kool-aid doing Sun, AIX and System V system programming for 10 years. Come to think of it, a lot of pizza too. Not an enviroment that gives joy to me any longer. With +97% market share on their own platform, it will not be vaporware because you say so.

>If Apple can prevent Mesh from getting on the iPhone then it is >definitely a problem for Microsoft.

Apple needs good apps and platforms more then MS needs iPhone users.

>What does MOE stand for?

We both know. It is still a runtime - not an OS. But you knew that.

>Android is definitely an OS.

Who cares about Android (it needs a lot of work) and why switch topics?

And it also uses proprietary .Net technology, no?

> BitTorrent is not meshy

How about when you combine it with XMPP? Meshy enough for you?

> the idea is to have open xml packets instead of binary streaming

Idea? What problem is Mesh trying to solve that cannot be solved with existing solutions? It seems to be a solution in search of problems that people don’t have or have only because they are using MS products.

> This keeps communication consistent across platforms

Dude, you need to stop drinking the kool-aid before its too late.

Mesh is vaporware as far as I am concerned until it supports other desktop and mobile operating systems.

> Mesh will support Apple and plans for linux

Have any evidence of that besides press releases?

> Apples terms of service is Apples issue - not Microsoft’s.

If Apple can prevent Mesh from getting on the iPhone then it is definitely a problem for Microsoft.

> If they what to open it, they should.

Apple does not want to open the iPhone for Mesh, Silverlight, Flash, Java, Android, Firefox or anything that resembles a platform. Jobs wants to lock that sucker down so he can drive the market.

Another prediction: Android will force Jobs to open up the iPhone and App Store.

> It is not an “Internet OS” and never will be.

What does MOE stand for?

> The browser is not an OS. NFS is not an OS. FolderShare or
> Google Drive is Not an OS. Combining them is not an OS.

Correct. However, Android is definitely an OS. Android already runs on Linux, Windows, and OS X. How hard do you think it will be for the OSS community to create Android browser plug-ins? Java applets will rise from the dead with new powers.

TMK, Mesh will support Apple and plans for linux. Apples terms of service is Apples issue - not Microsoft’s. If they what to open it, they should. There is some guy doing an “open” version of Mesh (ala Mono like) I can’t remember the link.

It is not an “Internet OS” and never will be. I don’t even know what that means or why people say that. The browser is not an OS. NFS is not an OS. FolderShare or Google Drive is Not an OS. Combining them is not an OS. Mesh is a platform - period. A way to share information between devices. Your OS is your OS, the network is the network.

Protecting the Windows monopoly at all costs.

I was not referring to file formats. I think MS has shown some good will here recently. ODF support in Office and leveraging Atom/APP are great moves.

> What other tried and tested client-server or P2P apps are you refering to

client-server: HTML/XML over HTTP
P2P: BitTorrent

> I have not had any issues so far with demo apps

Can I run the demo apps on my Apple laptop and my Palm smart phone? Does Mesh meet the iPhone App Store’s terms of service? Any plans for MS to support Mesh in Android?

I do like FeedSync as a feature. I just don’t have a lot of faith that MS will be able to position Mesh as an Internet OS.

2) User has the option (currently required in beta) to sync all feeds to the cloud. In RTM, user can P2P to devices and sync certain things to the cloud. So you have best of both.

3) With RSS, xml, and atom as the transports, what proprietary game is this?

4) The PM was one of the major devs in the new TCP stack in Vista and is very sharp.

5) Tooling for diag and test will be important I agree. But that is important or any company that does this. MS is good at platforms and tools (i.e. VS) so I don’t see that as problem far after V1. APIs will be callable from PowerShell (as all .net objects) so admin scripts can be rich.

6) What other tried and tested client-server or P2P apps are you refering to. Do they have any of these issues you refer to?

7) It is being built to handle MSN and larger scale-out.

I have not had any issues so far with demo apps - Live Desktop, file sync, remote desktop. They just work. I want other features, but that is always the case.

I am willing to wait and help in the beta and wait for RTM before saying it is not going to work or jam them on issues that just have not appeared yet.