.
/v3-uk/news/2002324/axing-winfs-signals-web-focus-microsoft
05 Jul 2006, Robert Jaques , V3
Microsoft has uncharacteristically "missed an opportunity" to spin its recent decision to cancel the Windows File System (WinFS) as a positive step that will allow it to focus more effectively on the web.
A recent advisory by Gartner vice president and fellow David Mitchell Smith stated: "Microsoft missed a significant opportunity to turn this potentially negative announcement into a positive.
"It could have been chosen to show that Ray Ozzie as the new chief software architect is beginning his drive towards Windows Live by terminating projects that do not align with it.
"Despite the company's reluctance to point it out, we believe that Ray Ozzie is having an immediate positive impact by making the decision to end WinFS. He should gain respect internally and externally as a result."
According to Gartner, WinFS is a "monolithic software component that has come to an end, though the vision for integrated storage has not".
The analyst firm noted that WinFS was to have had an application programming interface as well as search capabilities, but Microsoft has had difficulty describing a compelling need for WinFS beyond the 80 per cent 'good enough' offering via simple search.
An equivalent 80 per cent 'good enough' solution is now available from desktop search solutions from Microsoft, Google, X1 and others, and is in Windows Vista beta.
Overall, the end of WinFS will benefit Microsoft and users of the company's products, Mitchell Smith believes.
"The vision around unified storage has distracted Microsoft going back to the 1990s operating system, codenamed Cairo, which never shipped," he said.
"The unified storage issue also contributing substantially to problems that led to the 2004 reset of Longhorn. When Microsoft announced it was removing WinFS from Longhorn, Gartner said there was a 30 per cent chance WinFS would never ship."
The decision to axe WinFS was made public on 23 June 2006, when Microsoft said on an internal blog that it will to deliver separate beta versions of WinFS, but will include more mature WinFS technologies in the next release of SQL Server, code-named Katmai.
The vendor also said it has no concrete plans for WinFS as part of its desktop operating systems.
Microsoft customers do not need to do anything at this time, Gartner advises. The analyst company noted that few if any companies had actively planned to deploy WinFS, so this announcement should have minimal impact.
"Microsoft will focus more on delivering offerings that better fit with its Live vision and the web, as opposed to projects that reflected Bill Gates's vision for the PC operating system," said Mitchell Smith.
Do you agree?
Gartner are missing the point
WinFS was not really an end user feature - apart from the search capabilities. What would have been an end user feature were the fantastic applications that developers could have built with it.
Developer's spend vast amounts of time trying to hack into Microsoft's unpublished data stores (e.g. Outlook's PST & OST) and reinventing the wheel with their own new data formats when writing applications. Most of which are not open and are unpublished and therefor unusable by anyone else.
Finally developers could have easily accessed contacts and emails and any other file from any software supplier.
I believe WinFS would have spawned genuinely amazing products on the desktop.
I don't think the loss should be understimated. Check out Microsoft videos for the tip of what could have been a revolutionary set of applications.
Posted by Laughing John, 05 Jul 2006
WinFS will be replaced by ...
Laughing John,
You say, "I believe WinFS would have spawned genuinely amazing products on the desktop." Well, I believe you are right, up to a point. The amazing products you imagine will come about, but through a different technology.
IMHO WinFS was, typically, misconceived and was only part of the required solution. The replacement will be something else, something that is much bigger and more elegant than WinFS. So don't worry, the "revolutionary set of applications" are coming.
(BTW, I am not going to say what I believe the replacement to be.)
thanks.
Posted by cyber sammy, 10 Jul 2006
Oooh you're a tease ....
Cyber Sammy,
you can't say all that and then not tell .... :o)
I agree with what you are saying - I'm sure this concept won't die, and there seems to be a lot of debate around a more web-centric solution, databses etc.
To be honest I don't care what it is, but I do want to be able to easily access and send emails, read and create contacts and set up scheduling and other information. Ideally all in one place (or at least a sync'd copy). I want that data available across an organisation or even across the internet.
I want to be able to create data stores and easily sync data (on-line/off-line & client/server). I want automated back-up and recovery.
I want it built into the operating system or software stack (internet?) so I know I'm going to have access to it.
I want a database where I can add data columns "on the fly" (which can exist on some but not all rows). (I know the XML and variant types go some way towards this, not sure WinFS did). I want a flexible schema so that I can define my datastore and understand others.
I want to be able to easily integrate my software with other peoples (not just MS), both sharing my data and gaining access to others peoples.
Ideally I want it to be a standard, so I can use it across operating systems and I want it to be secure.
There's nothing new here - it's stuff we do all the time, I just want to take the pain away, stop reinventing the wheel and share the data around.
And what's more I want it yesterday, not in 5 years time... I don't ask for much ;o)
Posted by Laughing John, 18 Jul 2006