Showing posts with label VISTA. Show all posts
Showing posts with label VISTA. Show all posts

Thursday, 14 August 2008

Will Midori have Brains as well as Brawn?

I recently blogged about the arrival of Vista. The gist was that Vista was suffering from the bloated consequences of Microsoft’s golden rule of compatibility. The “new hope” being Midori which apparently drops this mantra with a fat free Operating System embracing the latest technology and paradigms of “manycore” computing etc.

Great! This will definitely have a positive impact on security, performance, scalability etc. But in looking for a new operating system I still want something more than just a young fat free athlete. I want it to have brains as well as brawn. I know I’m demanding, but then Microsoft does have the resources to deliver.

What do I mean by brains? Well to be honest I don’t have a nice one line explanation. But let me start by giving you some context....

In January, Bill Gates was over in the UK doing what I’d describe as his “goodbye” tour and I was lucky to get invited to one of these events. Even luckier still, I entered a question which got chosen to be asked of Bill Gates. This was my question....


Is the pace of innovation slowing down?

"There have been many technological and software innovations over the years with the Internet, Client/Server, Web applications and .NET etc., but is the pace of innovation slowing down? .net was a major innovation but since then many would characterise Microsoft releases as user experience enhancements? Are we likely to experience this consolidation until the next leap, perhaps to AI platforms? What work is Microsoft doing in this area?
"

Bill’s answer wasn’t a one line answer either, but the gist was that Microsoft can only move as fast as the market will allow etc. True of course, but who do you choose as the pace-maker? Some organisations are still running DOS and Windows 3.1 whilst others are more than happy to lead with the latest (and yes, of course, a lot of organisations sit in between). The point is that you need to cater for both with the leaders helping to mature the new products that the followers later upgrade to. So whilst I accept Bill had a point, I don’t accept it fully.

So back to my point about the next operating system. I guess some would argue, that Microsoft breaking compatibility, is in itself, a momentous decision (if they do make it!). However, it won’t be that dramatic, since either Midori will run in a VM under windows or windows rill run under a Midori VM; or both concurrently under VM. So not really as momentous as we might think.

So what does a windows operating system - with brains - look like? Well here is just a taster (perhaps you can add some of your own)...
  • It should be fully integrated to the cloud with security implemented as standard.
  • Data/Apps stored online with backups to the pc/laptop rather than the other way round.
  • Solutions – not simply software – should be delivered jointly through install and services adding greater value.
  • Removal of data silo’s through xml standards and online data hubs.
  • Improved meta-data (as well as standards/hubs) to aid search.
  • Not having to rebuild your PC/Laptop every 6-12 months to restore performance.
  • Application installs that do not pollute the Operating System: integration rather than invasion.
  • Ability to restore your environment online to any laptop/pc.
  • Software that adds to the whole rather than simply a point solution (this requires the Operating System to have the right architecture).
  • An Operating System capable (really capable) of self repair and that can diagnose (and fix) automatically and not come up with dumb suggestions like "the cable isn’t plugged in" when you are trying to connect via wifi.
  • An Operating System that supports a degree of artificial intelligence which can be leveraged by applications - and I don’t mean pathetic “office assistant” type gimmicks.

If pushed, I would define the new O/S as being: Secure, fully integrated to the cloud, AI Powered, Data and user centric. e.g a long way from where we are today.

For the majority of users out there who don’t understand about firewalls, service packs, antivirus databases, backups etc (and quite frankly they don’t want to) this would Rock their World. Oh and yes, they wouldn’t go out and buy a new PC each year because they don’t know how to fix the one they have that each year gets infested, slow and unusable.

Like I said, Microsoft could deliver on this. If not I can see both Apple and Google eying up the possibilities of giving birth to this new type of computing.

Monday, 11 August 2008

The Vista Saga

There has been much noise on the arrival of Vista in our lives with deep emotion and drama. Having used Microsoft's latset Operating System for over 6 months I can relate to this emotional experience!

To me, an operating system is much like a car; it’s a means to an end. We expect it to do the business; start first time, get us to where we want to go safely and quickly, and without breaking down etc. We don’t celebrate that it achieves this, but get very emotional when it fails, as we rely on it, day in day out, to get us where we want to go.

However, emotions aside, it’s time to take a dispassionate view of the Vista Saga. I’m not one who views Vista as a great product marred by poor perceptions (as Microsoft would suggest) or that Vista has transitioned overnight from a Skoda to a Rolls Royce with the arrival of Service Pack 1. But nor do I believe Microsoft was invaded by gremlins that had too much food after midnight.

What I do believe is that Microsoft are finding it harder and harder to build on top of windows. Vista, of course, is not a single operating system, but just one more incarnation of NT which had to inherit legacy windows which itself was built on DOS (and with many bits, 16, 32 and 64). It has also had to cope with many files systems from FAT to FAT32, NTFS (and until it was jettisoned from the plans, WinFS). It has also had to cope with the myriad of configuration architectures, from .ini to .reg to metabase to .xml. The list is not exhaustive but you get the drift.

The Microsoft Windows dynasty, whilst hugely successful, has got fatter and fatter and is now not just clinically obese, but morbidly so. This overweight geriatric, whilst sporting beautiful clothes, is about to croak.



Vista stuggling to delete a 443 byte shortcut in under a minute

Sorry, I said I’d stay away from emotions.

Microsoft’s mantra of compatibility has lead to windows inheriting everything from the past. This has resulted in the bloating of windows and critically, 3 increasingly impossible missions: Performance, Quality & Security.

Windows development and testing is constrained by the many "personalities" that exist within it and the many compromises that Microsoft have had to make to ensure compatibility.

Trying to ensure performance, quality and security whilst not impacting on compatibility must be like trying to tap dance through a mine field. And of course testing all the different components, personalities, configurations, files systems is a gargantuan undertaking that in reality could never be completed fully.

So the real problem isn’t the developers or testers but the impossible mission that Microsoft set itself in the name of backward compatibility. Fortunately it looks like Microsoft have an alternative plan to develop a compatibility free, fat free replacement called MIDORI.

The only question is whether Microsoft will try and put windows on an Atkins diet to extend its life or accelerate Midori to enter in a new dawn.

Whatever they do, I just hope it does the business.