BT

Windows Server 8 Marks Shift Towards GUI-Less Future

by Jeff Martin on Oct 03, 2011 |

Microsoft has alerted Windows Server developers and administrators that the platform's future will be one in which the traditional GUI applications will no longer be supported. The Server Core installation option, originally debuting in Windows Server 2008, will become the preferred installation option for Windows Server 8. Server Core offers increased security and performance benefits over the historical full server with full graphical shell option.

Microsoft's Jeffrey Snover, Distinguished Engineer and Lead Architect of Windows Server, and Andrew Mason, Principal Group Program Manager presented the session “Windows Server 8 apps must run without a GUI - learn more now” at Build 2011. This new focus on a GUI-less future for Windows Server was prompted by several goals, including improved security, reduced storage requirements, and easier remote server administration.

Before Windows Server 2008 the full GUI environment was the only install option available for Windows Server platforms. Windows Server 2008 saw the introduction of Server Core, which allowed an administrator to decide between that or the full GUI at the time of initial installation. Unfortunately this meant that if a server's role changed in the future, a fresh installation was needed. With Server 8, this is changing as administrators will be able to switch between a full server UI install or a Server Core-based install at any time.

Despite the shift, Microsoft embraces of the uses of GUI-based management systems, but intends for those systems to be run on a client machine separate from the server. This separation allows administrators to benefit from the benefits GUIs without introducing greater security risks to the server platform. Examining the performance of a Server Core installation versus a traditional full GUI install under Windows Server 2008, Microsoft found that the need for critical patches was reduced 50%-70%. Microsoft feels so strongly about the improved security with Internet Explorer's removal that it has also removed User Access Control from the Server Core installation.

A recent study by Danish firm CSIS researched the vulnerabilities introduced by typical desktop applications, indirectly demonstrating how preventing their execution on a server would increase security. Common GUI-based applications Adobe Reader and Adobe Flash, for example, were responsible for 48% of the most common exploits used by malware writers.

Beyond the security benefits, the Server Core installation option simplifies multi-machine automation, lowers the installation footprint, and improves performance. To assist administrators with this change in philosophy, PowerShell support has been expanded to include over 2300 cmdlets with Windows Server 8.

Hello stranger!

You need to Register an InfoQ account or or login to post comments. But there's so much more behind being registered.

Get the most out of the InfoQ experience.

Tell us what you think

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Email me replies to any of my messages in this thread

You mean... like Unix? by John L

So nice to see Microsoft finally starting to catch up to Unix.

I feel sorry for those folks who feel the need to remote desktop to a server to get anything done. (Not.)

Re: You mean... like Unix? by John L

(Yes, yes, I realize that ssh/telnet is the Unix equivalent of remoting to a server, but there's quite a lot that can be accomplished with just exported filesystems, since everything goes in .rc files.)

Microsoft doesn't need to be a copycat... by Clinton Begin

So seemingly so, Microsoft's current strategy is: Be like Mac on the desktop, be like *nix on the server, be like Google on the web.

If anything is to ever be Microsoft's undoing, it will be this. For decades, love them or hate them, Microsoft has succeeded by doing what they think, what they believe and what they want. Absolutely they borrowed ideas from competitors, but they still did it their own way. Sure they didn't always do the best job... but they were still Microsoft and did it their own way.

Companies that follow, fail. Even though Microsoft may have been copying, they weren't necessarily following. They were making smart decisions about implementing that which was known to work, in addition to trying a few things that were known to not work. They succeeded in some and failed in others. But they did not follow.

For the first time, I see Microsoft as truly following. From Windows 8 Metro, to Bing, and now to this. What they're losing sight of is that their core has always been business. They should remain business first, and muscle their way into the home on the back of business as they have always done.

Currently neither Apple nor Google have a viable alternative to Windows and Office in the workplace. And I'm not sure they even care to. Both companies are after the general consumer. In Apple's case, they seem to be moving further from business, not closer. And in Google's case, they're approaching it with all of the vigor of any of their other 10 year betas.

So the immediate future will see you owning two computers and two phones. The Mac and the iPhone you bought for yourself. And the PC and the BlackBerry that your company bought for you.

Now we all know RIM is dead. That is a virtual certainty. No hope.

This leaves an immediate gap for Microsoft to fill: Corporate targeted phones. Make them business first, but then add the flash, flair and fun to them second (and without threatening the business side). They don't even have to be as good as the iPhone... just close enough to make it not worth the extra $200 + $50/month for a personal iphone. Heck, if I were in their position, I'd buy RIM, keep the hardware, put Windows Mobile on it, hook it up to Active Sync and give phones away to everyone who buys exchange server for the first year.

Computers are the same way. Business first. Add the fun second. And when portability finally dominates in the workplace, then make the computer the company bought for its employee just "good enough" to make the $1000 for the MacBook Air not worth it.

Microsoft has to let Mac be better. Let Google be better. Let *nix be better.

And then they have to make it not matter, like they have always done.

If they follow, they're admitting defeat and failing because they'll be chasing someone else's business plan from 5 years ago to attempt to be the market leader 5 years from now (~Tom Peters).

IMHO
Clinton

Re: Microsoft doesn't need to be a copycat... by Jean-Jacques Dubray

>> They should remain business first, and muscle their way into the home on the back of business as they have always done.

but, Clinton, this is overwhelming false, today, computing has become a "consumer" market, devices are built for consumers first and maybe the business might find a use for them. Apple has started to focus on consumers in the early 2000s and it paid off. Facebook made it a de facto strategy and Google jumped on the bandwagon later. Today, either you compete with IBM and Oracle, or you compete with Apple, Facebook and Google. Thinking that you'll attract consumers with business gears, tools and goodies, is IMHO the wrong strategy. Nobody cares about Windows and Office, any longer, and the people that do are quickly becoming the minor minority (<5%).

So either Microsoft splits itself to compete on both markets effectively or it will fail, because businesses will not necessarily want what consumers want and vice versa (both on the client(s) and the server)

Re: Microsoft doesn't need to be a copycat... by Clinton Begin


but, Clinton, this is overwhelming false, today, computing has become a "consumer" market, devices are built for consumers first and maybe the business might find a use for them.


I said nothing to the contrary. In fact, I said exactly that. I said that Apple and Google etc. are focusing on the consumer first. I said that in plain English. So I'm agreeing with you there.

Where we disagree is that business likely will NOT find a use for such things, anymore than they'd find a use for an XBox or a DVD player in the office. In fact, many companies block Facebook, some even block Google (hell, some COUNTRIES block Google). And Apple might as well be considered banned from the workplace -- if you don't believe me, get caught texting, watching videos or playing Angry Birds on your iPhone at work.

My point is exactly that nobody is offering a compelling alternative for the business, likely because there is way more money in consumer oriented devices. Apple knows this, and they're smart to focus there.

My point was furthered by my suggestion that rather than compete with Apple, Microsoft would do better to focus on one thing that everyone is losing sight of: business. And yes, if they manage to retain some household market share because the workplace provided devices are good enough, then great.

In any case, those were only supporting points. And this thread is more about the server side and *nix, I was only using those others as examples of where this is already happening -- mostly because we can only speculate on this apparent GUI-less Windows future.

My global point was that it was disturbing to see Microsoft giving up on what they believe to follow Apple, Google and Linux...

One case in point regarding how they can succeed without following in the consumer space: XBox. They did everything different there. Commodity hardware (yes, here commodity was different), free development tools, innovative controller, and most recently: Kinect. Here they've shown that they need not copy and need not follow... or even if they do, they can succeed with innovation. I'd like to see more of that kind of thinking in their OS and business space.

We don't need MS Android, MS iOS, MS *nix or MS Facebook. We need Microsoft.

Clinton

Re: Microsoft doesn't need to be a copycat... by Jean-Jacques Dubray

>> My global point was that it was disturbing to see Microsoft giving up on what they believe to follow Apple, Google and Linux...

yes, this is exactly what kills companies. Sun got killed because it decided to react to every move that Microsoft did. Microsoft is a bit better at that game, they have kept an eye on the financials, but eventually this is how you become a bureaucratic shadow of yourself, PMing instead of innovating. Google seems also slipping into that mode chasing both Facebook and Apple and with search becoming less relevant to their top line.

The overall problem of these companies though is that when you become so big, any innovation looks like a distraction, bringing only a few 10 or 100 million initially until they reach the billions they are all chasing.

Re: Microsoft doesn't need to be a copycat... by Clinton Begin

>> Google seems also slipping into that mode chasing both Facebook and Apple and with search becoming less relevant to their top line.

You and I agree more than not, I think! :-) I also feel this way, and you're one of only a few others that I've encountered with this opinion.

Perhaps specialization is what's best. These mega-corporations are trying to do way too much. It's hard to say though, because one in a hundred of their gambles pays off. Google's Android gamble is working (patent battles aside), Microsoft's XBox paid off, and of course Apple's iPod and iPad all took off -- they're on a streak.

So maybe it's unfair of me to think they're doing too much, when really it's just a giant gambling game, whereby they increase their odds of a success with a shotgun, rather than a marksman approach.

I think the consumer ends up paying the price though -- in time, money and confusion.

In these times, I suppose it does not pay to be an early adopter.

Cheers,
Clinton

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Email me replies to any of my messages in this thread

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Email me replies to any of my messages in this thread

7 Discuss

Educational Content

General Feedback
Bugs
Advertising
Editorial
InfoQ.com and all content copyright © 2006-2014 C4Media Inc. InfoQ.com hosted at Contegix, the best ISP we've ever worked with.
Privacy policy
BT