Why Windows Embedded 7 (WES7) should be part of SA, and how to force the issue with Microsoft

Last week there was a blog post and interesting discussion on the use cases for Windows Embedded Standard 7 thin clients (WES7). Outside of the usual "thin client vs.

Last week there was a blog post and interesting discussion on the use cases for Windows Embedded Standard 7 thin clients (WES7). Outside of the usual "thin client vs. other options," a key issue is the fact that Microsoft wants you to pay for the WES7 privilege. It's important to educate the community that this is complete BS and that we need to do something about it as opposed to rolling over and letting this sleep while the monopoly forces more cost down our throats.

It's a really bad idea to just pay Microsoft for the privilege of using WES7. If you want to do this, you need to refuse and stop letting them push you around. If you're naive enough to lock into Microsoft with Hyper-V for RemoteFX, Systems Center for management, MDOP for App-V, and SA, then it's time for a wake up call to understand that when dealing with Microsoft you always need leverage or they will screw you.

In today's world of Windows XP, WinFLP (Windows Fundamentals) is given to SA customers for no cost. WinFLP is essentially a stripped down version of XP that you can deploy on commodity hardware to get the advantages of a slimmer Windows build with the flexibility of local execution as needed for your desktop environment. This is great if you don't want to buy new client hardware for your desktop virtualization project since you can redeploy your existing PC fleet with a thinned-down XP build to build a better ROI story. But with Windows 7, Microsoft has not provided clear direction on what their new WinFLP-like solution will be. Their answer of WES7 helps, but unfortunately they want to charge for it! So what can you do about it?

The answer back to Microsoft is very simple: F*** you! Tell them the value of SA has been diluted and that you're not interested in signing up for it again. Even if that's a few years away for you, tell them now that you're not interested in upgrading to Windows 7 until WES7 is given to you as part of SA just like WinFLP is today. They REALLY care about Windows 7 adoption now so now's the time to put the brakes on it. While you're at it, tell them that MDOP is not something you want or are willing to pay for unless they make it part of your SA agreement with VECD. Then tell them you won't pay for that either until a user-based VECD license is made available. To help yourself with this, avoid App-V or don't fully deploy it, let them know all the gaps with App-V and point out how slow their products innovate.

That's a direct challenge to the value of SA and MDOP, two things they really care about. Refuse to use MDOP, let Microsoft know that it's garbage, remind them that App-V doesn't work with IE 6, and now you have several more reasons not to go to Windows 7/SA/MDOP/VECD.

Maybe you can use this momentum to get off Systems Center as much as you can now too. If you already have it, relegate it to simple functions like inventory (which is what I bet most people use it for historically due to SMS) and some software distribution. Try to introduce other products and create even more leverage here if you can. If that's not a current option, review your whole management strategy as part of your future Windows 7 plans and actively look to get off Systems Center later.

And don't forget about RemoteFX. You should also refuse to entertain that since it will lock you into Hyper-V (another thing they really care about). Let Microsoft know that RemoteFX doesn't offer enough value on it's own and that you need them to explain to you why it's a good idea for them to marry Hyper-V with RemoteFX.

While you're thinking about your Microsoft value, you should under no circumstance use the MS VDI solution. A HUGE benefit of desktop virtualization solutions is that they allow you to use other platforms. With mobility becoming a greater part of the ecosystem, Microsoft will simply not move to embrace non-Windows options, so stay away. It also makes no sense to let Microsoft lock you into Windows/Hyper-V (and charge for it) and then try to charge you again for all the management tools to make it work! These are all increasingly commodity items, and Microsoft has to include or add a lot more value to justify their current business model.

But I digress. Back to WES7...

Deploying WES7 is a great option as opposed to a true thin or zero client. It's a myth to claim that WES7 is terribly difficult to manage if you design a slim build and simply point it to your existing WSUS infrastructure (or the public SUS servers). If you avoid putting other software on it, your attack surface stays small. WES7 is also a great way for Microsoft to keep Windows on the client so others can enable features like reverse seamless to take advantage and add value to desktop virtualization solutions that need remote and local execution flexibility.

Of course when comparing WES7 to a true thing client, I agree the OpEx is not a low with WES7. (At least not on paper.) But the CapEx argument for thin clients is not true, because cheap PCs cost marginally more, also offer low power models, and as soon as you try to add a second monitor you are not screwed. The cost savings of a thin/zero client quickly comes back to haunt you when the user experience is compromised and you have to run something local. Windows clients (including WES7) will provide you with the most support for Windows hosted infrastructure.

So wake up and force the issue with Microsoft now. Put the brakes on your Windows 7 deployment, revaluate all the Microsoft management tools that lock you in, and demand more value for SA. Microsoft doesn't understand what people want and can't react fast enough to the needs of the future desktop. So keep your fingers crossed that they can make the one small gesture of providing WES7 free to SA customers to at least make SA on Windows 7 equal to what it is today.

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

Just tell Micro$oft you are ditching their bloated and crappy Office cow and moving to Google Apps. I bet that will light a fire under your M$ sales rep. There is nothing like sticking it to Micro$oft, they overcharge for mediocre crap and zero innovation. I promise you Google Apps is the way to go if you want to wrestle anything free from M$ including WES7.



Google Apps is poop - love the idea though.

They should stick to Google Wave... oh wait :) Seriously though - they have some way to go and M$ know it.

However the point I think your making is valid - there are alternatives to threaten to move to or in some cases, use!

@AppDetective... Entertaining as always. I like the stance and can vouch it works a lot of the time. What about smaller Businesses though? M$ could think "f**k them - small fish". Medium to Large size organization can certainly get away with this level of cheek :)

Hyper-V/RemoteFX argument is for another day :)



I guess there is one major thing missing here and that is the history behind both solutions you mentioned in this article (WinFLP and Windows XYZ Embedded).

These are two completely different beasts in several levels and as such they target two completely different audiences.

Windows Embedded (NT, XP, 7) came to the market as a solution between Windows CE and the full blown Windows OS at the time (NT as NT Embedded was the first full blown one to be embedded). This allowed partners to leverage whatever plugins/clients they had on the full Win32 platform to the Embedded one, without having to deal with Windows CE.

That is the reason why you find on the embedded systems control mechanisms like the filter driver that prevents writes on WinXP Embedded/WES7 (enabled out of the box). So this solution was really designed/provided to OEMs delivering thin clients to the market.

WinFLP (if you guys remember "Eiger" and "Mönch", that is it) came as a solution for companies willing to reuse existing hardware that became too underpowered to run the latest and greatest Windows OS, mainly as a client for RDP/ICA. It was born as an XP derivative, meant for slow hardware. The target audience here was/is completely different than the Embedded solutions.

This makes a huge difference in terms of what you do to update/maintain/support the platform.

Why you do not see these fancy new thin clients with higher end video cards capable of RemoteFX, Citrix HDX and so on running WinFLP? You need a platform that supports all this and is optimized for that. With several cards now lacking WDDM drivers for example, it is clear the option for OEMs is to use the embedded systems. As WinFLP is meant for older HW it does not require for example Direct3D 9 compatibility what WDDM requires.

The bottom line is very clear: for real thin clients or real embedded systems, Microsoft wants you to use WES7 and the likes. For older stuff they want you to upgrade the machine, not keep using it with WinFLP. Simple as that. It is good for them money wise and good for the whole Wintel ecosystem (as HW vendors will sell HW to replace aging machines). Does it mean it is good for the customer? Hard to say but the main reason why we are paying $300 for full PCs today is because of economics of scale. If we all decide to keep PCs for 10 years may be we all will have no job in 5 years from now. :-)

So MS/HW vendors definitely want you moving their products and not reusing old stuff with WinFLP. And if you are thinking about using WES7 then, well in this case they want you to use real thin clients what again brings us to more sales on the embedded ecosystem (thin client vendors).

My 2 long cents I guess. :-)


Thanks for the article @appdetective. When I was I was entertaining the idea of using WES7 as a potential target device for VDI, or even RDS for that matter, I was severely taken back by the fact that VDA  license was not waivered for SA customers.

Now, I understand where you’re going with WInFLP comparison with WES7, but really they only compare half-and-half. WES7 is half an update of WinFLP, half of XPe. XPe in contrast to WnFLP was targeted for “embedded” devices within the manufacture – XPe based “thin clients” within this context. It never was available for the public, SA or not, but rather by some DEV/OEM licensing scheme (if I’m not mistaken).

Regardless, now that WES7 is indeed a hybrid model and available for general consumption, it ought – no, it SHOULD be part of SA licensing. I ask again, have anyone had any success with negotiating with their MS contact on this subject. Either way, successful or not, I regret the official MS stance on this as now, especially if they want to have any tangible offerings in this rapidly evolving and competitive market. Also, as others already said, neither am I to bring in the topic of RemoteFX at this point – suffice to say is that it’s there lingering, waiting as an integral part of the discussion, later…


I'd like to see the community take the same approach with respect to VDA/VECD licenses.  Tell MS to quit standing in the way of innovation by charging for additional licenses even though they are not providing any new functionality to the stack.

For example if i have SA (aka already paid for the client OS) and now I want to access my OS from a zero client or any non-MS client I have to pay an extra $100 to do it?  WHY WHY WHY?

Why does it


So...  I'm reading "don't lock yourself into any more MS products" and then "Microsoft [WES7] is great - we should all us it".



Oh yeah, Microsoft VDA and SPLA licensing is another "interesting" topic, as it is no-no, I haven't even bothered about the WES7 specifiks in this... Anyone from or close to Microsoft care to comment on the generic topic as such and maybe the SPLA specifiks as an added bonus?


If WES7 falls under SA then that gives more fuel to type-2 hypervisor solutions.

I think WES7 will be the MS answer to type-1 and type-2 client hypervisors. Maybe down the road they will integrate hyper-v in it...


@daniel Bolton I think smaller customers have more ability to go to Google apps as opposed to the big guys where Office is entrenched. That is the small guys leverage with MS pricing. The big guys should think about getting leverage with Exchange, Sharepoint, Sql Server and OCS. Mix that up a little to upset the MS balance.

@Claudio the bottom line is that I want a real thinned down Windows client to connect to hosted desktops with the full Window 7 hardware abstraction layer (HAL) so I can use it on any piece of commodity hardware. The virtual desktop clients all provide better user experience on Windows clients. I'd love to say that MAC and Linux are there, they just are not and that means I have to either wait or make compromises for my users. MS should support more Windows everywhere as a result, I don't they understand the problem and opportunity.

@Kimmo, when the time comes, I will not be paying for WES7 if MS wants to a  renew their SA agreement with me. I want more value not less that what I already have.

@SE I think MS has to protect revenue which is device based, hence VECD etc. What they need to introduce is user based licensing to really solve the problem. I am willing to bet that moving forward there will be more devices than people, and hence this force the trend or customers will pay through their noses.

@Stephan. Yes, but unfortunately we are already locked into the Window OS, and I need to provide a new experience on that. To do that Windows endpoints is my best point. I'd love to say Linux, but it's not there. That does not however we need to lock into everything MS, which is the point I am trying to make. Diversity creates leverage, and OS is becoming more commoditized over time so forcing diversity will force them to add more value and innovate or they are dead just like the Roman Empire.

@Kimmo. I wish there was a Windows 7 SPLA, because that would a real step towards user based licensing.

@Icelus. That a good thing for MS. Shipping bloated Windows every-time is an old idea. Not quiet JEOS, but a step in that direction to accomplish the task at hand. MS 100% will move on client hypervisors once/if the market matures. It's the classic MS way. Do nothing, then buy your way in or copy with an army of people and become good enough over time by eating away the incumbents low end customers and slowly move up. I certainly expect they will try to do something if they think it will help people move to Windows 8 faster way as opposed to the junk that is Med-V today and nobody in their right mind uses.

@all, thanks for the comments. Now go give MS some $h1t and have them explain to the world why this obvious thing is getting no traction. I am sure they are listening, so for once do something to solve the problem to help those of us dealing with Desktop Virtualization and keep value at the very least the same. Pretty please MS.....Oh F it, this is how you keep Linux irrelevant/marginalized on the endpoint for Desktop Virtualization MS. Now that is something you care about you greedy b@st@rds.....



I am not sure where WES7 will go. I believe the future will be a combination of fat clients with type-1 client hypervisors and thin clients.

Also, the future should involve managing OS instances, App instances, and Device instances in one console or multiple MMC snapins that way you can have your User instances there as well.

"Reverse Seamless" is solely a server based VDI feature, and if the computing is on the client you can just stream the app to the local VM and then even use "Secure Application Sharing" to share the corporate app to other non-corporate VMs.

Also, the recent announcements of Intel purchasing McAfee indicates a win for client type-1 hypervisors and it will introduce more service VMs in addition to Citrix Receiver on XC.

Innovation is key, Windows on the endpoint will kill innovation. We need a very small open platform that sits directly onto hardware to drive 3rd party vendors to make innovative service VMs.

VMs are the future. My opinion, MS won't care what's on the endpoint (not until they port hyper-v over) nor can they control it, it's the hardware vendors that have the control. To add value to desktop virtualization MS will have to develop their Windows architecture around the VM delivery model to make it more resiliant to the dynamics of it.

VDA sucks. Either SA or FU.


The future will evolve but today there is a need for simpler Windows endpoints. Reverse Seamless is about handling exceptions without disrupting your work when you get all the benefits of “centrally” hosted desktops but there is that one app or multi media format that can’t be remoted. It’s a key feature that is needed in the solution stack. It’s not a product.

The Intel thing is interesting, but according to a Forrester blog I read today, it’s there to support their mobility strategy. I agree this will also help things like XC, but I agree as the blog also stated that useless features like vPro for security will not mean customer uptake. What will Symantec do is a killer question over the next few months.

I agree with innovation on service VMs but the horsepower to do this is heavy and not suited for all device types. For ultra thin devices there will be a compromise for user experience, so need the Windows option. I don’t see the remote protocol vendors bringing Linux client features to market before Windows. Just not going to happen.

I disagree with your assertion that MS does not care what’s on the endpoint. They care about Windows on the endpoints, and are stupidly allowing Linux clients to become prevalent when actually they are not better, even for a security which  is a huge myth. Lot’s of idiots that just don’t get Linux/thin clients are going to screw them eventually and are married to sunk cost ad decisions.