David M Smith

After I just posted my previous entry titled Silverlight Steals the Show! I realized i had also buried the XRM non-news. So I was guilty of burying it almost as much as Microsoft is. So below is the last paragraph from the post, highlighting the non-news. Don’t get too non-excited about the non-news…

XRM is the underlying platform technology underneath Dynamics CRM Online.  It is analogous to force.com’s origin underneath the Salesforce.com app.  Other than a brief mention of it as something that can work with Sharepoint and one session at the very bitter end of the conference, XRM was missing in action.  It is one of the more silent  but promising pieces of  Microsoft’s cloud technology. XRM is Microsoft’s only fully multitenant application platform as a service technology that runs on and off premises).  I waited until the end of the day 2 keynote to write this to make sure but for now it looks like it will continue to remain as such for a while longer.

At day 2 of Microsoft PDC09 in Los Angeles today. Time for a quick set of highlights. More to come in a lot of research in the coming days.

The overachieving announcement: Silverlight 4

The big news today is Silverlight, Specifically the beta of version 4. Siverlight continues its march towards providing moreand more of the functionality of .net, full WPF and Windows.

Version 4 includes many features for media use such as  webcam support, microphone support, offline DRM, Live streaming (including a way to stream to iphones)

It also has many features targetted at enterprise LOB uses such as printing, rich text (including rght to left display for those kind of languages), clipboard, right click,mouse wheel support, drag and drop,  hosting HTML (and other plugins like Flash). And a really big deal is the ability to compile once and deploy on Silverlight 4 and .Net 4.Tthis will open the door for many uses within enterprises.

More out of browser support. More browser support (Google Chroime). and all still in a 5MB download.  Very impressive.

So much developer emphasis with Silverlight really is in contrast to so little for Windows 7. Don’t get me wrong. Windows 7 is great and will be
very successful. But the difference between Silverlight and Windows is continuing to shrink.  Silverlight 4 also even supports multitouch as well as lots more direct hardware access.

The underachieving (non) announcement: XRM

XRM is the underlying platform technology underneath Dynamics CRM Online.  It is analogous to force.com’s origin underneath the Salesforce.com app.  Other than a brief mention of it as something that can work with Sharepoint and one session at the very bitter end of the conference, XRM was missing in action.  It is one of the more silent  but promising pieces of  Microsoft’s cloud technology. XRM is Microsoft’s only fully multitenant application platform as a service technology that runs on and off premises).  I waited until the end of the day 2 keynote to write this to make sure but for now it looks like it will continue to remain as such for a while longer.

At Microsoft PDC09 in Los Angeles today. Lots of big news and interesting things to write about but the big word is fabric.The introduction of the AppFabric brand by Microsoft for its appserver is the tip of the iceberg. First of all, the idea of a fabric for running distributed applications is not new, but its applicability to better explaining Microsoft’s vision of running apps on premises and off premises using a consistent framework is new and is a good step forward.

So, as Microsoft is introducing a name for its appserver layer around 10 years after everyone else, it’s fitting that I had selected (completely at random) a shirt I’ve had for around the same amount of time.  Its an old favorite from The Territory Ahead and it certainlyis unique. but i had no idea that talking about all the fabrics while wearing it would make me such an easy target…

for those of you keeping score at home, there are  now  4 distinct meanings and uses of the term "fabric" just by Microsoft alone:

1. Azure fabric controller (a management component, gives the elasticity to the VMs in Azure)

2. Appfabric – this is the new appserver brand.

3. Systems center also uses the term fabric as a management construct
and,

4. Live mesh also makes heavy use of a fabric.

Appfabric is an improvement over the Azure Services Platform and .net services (and the hopelessly confusing terms Windows Azure and the Windows Azure platform – yes they are different. Who knew?), but not without introducing new confusion.

So just when you thought nothing could be more confusing than cloud, here comes fabric.  Hopefully it wont be as bad. But just in case,
I think I will trademark FOA. After all, Fabric Oriented architecture can’t be far behind

Consumerization has long been a good thing and a not so good thing for Microsoft. As one of very few companies with a significant presence in both consumer and enterprise markets, it is positioned to take advantage of that crossover.  However, as it has gained the trust of IT, it is reluctant to be seen with “the smoking gun” of empowering individuals just a bit too much.

Today’s official consumer launch launch of Windows 7 represents another example of this. The company has in the past been a recipient of the benefits of consumerization.  Windows 95, for example, was very much driven by this phenomenon and is a classic example of it.

It happened, to a lesser degree, with Windows XP.  Microsoft would love for it to happen with Windows 7. It reinforced the potential benefit when it  didn’t happen  with Vista.  Consumers didn’t get excited about
it and didn’t drag it into businesses. In fact, their lack of enthusiasm likely slowed down business interest. While we won’t likely see long retail lines like we did with Windows 95, now we get launch parties and mass media advertising in order to stimulate consumer demand.  Having a very good product this time will help as well.

However, the other big thing going on these days in the consumer PC space is the growing popularity of netbooks. These pint sized (and pint-priced) machines have started to become used in businesses as well.  But the real impact that will be felt is on pricing.  Microsoft has had to offer very low priced versions of Windows (XP until now) for netbooks in
order to compete with free Linux offerings.  The threat is that there will be a general trend to lower prices of consumer versions of Windows to the point where the difference between what businesses pay and what individuals pay will be quite large.

In that case, how long before we start to see businesses consider the use of consumer versions of Windows?  Yes, of course, they wouldn’t get
some capabilities (such as Active Directory), but how much will that be worth.  It will be interesting to see…

Controversial issues tend to attract the lunatic fringe to the poles of those issues. We see it all the time in politics. We also see it in cloud computing.    In the cloud case, the poles are public cloud and private cloud.  And the extremes bring out the lunatic fringes on both sides.

First the public side. Purists (often called lunatics) will go as far as to say that private clouds don’t exist.  But the real danger is the loss of potential benefits when people overreact to limits in pure public cloud environments.  In such cases, often there will be a showstopper, probably around security  or compliance.  The lunatic overreaction is to say that all cloud computing can’t be used. And the potential benefits of less pure public and hybrid and virtual environments lost.

On the private side, there is incredible abuse of the term.  No shortage of vendors are claiming whatever they were selling last year is now private cloud.  There are also IT people wanting to claim they have them too. whatever they are. The danger is lunatic overreaction to stretching the term.  So the potential benefits that could be derived from using cloud concepts internally can be lost due to lunatic skepticism.

I wonder when we’ll see the private cloud lunatics lobby against any public option

Yesterday I blogged about this and I had a couple of further thoughts.

First, my colleague Nick Gall brought to my attention that I may have implied accidentally about the capabilities of GCF (Google Chrome Frame).

GCF does not replace the IE renderer as the default renderer, it merely adds GCF as an optional renderer that is only invoked if explicitly requested by the web page creator. So GCF is not going to solve ANY problems until a substantial number of websites include the GCF tag.

Then I read some more postings about it including which say that Google is being very clever.  Actually, if they were really clever, what they would do is the opposite: an IE6 rendering engine (by a type of wrapping of the existing IE6 DLLs) for Chrome for Windows 7.  I don’t know how hard it would be and if it would cause licensing or other issues. But it would be clever if they could do it. And valuable. and would solve the “IE problem” for many.

Well, it really depends on what you think the “IE problem” is. If you haven’t heard about Google’s Chrome Frame, it’s a plugin for IE (6, 7, and that basically replaces the IE rendering engine with Google’s Chrome.  It has gotten a lot of attention as an “IE killer”. It allows IE users (especially those on IE6) the ability to run modern web apps much better.  I’ve started getting questions about whether it solves ‘the IE problem’. 

Many enterprises have IE6 apps that don’t work properly on other browsers. IE6 works only on Windows XP, effectively keeping these users hostage to XP.  Not only can they not move to other browsers or OSes, they can’t move to Windows 7 since IE6 does not run on Windows 7. Thus far Microsoft has not provided a way to run IE6 (apps) on Windows 7 except for a very heavyweight “XP mode” virtual machine.

So, you may think of your IE6 problem as one of two things:

1. If it’s “I have to have IE6 and I’m on XP but I want a modern rendering engine”, then yes, Google Chrome Frame is a potential solution.

2. If it’s “I want to go to windows 7 and I need to run ie6 apps”, then no, I’m afraid it’s not much help.

Most have scenario 2 as the issue.  And it’s this that will haunt Microsoft by delaying Windows 7 upgrades in enteprises for a long time unless they address it.  And it will be a source of pain to enterprises as well.

Even in scenario 1, if desktops are truly locked down, then users likely can’t download and install a plugin any more than they could just install the Chrome browser or another browser. So it doesn’t really solve too much.

So, while Google is at least doing something here, it isn’t going to help with “the IE6 problem” very much.

Not that there isn’t already too much in this world, but hey its a cool headline…

Recently there was another short gmail outage. and it sparked the usual outrage over how unreliable cloud computing is and how OMG nobody should use it etc., etc. Of course people wasted more time complaining about the outage that actually experiencing it, but I digress.

There was also recently an internal email outage that affected us here. What’s notable is that there was no outrage. No claims that ‘on premises email is unreliable’ and how everyone should move to the cloud. I’m not saying there should have been outrage, just noting that there wasn’t any that I saw. Every system has outages. Ours here is generally very reliable and I don’t complain when there is a glitch. But then I don’t complain about a minor outage with a cloud service either. Do you? When your internal email is down for an hour, is there outrage?

I’m not our email or high availability expert. I’m “just a country analyst” and dumb user (of course I’ve been called worse). But some digging shows that most enterprises expect an SLA of about 99.5% (maybe even less), which means downtime of about 3.5 hours per month. Can you imagine the blog traffic if gmail were down for 3.5 hours per month? Twitter would croak (oh wait, it already is croaking..).

Typical uptime for cloud email services is in the 99.9% range. Last time i checked, that looked a lot more reliable. Is there a double standard here? Is internal IT cut a lot more slack? Are expectations unrealistic of cloud providers?

According to one of my favorite philosophers, Yogi Berra, Its hard to predict, especially the future”. He’s right but it doesn’t stop many people from trying. In fact predicting the future is essential to many aspects of our lives – in business, and beyond. Many professionals have the need to accurately predict outcomes of the future to be successful in their jobs. And many have occupations where predicting the future actually is their job, one way or another. As an analyst at Gartner, I am of course a good example of this.

Some of this is common sense. Some is controversial. Some goes completely against what most think and against what people are taught even at organizations who train people to do predictive type jobs. But it works for me.

Here are my ten guiding principles for accurate prediction:

1. Care about being right. This sounds obvious but circumstances and other requirements often get in the way. Professionals whose job involves making predictions face pressures to have an opinion, no matter what, and to generate visibility. This can lead to quickly formed opinions and overstating and over hyping things. While these things may in fact need to be part of a strategy, they do not have to be the primary goal. Tempering such behavior by placing the goal of being right at a higher priority is one of the real keys to accurate prediction. You can’t be afraid to be wrong, but you can’t place being right at lower priority and expect to be good at predicting.

2. Be an “innumerate”. Be extremely skeptical of any numbers. Many believe that numbers don’t lie. They don’t of course, but people do. And they state the numbers that they want to state to make their case. And they get things confused. Numbers are more useful in looking back at history than in predicting (looking back at history is helpful and numbers can help). Be especially wary of survey data. Often the questions are poorly formed and the respondents not necessarily knowledgeable. There is no substitute for talking directly to people to make sure that you understand context and that they understand the question. And follow-up is possible.

3. Ask yourself “Why are they telling me this?” Understand the motivations of sources of information. Everyone you meet has some type of agenda. Sometimes it is truly to educate you, usually not. It is critical to understand what the source of information wants you to think to put the information into context.

4. Ask yourself “What would I do”? Put yourself in the shoes of the CEO or key decision maker of the entity if possible. This is a key tool to predicting how companies and organizations will behave. If the prediction is about that company, this is the major key. If it is more general, putting yourself in the shoes of multiples and playing out scenarios is helpful.

5. Recognize that most of the time, you will know less than your sources. The world is full of specialists. Depending on circumstance, you may know as much as your sources but there is almost always someone who is more of an expert than you. So you need to develop strategies for assessing the credibility and honesty of a source. A useful tactic is to lead a discussion towards an area in which you do know a lot and test the source’s honesty and credibility. This can help determine what weight to give the source

6. Don’t jump to conclusions. Whenever possible take your time. When pushed for an opinion, it is best to say “if I had to have an opinion I would lean towards x”, but not highlight these types of things as “predictions”.

7. Find “bubbles”, conventional thinking and poke at assumptions. Try to understand why most people have a certain belief and figure out what assumptions they have. Look for misunderstandings, confusion, motivations and social trends.

8. Get information you’re not supposed to have. Basic networking is essential to knowing your subject and to getting information you’re not supposed to have (Obviously those subject to “insider trading” types of issues need to tread carefully here). Listen for slip ups. Put the pieces together. Fill in the holes. Speculate.

9. “You’re only paranoid if you’re wrong”. Explore conspiracy theories. While they usually won’t be the prediction, the exercise of examining possible conspiracy theories often is fruitful. Remember At the very least there is bound to be some aspect of the theory that has some truth to it and may point the way towards a good prediction. However, it is far more likely that stupidity or laziness, rather than conspiracy, is the cause.

10. Constantly test, validate and refine. Every chance you get to talk to a person whose opinion you respect, test new theories. Every chance you talk to a source of information, test your theories and gauge their reactions. Be open to tweaks.

Some other principles that I always test against:

- Simple is better than complex
- People (especially politically oriented people) have to be able to declare victory
- Understand when there is “skin in the game” and when logical reasoning will not explain everything
- Fear and Greed are the ultimate forces driving everything

Note, I wrote this several years ago as part of an internal exercise here at Gartner. At the time, there wasn’t an appropriate mechanism to share this. Now, with our blogs, the time is right.
Enjoy and don’t be shy to comment.

I’ve come to view “software upgrade” as a term that meant, yes, “a software upgrade, but you will really need a hardware upgrade if you hope to actually use it even for what you used to use it for”. Especially with new OS versions. It has been a given that new versions of OSes would drive new hardware sales.

I’m testing today two new OSes – Windows 7 and Apple’s Snow Leopard. I’ve been using prerelease versions of the Microsoft offering for some time and am not surprised to see that it still has the property that it runs much better on the same machines that its predecessor (Vista) did. I had heard that Snow Leopard would have a similar property. But I wanted to see first (it does).

In addition, I recently updated my first generation iPod Touch to the latest OS (3.0).

Every one of these upgrades have given more and better functionality with better performance on the exact same hardware, much of it 2 years old or more.

OK, being better and faster than Vista probably wasn’t that hard to do. And if you talk to Apple a lot of the enhancements in Snow Leopard are designed to clean up the OS to prepare for future enhancements.

The question is: is this all a coincidence and the next upgrades will return to “normal”. Or is there a “New Normal” for upgrades as well? After all, beta doesn’t mean what it used to…