The Advancement of Windows: Ales Holecek - Windows Shell (Windows Explorer, Desktop Search, UAC, Aer
- Posted: Oct 12, 2006 at 4:42 PM
- 34,092 Views
- 35 Comments
Download
How do I download the videos?
- To download, right click the file type you would like and pick “Save target as…” or “Save link as…”
Why should I download videos from Channel9?
- It's an easy way to save the videos you like locally.
- You can save the videos in order to watch them offline.
- If all you want is to hear the audio, you can download the MP3!
Which version should I choose?
- If you want to view the video on your PC, Xbox or Media Center, download the High Quality WMV file (this is the highest quality version we have available).
- If you'd like a lower bitrate version, to reduce the download time or cost, then choose the Medium Quality WMV file.
- If you have a Zune, WP7, iPhone, iPad, or iPod device, choose the low or medium MP4 file.
- If you just want to hear the audio of the video, choose the MP3 file.
Right click “Save as…”
- Mid Quality WMV (Lo-band, Mobile)
- WMV (WMV Video)
We recently had a conversation with Ales Holecek, Director of Development for the Windows Shell team (Raymond Chen, the famous Win32 blogger, works for him...). Ales spends most of the time answering a
single, broad question: What is the Windows Vista shell?
Of course, you can't talk about Windows Vista shell without talking about UAC (User Account Control), Windows Explorer, Aero, and Desktop Search, right?
Of course, you can't talk about Windows Vista shell without talking about UAC (User Account Control), Windows Explorer, Aero, and Desktop Search, right?
Comments Closed
Comments have been closed since this content was published more than 30 days ago, but if you'd like to continue the conversation,
please create a new thread in our Forums,
or
Contact Us and let us know.
Follow the Discussion
Interesting Conversation. I'd love to hear more from Tjreed Hoek about his thoughs on Areo in the shell, and his thoughts behind the design.
Great video tho.
I gather from this video, that MS went to the direction of writing new OS (VISTA), because the current model in XP and 2003, is unupdatable, because the code base is old and is not organized in a way to allow for future improvements. SO when he said that Vista is "a mile stone for us", he meant that Vista will be a new platform OS that will enable them to move to the future without the strings that were in the previous implementations of things in XP and 2003.
So when Windows Vienna comes, its going to be breath-taking. That is why I am not so much excited about vista, but about what is going to come after vista as a progress ontop of the new foundations that were implemented into Vista. That is why I compare this stage as going from DOS to Win95.
Intersting analysis. The hypothesis(first paragraph) is a compelling one. However, your conclusion, which appears to miss an understandably implicit point: data created as a direct result of your interactions with the system, how you use it and how it is capable of making your intentions achievable in a predictable and reliable way, makes proving or disproving the hypothesis empirically impossible(we think we may be on to something theoretically, but theory is purely abstract).
It's nice to see that people are beginning to realize that Vista is a new OS, one that is also intelligent: composed of subsystems that are capable of not only learning, but also prone to interacting with newly gained knowledge in a cybernetic way.
Vista sows the seeds for a future Windows that becomes not only intelligent, reliable, safe, performant, usable, but also predictable, composable, homeostatic. Evolutionary acceleration of Windows will remain static without high use of the system by real people.
C
hate to be pedantic here Charles (what am I talking about, I'm a geek - I live for pedantry!) but it's impossible to prove anything empirically.
You can verify your hypothesis, but you cannot prove it, as in order to prove a hypothesis empirically you'd have to observe all possible permutations and outcomes, (and as you're being empirical you wouldn't be allowed to inform your observations from your theory, as you haven't proved your theory yet! So you'd have to literally observe all possible permutations ever ever ever.)
In science, a Theory (with a capital T) is another way of expressing an Hypothesis that has been "proven" due to direct observations, or measurements, that show the conjecture to be factual.
I postulated that above, because in my experiance, back in 95 and 98 and then in XP, I found it hard to really have freedom in programming new implementations of things. You feel as if you are trying to write a page with 4 people holding your hand or so. Its kind of hard to think clearly, and hard to implement your ideas. Now once .NET came, some of these hands were lifted and you can move freely, so to speak, and have your own implementation under the new standard. The obvious next step was to get rid of the old OS, and re-write the Windows OS in a new and more innovative way, where I dont have so much hands tying me down (All the useless APIs in the old OS). It brought a new way to think clearly, and to implement cleanly any idea you might have. In that sense computer programming becomes as powerfull as magic, and the sky is the limit as they say.
I only wished that MS made this move back in the 90s , instead of making XP, we could have had Vista, and now Vienna. Having said that, maybe things were not as ready as one would like, but its all for the best I suppose.
If memory serves, Laws also have to be consistent under all conditions. i.e. if some set of values (the mass of two bodies, and their seperation for example) then some other effect (the gravitational attraction between them) is always the same.
But you still can't (logically) prove anything empirically. Seriously, in fact it's nigh-on impossible to (logically) prove anything at all, in terms of real world behaviour. (The most "proven" set of laws in Physics is called Quantum-Electro Dynamics, and even then they're only 99.9999999% (can't remember the exact numbers) sure it's true.)
If you want to dispute that, then go and read a really good book called "What is this thing called science?" I think the author is called Alan Chambers, but I could be wrong. It'll show you that in actual fact Theorys are established not by proof, but by everyone failing to disprove them.
Well Then what are Laws? I learned that Hypothesis is a postulated explanation of natural phenomena. If the hypothesis explains the phenomenon under varying conditions, it becomes a working theory. If this theory is yet demonstrated to , over time, explain the fundamentals of the phenomenon, and all outcomes are accounted for, then it becomes a Law of nature. Like the Gravetational laws, laws of motion, Newtons laws etc..
So , generally, in science, we never use the word "Proven", we say the hypothesis was not demonstrated to fail, yet or something along those words.
Agreed. Though in the case of Quantum-Electro Dynamics, or in general Quantum mechanics, the Heisenberg uncertainty principle, makes it impossbile for us to generate reliable measurements, much less formulate theories. Its like walking on unstable grounds and trying to measure the speed of falling objects. So ya I agree completely that in science, we never say this is proven, we say that this failed to be disproven, or no reliable demonstration succeeded to render a theory or even a hypothesis false, yet.
Well it appears there was no thought that went in to it. Havn't microsoft paid any attention to Apple? Why, oh, why do they have these discusting '90s style drop-shadows? They are WAY to strong. Please give us sliders where we can control the size and transparency! It couldn't be so hard. That or allow us to turn them off.
And there are wide inconsitencies accross the system. Why not make everything in the same style.
To the designers: LESS IS MORE!!!
Do your animations and other cool stuff, but don't overdo every single effect just so people will notice them. Subtle effects! Again...look some at Apple.
Also, missing some wallpapers at 1920x1200.
"Evilware" that onstantly pops up a UAC security dialog all the time. [6]
That's not entirely true either. If it were, I could invent ideas all day and have you busy disproving them, or trying to. I could claim that with my mind I can make desks float around the room, and if you spent the next 30 years trying to disprove that, that wouldn't mean that my theory has any credibility.
It comes down to probability too. For instance, in math we have infinite theorems such as "All even numbers will be divisible by 2". Now, I cannot prove that...why? Because I have not tried every possible even number to see if it is indeed divisible by two. But, the probability is so astronomically high that this is considered a fact.
A theory doesn't get accepted just on the fact that it's been around for a while - it's got to demonstrate its predictive power to the masses.
Ok, for a start - in Maths theorys can be proven. Maths isn't a science per se. it's a branch of logic. As such it is constructed of a limited set of concrete rules, and things can be proven absolutely true one way or another. (But never EVER are they proven empirically. Fermat's last theorem had empirical-ish proofs which went up to hugely large numbers, but it wasn't considered proven until a general proof was established.)
I defend my statement, Theorys are establish by everyone failing to disprove them. Your "I could make desks move with my mind" theory would last a very short time, scant seconds after someone says "prove it".
Now your last sentence is much much more relevant. In science, the measure of how good a theory is comes down to two things: "Does it explain what we already know better than the current theory?" and "Does it predict something new?"
The second part is crucial, as that's where theories derive their strength. If a new theory is simply made to fit the facts, and doesn't provide any new predictions it's considered a rubbish theory. The best theories create weird new predicitions, (like Quantum Mechanics did - really crazy predictions about certain things.) that would be really easy to prove wrong. It's simply that no-one has which means they've been accepted.
As a quick example, Einstein's general relativity predicited that the position of the stars would appear to change when the sun was near them. Now you had to have a solar eclipse to see the stars, because he said they'd be the ones really near the sun. But still, a pretty easy theory to shoot holes in - "the stars move! you're mad!" And yet when they measured it, he was right - General Relativity had its first proof and the rest is History (and Gravity, and Space as well! Boom boom!)
YOW! talk about getting off topic !!
And just think! If I'd made a comment about Linux, instead of one about the logical impossiblity of empirically proving something - where would we be now?
Sorry, for derailing the thread, you can have it back now.
I am not following you here. AFAICT, Vista has a ton of new improvements and brand new features, but it is not a total rewrite of Server 2003R2 from scratch (check me on this Charles). You still have win32 and existing core and code bases. It would be interesting to get a Swag on percent of new code and percent of code that has changed. Maybe an easier number is what percent of 2003 has *not changed?
I am also going to enjoy the new lock primitives in Vista, such a Reader/Writer locks, Condition variables, Lazy-init, Improved Thread Pool, new InterlockedXXX functions, and application deadlock detection apis. When you start adding all the stuff up, there is a ton of hard-core goodness going into this product (this is the very hard-work stuff you don't really see).
Agreed. Even in Mathematical theories, the Pythagorean Theory, is a theory because no one brought forth a demonstration that will show its false, like having a right-angeled triangle where the sum of the squares of 2 sides does not equal to the square of the hypotenuse. If there is such an example then that thoery is demonstrated to fail, demonstrated to be false.
So Massif is correct with respect to the scientific terminology as well as logic of it as it pertains to the scientific process.
You just proved my point. A theory isn't established simply because somebody cannot disprove it - and you demonstrated that when you said "Prove it". A theory is established not for fighting off critics, but for affirming, or explaining something. So you cannot say, "Theorys are established by everyone failing to dispove them", and then counter my claim that I can float desks by saing, "Prove it". Because by your own words, I dont have to prove it, you how to disprove it. And if you can't, then the theory that I can float desks is possible. But that is nonsense, of course
Its totally new. written from scratch, or in some cases some win32 were replicated in new code. Old code was archived for the most part. So its totally new OS. Just like Going from DOS to 95 with new implementations. That is my information that I have. Its like rewriting win32 (preserve the APIs and features, but different implementation in code to fit in with the rest of the stuff).
Backward compatibility is one hindering factor. If we can get rid of backward compatibility all together, things would be more bright. I am all for having people who have old apps to stay on XP, and Vista would require rewrite of those apps to run. Then we go from there as a fresh start. But I guess its the promise of MS to its customers that make this backward compatibility appear in every new release of the OS.
I think what he meant was to show that your theory 1) explains the phenomenon. 2) Is able to make solid predictions with respect to different situations and outcomes. 3) That no situation is brought forth, that demonstrates that the theory fails to explain the observable set of circumstances. The terminology is important, when you demonstrate that the theory fits 1 and 2 it does not mean you proved it. It means you postulated an educated guess, that explains the main aspects of a phenomenon, and also if its a good theory, would predict correctly new outcomes under different situations, and that no situation existed yet, that would show the postulated educated guess is false (a situation where the theory has to be updated , discarded, or refined so as to explain the new situation under which it failed as written). So Massif is correct.
To get more on this, read the about scientific method (Hypothesis,Theories,Laws).
There's a lot to talk about here. Did any of you scientific philosophers actually watch the video? Do you not find the changes interesting enough to discuss?
Thanks,
C
Where are you getting your info? Don't you remember "the reset" in late 2003?
http://en.wikipedia.org/wiki/Development_of_Windows_Vista
Neither the first "xp vista" nor the "2003 vista" started from scratch however.
Do you not understand anything about the NT architecture of Windows and basis of the underlying core OS design and philosophy?
In terms of the core OS, sure NT has been updated, but it also is very much as it was, and this is a testament to the strong and extensible design of the NT design architecture that is now almost 15 years old.
I think you are seeing Windows too much as the Win32/Win64 subsystem, and do not take into account or seem to realize that these are JUST subsystems running on the NT kernel architecture.
That is why there is a Win32 kernel for that subsystem but also an NT kernel that is beneath ALL subsystems. The nature of the NT architecture and the concepts of subsystems is fairly unique and why you can run a full UNIX subsystem alongside Win32 that is a full OS implementation and not an emulation environment.
If Windows was 'not' updatable, then a new kernel technology and architecture would of had to created, instead because of the extensibility of the NT model, all these great new ideas and technologies were easily added to Windows without a core or architectural change even being necessary.
There will come the day that the NT model does reach its limits of extensibility, but that is several years in the future, and will more of the boon of a new kernel technology that will probably have many of the NT aspects even then.
I agree that Vista is a large leap in terms of R&D implementation, but it is very much STILL built on the same OS model and concepts that NT first introduced. Go read up on NT and what is underneath Vista's subsystems, you will be surprised how strong the underlying technologies are, have been, and how extensible these concepts developed in the early 90s are even today.
Vista has a tremendous amount of technology that is not only new but also a result of the implementation of features that have existed in NT since it was first created.
XP also had stronger security than most people expect, but for application compatibility from the 9x age where all software expected to have administrative rights, MS took the route of compatibility instead of enforcing the NT security model at the level it could have been enforced. Vista not only enforces the NT security that has always been there, but adds to it.
Vista is about new technologies that are stacked onto the Windows architecture, in addition to changes in the architecture.
As some people have observed, a lot of the features are behind the scenes, making the sale of Vista to novices a bit harder. Just look at XP, there are tremendous advantages and new technologies it offered over Windows2000, yet you will find people that think it is just Windows2000 with Themes for the GUI.
MS will have trouble, even as it had with XP, in explaining the extent of the new features and 'intelligence' that works in Vista, but through the showcasing of new software concepts in the next year this will become easier, as Vista will do things that no other OS can even begin to conceptually implement.
Vista will also shine when users and non-tech journalists realize that existing applications not only run better, but significantly faster. (Take a Vector illustration application like AI or CorelDraw, they are extremely fast on Vista do to the new WDDM acceleration and composer.)
However to try to sell Vista a complete new OS does not do justice to MS nor is it factual. Your assumptions about the ‘existing’ Windows code base being poorly written or non-extendible are also wrong and misleading.
Vista is full of new features, but the sound concepts and brilliant architectural concepts of the NT platform are very much still alive in Vista and at the core of making Vista what it is.
Anyway, with regards to UAC, is there a reason why (legacy) applications which require “administrative” privileges don’t get “sand-boxed”, essentially being run in there own “app-domain”. Something akin to what Windows 3.1 first did with DOS applications, but extended to include all of the stuff that UAC is overseeing – a virtual Win32 subsystem for each offender:
Obviously it would not be easy to implement, considering things such as state (remembering, merging , etc). But I imagine that it would provide a better end-user experience, given the aims of the initiative. Yes/No ?
PS: I like the Freudian-slip at 38:12 "With respect to shame, ah, shape of our name-space"...
What about some demos guys? And what happened to those interesting questions. Thank goodness, I knew how to fast forward on WMP.
Nobody forces you to watch this. If you are not interested in the topics discussed in this video (pretty clear from the title...), we have plenty of other media content for you. You want a demo of the Vista shell? Why don't you install RC2 and play around with it? Channel 9 interviews are conversations with the people who make our technologies. You're not going to see demos in every one of these.
C
One thing I do want to chime in on though is the confusing "inheritance" model in the desktop and start menu. The problem I'm having is that icons can come from either the the user's own folder or from Public (on the desktop) or Default shared folders (in the start menu). Just by looking at it, it isn't at all clear what's coming from where and it's really rather confusing. It can also make it hard to do relatively simple things. For instance, I like to remove a lot of the marketing verbage from Start Menu entries to keep it clean (e.g., "Windows Media Play" -> "Media Player" and so forth) and I'm finding I have to do a UAC elevation each time.
I'd like to see this move to a more prototype-based model in future versions--e.g., rather than merging a user's icons with the Default icons, give the users local copies of the icons that they can modify or remove as they see fit. Default could be seen as a sort of "broadcast" folder for admins and new applications that install for all users. Of course, there could be GP policies to restrict user control when it really isn't desirable (e.g., public computers). Anyways, just a suggestion...
Also a few questions: Since "Documents and Settings" is now "Users" could you rename "Program Files" to "Programs" or maybe "Applications"? And why, in "Public" are the folders named "Public Documents", "Public Music", etc. instead of the more obvious "Documents", "Music", etc.?
Thanks.
The "Story of Vista," as I understand it, is that the vast project went out of control in 2003 because unintended and unanticipated component interactions were occurring and the OS could not be stabilized given time and money constraints. So, it was decided to ditch, for the time being, one of the major new features, ditch some quantity of new code, and restart development from a stable platform like Windows 2003 Server and/or Windows XP.
An earlier Channel 9 video features a group of people who are working to better understand how all the parts of Windows interact. It is sad but true that complex systems are hard to understand in all their complexity. The market-driven need for backward compatibility adds to complexity.
Sorry to carry on this tangent. It might be called the Pythagorean Theory (or Theorem to be correct) however that is not because it hasn't been proved. It's just a legacy term that's stuck. There are various proofs of this theory which completely validated it over a 100 years ago.
http://en.wikipedia.org/wiki/Pythagorean_theorem
I would surely like to have more customization ability of the UI in vista. Why not release a tool for modifying the bitmaps which form the basis of the UI. Atleast don't encrypt and hide them away in some signed dll. Just put them as resources so that one can edit them with a resource editor. Then enthusiasts can atleast be saved from the texas style drop-shadows which are way to big.
If the dropshadows aren't bitmaps but created on the fly then give us strength, size and bluriness sliders. PLEASE!!! Pretty please with sugar on top. Maybe as a powertoy or as registryvalues or something. Please! Save us who don't want dropshadows from the photoshop play-land.
As someone from microsoft stated over there at ShellRevealed, the drop shadows couldn't be changed because of compatibility issues and time constraints but they will change it in a future release.
Remove this comment
Remove this thread
close