Slashdot Mirror
Microsoft Officially Shows Longhorn, WinFX
Theaetetus writes "Microsoft today unveiled its most detailed look yet at its new OS, Longhorn, due in 2006, during Bill Gates' keynote speech at the company's Professional Developers Conference in Los Angeles. An article at Internet Week describes some of the goals: avoiding viruses, worms, and 'building apps that are as smart as Outlook.'" The company "also unveiled 'WinFX,' which it described as a new application programming model for Windows that is the evolution of its .NET programming framework."
"building apps that are as smart as Outlook."
I was hoping they'd shoot higher than that.
Ok, so they've taken over the bottom of the screen with their explorer bar, now they're taking over the righthand side to show off stock reports? A few more years of this, there won't be any room left on the screen for apps.
Karma Whoring for Fun and Profit.
These announcements are nothing more than vague future directions...
You are in a maze of twisted little posts, all alike.
Yes, the FX comes from effects, I can buy that on a video card (going for video effects) but how does that tie in to an application framework?
"We need your feedback. We need your involvement to get this right."
Go open-source !!
Hey, that scum in your septic tank does a pretty important job... don't start comparing it to outlook to it, that's just mean....
The goals of this OS seems pretty much the same as the last one. The productivity gains of having a "sidebar" are probably the same as the MSN website sidebar, which is kinda like having a billboard blinking outside your bedroom window all night : a distraction.
An XM-based FS is going to be a meta-data nightmare, with more churning than one thought possible. The pagefile size will need to be quite large to cache all that crap. But they'll use the extra-speedy Intels to compress is on the fly anyway.
Most of *any* speech recognition is going to be from research done on [cough] *nix machines of the past decade.
Revamping the graphics system is just what the DirectX doctor ordered: new APIs! Everything can be antialiased, from busy dancing icons to cursors to controls. yawn.
By keeping everyone busy adopting the new platform, form ignores function and we get the same stuff in a new box. I hope they keep pushing it out. Then again, we're talking about people who confuse an OS with their desktop images.
mug
I would wager some money on the fact that this new WinFX is basically .NET with new APIs and some kind of code signing technology with enforced DRM to finally kill Project Mono. It was only a matter of time before they pulled this kind of thing.
.NET apis have changed, and the .NET runtime no longer runs unsigned code, 4 years of work on Mono will be down the shithole.
After all, you didn't honestly think that they'd let that continue for much longer, did you? This way, when Longhorn debuts in 2006, and all the
"A demonstration of WinFS featured a method to "stack" documents by author in a window, with the heights of the stacks corresponding to the number of documents, as well as file views that showed snapshots of documents, rather than just file names."
... In addition to those snapshots posted of Longhorn over the weekend, isn't it a bit odd that Longhorn is essentially using the brushed metal look from OSX 10.3? The only difference being that MS made the grey a bit darker. Kudo's to the MS UI team.
And ten years before this, Apple patented Piles:
"Apple holds a patent on this one. Developed by Gitta Salomon and her team close to a decade ago, a pile is a loose grouping of documents. Its visual representation is an overlay of all the documents within the pile, one on top of the other, rotated to varying degrees. In other words, a pile on the desktop looked just like a pile on your real desktop.
To view the documents within the pile, you clicked on the top of the pile and drew the mouse up the screen. As you did so, one document after another would appear as a thumbnail next to the pile. When you found the one you were looking for, you would release the mouse and the current document would open."
The magic 8-ball says: "Outlook not so good"
Bart: Wow, it does work!
...just my 2 gil.
I am not one of those people who go around professing the evilness of Microsoft. I did, however, come across this logo on news.com.com that does look pretty evil. I doubt that it is official or anything
Evil Logo
Not everything is analogous to cars. Car analogies rarely work.
From the article,
.Net code, which is supposed to help prevent developer errors that can lead to unsecure applications, according to Microsoft."
"Everything that gets written for Windows will be
Everything I read in the article from this to the talk about the file system and how it allows "searching for an array of files...strewn across ever-larger hard drives" and creating all these "smart" programs that "automatically sniff out network connections," really makes me wonder how secure this OS is going to be. Am I the only one who reads this stuff and thinks that a security vulnerability at any point has the potential of corrupting an entire system or even LAN?
Having easy lines of communications between the OS, apps, files and networked resources is great, but who's doing the gatekeeping between all these resources to keep them secure? And how is it being done? Or is it just assumed that once something is "trusted" its trusted to do anything it wants?
Or am I just paranoid?
The WinFX announcement confirms something that I had suspected for quite a while, and that is that .NET was meant to be a replacement for the Win32 API. Win32 is the "familiar" application framework for Windows, but as many have noted (and most Win32 developers know), it is a complicated, cumbersome beast. Give me a choice between Win32 and raw Xlib and I'd take Xlib, thank you very much (but Win32 is a full blown C API with windowing functions just one of many facets, so don't read into this comparison too much.)
Anyway, Win32 is implemented as one of many subsystems on NT and all its successor operating systems. .NET, and now WinFX, are/will be implemented in the same way, as just another set of APIs. But this is significant, because Microsoft hasn't done this just for kicks. I believe they are on the way to offing Win32. Why?
1) It's 32-bit, and the IA32/x86 market has its days numbered now. Honestly, not many of us need 64-bit computing, but at some point, killer apps will appear. As we all know, Microsoft's preferred method of forcing an OS "upgrade" down people's throats is bundling it with hardware. Aha.
2) It's not portable. This ties into the first point, but why might Microsoft be interested in portability? I don't just mean hardware, I'm talking about OS portability. Microsoft wants a contingency in case Windows (NT/2000/XP/2003/Longhorn...) finds itself becoming a legacy system (I think it already is, but that's just my opinion.) Maybe it's finally dawned on Microsoft that a VMS-based kernel with heavy process invocation fees isn't going to be able to win benchmarks while Linux keeps getting faster and better. Microsoft is only winning server benchmarks by virtue of building their SMB/CIFS and HTTP daemons into the kernel, you know. Who cares about stability? Benchmarks sell software to IT-ignorant PHBs.
3) Win32 is messy, and most Windows C(++) programmers avoid using Win32 directly at all costs (that's what MFC and ATL are for). Microsoft likes DRM, and DRM requires kernel/subsystem-level API calls. Likewise DirectX, which Microsoft is truly investing in; they know multimedia is their strong point and that the enterprise server market is something they can never corner. SMEs running VB apps using MS SQL, maybe, but not Fortune 500. So, they want a framework that is as "open" and "powerful" as Microsoft believes it can be, without opening up the source, of course.
So... whew. There you go.
"I am root. Bow before me." To this I say, "You are root, and you bear the sins of the world upon your shoulders."
It looks like Microsoft is already playing catch-up with Linux in some respects. The "sidebar"? What about Windowmaker's dock apps? What about gkrellm? What about the various panel apps for Gnome and KDE? I haven't seen any details about the WinFS file system, but I'm betting that whatever Microsoft comes up with could easily be done with some combination of MySQL, OpenOffice.org's document architecture, a pretty GUI and some glue to hold it all together. (It's an obvious point, but in case anyone has forgotten, developers have choices choices choices with open source: the GUI could be motif, Tcl/Tk, GTK, Qt, OpenGL,
In brief, unless Microsoft has a huge ace up their sleeve, whatever they want to do or come up with has already been done or can be done quite quickly with the enormous, comprehensive open source infrastructure that is available today.
The trick is that everybody wants just one more feature, but not everybody agrees on what that one more feature should be.
If you add the most frequently requested features... "OH MY GOD IT'S BLOATWARE! The preferences are so confusing! It takes so much disk space / memory / time to load!"
If you leave anything out... "WHAT? I CAN'T BELIEVE THEY SHIPPED THIS PIECE OF CRAP WITHOUT IT! They must either be retards, or they think I'm too stupid to want it, or they think they're smarter than me!"
Even if you try to find a balance, there's gonna be some guy who is pissed off that you omitted his pet feature and kept a bunch of crap he doesn't want.
Bill Gates just made the Adam Osborne mistake. He announced "WinFX", whatever that is, as the improvement to
Adam Osborne's company made an early personal computer. Adam announced a new model long before it was ready. Sales stopped because everyone wanted to wait for the new model. Adam's company went bankrupt.
It was amazing watching the bankrupting of the company on TV at the time. Osborne's company went from being one of the fastest growing to having insufficient money for operations in about two months.
It was a sobering lesson. Computer companies sometimes die extremely fast. Novell, WordPerfect, Corel, Fifth Generation Systems, and Central Point are examples. There are many others.
Microsoft has not been managed well. The company survives and profits because of having a virtual monopoly on operating systems and on office suite file formats. Think about it, suppose someone had a monopoly on water. That person could soon be much richer than Bill Gates.
For most businesses, the free Open Office is all they need. There are significant benefits to Open Office. It is much less quirky than Microsoft Office, for example. Most people are not very observant about the software they use, and they hardly notice the difference between Microsoft Word and the Open Office word processor.
Right now, many businesses use software that runs only under Microsoft Windows. However, there are many desktops that only need software that is already available for Linux. Those can benefit from the increased stability of Linux.
People don't care about the cost of Windows. The cost is only a few dollars of the cost of the computers they buy. The biggest issue against Microsoft is its adversarial behavior toward its customers. Using Linux means never having to say "My operating system company is partly my enemy."
Microsoft is on the way down. Most people don't realize that yet, however. Microsoft is one of the biggest management failures the world has ever seen. If the company could make a few changes in its behavior, it could stay profitable. However, it seems that abusiveness is more important to Microsoft than money.
Note that WinFX is someone else's trademark. WinFX is the most cracked and cheated program I have ever seen. There are 50 times as many links to cheats as there are to the product!
Microsoft has scheduled an MSDN TV program about "WinFX" for November 6 (Subject to change by Microsoft, of course.)
Microsoft claims that WinFX is their trademark. (The link is to a Google conversion of a
Microsoft has a history of picking inappropriate trademarks. "X" means unknown. It was inappropriate to use the letter X in conjunction with "Xbox" and "ActiveX". Aside from being someone else's trademark, WinFX sounds too trivial for use with an extensive programming product. Traditionally, "FX" has been used to signify "effects".
1) Managed directX has, at worst, a 10% performance penalty against the exact same C++ code. People are always complaining about how we have an excess of performance in todays' CPUs. This seems like a good use of it to me, thanks to #2:
2) Managed code does not have buffer overflows. How many bugs in Windows and Linux, especially rootable bugs, are a result of a buffer overflow? 50%? 75% 90%? I don't know, but it is a lot. Dotnet code has zero buffer overflows.
3) Managed code avoids DLL hell: the GAC and side-by-side execution ensure that programs will continue to run on versions of libraries that they are designed to support, since minor/major version upgraded files will not be fed to these applications (although revisions still can for bug fixing reasons.) Neither the user nor developer need to even THINK about these issues - the runtime simply takes care of them.
4) Managed code upgrades to 64-bit in a neutral and architecture-independent way. Apps that are "bit neutral" will run on a 32-bit system JIT'd for 32-bit mode, and those same EXACT EXE files will run in 64-bit mode on a 64-bit system, including making use of new registers and other such things. No recompiles - the JIT takes care of it. This also means that much of the code Microsoft writes - mountains of it - to handle all kinds of things from Office to [insert favorite feature here] can be transported across 32/64 bits and architectures. No more Mac version of Office if they want - Abstract any platform-specific calls into one or two classes and have everything else be managed bit-neutral code. Notice that no one is being silly enough to suggest write-once-run-anywhere for useful apps; that is and always was a pipe dream.
I would not doubt that the dotnet runtime on Longhorn is not going to call the Win32 API much; They might just be doing it internally and only using the Executive (NT/2K/XP's kernel native API) when necessary. That would explain part of the time length. Not only do you have to upgrade your existing code to C#/VB.NET/Managed C++/whatever other dotnet language, but you need to rewrite the new runtime to completely rid it of any dependance on the Win32 API. In this way, you also make the runtime a little bit more platform neutral, vs having to convert it from Win32 to Win64 for other platforms. But this is just a guess.
Natural != (nontoxic || beneficial)
"building apps that are as smart as Outlook.
Too.....many.....jokes!
So lets go over this:
1. User gets email.
2. User clicks email to view it.
3. User is infected with virus.
Explain to me how its the users fault again? Maybe they should have been running some 3rd party antivirus software?
Oh wait, if VBS scripts didn't have the inherent ability to automatically launch scripts, it would be a non-issue.
Ok, that came off a little more condescending than I thought but the point stands: How in the *world* is that the users fault? Should they just not read email?