Preview sports new file system, gee-whiz graphics, and some security fixes. The next version of Windows, code-named Longhorn, is still in the early stages of its journey to the retail corral, but PCWorld’s hands-on look at a preview reveals features we’d love to have now–while raising some intriguing questions.
There aren’t any questions in the article at all. Only in the title. In fact, the title is even misleading, since it doesn’t even attempt to answer if Windows 2006 would be your next OS.
What a fluff piece!
if not, then I will stick with windows XP and my Mac.
maybe if OBOS becomes a good platform for multimedia and electronic devices.
To me, all this hype about Longhorn just seems a bit much for something so far off. Furthermore, what steps the company is taking with the number of (I’m surely) well paid developers it has seems to be few. Another overhaul of their GUI… well, at least it looks a tad better than XP’s did. The WinFS thing seems minutely interesting — but I’m wondering what type of performance hit will be taken because of it. Also, some big questions I got is how much of this resides in kernel space? Assuming that WinFS is all it says it is, how much of the work in terms of what it can do as a part-DB/part-Natural FS is done on the kernel level. If all the database functionality is employed on a kernel level are we looking for increased instability and insecurity?
What I’d like to see come out of longhorn is a truly refined GUI that simply doesn’t have all the bells and whistles they seem to think users want. Something nice to look at but not too resource intensive.
I’d also like to see a rewritten command line interface, while Microsoft may see the command line as something completely and utterly unnecessary for the every day user, I think more and more they have to start appealing to the geek community. One of the major strong points I find with Linux is how well developed the command line is, in terms of what applications are available for it, but also in terms of how powerful it can be.
Last but not least, clean up the registry idea or employ it completely different than it is now. The registry doesn’t need to hold EVERYTHING, and for that matter, it shouldn’t. The core configuration of applications and system information is fine, but some of the random stuff I’ve seen lying around in XP’s and earlier’s registry is just bleh.
Longhorn: The most hyped OS for the year 2003.
Those features are nothing out of this world, nothing that can not be implemented on another OS.
Like they did before, this is nothing else than vaporware just to keep competition out of the scene.
But this time isn’t going to work.
I’d also like to see a rewritten command line interface
I thought I heard MS was doing this, but shell scripts look more like C# programs…
Personally I am not a big fan of CLIs, I hardly ever use the one in Windows and only use bash in Linux because there isn’t a GUI version of the program I want to use. I agree that OS/user-level scripting languages are good for quickly automating tasks, but I don’t want to do this kind of thing on an interactive basis. The problem with many GUIs is they are too clumsy so they make an archaic command-line look quite good.
Command lines aren’t archaic, they are just good at different things. For certain tasks (file management, for example) they can be much more efficient than GUIs. GUIs are good for making simple things easy, or managing visually-complex things, but CLIs are often great for making very complex things easy.
The power of the CLI is why you see things like AutoLisp in AutoCAD, and MEL in Maya. The CLI (and scripts) really takes advantage peoples’ well-developed abstract linguistic skills rather than their concerete visual skills.
Let me give you a concrete example. In AutoCAD, you can specify the end-points of lines either visually or through an integrated CLI. In fact, you can start part of the process in the CLI and finish it up in the GUI, or vice versa. The fact that I was comfortable with the CLI meant that I had a major leg-up in my computer drafting classes.
In my opinion only news with Longhorn is that MS goes further in bundling all its applications together in order to prevent third party applications from being installed in Windows and to prevent Office files from being opened in other operating systems – in Linux, that is.
But why would anybody take Longhorn because of that? In 2006/7 Linux market share in desktops will be around 3-5%, even higher in corporate world, and MS will have hard times ahead if it tries to introduce new systems that shut the rivals out.
The article itself was just a paid advertisement, with no content and no message.
You linux guys sound very tough until you find out you can’t play it on your linux box.
You’ll dual boot with longhorn the same way you dual boot with xp now.
Microsoft’s your daddy.
– Microsoft Fan
PS) Happy new year
So, why would id do all their development cross platform, and on Macs? Like all id games, if there is a doom 4, it will be on Linux/OS X/Windows.
My next OS will be Panther.
I heard once that Microsoft is completely rewritting its Win32 API in .NET. Is this true?
I don’t dual boot with windows and never will. If someone wants to send me information they better use a format that the rest of the world does and not just the ones sucking on redmond’s teet.
If not being able to open files that microsoft creates is a sacrifice, then so be it. It’s a small sacrifice to pay for having software freedom.
By 2006, I will have linux as my OS for my desktop. My servers are mostly converted to Linux. Just have one more to go. Longhorn won’t be on my PC at all.
Nothing New..I read that article last Month
I heard once that Microsoft is completely rewritting its Win32 API in .NET. Is this true?
I think .NET is going to be a wrapper around Win32, but not sure. Anyway, I sure hope they are going to rework the CLI, as bash is one thing I miss from Linux.
You give the example of AutoCAD. Although I have not used this program, I appreciate that CLIs integrated into programs can be accurate and flexible vs. a GUI. When I said ‘CLI’ I really meant UNIX-style command lines. IMHO UNIX/Linux command tools are far too complicated and have too many feature and options. Sometimes I try to find a simple option but have to wade through massive man pages. This is far too time consuming. And a simple mistake can easily cause crazy or disastrous results!
I agree that a well-designed CLI or scripting language aimed toward a certain task/problem/application can be more efficient than a GUI and reasonably easy to use.
“You linux guys sound very tough until you find out you can’t play it on your linux box.”
Well, since Doom 3 is being released on Linux by ID Software, as well as past games by them, I would exoect that Doom 4 will be released for Linux as well.
Happy New Year to you as well.
In 2006 I’ll finally get rid of my Windows partition.
I still need a couple of win32 applications.
I heard once that Microsoft is completely rewritting its Win32 API in .NET. Is this true?
Win32 will be there for compatibility only. The core of Windows will be written in WinFX, which is the next generation Windows API. In following the scheme (Win16, Win32, etc), you’d expect Win64, but Microsoft already has tons of info about WinFX.
The front end of nearly everything will be XAML, which I have to say, looks cool.
Touche! LOL!!!!
Keep Longhorn in people’s minds and they may not look at the current and near future alternatives. At least, that is my take on the current Longhorn hype. When all you have to offer your consumers for the next 2 years is SP2 for WinXP, and contracts expire before the planned release, you have to keep reminding them that you will have something “better” to offer.
Personally, I could care less about Longhorn. I’ll milk Win2k and WinXP for as long as I can. Otherwise, I know there Linux and someday, there may be good music support for it.
If people don’t want bells & whistles why aren’t we all still using DOS? plz don’t try to tell me the most succesful company in the world doesn’t know it’s customers, you have to realise that sometimes what we “think” we want isn’t what we really want, phew…HEAVY!.
Secondly why are people still so concerned about resources? resources are there to be used(afterall isnt that what resources literally means?) the important factor is how well resources are managed.
Take it ez folks.
why aren’t we all still using DOS?
Who said we don’t?
I heard once that Microsoft is completely rewritting its Win32 API in .NET. Is this true?
Take a look at the following article for a nice explaination:
http://www.ondotnet.com/pub/a/dotnet/2003/11/24/longhorn_01.htm
2006 the year of HDTV. MS finally achieves the vision of displacing the propaganda box on every TV stand and in every home.
If people don’t want bells & whistles why aren’t we all still using DOS? plz don’t try to tell me the most succesful company in the world doesn’t know it’s customers, you have to realise that sometimes what we “think” we want isn’t what we really want, phew…HEAVY!.
You are a bit off there. While a GUI is a very useful thing for quite a few purposes – I for one avoid CLI mostly – filling a GUI with useless bloat is stupid.
Another point: I sure am not perfect, neither are other people. But if sometimes Microsoft knows what we really want (which I actually doubt) what about the other times where we don’t want stuff and still can’t turn it off because Windows does not allow us to (just look at the IE-integration which is getting harder and harder to remove by 3rd party tools).
Secondly why are people still so concerned about resources? resources are there to be used(afterall isnt that what resources literally means?) the important factor is how well resources are managed.
Most people use resources with that meaning. Bloat = uses more ressources than wanted – thusly managing your existing resources badly – but you are unable to turn it off.
“who said we don’t”
Unless you using DOS as your primary OS your reading/replying out of context.
I actually agree with you on the first point, but that’s a fellow geeks point of view, no business in it’s right mind would limit sales by focussing it’s attention on the minority.
I don’t really understand the second point you make,what can’t you turn off? if you wanna take a swiss army knife to Windows XP it will run under 32Mb
Well, CLI’s in general are used for efficiency, not intuitiveness. The non-computing equivilent of the CLI (reading and writing) is unintuitive, takes years of practice, but is very powerful. CLIs like the ones you see in AutoCAD don’t need to be easy, people get trained in using them. Their major benefit is increased productivity after the training is complete.
“The front end of nearly everything will be XAML, which I have to say, looks cool.”
XAML may look good to you, but it is copied from XUL, wich really exists, and isn’t purely vaporware. You can actually contruct completely cross-platform GUIs, which you’ll not be able to do with M$-XAML.
Amen to that! I do exactly the same thing. I someone sends me a doc-file I return it with the words: Do it again, do it right.
I havent dual booted (to Win that is) for over a year, and I know I will never do it again either.
people don’t want bells & whistles why aren’t we all still using DOS?
You are a bit off there. While a GUI is a very useful thing for quite a few purposes – I for one avoid CLI mostly – filling a GUI with useless bloat is stupid.
Maybe a better explanation would be that the DOS CLI was horrible, and many people who remember it (or just heard about it) do not understand that there are CLIs that actually work?
I wouldn’t want to be forced to use either a GUI or a CLI exclusively, at least not when it would mean using a current WIMP-based GUI or a current sh-based CLI. Both have to be improved. Longhorn, as far as I understand, will have a new CLI, but what I heard about it didn’t impress me all that much – we’ll see when Windows 2006 is released in 2010.
This file system that MS is advertising so much with database capabilities, what exactly is that? Does the new Reiser4 system that Lindows will be sporting have similar database functionality?
People have been predicting that every year for the last 4 years and more, that Linux would be coming to the desktop NEXT YEAR.
Well, while it’s made strides. Today, it’s STILL not as polished, fast or usable as Windows 95.
The underlying OS is MUCH more sophisticated, stable and powerful. Yes.
The Desktop is still slow, unfinished, inconsistent, and needs a high degree of skill to maintain.
Lots of strides have been made in this direction. But, until developers get real about making either KDE or GNOME fast and polished…
The OS most people will be running in 2006 will be Windows based.
Past history shows that all these hopes for an OpenBeOS will be unfulfilled.
People have been promising replacement OS’es for a decade, and none of them have ever gotten to a stage that is usable.
I’m following ReactOS, AROS, and other OS’es closely.
But, most OS’es loose steam, and the maintainer has to change priorities before it’s done and usable.
FreeDOS is probably the most successful OS replacement project ever. And it’s STILL in development.
I don’t think 2 more years is going to advance BeOS Clones, and other OS Clones that much.
It’s just reality.
We may be at a 2.8 Kernal in 06. We may have newer versions of KDE and Gnome…
But, a basic design decision in the Linux world is going to have to change in order to unseat Windows as king of the hill.
And that is… Geeks need not apply. That Linux is for the Aunt Minnies of the world. For users who don’t know Crontab from the Diet Soda TAB.
They don’t want to mess with stuff like that.
Lycoris, Lindows, etc are getting close. But a LOT more work in smoothness and UI consistency needs to be done.
Just my opinion.
1) this article felt like fluff. I mean seriously wheres the meat, tell me more!! I want to know more! lol. Theres stuff in it I’ve already heard about. I wish there was more to it.
2) 2006 is a long ways away, and I will be patiently waiting for the next Huge OS war, if there is one by then. You know MS will market Windows Longhorn up to this point like no tommorrow, however in taking its time this gives chance for linux to become better and it might actually be better.
3) As far as command line versus Guis go. Command Line does make you focus more and increase productivity only when the GUI is too fancy for the person to want to stop staring at how pretty it looks. It is not intuitive and looses its purpose when dealing with internet things such as web browsing, unless your one of those people thats crazy enough to use a text browser lol.
4) As far as using Linux Only I am that way. However, my zealotness has dimished over the years, seeing as how it kind of is annoying not being able to help those. When I was only using Mac OS 9, since I only had a mac, I felt powerless when people looked to me to solve there windows problems and I didnt know how. I feel by far the best computer people are those that know how to handle and run as many things as they can, even though MS may suck its still important to know it, otherwise your pretty much inferior in my opinion. What use is a person who knows linux to those 90% who want their windows machines back up? None.
“I heard once that Microsoft is completely rewritting its Win32 API in .NET. Is this true?”
I think .NET is going to be a wrapper around Win32, but not sure. Anyway, I sure hope they are going to rework the CLI, as bash is one thing I miss from Linux.
Instead of assuming, how about getting the information straight from microsoft?
news://msnews.microsoft.com/microsoft.public.windows.developer.winfx.* has all the technologies that appear in Longhorn.
I think that XUL is a really interesting idea, but it seems like the uptake hasn’t been as rapid as it might be if it is to pre-empt XAML when Longhorn is finally released. Someone aside from Mozilla and Luxor really need to get behind XUL for it to be successful. The other best hope might be other XML- related technologies such as XForms and XHTML.
“But, a basic design decision in the Linux world is going to have to change in order to unseat Windows as king of the hill. And that is… Geeks need not apply. That Linux is for the Aunt Minnies of the world. For users who don’t know Crontab from the Diet Soda TAB.
They don’t want to mess with stuff like that.
Lycoris, Lindows, etc are getting close. But a LOT more work in smoothness and UI consistency needs to be done. ”
AMEN! It’s nice to see that some people have open eyes on reality. Their is much work to be done with Linux, both in the OS and on the software that runs with it. Until DELL, Compaq, HP, IBM, … finaly start to bundle PC with Linux, I’m not sure that Joe user will be using Linux in 2004,2005 or even 2006.
>This file system that MS is advertising so much with database capabilities, >what exactly is that?
Dare I say, if you want to see what WinFS will be like then install BeOS
:-).
The basic idea is that a file is not just represented by it’s filename and location (directory) but also by other attributes. Then you improve the search features so that you search on this extra information.
The simplest example that most people use is music files. If you add attributes like “Song Name”, “Artist”, “Album”, “Album Year” to the file then you can run a search like “Music files where Artist = ‘Prince'” and regardless of where the files actually are you get what you asked for.
This is of course all nicely integrated into the GUI so the result set looks and acts exactly like a directory to the end user.
Detractors argue that they already have id3 tags and specialised software to do this already. But of course by arguing against the example they are missing the point, attributes can be attached to anything, you are generally limited only by your imagination.
Of course the real issue is that if you are not the sort of person who is prepared to categorise your files then you will not benifit from this sort of technology.
However, if XML becomes all pervasive then you will not have to, since XML is designed to be searchable. So if there is a XML-WinFS interface then you can expect to be able to query against what the filesystem thinks are major attributes of the XML file.
BeOS originally uses a real database for this functionality but found it was too slow. They eventually built it into the filesystem in a different way. Microsoft figures that their database will be fast enough or the minumum hardware will be fast enough.
>Does the new Reiser4 system that Lindows will be sporting have similar
>database functionality?
I don’t know but many filesystems are expanding in this direction.
Cheers
David
Linux 2.6:Performace increase
Apple OS 10.3: Performace increase
FreeBSD 5.x: Performace increase
Windows Longhorn/XP: Perfoamcance goes down the crapper.
Why are you comparing 3 full-blown operating systems with the Linux kernel? I would imagine that when you throw the desktop enviroments on top of Linux, it’s about as slow/fast as anything else. Though I can’t speak for FreeBSD, but OSX was so damn slow to begin with, a performance increase was kind of a no-brainer.
No to my knowledge,
Reiser is just a really solid journaling fs.
However there is gnome storage which looks very promising:
http://www.gnome.org/~seth/storage/
You linux guys makes me laugh a lot !!!
“i’m not gonna use this nonsence longhorn … ”
Just use the best plateform that provide the best work for what you need.
You need a fast OS for multimedia works ? Web surfing ? email ? chat ? pear to pear ?
Then Use BeOS !
You need to work on hard network application ? big database systems ? Webservers ?
Then use Linux !
Need to …. play ?
Then use Windows !
The best os to me would be the filesytem of BeOS, the GUI of OS X, the network very easy to use Apple RendezVous, the network security of a BSD and … the sofwares list of Windows
In fact, actually, a Gobe 3.0, a RendezVous + a 100% supported USB on my PhOS would be far enough to me (and i think for at least 60% of computer users) !
Well your obviously quoting kernel speed increases in the first 3 and gui for MS(surprise surprise), Windows XP kernel is much faster than previous versions also.
overall as a complete OS they’ve all got slower.
Sorry Darius you pipped me to the post ^
I actually agree with you on the first point, but that’s a fellow geeks point of view, no business in it’s right mind would limit sales by focussing it’s attention on the minority.
That is true, but they should at least allow people to switch off all the fluff and bloat that they don’t want to have. A good example would be the Luna theme. A bad example is (once again) the Internet Explorer. It requires third-party-tools (www.litepc.com for example) to get rid of that thing. It would be nice if you could just switch it off like you can stop Nautilus from drawing your desktop in GNOME – although I have to admit it is also not that easy to find there, however easier than in Windows.
I don’t really understand the second point you make,what can’t you turn off? if you wanna take a swiss army knife to Windows XP it will run under 32Mb.
Sure, but can the non-geeks do that?
To sum it up: I would prefer either having fluff and being able to turn it off piece by piece or have no fluff and be able to add it piece by piece. I’d prefer the second one, though.
i heard (from reliable sorces) that in longhorn there introducing a new type of certificate intergration for all .exe you run.
simply this means that if u dont have a valid and unique certificate from the program distributer themselves or microsoft, that u will not be able to run it.
Thsis means no pirate software ppl!!!!!!!
Are you by any chance talking about DRM, Palladium and TCPA? If so, then we all should know it already.
Is that a bad thing?
Yes these are a little old and the XP service pack may have changed some of this. You can ignore the linux comparisons here and look directly at the comparisions between W2K and XP. To me it does not look like the XP kernel is faster.
http://www-106.ibm.com/developerworks/linux/library/l-rt4/
http://www-106.ibm.com/developerworks/linux/library/l-rt5/
http://www-106.ibm.com/developerworks/linux/library/l-rt6/
Windows XP PRo VS server 2000
One of those windows OS is server optimized the other isn’t.
Not fair to compare disk I/O between those 2.
Windows XP (NT 5.1) VS Windows 2000(NT 5)
Those are not comparing disk I/O, but basic kernel features such as interprocess communiction methods(pipes, sockets) and syncronizaton primatives(semaphores, mutexes). These are basic and very important kernel pieces. (I have been looking for confirmation on this next bit, but have not been able to find it)W2k Pro and Server (not Advanced Server) use the same kernel. It would seem that this is a very fair comparison since XP Pro and W2k Pro are aimed at the same people.
Mac OS 10.4. I’m going to install SuSE Linux (as soon as they deliver it) on my work machine, but I dont consider it “my” operating system since the company owns the computer. No, I’ll stick with Macintosh. Nothing about Windows 2006, 2007 or 2008 will interest me, whenever they get it released.
Will likely be DragonFly on my desktop, but I may get Longhorn if I decide to buy one of those Tablet PCs around the time it comes out. The current crop are too small and too slow for my liking, but not an entirely silly concept.
..
frankly, i still think XP is the idea os of mine. its more stable, secure than its predecessors.
easy of use, no hardware support issue, has the widest range of software selection(no doubt about it).
just as things are getting better and better, here comes NGSCB: a ‘benefit’ i would never want for a os.
so my main os will be narrowed down to either osx or linux.
typing in windows xp.
You need a fast OS for multimedia works ? Web surfing ? email ? chat ? pear to pear ?
Then Use BeOS !
Unless you by multimedia works mean listening to MP3s then BeOS isn’t the OS, it has practically no apps for media creation, and it’s really sad. Besides it can’t even be used to watch movies unless you use VLC which is multiplatform anyway. Even linux is a better OS for media creation. BeOS (still) has the potential but it’s not there, sorry. And browsing is really slow in BeOS unless you use Net+ which is really dated and doesn’t shine on newer hardware compared to browers on other systems.
BeOS does have a kicka-ass environment though, and it’s easy and relaxing to use, which is the reason I prefer it. It’s a great OS for development IMO.
Will windows 2000 anything be my OS? No. We’ve gone into that one at great length. Is beos great for media? back in the day, it was, and it will be again. It will also eventually be able to support archtectures never before supported, if things go well. But it will be Obos. Not BeOS. Be inc is dead. Microsoft Killed it. I believe that it will come back to haunt microsoft later, if not sooner. Whatever you think or say, know that “That is business”.
You can say it’s a bit of a paradox. Leave the fold of microsoft, knowing they haven’t offered you decent product for the price. Go to someone else. if they are a corperate entity not under control of microsoft (See: Microsoft Virtual PC for the Mac to understand that Apple is very evil along with The whipping boy of the month winner, Microsoft), then chances are, they are dead meat, if microsoft can see their way to making it happen. Unfortunatly for be inc, they were too small, and they couldn’t get intel, sun, ibm, and others to come to their corner of the ring, they were the underdog, and I believe americans in general stopped rooting for the underdogs.
I remember the patriots in the superbowl in the 80’s. Talk about underdog. I’ve got a cinderella story for microsoft, but it comes from the sadists guide to children’s stories.
First, it’s a well-known fact that, given a finite set of resources, apps, drivers, and system components will gradually grow to consume all of those resources. Look at the history of every operating system that has ever existed and you will confirm that fact (it also has parallels in the physical world. Think of how the world’s population has grown from a tiny outcropping in Africa to fill every continent). So, stop your griping about the “bloat” of operating systems and deal with reality. I, for one, am glad that app developers are thinking of new and innovative ways to use the resources that they have available to them — even if those ways consume more memory. What’s more important is the usability of the operating system. I don’t think it’s credible or reasonable to argue that Windows isn’t currently usable, even when resource consumption has grown over the years. Sure, you could argue that Win2K doesn’t run well on a 16MB Pentium 100 — and criticize it on that basis — but that isn’t a reasonable criticism. It wasn’t designed to do that.
Second, many of you are suggesting that Longhorn doesn’t have anything to offer. You’re seriously deluding yourselves. Longhorn is going to change the paradigm for how most apps are written. By 2006, an increasing number of apps will be written in “managed code”. Microsoft certainly didn’t invent managed code, although its p-code foundation for VB/VBA certainly predated and inspired the makers of Java. It would not surprise me if, eventually, Win32 apps were run solely under emulation. Too slow, you say? Maybe today. But we are talking about hardware which is 3 years away. Moore’s Law will not only make this possible but virtually guarantee that it happens. There will always be a reason to code to the metal (ie. drivers, kernel, etc) but the reasons are becoming more limited and flimsy over time. Most of us wouldn’t think of writing in assembly language anymore. Longhorn also offers many visual improvements in the shell. KDE/Gnome are constantly in catch-up mode with Windows. Even with MS releasing Longhorn in 2006, KDE/Gnome aren’t going to outpace it anytime soon. They’re slower, more klunky, and stymied by bitter divisions over which should be the Linux GUI.
Third, many of you don’t seem to understand that Longhorn is virtually guaranteed to have significant market share, even if you don’t like it. Windows XP goes out the door on virtually every new desktop PC sold. We’re talking about millions of PCs. By 2006, the majority of operational PCs will be running XP, not Linux. When MS releases Longhorn, OEMs stop offering XP and start offering Longhorn. People aren’t going to stop buying PCs. XP will be nearing the end of its useful shelf life and MS will cut support. OEMs will naturally move to Longhorn. And so the story goes…
Fourth, I would suggest that many of should waste less time debating over why Linux isn’t gaining traction on the desktop and, instead, focus on how well it IS doing on the server. That isn’t anything to complain about. Every product has a niche. Trying to force products into niches they weren’t designed for is generally a bad idea. I think the biggest reason that Linux hasn’t been adopted by more desktop users is that the Linux community has been studiously listening to ITSELF — NOT END USERS!!!
Fifth, an OS (and software/hardware, in general) isn’t a substitute for a life. Live, people. Get out of your parents’ basements. Get laid once in a while (yes, even you Linux geeks with the bad skin, teeth, and dirty hair). There are other foods besides Pizza. Exercise. Drink water. Eat vegetables. Instead of giving your services to open source projects, try donating your time at a soup kitchen or be a Big Brother. Don’t kid yourself: Your software will be obsolete within a few years. But human beings last 75 years. Make a difference that counts. Invest in humanity. Not machines.
Well, Gates, did it again. Instead of Hiring a Graphics Artist to do their GUI, he just directed the development team to copy Apple. Cheap Gates, how do you think he got so rich.
But, at least I won’t have to look at those Clown colors in XP any more.
Thank you Apple.
Seriously, by 2006 shouldn’t Microsoft start writing that Linux Compatibility Mode API?
“I would imagine that when you throw the desktop enviroments on top of Linux, it’s about as slow/fast as anything else.”
Nope, or at least not in my experience on the same computer over time, with no hardware upgrades. KDE 3.2 with the 2.6 kernel runs faster on my computer than KDE 3.1 with either 2.4 or 2.6 kernel, which in turn is faster for me than KDE 3.0 with 2.4 kernel, which again in turn was faster than the final 2.x KDE release with the 2.4 kernel. Every KDE release has performed better for me than the previous, and the 2.6 kernel gave even more of a speed increase.
Heck, I’m still running Win2K on my desktops. Got a WinXP install for gaming and WinXP on the laptop mostly due to drivers issues. I don’t think Longhorn will be my next OS for the next four or five years and by that time I might not only be running OpenBSD on the server…:-)
that being yelled out, I have to say that windows was always about being a platform for me.
a platform to run other crap.
ahhhh. the early days. i remember when i ran DOS to play games, Winnt 3.51 for photoshop, and Win95 for multimedia.
what a joy. 3 operating systems to get it done.
later on, i booted win98se for games, winnt for sysadmin of the nt4.0 domain, and win2k for photoshop and video editing.
now i’m running xp for all those windows related apps.
about F’n time.
so lifes good right?
with XP came some serious baggage, and serious bloat.
and while a good 20 minute session on a new XP box, and I can have much of the crap cut out…..i.e. services disabled, outlook/msn messenger removed, firebird installed, putty installed, etc….it’s not looking good. microsoft finally provided the platform that would allow my third party apps to run on the same setup.
but to stop there would be insane for Microsoft. Longhorn is going to be a big, ugly bloated piece of crap.
It will take me 3 times as long to turn off the eye candy and disable the useless services, install antivirus, install anti-keyloggers, install anti-spyware, install anti-adware, remove IE, or whatever BUILT-IN crap that is lined up.
there are people TO THIS DAY, who hate XP, and will live and die by Windows 2000.
There will be a similar group for XP.
then add to that the COUNTLESS millions using Win9x.
i’ve got some serious BAD NEWS for microsoft.
their user base is becoming fractured like never before.
and believe it or not, the NT4.0 domains, WILL NOT DIE.
it’s all a bad dream.
and i don’t think longhorn is going to save them.
This Microsoft product is going to fall flat on it’s face. The company is a disgrace to the computer industry and it has too many enemies. Microsoft is carrying too much baggage and it’s beginning to drag everyone in the industry down the drain.
Microsoft is a threat to humanity because it violates human rights and freedoms to open standards and a competitive marketplace for information technology. Microsoft makes every choice to abuse their monopoly position, their product is designed to rely on private binary formats that tie users with no choice, to their expensive tools. It is not logical to design software in which one application has intimate knowledge of the implementation of another application. These methods are used only to take advantage of the ignorant end user. Microsoft is a disgrace to the United States of America.
The timegap is too big. There are too many competitors out there now that can more easily take a slice of Microsofts pie. And the DRM lockin system will not save them because by 2006 most of the world excluding America will be using open document standards.
Most of what you said is just bullcrap. Don’t give excuses for bloat. It’s greed a laziness. MS doesn’t want anyone to write real software ever again. They want you to keep buying their ever-expanding tools to create megabyte software. The fact is most of the world cannot afford the kind of computers required to run their bloatware (Linux and related software are almost as bad). The truth is, A P100 is a bloody fast machine. Compare it to a 6502 or Z80 and you will see that almost all of those extra cycles are just being wasted. The kind of thing people want to use computers for only needs tens of kilobytes at most. No company in the software industry can do this. No open source project can do this. That is unless we drag ourselves away from the 1970s, dump our FORTRANs and start over.
“The kind of thing people want to use computers for only needs tens of kilobytes at most. No company in the software industry can do this. No open source project can do this. That is unless we drag ourselves away from the 1970s, dump our FORTRANs and start over.”
So… you’re saying that both Windows & Linux are written in FORTRAN?
Seriously though, You’re not going to see tiny programs that do what people want because people want to write reasonably portable code in a reasonable amount of time. Writing a web browser, email client, kernel, UI, file manager, office suite etc in assembly just ain’t gonna happen. By the time you’re done, the chip you wrote the things for will be EOL.
So… you’re saying that both Windows & Linux are written in FORTRAN?
C is effectively FORTRAN. It translates formulas and turns them into machine code. I’m not suggesting to write web browsers in assembly either, but… (I might come back to this later if I’m brave enough)
A lot of the software we have solves artificial problems. For example, you mention file manager. The prevailing mindset is every computer needs a file system. It’s a pretty awkward way of storing information. Would prefer something record-based, with proper searching facilities so I don’t have to bother managing files. Most tasks do not even need that, just a nice way of flushing variables to disc.
By the time you’re done, the chip you wrote the things for will be EOL.
That’s kinda part of the problem, because it makes business sense to write for portability (in the non-x86/Windows world, that is). We never have time to write software that will run well on any particular machine. And most software is too big for anyone who holds their sanity dear to start writing optimised routines and cutting out the entangled bloat.
It seems Microsoft and Linux indeed has something in common…
Fortran is a niché language. It is extensively used, tried and tested for scientific applications using numerical methods.
You don’t write a Web browser in fortran, it is however excellent for writing a numerical solver for systems of Differential Equations.
Bringing up Fortran here is totally irrelevant. Most of the best scientific libraries are still written, and continue to be written in Fortran. They are definitely not bloated, unstable or anything. Rather they are precise, thouroughly tested and efficient.
I could care less about win longhorn , if i want play games i use a console (no NOT xbox) webserver? on FreeBSD …and if i was going to play games on a comp id rather play it on
AmigaOS4 on a Aone board….
I didn’t mean to bash this language particularly, but the idea of generating code from infix equations. I appreciate why this is preferrable in a mathematical/scientific community where the notation has evolved over centuries. It may be just what people need to crunch those time-consuming mathematical tasks.
I do not, however, think this is the best way of expressing modern computational tasks. OSes like Windows, to get back to the article, are interactive, not batch cycles; they are more memory-oriented than arithmetic-oriented.
OK, I admit to being a Forth fan. Sorry I am annoying you all I like to think of it as the ultimate macro assembler where you can still be close to the machine, but abstract some of the messy details that make assembly programming so hard. And eye-blink compiles.
To those C programmers out there, I like you language actually, in a sick kind of way. It is quite possible to write good software with it, but features like data-typing which are supposed to be there to stop you making mistakes can blind you of many opportunities to make your program simpler and faster
You’ll be playing Doom 4 on longhorn.
You linux guys sound very tough until you find out you can’t play it on your linux box.
You’ll dual boot with longhorn the same way you dual boot with xp now.
You should at least know what you are talking about before you puke out something like that.
id has been very good about supporting Linux users with most of their previous games, as they will be doing with Doom 3 (4 does not exist). They even release their older engines (Quake and Quake 2) under the GPL. Don’t be an idiot.
There has been ever increasing support for major game titles with native Linux installers from developers. Not to mention increasingly good emulated Direct X support from WineX.
-yutt, resident gamer
ZX80 P100 and other much slower but more predictive, no speculative execution, register renaming, weird cache hierarhy. AthlonXP/Pentium4 RAM access take > 100 CPU ticks, CPU perform 3 op/tick. VM is really Evil of 21 century, you just move mouse over button and can not predict that it is time to commit HD journal, no free RAM and ( surprise ) tooltip handler start to load strings from HD. I hope Intel, AMD, or anybody else ( Red Dragon ?) do some deep investigates and return to us good old times of predictive, guarantee programming enviroment.
I scare Longhorn because it whill utilize hi-end 3D technology, and thrust me – Bill do it wery well, if no he buy somebody who do it. We still make “3d look” in GNOME and KDE by pixel based hand painting and share art in .png. I guess HD, RAM and CPU capacity in 2006, but who know how fast whill be Radeon 2006 and what task it can do whith it 64-way shaders and 8-port 1 Gb memory ? Do YOU can write good XFree or whatever driver for it ? And were that army of brave hackers that spend all free time learning Blender ? May be we got real kerning in GNOME 3/4 or fast gnome-terminal, but it whill compete with terrific smooth 2d/3d enviroment of Longhorn. Disband core xfree team do not help a lot, it is too late.
Quality.
I’d like to see microsoft components be tested with another Magic Number. 300 isn’t what I consider enterprize quality.
I’d like to see the Microsoft QA team,
1 — actually exist
2 — use 3,000,000 records as the test file in their apps.
Sounds like a Flame.
Sorry, but this is my biggest problem with Microsoft.
Years after all you early adaptors lie to the world and tell us how great Microsoft Software is, I try it out, and find it breaks all the time, except for that 1 man Dentest office in Royersford, Pennsylvania.
One will find a good unbiased review or Longhorn pre-alpha on http://www.extremetech.com (it’s a couple months old so you may have to search).
As always, PCMagazine is a hype box. Why does anyone still read that magazine?
From the review’s I’ve heard, one from a M$ Windows developer who runs it himself, it runs like a dog on anything less than a 2+ GHz P4 with a high end graphics card and at least 512MB of RAM. Your E-machine isn’t gonna cut it, and at the current salesmode neither will any other new brand-named computer you buy unless you buy the graphics upgrade. All those millions of computers with onboard graphics will most likely choke on Avalon; unless they get it together and quit this z-buffering non-sense.
WinFS for My Documents. Currently, My Documents comes with a couple of folders by default: My Pictures and My Music. Now since WinFS is little more than SQL for documents, which is a great idea…FOR DOCUMENTS. However, for people’s music and pictures, bad idea. According to extremetech, it took them 45 minutes to move 300MB of media into the WinFS spot. That’s pretty bad guys, I think I will stick with my Reiser partition.
Sidebar…Good idea, it had better come with some applications by default though. I don’t want to pay extra for a calendar.
Speech recognition, this is something I haven’t seen done well yet. If Microsoft does it first, lovely. But last I used was a Dragon Speak program, and it was a real dog. I don’t want to spend a week teaching it to understand me. I’m sorry, but a mid-western “twang” does not warrant a week of reading to it. Maybe if I was originally an eastern speaker learning English, but not as someone with little accent.
They’ve got 3 years left, let’s see if they get anything done in that time.
Why can’t we understand that the majority love easy to use beautiful things? Its just a part of human nature and Longhorn is a part of that, choosing whats easy. I can’t see Linux offering any consistency interms of the interface and applications built for it.
Like someone said before, company’s such as Dell, HP, IBM etc… do not see a market for Linux on the desktop, their is no incentive to put it on their machines. Once consumers see that ugly KDE/GNOME interface for the first time and the lack of compatibility, they will run to the local computer store and buy a copy of Windows, format the drive and install it.
The only market I see Linux having competitive power is the Server market and its not even against Windows, its Unix. The market share that Linux has today in the Server arena is still dwarfed by Windows Server.
As for Apache being the web server of choice, thats nonsense, Apache Servers are closet servers. People, we just have to understand that Windows Longhorn will be the winner in 2006, Linux will be 200 steps behind, OS X 199 steps behind.
Just make up your mind and double boot, you know you Linux heads love Longhorn and you certainly cannot wait to use it.
“Just make up your mind and double boot, you know you Linux heads love Longhorn and you certainly cannot wait to use it. ”
It’s dual boot, not double boot.
And there’s no reason to pull your wallet and install longhorn. I’m happy with Linux. If you’re happy with Microsoft, then that’s fine with me.
I dont need Windows anymore, Linux is better and more fun.
If you have problems with MS quality why are you Microsoft Partner?
http://www.atsva.com/partners/index.cfm#Microsoft
Yawn again
The company is a Microsoft partner.
Does a Microsoft Partner get kickbacks.
Top Mgmt doesn’t code or do maintenance on the project do they?
In my last consultancy, That mgmt actively attempted to convert all Java prospects to Windows solutions.
Upper mgmt was clueless about the quality of the API’s coming from Microsoft. We also got clients who requested only MS VB solutions.
I remember one time where the client requested a VB server application, with a vb Client using the RichText control.
The RichText control crashed after 300 records, ( you might remember that ), and the server started losing VB threads after 300 were issued. My mgmt took their best developer off another project to check my work. AFter three days, he came to the same solution, we couldn’t use MS VB components to write this solution. It was converted to C++.
“Yawnhorn”
How about: “tootourhorn”
I am slowly getting rid of MS. I don’t care what Gates et al comes up with. Within a year MS will be out of my life. I can only speak for myself. I do it for political reasons. I am a capitalist, and MS represents an abject failure of capitalism, Gates being the personification of that failure. Something was wrong in the market place for this company and person to have grown so rich and powerful. “Luck” was also a factor. In a healthy market place no one company can get this big without something unhealthy going on.
I don’t know if Linux will succeed in developing a large desktop market share, even by 2006, but it seems a good way for me to go in order to rid myself of the Gates menace. I run Windows now, but having just done a clean install I will not reload MS Word, but will use Open Office. All my text files are already kept as Rich Text, and I try to keep all MS proprietary file-types off my computer as the first step in switching.
PC World needs subject matter, perhaps having given out all the XP tuning tips it can get away with.
The RichText control crashed after 300 records, ( you might remember that ), and the server started losing VB threads after 300 were issued.
VB Threads ? There is no such thing as VB threads. It is possible to multithread in VB, but that requires building your project as an ActiveX EXE (COM Server) and you’ll be following the threading rules according to COM.
If you pulled it off any other way in VB you deserve the hell that came upon you.
My mgmt took their best developer off another project to check my work. AFter three days, he came to the same solution, we couldn’t use MS VB components to write this solution. It was converted to C++.
So basically a lack of experience cost your company a few bucks ? Sounds normal.
VB Threads ? There is no such thing as VB threads. It is possible to multithread in VB, but that requires building your project as an ActiveX EXE (COM Server) and you’ll be following the threading rules according to COM.
So! This didn’t work! Try it out yourself.
After 300 threads, the server starts losing them,
they fall asleep and never awaken.
Typical Microsoft Coder, shots off his mouth,
but doesn’t test anything.
If they make Longhorn cheaper, faster, smaller, more stable and more secure than Windows 2000 then I may buy it.
If it’s expensive, slow, bloated, unreliable and less secure than 2K, then I will not.
I can’t think of a single new feature I have seen in the various Longhorn ‘reviews’ that will make my work any easier.
As I only run a single app on Windows, with it’s own widget set, the entire OS is pretty much relegeted to being a very expensive dongle.
I will be using Windows 2006, Windows 2000 Pro/Win XP Pro happen to be the most stable, robust, and fast operating systems I have ever used. It will be successful and plus be 64bit versions, it will be very nice indeed.
<<<Most of what you said is just bullcrap.>>
Rrrrrright, Luke. If the software industry were being run by guys like you, there would be no innovation. Software would stagnate, nobody would add any new features (what did you think — new code comes without any footprint penalty?!? Geez. Get a clue.
You should really open the curtains tomorrow.
you guys dont seem to care what drastic effects TCP/DRM, Palladium is going to do to our computing experience….
I find it incomprehensible that some of you out there are actually waiting axiously for this cancerous relase of windows.
Frankly, if your ignorance is clouding your perspective to such a degree that you fail to see the danger ahead, then perhaps you deserve the consequences..
Can you tell me for exact what is the effect of Microsoft’s implementation of DRM? No? What about how is Palladium going to be implemented? Nope? I find it hard to believe that you would know anything about Palladium while nobody outside Microsoft and their partners – even crazily pro-Microsoft guys, don’t know much about it.
And if you see the white paper Microsoft released, it is nothing more than today’s DRM in Windows XP with hardware underpinings to make it more secure. To the average consumer, Longhorn should be no more as restrictive to them as Windows XP.
I really hope Micro$oft dubs around for two years and then comes out with another cobbled-together WinMess.
I got a new Sony Vaio laptop for Christmas, booted into the bios setup, set it to boot from cd and installed Debian Libranet- which I then cleaned up to run “Sarge”. XP “Home” makes a good coaster…
Seriously, despite all the “Linux ready for the desktop?” blabber you read, Linux not only is ready,but it’s THERE.
I’m browsing wirelessly, running a printer hooked to my wifes
winXP box, and playing tunes- all from my LazyBoy, by the gas fireplace…
I think as more (and better) Linux distros are released and hardware compatibility becomes less of an issue, we’ll see more people grabbing (“lindows”) boxes. Once the base OS becomes the standard, we’ll have a new generation that will shudder and roll their eyes at the mention of..well you know..
>I’m browsing wirelessly, running a printer hooked to my wifes winXP box, and playing tunes- all from my LazyBoy, by the gas fireplace…
Let me tell you a secret: you can do that with your original Win XP you paid Microsoft $50 to.
Thank you for supporting Microsoft.
Tell us what you, as a home user, can do with Linux you can’t with Windows- that would be much more interesting.
If nothing comes to mind, then in 2007, 99.99% of people who buy their PC will turn it on, boot into Windows 2006, and would not look for alternatives.
Tell us what you, as a home user, can do with Linux you can’t with Windows- that would be much more interesting.
I think the question would be better if you had asked what can he do with windows that he can’t do with linux? THAT is the question at hand. If a person that uses their computer as a desktop PC can do everything in linux that he/she used to do in windows, then that would, in my mind, be proof that linux IS ready for the desktop.
I can tell you what I CAN do in linux that I can’t do in Windows. Browse the net without a million pop-up ads pissing me off. How about not being worried that I’m going to get a virus every time that I boot up? How about trouble-shoot the GUI if there is something that is preventing a boot-up of it. In windows you’re pretty much stuck with the GUI at all times, unless you want to attempt to use that poor excuse for a rescue disk that is built into the XP installer.
If everything works as well, or better for him in linux, then all the more power to him. Personally I run Libranet as well (upgraded to Debian Sid) and everything works great.
..
http://www.microsoft.com/resources/ngscb/default.mspx
No I don’t think people will be in a hurry to upgrade to Longhorn when it finally gets released. The adaptation rates of new Microsft-OSes since winNT4 have bin slower for each new release. Why? NT4/2k/XP already fill the requirements of most organizations.
The cost of an upgrade is not just the cost of the licensing of the new version, but also the cost of the labor reqired for the upgrade, not to mention potenital loss of productivity during the upgrade process. The main reason for upgrading will be that MS decides to end life of their product. That is the upgrades will be forced.
As the main cost of an ugrade is in labor and loss of productivity companies will likely look for solutions where they can be in control of the upgrade process. The obvious way to do this would be use open source. The fact that this lowers the licencing costs just makes the choise even better.
And as companies will delay upgrades, as long as possible chanses are that opensource solutions have improved a lot when upgrades are needed making the open source choise even esier.
Even though we can expect the rate of improvements to open source software to slow down in the future, this is because it gets more and more expensive to develop useful functionality the more perfect a system becomes. Of course this rule also applies to Microsoft. In fact even more so as they have to carry all development costs by themselves, while opensource projects can be cooperative efforts where the costs are shared among many participants.
It doesn’t look good for Microsoft. Their situation is a little like the one Apple had pre MacOS-X, perhaps even worse. The old MacOS was seriously out of date and there was a great incentive for them to upgrade to a new and better OS. This isn’t the case with XP compared to Longhorn. But still, I wouldn’t be surprised if MS tried to go the open source way. Perhaps it will be something BSD based just like Apple , if they can’t get their act together with Longhorn.
“People have been predicting that every year for the last 4 years and more, that Linux would be coming to the desktop NEXT YEAR.”
Just like, people was talking about Linux on servers for years before it finally appeared. Linux was ready for the server long before it was used in this field. What was missing was applications. The situation is the same for the windows desktop. OpenOffice.org is a start but it need to be followed by a few more before we see any mass migration to the Linux desktop. The interest in Linux from many governments could accelerate the process. And you have to
remember that 2006 is 2 years away, and the quality of the desktop user experience will almost certainly exceed that of windows XP by then.
“Well, while it’s made strides. Today, it’s STILL not as polished, fast or usable as Windows 95.”
Sorry, but I think your view of windows Windows 95 has passed through som nostalgia filter that paints brighter than it actually was. It could take a day before you even got the thing installed and running getting hardware drivers from all over the internet. Not to mention that internet was an extra add on to the original win95. Remember the limitaions in naming of files (even though much better than 8+3) Remember how hard it was to install and remove software, compare that to yum, aptget, or any other modern Linux software package management system.
Compare the possibilty to run switch between several logged in users on the same or other machines (this is not even possible in XP as fast user switching cant be done on a network). Compare the possibility to create thin client desktops that can be easily managed.
“The Desktop is still slow, unfinished, inconsistent, and needs a high degree of skill to maintain.”
It is true that the Linux desktop requires higher skills to admin. But it can be done remotely, and it is easy to script this means that one Linux admin can handle far more users than his fellow windows collegue. Being more configurable, it is also possible to configure the user desktop to better fit the business process of your company. All this makes Linux a better choise for the corporate desktop. Your point may still hold for home users.
“Lots of strides have been made in this direction. But, until developers get real about making either KDE or GNOME fast and polished…”
Both Gnome and KDE gets faster and better each release and both sides make real efforts to make them interoperable.See http://www.freedesktop.org. And by 2006 we can expect two major new releases of both KDE and Gnome. The new Linux 2.6 kernel have features that are designed to increase the speed of desktop applications, so I thingk that you will see major improvements even in a few months from now.
“The OS most people will be running in 2006 will be Windows based.”
This is probably true. But it will be windows 2k and windows XP, not Longhorn. The upgrade rate to Longhorn once it has bin released will be slow, until the end of life of win2k and windows XP. Nowdays, companys want to have some expectations of a ROI before they upgrade their systems. The days were you upgraded just because there was a new version is long gone.
“But, most OS’es loose steam, and the maintainer has to change priorities before it’s done and usable.
FreeDOS is probably the most successful OS replacement project ever. And it’s STILL in development.”
A project that is not in development is a dead project.
Besided Linux have a lot of core developers, even if Linus was to be run over by a bus tomorrow there would be lot of people that could take his place. In fact for the last versions of Linux the shipping versions have bin maintained by other people than Linus. While Linus have bin working on future versions.
“I don’t think 2 more years is going to advance BeOS Clones, and other OS Clones that much.”
The problem for BeOS is that it is a new OS, while unix clones like Linux benefits from already having users knowing how to use them, and developers that have some idea of how to port programs from other platforms.
“We may be at a 2.8 Kernal in 06. We may have newer versions of KDE and Gnome…
But, a basic design decision in the Linux world is going to have to change in order to unseat Windows as king of the hill.”
You almost sound like the KDE and Gnome usability groups.
“And that is… Geeks need not apply. That Linux is for the Aunt Minnies of the world. For users who don’t know Crontab from the Diet Soda TAB.”
I don’t think Linux is for Aunt Minnie. At least not if Aunt Minnie is a home user. Unix/Linux is designed to be a good multiuser network centric OS. This fits like hand in glove for most corporate situations, where the ease of remote administration and scriptability makes a difference. But Aunt Minnie probably couldn’t care less that it is just as easy to admin 1000 boxes as it is to admin the one she has on her desktop.
“Is that a bad thing?”
No, its a good thing. But it will also mean that people will evaluate their needs more closely. Companies will have to provide software for their employees so that they can work from home. Nowdays many companies turn the blind eye to software piracy from office to home environments.
But this also means that the knowledge of how to run windows software will be less abundant. Students and other people with not so strong economy will use other solutions, e.g. OpenOffice instead of MS-Office. When thes students graduate and get businesses of their own they will be less likely to use MS solutions. Today software piracy probably do more good for software industry than it damages it, as the userbase of the products increase to include people that never would have used the software if they were to pay for it. And when they have a business, that can pay for the software they buy what they are used to.
…