The intuitiveness comes from the fact that you can easily navigate menus and dialogs to find the desired settings. With *nix, you've got to consult pages of documentation (either man pages or the web), find the right text file, and modify accordingly. Even customizing gnome/enlightenment/favorite window manager with point-and-click interfaces loses to Microsoft or Apple, because Microsoft and Apple have user interface folks whose sole purpose in life is to conduct user studies then design and test interfaces.
I've had similar problems with getting video to work recently with Windows. My solution was to try a little hunting on the web, finding an appropriate codec package, and installing it. That's a lot of work for an end-user, but I submit that the experience was no worse than a typical scenario would be under Linux.
You just don't see installing an app or applying a patch breaking other, unrelated things in the *nix world.
This is an unfortunate side-effect of re-use of Windows components. I think re-using those components is a very good thing however. For example: Windows explorer (the shell, not the browser) uses the same html rendering engine as IE for displaying text and widgets. The IE guys did all the work for that. All the shell guys need to do to display stuff is generate some html and send it down the pipe. So does Outlook and Outlook Express--all that pretty html formatted email is drawn with the same engine.
The video codec thing is another great example. Those third-party players you downloaded? They're all just wrappers around the standard OS video playing components. The authors of those apps were able to crank out the software very fast, and could count on the video working at least as well as Windows Media Player.
So, if you wanted to write a new codec, just make sure it conforms to the standard windows codec API and have your installer register it with Windows. Now anyone can use your new codec without having to install a new player (or RH).
This saves a lot of dev time for everyone involved. And the headache with patches and codecs will get better over time.
Did you turn on CD-ROM booting in your system's BIOS? It's worked quite well in my experience--I haven't needed a startup disk since I was running Win98. Even my poor old 440BX machine boots from the CD-ROM...
Admittedly, my sarcasm was extreme. I wrote from personal experience though. I've run several versions of RedHat, including a real old one, 6.2, 7.0 and playing with my girlfriend's 7.3 machine. I've also run OpenBSD 2.9 and 3.1. Older versions of Linux and BSD were absolutely horrible to get a decent desktop running (fighting with X, getting sound to work, etc). And if things did work, then one needs to find the right.conf file and look up the correct syntax and settings. 7.3 is better than I had used before, but my girlfriend still has issues with TrueType fonts, to name one example.
Windows and Macintosh (inc. OS X) just work, and tweaking settings is very intuitive. Microsoft and Apple put more money and developer resources into their products. As such, the fit and finish is much better.
More importantly, it does what YOU want it to do rather than what some marketdroid thinks it should do
Amen to that! I love spending a week fighting with my Linux install trying to get it to do something that Windows does out of the box. And compiling my own kernels! Just wait until I show grandma! She's afraid of computers, and can barely manage email and Word on her PC, but I think I can convert her. Hopefully I won't shoot myself in the head out of frustration while I'm trying to explain Window Managers to her. Wish me luck!
Seriously, without monetary incentive to improve your product, the software remains crap. How many OSS programmers just love to stay up all night to make sure Ctrl+Tab works perfectly in their app? Or spending a lot of time making really nice icons? Sure, Linux might have a great feature-laden kernel, but it's a real pain in the ass to configure and use if you're not already a guru. OSS software lacks the polish to make it really useful to end-users (because OSS programmers just want to work on the fun stuff).
This is true because Star Trek is science fiction. Anyone can go to the sci-fi section of a used book store and see a whole wealth of creativity and fresh ideas about humankind's future in space. The "they're just out of ideas" argument just doesn't hold water. The powers that be in the Star Trek business seem to have gotten themselves into a rut, but that doesn't mean the potential is gone.
You're very right. Dare not say anything good about Microsoft around here. It'll only get you attacked. I often wonder if Slashdot is about News for Nerds, or News for Linux Zealots. More often seems to be the latter. Bleh.
Before I get flamed... M$ is crap! Anyone anywhere using Windows is stupid! All of M$'s thousands of very intelligent, very motivated and highly paid developers are all incompetent!
And for good measure: (despite this being blatantly OT) Down with the DMCA!!!!
That's true to a degree. Closed source has the benefit of having a manager that can arbitrate these sorts of fights. Or decide in favor of one side or another.
Open projects who have a decided leader (Linux w/ Linus for instance) can also work this way, though I doubt Linus wants to arbitrate every argument:)
Microsoft typically uses typedefs to ensure the right sizes of ints and such. UINT and DWORDs are 32 bits long. I think there are also INT16, INT32 and INT64 types. When porting to a new architecture, they'll just make sure the typedefs are right.
Dude, you can't honestly tell me that adding a heatsink and fan was original innovation and that Intel stole it. If I were a chip engineer and I had to deal with excessive heat, then the voltage reduction would be obvious. Even more obvious (even to laymen), is cooling it off physically.
Unfortunately, that's a symptom of maturing software. Companies add features with every version. All those features (and fixes for bugs they create) make a lot of code. And you can't remove features from version to version...that's a real quick way to piss off all your users.
Though I'm not sure if you could blame poorly written code, or the fact that it does so much more than older versions. Both probably.
Intel was the only x86 chip contender at the time. And they weren't marketing a 16-bit chip that was at all comparable to the 386. This time around, there are very attractive alternatives in the P4 and Barton chips.
I'm pretty sure that Athlon64 will sell like hotcakes when they do release it. But I don't think that waiting six months is a bad decision.
90% of desktops out there run Windows. Releasing a 64-bit desktop processor now (or soon) means that OEMs will put 32-bit Windows on it. And as far as I know, there are still very few x86-64 apps out there as well. Since no one in the desktop world can take advantage of 64 bits, why release the processor (and spend the money marketing and supporting it) right now over a relatively cheap update to the current Athlon? I know that I won't buy an Athlon64 right now if it's only benefit over a P4 or AthlonXP is 64 bits.
Once MS comes out with x86-64 Windows, people will start making apps for it. Athlon64 will look a little more attractive then.
"It just takes one machine to get going," he said. "At any given point in time, it is hard to be 100 percent patched with any machine. We are working hard to make patch management easier. But 100 percent is a high bar and in this case we are not there."
He said that it's a tough problem and that they're working on it. Then admitted MS's deficiency.
Oh wait, sorry! I forgot my blatantly-closed-minded-rabid-Linux-zealot glasses. I see your point now. Silly me. This is Slashdot, after all.
Why did you post that? Just to attack somebody and feel cool? Fine. That's a mighty big penis you've got there. Feel better?
For everyone else: Oh, it's too hard, that's it.
Rick Devenuti didn't say that it was too hard. He said that patch management is hard, and that the current state of things is bad. He also said they are working on it. Microsoft is learning from this incident...rest assured that it'll drive some serious improvements in the software. That's the way software engineering works.
Yeah, I guess two. It seems like years...I guess life is like that:)
But Windows still hasn't blue screened on me. By and large, almost all blue screens are caused by bad hardware drivers. I'm sorry your particular configuration of hardware and drivers causes your machine to crash.
That's a danger when your operating system supports thousands and thousands of devices. Microsoft and/or hardware companies simply cannot test all configurations of hardware, OS versions and drivers. They do the best they can, but sometimes folks find something that was missed. The alternative is to limit the hardware: Apple does this. They know exactly what goes into every box, and can really test the hell out of those configs. I like windows supporting anything I stick in my box, and don't think MS should back down from that goal.
MSResearch got something like $5 billion last year. Their work goes into all kinds of stuff: Windows, the spam filter for MSN8, SQL server, etc, etc. Some of the stuff MSResearch is working on will never make it into a product, and some of their work is just publish (so everybody uses it). Cartoon cel shading for 3D games first came out of MSResearch, and I think that the algorithms are public domain.
The budget for MSResearch doesn't include the budget to pay the development teams for the work on new products. I can't quote you specific numbers for the budget for just Windows, but it's in the billions.
God I love regedit. I distinctly remember one incident where my girlfriend had a program running (revealed in the Ctrl-Alt-Delete dialog) that she couldn't identify. I don't remember what the specific name was--we'll just call it abc.exe. So, she fired up regedit, searched for abc.exe, and indiscriminantly deleted all the keys that showed up. I almost shit my pants. ("Are you absofreakinglutely sure you want to do that?!?")
The program turned out to be BackOrifice, so it was a good thing that it was killed utterly. Regedit is so cool.
It's sooooo painful to debug on Unix. It makes me want to kill children. At work, I use Source Insight for my source browsing, and VS for debugging.
If you can spare the cash, I suggest you get Source Insight--it has a parser that looks up symbols while you work. You'll fly through source code like a mutha. There is a free trial version to download and try. If you do, you'll say "Holy Crap!"
If you ever start doing development work on Linux/Unix, you'll wish you could go back to Windows. gdb kinda-sorta works, but you'll probably end up doing all your debugging with printf's anyway. The VS debugging capabilities are the best, period, amen.
If there is an open source debugger that works as well as VS, please let me know. However, the last time I did Linux programming (about a year ago), it was printfs.
I've been running 2000/XP for years now, and I can probably count the number of blue screens I've seen in that time on one hand. I'd really like to know how the myth of ever-present blue screens keeps getting perpetuated. Is it just everybody out there is still running Win9X and doesn't know how to keep the programs loaded at boot at a minimum? (I ran a very stable 98 box prior, but I had to rule over it with an iron hand and regedit)
If you're still running Win9X and you hate how instable it is, go and get an upgrade. You can probably find Win2k CD's cheap online.
The Outlook object model is accessible by any executable on the system that cared to talk to it. Palm syncing used the OM, as does everything that interfaces with Outlook. Outlook doesn't know the difference between a virus executable and a good one--unfortunately, virus writers don't sign their executables "THIS IS A VIRUS".
From another post: The security update DOES do that. When you install it, it does two big things: blocks all access to executable attachments, and requires permission before an external program can gain access to your Address Book or use Outlook to send email. Yep. Unless the user lets the virus do its dirty work, it's stopped.
The devs and security PM's in Outlook decided that this was the most secure way to still allow OM functionality. It was not decided on by higher-ups for competition reasons.
Slashdot Mirror
Amen to that. I think that if I start a hacker group, I'll call it Bob. Or maybe Alowishus.
I'm sure that the FBI's computer crime division's ears really pick up when they hear about a group named 'm4D #4X0r5'.
The intuitiveness comes from the fact that you can easily navigate menus and dialogs to find the desired settings. With *nix, you've got to consult pages of documentation (either man pages or the web), find the right text file, and modify accordingly. Even customizing gnome/enlightenment/favorite window manager with point-and-click interfaces loses to Microsoft or Apple, because Microsoft and Apple have user interface folks whose sole purpose in life is to conduct user studies then design and test interfaces.
I've had similar problems with getting video to work recently with Windows. My solution was to try a little hunting on the web, finding an appropriate codec package, and installing it. That's a lot of work for an end-user, but I submit that the experience was no worse than a typical scenario would be under Linux.
You just don't see installing an app or applying a patch breaking other, unrelated things in the *nix world.
This is an unfortunate side-effect of re-use of Windows components. I think re-using those components is a very good thing however. For example: Windows explorer (the shell, not the browser) uses the same html rendering engine as IE for displaying text and widgets. The IE guys did all the work for that. All the shell guys need to do to display stuff is generate some html and send it down the pipe. So does Outlook and Outlook Express--all that pretty html formatted email is drawn with the same engine.
The video codec thing is another great example. Those third-party players you downloaded? They're all just wrappers around the standard OS video playing components. The authors of those apps were able to crank out the software very fast, and could count on the video working at least as well as Windows Media Player.
So, if you wanted to write a new codec, just make sure it conforms to the standard windows codec API and have your installer register it with Windows. Now anyone can use your new codec without having to install a new player (or RH).
This saves a lot of dev time for everyone involved. And the headache with patches and codecs will get better over time.
Did you turn on CD-ROM booting in your system's BIOS? It's worked quite well in my experience--I haven't needed a startup disk since I was running Win98. Even my poor old 440BX machine boots from the CD-ROM...
Admittedly, my sarcasm was extreme. I wrote from personal experience though. I've run several versions of RedHat, including a real old one, 6.2, 7.0 and playing with my girlfriend's 7.3 machine. I've also run OpenBSD 2.9 and 3.1. Older versions of Linux and BSD were absolutely horrible to get a decent desktop running (fighting with X, getting sound to work, etc). And if things did work, then one needs to find the right .conf file and look up the correct syntax and settings. 7.3 is better than I had used before, but my girlfriend still has issues with TrueType fonts, to name one example.
Windows and Macintosh (inc. OS X) just work, and tweaking settings is very intuitive. Microsoft and Apple put more money and developer resources into their products. As such, the fit and finish is much better.
More importantly, it does what YOU want it to do rather than what some marketdroid thinks it should do
Amen to that! I love spending a week fighting with my Linux install trying to get it to do something that Windows does out of the box. And compiling my own kernels! Just wait until I show grandma! She's afraid of computers, and can barely manage email and Word on her PC, but I think I can convert her. Hopefully I won't shoot myself in the head out of frustration while I'm trying to explain Window Managers to her. Wish me luck!
Seriously, without monetary incentive to improve your product, the software remains crap. How many OSS programmers just love to stay up all night to make sure Ctrl+Tab works perfectly in their app? Or spending a lot of time making really nice icons? Sure, Linux might have a great feature-laden kernel, but it's a real pain in the ass to configure and use if you're not already a guru. OSS software lacks the polish to make it really useful to end-users (because OSS programmers just want to work on the fun stuff).
There's still some life in the old girl.
This is true because Star Trek is science fiction. Anyone can go to the sci-fi section of a used book store and see a whole wealth of creativity and fresh ideas about humankind's future in space. The "they're just out of ideas" argument just doesn't hold water. The powers that be in the Star Trek business seem to have gotten themselves into a rut, but that doesn't mean the potential is gone.
You're very right. Dare not say anything good about Microsoft around here. It'll only get you attacked. I often wonder if Slashdot is about News for Nerds, or News for Linux Zealots. More often seems to be the latter. Bleh.
Before I get flamed... M$ is crap! Anyone anywhere using Windows is stupid! All of M$'s thousands of very intelligent, very motivated and highly paid developers are all incompetent!
And for good measure: (despite this being blatantly OT) Down with the DMCA!!!!
That's true to a degree. Closed source has the benefit of having a manager that can arbitrate these sorts of fights. Or decide in favor of one side or another.
:)
Open projects who have a decided leader (Linux w/ Linus for instance) can also work this way, though I doubt Linus wants to arbitrate every argument
Microsoft typically uses typedefs to ensure the right sizes of ints and such. UINT and DWORDs are 32 bits long. I think there are also INT16, INT32 and INT64 types. When porting to a new architecture, they'll just make sure the typedefs are right.
Dude, you can't honestly tell me that adding a heatsink and fan was original innovation and that Intel stole it. If I were a chip engineer and I had to deal with excessive heat, then the voltage reduction would be obvious. Even more obvious (even to laymen), is cooling it off physically.
Unfortunately, that's a symptom of maturing software. Companies add features with every version. All those features (and fixes for bugs they create) make a lot of code. And you can't remove features from version to version...that's a real quick way to piss off all your users.
Though I'm not sure if you could blame poorly written code, or the fact that it does so much more than older versions. Both probably.
Intel was the only x86 chip contender at the time. And they weren't marketing a 16-bit chip that was at all comparable to the 386. This time around, there are very attractive alternatives in the P4 and Barton chips.
I'm pretty sure that Athlon64 will sell like hotcakes when they do release it. But I don't think that waiting six months is a bad decision.
90% of desktops out there run Windows. Releasing a 64-bit desktop processor now (or soon) means that OEMs will put 32-bit Windows on it. And as far as I know, there are still very few x86-64 apps out there as well. Since no one in the desktop world can take advantage of 64 bits, why release the processor (and spend the money marketing and supporting it) right now over a relatively cheap update to the current Athlon? I know that I won't buy an Athlon64 right now if it's only benefit over a P4 or AthlonXP is 64 bits.
Once MS comes out with x86-64 Windows, people will start making apps for it. Athlon64 will look a little more attractive then.
"It just takes one machine to get going," he said. "At any given point in time, it is hard to be 100 percent patched with any machine. We are working hard to make patch management easier. But 100 percent is a high bar and in this case we are not there."
He said that it's a tough problem and that they're working on it. Then admitted MS's deficiency.
Oh wait, sorry! I forgot my blatantly-closed-minded-rabid-Linux-zealot glasses. I see your point now. Silly me. This is Slashdot, after all.
Why did you post that? Just to attack somebody and feel cool? Fine. That's a mighty big penis you've got there. Feel better?
For everyone else:
Oh, it's too hard, that's it.
Rick Devenuti didn't say that it was too hard. He said that patch management is hard, and that the current state of things is bad. He also said they are working on it. Microsoft is learning from this incident...rest assured that it'll drive some serious improvements in the software. That's the way software engineering works.
I believe that Windows 2003 Server will have it. And I'd imagine that the SQL Server team will strongly consider it for their next release :)
I'd guess that all MS products will eventually move to an auto-update feature like XP's (optional like XP's of course).
Yeah, I guess two. It seems like years...I guess life is like that :)
But Windows still hasn't blue screened on me. By and large, almost all blue screens are caused by bad hardware drivers. I'm sorry your particular configuration of hardware and drivers causes your machine to crash.
That's a danger when your operating system supports thousands and thousands of devices. Microsoft and/or hardware companies simply cannot test all configurations of hardware, OS versions and drivers. They do the best they can, but sometimes folks find something that was missed. The alternative is to limit the hardware: Apple does this. They know exactly what goes into every box, and can really test the hell out of those configs. I like windows supporting anything I stick in my box, and don't think MS should back down from that goal.
I work at MS, and people around here are shocked when I say I've got a BSD box at home. People on Slashdot are mean when I try to defend Microsoft.
:)
I'm glad somebody agrees with me. I hope you feel better soon
MSResearch got something like $5 billion last year. Their work goes into all kinds of stuff: Windows, the spam filter for MSN8, SQL server, etc, etc. Some of the stuff MSResearch is working on will never make it into a product, and some of their work is just publish (so everybody uses it). Cartoon cel shading for 3D games first came out of MSResearch, and I think that the algorithms are public domain.
The budget for MSResearch doesn't include the budget to pay the development teams for the work on new products. I can't quote you specific numbers for the budget for just Windows, but it's in the billions.
God I love regedit. I distinctly remember one incident where my girlfriend had a program running (revealed in the Ctrl-Alt-Delete dialog) that she couldn't identify. I don't remember what the specific name was--we'll just call it abc.exe. So, she fired up regedit, searched for abc.exe, and indiscriminantly deleted all the keys that showed up. I almost shit my pants. ("Are you absofreakinglutely sure you want to do that?!?")
The program turned out to be BackOrifice, so it was a good thing that it was killed utterly. Regedit is so cool.
Or you could do what I do: use XP's remote desktop to remote-display my IDE, and still use the vastly superior Visual Studio for debugging.
I don't mean to be combative. I'm just saying that this isn't an advantage that Linux can claim anymore.
It's sooooo painful to debug on Unix. It makes me want to kill children. At work, I use Source Insight for my source browsing, and VS for debugging.
If you can spare the cash, I suggest you get Source Insight--it has a parser that looks up symbols while you work. You'll fly through source code like a mutha. There is a free trial version to download and try. If you do, you'll say "Holy Crap!"
If you ever start doing development work on Linux/Unix, you'll wish you could go back to Windows. gdb kinda-sorta works, but you'll probably end up doing all your debugging with printf's anyway. The VS debugging capabilities are the best, period, amen.
If there is an open source debugger that works as well as VS, please let me know. However, the last time I did Linux programming (about a year ago), it was printfs.
I've been running 2000/XP for years now, and I can probably count the number of blue screens I've seen in that time on one hand. I'd really like to know how the myth of ever-present blue screens keeps getting perpetuated. Is it just everybody out there is still running Win9X and doesn't know how to keep the programs loaded at boot at a minimum? (I ran a very stable 98 box prior, but I had to rule over it with an iron hand and regedit)
If you're still running Win9X and you hate how instable it is, go and get an upgrade. You can probably find Win2k CD's cheap online.
The Outlook object model is accessible by any executable on the system that cared to talk to it. Palm syncing used the OM, as does everything that interfaces with Outlook. Outlook doesn't know the difference between a virus executable and a good one--unfortunately, virus writers don't sign their executables "THIS IS A VIRUS".
From another post:
The security update DOES do that. When you install it, it does two big things: blocks all access to executable attachments, and requires permission before an external program can gain access to your Address Book or use Outlook to send email.
Yep. Unless the user lets the virus do its dirty work, it's stopped.
The devs and security PM's in Outlook decided that this was the most secure way to still allow OM functionality. It was not decided on by higher-ups for competition reasons.