Please god no. The thing about that article that got me interested in Inkscape is the fact that the developers, and more importantly, UI designers are artists themselves, so clearly they want to build a UI that's focused around getting work done, and it sounds like they're doing a good job of that.
While I keep reading that the Gimp's interface was greatly improved with 2.0, when I've tried it, it felt as kludgy as ever. The Gimp does a lot of cool things, but create a smooth workflow it does not. For that reason alone, I feel it's better that this be a stand alone project. It allows them to build a much lighter system aimed at doing one thing and one thing well.
In general, if you're working with vector graphics, you're not really going to care about immediately working with raster. That said, I do think it'd be cool if someone could take the Gimp and strip it down to a very focused UI like Inkscape seems to be doing, creating a set of interlocking common programs like Adobe currently does with their Creative Suite. However, for this type of work, the plug-in-replacing-an-app mentallity is exactly what needs to be avoided because while it may work, an artist will usually be much happier with a lighter program aimed at doing what they want it to do, not ten thousand features they'll never need creating a cluttered and confusing menu system and obscure keyboard shortcuts.
I'd be one of them, thanks in huge part to Dropline Gnome. Excellent set of packages complete with a built in updater that lets me run a light, clean Slackware install, but not have to worry about updating and compiling huge packages - ie: Gnome or X.org myself. Todd does an awesome job keeping it up to date, and this combined with a 2.6 kernel (which drops right in on a default Slack install) make Slackware a really nice little distribution to use. But who am I to say? I'm only running it on 3 machines at the moment.;)
Personally, I'd take issue with the idea that the Gnome project is aimed at Windows users. If anything, it's aimed at Mac users, as many of its design cues come directly from Classic MacOS, down to the No/Yes ordering of dialog buttons. KDE, on the other hand, is a much more Windows-like interface.
Further, most everyone has used a computer at some point, but that doesn't mean they're comfortable with it. Other posters have pointed out that the spatial metaphor was the only way their clueless-whoever could finally keep track of files. Sometimes, it's not learning a new way so much as realizing the old way of doing something was broken or not up to snuff compared to the new way.
There's a reason I picked the XP start menu as an example. You're absolutely correct, the old layout was logical, and it was efficient - for its time. But let's make a few vital comparisons here. Whereas to get to apps you use often in the old menu would require you to either clutter your desktop with icons (which would require the extra step of minimizing open windows to get to) or digging through Start-->Programs-->Whatever. With the new menu, Windows tracks what you use most often and it's always with in 2 clicks for commonly used apps - Start --> App.
One step further, what if it's not one of your more common apps and you still have to dig through Programs? With the old menu, you had to hit start, and then move your mouse up a few inches, over other options. With the new style, you click start, and the "All Programs" option is at the bottom of the list, meaning you only have to move up a minimal distance, and that the option will *always* be located the same distance from the first click, which leads to a more consistant user experience.
And consistency is the key to what makes a good user interface. I absolutely agree that people shouldn't have to learn to use a new interface - an interface should be well-designed enough that it just makes sense to the user with at most a quick, simple explanation. The problem is, we're all so used to the inconsistent, illogical bad habits of the interfaces that we have now that we *expect* them when we deal with something new. Sometimes it takes a bit of unlearning to realize that a new way of doing something may in fact be better even if we've already got another approach engrained into our heads.
For the eleventy-billionth time, what's so hard about middle clicking? Or setting Nautilus to revert back to the old behaviour?
Either one of these addresses this single point that everyone seems to be bashing 2.6 for, all the while ignoring that Gnome 2.6 is actually a decent amount faster than 2.4 ever was.
Myself, I like the spatial concept now, even though *at first* I was rather turned off by it. The key was I spent a few days using it and trying to understand it as a concept. Once you "get it," spatial is a great way to work, which does *not* require "new windows everywhere for everything" you open.
I think the main reason for these complaints is the fact that geeks spend 5 mins playing with the new Nautilus and then dismiss it because it's different from what they're used to. Same idea - how many geeks that work with XP instantly switch their setup to the classic style start menu after about 30 seconds, even though the new style start menu is laid out in a much more efficient and logical manner?
In the same vein, is there any way to get middle-click on a tab to close it under Linux? This is the one thing that drives me insane when working on one of my Linux machines, as it pastes the contents of your clipboard to the URL toolbar instead of closing the tab.
Any about:config hacks I need to know about to work around this behaviour?
My answer? Slackware 9.1 with a 2.6 kernel and Dropline Gnome 2.6. Todd at Dropline does a superb job keeping packages up to date, and Slack has just worked on pretty much any machine I've tried to put it on. It lacks some of the GUI admin tools of FC, SuSe, or MDK, but I like it and have found it to be solid for desktop use.
Good god man, you must consider a 15" 640x480 to be "good"? 1280x1024 is *lowest* I can stand using on a 17" monitor, and really prefer having 1600x1200. Some of us really need the screen real estate and have eyes that are good enough that it's not a problem. Just because you need a low res to be able to see doesn't mean it'll work for everyone.:)
From the article: "...MPEG 2 MPEG 4 WMV9 DiVX decoding and encoding, scaling, frame rate conversion, and anything else you'd like it to do for you..."
Both ATI and nVidia have had hardware decoding standard since the TNT2 days - it's the ability to do encoding and basically turn the GPU into a free DSP unit which is new.
Actually, the GeForce 6800 includes the hardware to do just that. I'm surprised no one else has mentioned it by now, as I thought it was one of the cooler features of the new NV40 chipset.
You have to keep in mind that recommended and minimum specs are two very different things. The average Windows XP machine, for example is in the 1GHz+ range. That said, I've personally run it on a PII 400 with no problems, it's just a matter of turning a few things off.
I imagine Longhorn will run fine on what is currently commodity hardware, but to get what MS considers the "optimal experience," it will take a machine about in line with those specs. Saying an OS needs 4-6GHz sounds like a lot now, but at the same time, saying one would need at least 1GHz to really be effective would have seemed just as ridiculous in 1999 when that PII was built.
Seems you misunderstood the nature of my post. I'm not a fan of pretty much anything that Longhorn will offer, and I'm a very big advocate of Linux and Free Software.
That said, the point I was making is the OP's presumption that neither him nor anyone he knows are currently interested in some of the computationally intensive tasks that a Longhorn-spec machine would be capable of is ridiculous. People aren't interested because it's currently difficult or time consuming. Forward moving tech will help to remedy that.
There's nothing in the Longhorn spec, other than possibly the 1TB HDD, which would be out of line for a new system around the time Longhorn is released. Dual core CPUs should show up in the next few month, 4GHz+ is almost a given within the next year, and with any luck, RAM will get cheaper as well. There's no reason why this would be anything more than your basic ordered-from-Dell system in a couple of years. Once people have that kind of power, they'll be willing to put it to use. That was the point I was trying to make.
1975: How many people do you know that care about writing papers on a computer? I don't, and know one I do does. Typewriters serve that purpose fine.
When the technology because cheap and readily available, people will be interested. Simply dismissing it because that isn't the case now is just silly.
The original poster is technically wrong, but as far as I'm concerned, in spirit they're about right. Comparing the ATI and nVidia Linux drivers is an absolute joke. With nVidia, you download a simple shell script, it checks for a precompiled module for your kernel, and if it doesn't find one, it builds one for you and installs it. After that, you chance *one* line in your XFree86 config file and you're done.
ATI, on the other hand, was a complete nightmare the last time I installed their drivers on a Linux box for someone. I'm fairly proficient in Linux, and he was running Slackware which is the distro I run myself day in and out. It still took us a couple of hours of playing around to get the drivers working properly due to a combination of quirky behavior and EXTREMELY poor documentation. I wouldn't mind doing it all manually, as long as the documentation is clear and concise and helps you get things done in a reasonable amount of time.
Personally, I do keep a Windows box around for gaming, but the parts from this get hand-me-downed to the Linux machines as I upgrade. For that reason, Linux drivers are important to me, and I'll be buying nVidia next time I upgrade. I can deal with spending 5 minutes on a shell script and a reboot to upgrade my video card - I can't handle 2 hours to do the same thing with an ATI card.
It's possible, especially given the 276 number, but it's actually a gentlemen's agreement between the Japanese car manufacturer's that has been pretty more or less ignored for the better part of the last decade. The twin turbo Supra MkIV, for example, was factory rated at 276hp because of this, but it's pretty well known that the car was actually putting out closer to 325hp from the factory.
Either way, I still make the distinction because even if the horsepower numbers are close in reality, the EJ25 puts out a good bit more torque in a smoother curve than the EJ20. The international STi is running somewhere around 17 PSI of boost, whereas the US STi is running only around 14.5 PSI to generate the same horsepower.
For the rally purist, the international STi is closer to the real deal, but the US version of the STi is definitely putting more raw power to the ground.:)
Like I said, it depends on where you are in the world. Most of the world *does* use the EJ20 for both, but in the US, the STi uses the higher displacement EJ25, generating a little bit more horsepower and a good bit more torque. Hence why I asked.:)
Slight nitpick, but is it using the engine from a WRX or an STi? They're essentially the same car, but the WRX puts out around 230 horsepower whereas the STi puts out either 276 or 300, depending on where in the world you are.
This is a really interesting project either way, neat hack.
Because it's absolutely ridiculous to have to go to a data sheet to install a freakin display? I'm a huge Linux advocate, 3 of my 4 machines run Linux, I like the way the OS works in general. But needing to tell XFree86 low level monitor specs has always annoyed the hell out of me.
I can generally blaze through the install of most Linux distros, but always end up burning 5-10 minutes looking up the blasted vertical refresh rate and other nonsense for my monitor. There's something wrong here.
Keep in mind that much of the ice caps are located on top of land masses - meaning that when they melt, while ice is less dense than water, you're adding more water to the oceans than was there in the first place.
Except now you're talking in the range of $4k/unit. I love Panasonic's notebooks as they're near impossible to kill, but money is one thing not readily available in education.
I already posted on this, but it's worth repeating again - Ft. Benning is HUGE. It takes up an entire country in GA. The fact is, even with several of these blimps, I doubt they'd be able to cover more than a fraction of the base's total landmass. Benning is mostly trees and forest, so it's not like we're talking about being in the middle of a major urban center, even though it does neighbor Columbus, GA (city of about 275,000).
I'm as big of a privacy advocate as anyone, but I see nothing disturbing about this, knowing the area they're talking about using this thing in.
Except it wasn't a surveilance balloon. The Japanese designed a number of balloons loaded with explosives, the idea being to release them into the gulf stream, which would carry them to the continental US. The one you mentioned was the only one that managed to do any damage, but that was just because the Japanese failed to correctly estimate some of the navagational factors involved.
I live in Columbus, GA, which is basically the city Ft. Benning is located directly next to. Benning kids go to Columbus schools, I have many friends that live on post, etc.
A little background - Ft. Benning is the largest permanent military installation anywhere in the world, taking up a full Georgia county. It's the home of the Infantry, and if you know anyone who's gone through jump training, regardless of branch of the military, they probably came through Benning. The place is huge, and is covered with trees, the majority of the land being undeveloped and used for firing ranges, survival training, etc.
I can see why they'd be interested in something like this; Benning is literally just too big to be fenced in like most military bases in this country. Depending on how much this costs to operate, it could be a very efficient way to keep the base at least a little bit more secure, which I see as a good thing.
Just thought I'd chip in with a local perspective.:)
Slashdot Mirror
Please god no. The thing about that article that got me interested in Inkscape is the fact that the developers, and more importantly, UI designers are artists themselves, so clearly they want to build a UI that's focused around getting work done, and it sounds like they're doing a good job of that.
While I keep reading that the Gimp's interface was greatly improved with 2.0, when I've tried it, it felt as kludgy as ever. The Gimp does a lot of cool things, but create a smooth workflow it does not. For that reason alone, I feel it's better that this be a stand alone project. It allows them to build a much lighter system aimed at doing one thing and one thing well.
In general, if you're working with vector graphics, you're not really going to care about immediately working with raster. That said, I do think it'd be cool if someone could take the Gimp and strip it down to a very focused UI like Inkscape seems to be doing, creating a set of interlocking common programs like Adobe currently does with their Creative Suite. However, for this type of work, the plug-in-replacing-an-app mentallity is exactly what needs to be avoided because while it may work, an artist will usually be much happier with a lighter program aimed at doing what they want it to do, not ten thousand features they'll never need creating a cluttered and confusing menu system and obscure keyboard shortcuts.
I'd be one of them, thanks in huge part to Dropline Gnome. Excellent set of packages complete with a built in updater that lets me run a light, clean Slackware install, but not have to worry about updating and compiling huge packages - ie: Gnome or X.org myself. Todd does an awesome job keeping it up to date, and this combined with a 2.6 kernel (which drops right in on a default Slack install) make Slackware a really nice little distribution to use. But who am I to say? I'm only running it on 3 machines at the moment. ;)
You mean the Admin Tools menu which can be added to the start menu such that it takes 2 clicks to reach? Start --> Admin Tools.
Thank you, try again.
Personally, I'd take issue with the idea that the Gnome project is aimed at Windows users. If anything, it's aimed at Mac users, as many of its design cues come directly from Classic MacOS, down to the No/Yes ordering of dialog buttons. KDE, on the other hand, is a much more Windows-like interface.
Further, most everyone has used a computer at some point, but that doesn't mean they're comfortable with it. Other posters have pointed out that the spatial metaphor was the only way their clueless-whoever could finally keep track of files. Sometimes, it's not learning a new way so much as realizing the old way of doing something was broken or not up to snuff compared to the new way.
There's a reason I picked the XP start menu as an example. You're absolutely correct, the old layout was logical, and it was efficient - for its time. But let's make a few vital comparisons here. Whereas to get to apps you use often in the old menu would require you to either clutter your desktop with icons (which would require the extra step of minimizing open windows to get to) or digging through Start-->Programs-->Whatever. With the new menu, Windows tracks what you use most often and it's always with in 2 clicks for commonly used apps - Start --> App.
One step further, what if it's not one of your more common apps and you still have to dig through Programs? With the old menu, you had to hit start, and then move your mouse up a few inches, over other options. With the new style, you click start, and the "All Programs" option is at the bottom of the list, meaning you only have to move up a minimal distance, and that the option will *always* be located the same distance from the first click, which leads to a more consistant user experience.
And consistency is the key to what makes a good user interface. I absolutely agree that people shouldn't have to learn to use a new interface - an interface should be well-designed enough that it just makes sense to the user with at most a quick, simple explanation. The problem is, we're all so used to the inconsistent, illogical bad habits of the interfaces that we have now that we *expect* them when we deal with something new. Sometimes it takes a bit of unlearning to realize that a new way of doing something may in fact be better even if we've already got another approach engrained into our heads.
For the eleventy-billionth time, what's so hard about middle clicking? Or setting Nautilus to revert back to the old behaviour?
Either one of these addresses this single point that everyone seems to be bashing 2.6 for, all the while ignoring that Gnome 2.6 is actually a decent amount faster than 2.4 ever was.
Myself, I like the spatial concept now, even though *at first* I was rather turned off by it. The key was I spent a few days using it and trying to understand it as a concept. Once you "get it," spatial is a great way to work, which does *not* require "new windows everywhere for everything" you open.
I think the main reason for these complaints is the fact that geeks spend 5 mins playing with the new Nautilus and then dismiss it because it's different from what they're used to. Same idea - how many geeks that work with XP instantly switch their setup to the classic style start menu after about 30 seconds, even though the new style start menu is laid out in a much more efficient and logical manner?
In the same vein, is there any way to get middle-click on a tab to close it under Linux? This is the one thing that drives me insane when working on one of my Linux machines, as it pastes the contents of your clipboard to the URL toolbar instead of closing the tab.
Any about:config hacks I need to know about to work around this behaviour?
My answer? Slackware 9.1 with a 2.6 kernel and Dropline Gnome 2.6. Todd at Dropline does a superb job keeping packages up to date, and Slack has just worked on pretty much any machine I've tried to put it on. It lacks some of the GUI admin tools of FC, SuSe, or MDK, but I like it and have found it to be solid for desktop use.
Good god man, you must consider a 15" 640x480 to be "good"? 1280x1024 is *lowest* I can stand using on a 17" monitor, and really prefer having 1600x1200. Some of us really need the screen real estate and have eyes that are good enough that it's not a problem. Just because you need a low res to be able to see doesn't mean it'll work for everyone. :)
It's probably an XFree86 configuration problem moreso than kernel. Make sure you have this line in the pointer stanza of /etc/X11/xf86config
Option "ZAxisMapping" "4 5"
If it's not there, add it, restart X, and it should work. HTH.
From the article:
"...MPEG 2 MPEG 4 WMV9 DiVX decoding and encoding, scaling, frame rate conversion, and anything else you'd like it to do for you..."
Both ATI and nVidia have had hardware decoding standard since the TNT2 days - it's the ability to do encoding and basically turn the GPU into a free DSP unit which is new.
Actually, the GeForce 6800 includes the hardware to do just that. I'm surprised no one else has mentioned it by now, as I thought it was one of the cooler features of the new NV40 chipset.
Given most of the software you named wasn't really around prior to Windows 95, I seriously doubt it. You're grasping at straws here.
You have to keep in mind that recommended and minimum specs are two very different things. The average Windows XP machine, for example is in the 1GHz+ range. That said, I've personally run it on a PII 400 with no problems, it's just a matter of turning a few things off.
I imagine Longhorn will run fine on what is currently commodity hardware, but to get what MS considers the "optimal experience," it will take a machine about in line with those specs. Saying an OS needs 4-6GHz sounds like a lot now, but at the same time, saying one would need at least 1GHz to really be effective would have seemed just as ridiculous in 1999 when that PII was built.
Seems you misunderstood the nature of my post. I'm not a fan of pretty much anything that Longhorn will offer, and I'm a very big advocate of Linux and Free Software.
That said, the point I was making is the OP's presumption that neither him nor anyone he knows are currently interested in some of the computationally intensive tasks that a Longhorn-spec machine would be capable of is ridiculous. People aren't interested because it's currently difficult or time consuming. Forward moving tech will help to remedy that.
There's nothing in the Longhorn spec, other than possibly the 1TB HDD, which would be out of line for a new system around the time Longhorn is released. Dual core CPUs should show up in the next few month, 4GHz+ is almost a given within the next year, and with any luck, RAM will get cheaper as well. There's no reason why this would be anything more than your basic ordered-from-Dell system in a couple of years. Once people have that kind of power, they'll be willing to put it to use. That was the point I was trying to make.
1975: How many people do you know that care about writing papers on a computer? I don't, and know one I do does. Typewriters serve that purpose fine.
When the technology because cheap and readily available, people will be interested. Simply dismissing it because that isn't the case now is just silly.
The original poster is technically wrong, but as far as I'm concerned, in spirit they're about right. Comparing the ATI and nVidia Linux drivers is an absolute joke. With nVidia, you download a simple shell script, it checks for a precompiled module for your kernel, and if it doesn't find one, it builds one for you and installs it. After that, you chance *one* line in your XFree86 config file and you're done.
ATI, on the other hand, was a complete nightmare the last time I installed their drivers on a Linux box for someone. I'm fairly proficient in Linux, and he was running Slackware which is the distro I run myself day in and out. It still took us a couple of hours of playing around to get the drivers working properly due to a combination of quirky behavior and EXTREMELY poor documentation. I wouldn't mind doing it all manually, as long as the documentation is clear and concise and helps you get things done in a reasonable amount of time.
Personally, I do keep a Windows box around for gaming, but the parts from this get hand-me-downed to the Linux machines as I upgrade. For that reason, Linux drivers are important to me, and I'll be buying nVidia next time I upgrade. I can deal with spending 5 minutes on a shell script and a reboot to upgrade my video card - I can't handle 2 hours to do the same thing with an ATI card.
It's possible, especially given the 276 number, but it's actually a gentlemen's agreement between the Japanese car manufacturer's that has been pretty more or less ignored for the better part of the last decade. The twin turbo Supra MkIV, for example, was factory rated at 276hp because of this, but it's pretty well known that the car was actually putting out closer to 325hp from the factory.
:)
Either way, I still make the distinction because even if the horsepower numbers are close in reality, the EJ25 puts out a good bit more torque in a smoother curve than the EJ20. The international STi is running somewhere around 17 PSI of boost, whereas the US STi is running only around 14.5 PSI to generate the same horsepower.
For the rally purist, the international STi is closer to the real deal, but the US version of the STi is definitely putting more raw power to the ground.
Like I said, it depends on where you are in the world. Most of the world *does* use the EJ20 for both, but in the US, the STi uses the higher displacement EJ25, generating a little bit more horsepower and a good bit more torque. Hence why I asked. :)
Slight nitpick, but is it using the engine from a WRX or an STi? They're essentially the same car, but the WRX puts out around 230 horsepower whereas the STi puts out either 276 or 300, depending on where in the world you are.
This is a really interesting project either way, neat hack.
Because it's absolutely ridiculous to have to go to a data sheet to install a freakin display? I'm a huge Linux advocate, 3 of my 4 machines run Linux, I like the way the OS works in general. But needing to tell XFree86 low level monitor specs has always annoyed the hell out of me.
I can generally blaze through the install of most Linux distros, but always end up burning 5-10 minutes looking up the blasted vertical refresh rate and other nonsense for my monitor. There's something wrong here.
Keep in mind that much of the ice caps are located on top of land masses - meaning that when they melt, while ice is less dense than water, you're adding more water to the oceans than was there in the first place.
Except now you're talking in the range of $4k/unit. I love Panasonic's notebooks as they're near impossible to kill, but money is one thing not readily available in education.
I already posted on this, but it's worth repeating again - Ft. Benning is HUGE. It takes up an entire country in GA. The fact is, even with several of these blimps, I doubt they'd be able to cover more than a fraction of the base's total landmass. Benning is mostly trees and forest, so it's not like we're talking about being in the middle of a major urban center, even though it does neighbor Columbus, GA (city of about 275,000).
I'm as big of a privacy advocate as anyone, but I see nothing disturbing about this, knowing the area they're talking about using this thing in.
Except it wasn't a surveilance balloon. The Japanese designed a number of balloons loaded with explosives, the idea being to release them into the gulf stream, which would carry them to the continental US. The one you mentioned was the only one that managed to do any damage, but that was just because the Japanese failed to correctly estimate some of the navagational factors involved.
I live in Columbus, GA, which is basically the city Ft. Benning is located directly next to. Benning kids go to Columbus schools, I have many friends that live on post, etc.
:)
A little background - Ft. Benning is the largest permanent military installation anywhere in the world, taking up a full Georgia county. It's the home of the Infantry, and if you know anyone who's gone through jump training, regardless of branch of the military, they probably came through Benning. The place is huge, and is covered with trees, the majority of the land being undeveloped and used for firing ranges, survival training, etc.
I can see why they'd be interested in something like this; Benning is literally just too big to be fenced in like most military bases in this country. Depending on how much this costs to operate, it could be a very efficient way to keep the base at least a little bit more secure, which I see as a good thing.
Just thought I'd chip in with a local perspective.