To some extent, it is logical why higher contract bids are preferred. This is generally taken as an indication that the contractor is going to be able to offer a certain level of service that a low-bidder may not be able to offer. I liked the example of the $1 coffee -vs- the $3 coffee.
Unfortunately, software development organizations also undertake similiar endeavors. Rather than leverage existing technologies, a development organization is generally content over-developing (and thus over-spending) to meet its goals. For some reason, the more LOC's written the greater the "quality" of the product.
So because it has more features that means it's "better"? That's a pretty lousy definition of better, IMHO.
My real complaint is the characterization that Vim is popular. There is nothing about Vim that anyone here uses that isn't part of vi. If it is so popular why are we scrambling to deconfigure this thing to make it behave like vi? We obviously don't want these features, so let's not call vim popular when it really is not.
OTOH, all of this wrecks havoc when you go to a box with a vim that's not compatible.
For my money, the original posters suggestion to remove system settings for VIM is the money-shot. Whenever possible I try to install a true-blue vi myself, can't stand any of the vim features; I've got nedit, jEdit and Emacs if I really need anything fancy.
TBPH, I don't think vim is "popular" at all. An informal survey of linux users here at work indicates that no one likes the "features" of vim and would prefer a vanilla vi.
Imagine if universities got into deals with book publishers that said they were going to use only books they published for educating students. Granted that some professors like to push books they have written, but a deal with a publisher would result in an extremely spotty curriculum and would ultimately produce a rather narrow education.
It's not so bad that C# is being taught, what is more devistating here is students will end up having to purchase Microsoft software to do their work. Academia has traditionally tended towards heterogenious computing environments (mainframe, UNIX, MS Windows, Mac, etc.), but with deals that force programming to happen with.NET and C# the students (and therefore ultimately faculty and staff) will be forced to gravitate towards this platform. While companies like Sun are entrenched in academia, at least a C/C++ compiler can run on NT, Solaris, Irix, HP/UX, Linux, etc. etc. etc.
Universities should wait until C# proves itself commercially and academically. There are plenty of language implementations out there, schools should be focused on those languages that teach programming principals and that have demonstrated themselves as historically significant.
You know what, when you reboot {n} times, that's {n} times that something can go wrong. Think about it from a users perspective; your machine is now rebooting with very little real feedback other than a notice that this is something Windows needs to do.
Now what? Do you remove the CD from the drive? Should you insert another CD? Maybe Windows needs a floppy to get the system started again... but you opted to skip that step.
It's silly. There's no reason for upteen zillion reboots when a modern operating system like Linux can perform the same task in a single step.
Another interesting quandry for Windows users is what to do when their media stops working. My Win98 CD no longer functions properly, can I just send this to Redmond and get a new copy in the mail? Seriously, what happens when MS stops supports W2K and your original media has a defect. You're not left with many options; once again, open source proves itself.
When your RedHat CD stops working, you can actually download and burn a new copy to get your install going again as quickly as possible. Try that with a W2K coaster. TBPH, I haven't yet gotten a RH coaster from any copy purchased; my only coasters have resulted from a flaky CDRW drive.
Incidentally, the same can be said for any new Windows CD. While my older discs are getting flaky, new copies of Win98, WinNT, W2K and XP have all been flawless - I've only ever seen people have problems with pirated copies of Windows.
Given the volatility of the ISP market, a national email infrastructure would have been a wonderful thing. You could maintain a permanent email with the US Post Office and not have to worry about what might happen to your address if your service provider should change.
Imagine not having to worry about @mediaone.com suddenly not working for you. Just about every major provider has undergone a substantial shift in how they process emails, resulting in everything from new domain names to new mail accounts. I can't tell you how many people I can no longer find @compuserve.com.
paradigm shift or natural progression
on
GUIs for Everyone
·
· Score: 1
I think there is a natural rift between CLI users and GUI users.
The GUI user tends to be an end-user who is looking for simplicity. This is typically someone who has a workstation and wants to run a few programs, like a wordprocessor and an email client, and doesn't want to be hassled with learning the commands needed to make these run. The CLI user, on the other hand, tends to want to glue various commands together to perform operations that would normally be tedious in a GUI.
These modes of operation do not compliment each other. It's like the difference between riding a motorcycle and riding a sports car, while they both accomplish the same thing you won't ever be able to have a sports car that's also a motorcycle.
I think the original author was trying to say something more significant than CLI -vs- GUI. Perhaps there is a more natural interface paradigm for the personal computer, using voice or maybe even thought to manage tasks on your PC. Imagine not typing at all, but strapping on a helmet and thinking your way through all of your tasks.
Ironically, computer programmers would be least-likely to enjoy any of these newfangled interfaces.
At any rate, once any product reaches a commodity status it is very difficult (read: nearly impossible) to provide consumers with any improvements on that design. There's a lot more we can do with the automobile, for example, but most consumers don't really care because they have a car that works exactly the way they have become accustom to it working.
Computers are the same way. As another poster pointed out, his grandmother switched WinXP back to the old Windows95 look. Why? Because she has become accustom to that interface.
For better or for worse, most people have seen Windows and expect a graphical computer to look that way. Breaking that with something as significant as a paradigm shift will be almost impossible. Complimenting that with a very simliar UI (aka: KDE) is much more plausible.
I think the open source community is right on target with mimicing existing windowing environments. Everybody copies from everybody else anyway, while we are in need of some UI guidelines to help maintain consistancy between applications on Linux it is at least some solace that Microsoft doesn't abide by its own UI standards anyway.
Somehow I doubt the commercial software establishment will be eager to adopt this kind of sales strategy. For that matter, why would academic / non-profit organizations have any insentive to buy MS software to begin with.
To some extent, the real problem is the version of Word that arrived on one machine is incompatible with the version that arrived on another. For that matter, the entire OS is incompatible between systems and these institutions are being forced to upgrade for even the most basic computing needs. The commercial software establishment has created this problem, I don't think piracy is the solution but there needs to be some leaniency with how licenses get managed by schools or non-profit organizations.
Perhaps a more realistic solution is to provide these user groups with a truly "open license" that presents them with a fixed cost to maintain all software in their organization. The granularity should not be at a per-seat level, but rather more like small / medium and large schools. Private schools should be treated differently than state or public institutions, and non-profit organizations should get a fixed-rate. This would get sent to the US government, which in turn would provide commercial software companies with tax incentives and/or cash infusions.
Of course, given the complexity of implementing any reasonable licensing scheme with commercial software, it's ultimately going to be easier to adopt open-source solutions like Linux.
I'm not convinced the user really gets a choice here.
My gaming system got nuked by lightning last night. It was a cheap box with a cheap UPS and Win98 (it wasn't even turned on!). At any rate, while shopping for a computer the popular vendors (Dell & GW2K) offer pre-installed XP with either MS Works or a varient of MS Office XP.
No option. Can't opt-out of the Office or the XP install. What kind of *choice* is that? I'm not surprised that a Microsoft os would be mandetory, but I am surprised that Office XP is pretty much a pre-requisit as well.
I've already bought XP and don't want it, my copy of Win98 is perfect for this system.
To make matters worse, the only option seems to be XP or W2K Pro and I *know* that my games won't work with either of this. Why would I want to pay for something that I know won't work for me?
The fact is, most consumers end up doing this and this perpetuates the "popularity" of Microsoft applications. While they certainly write some neat looking programs, I can think of quite a few apps offhand (WordPerfect, 123, Harvard Graphics, Ami Pro, Lotus Notes,....) that all did a better job than any contemporary MS app.
Let's face it, MS won by cheating. There's nothing wrong with that, but let's not say that consumer had a choice in the matter.
Rasterman is right. Windows has won the desktop on the Intel i386 platform. It will be interesting to see if they can also win the IA64 platform, they are losing on PowerPC and the IA32 is showing its age. I believe MS will find it impossible to get beyond the i386, consumers don't want to repurchase software again and closed-source solutions make this a requirement.
So it'll cost me $350 for 1GB of online storage? Last time I checked, 1GB was selling for about $2. Even after factoring in the cost of putting this online (say $50 a month for high speed Internet access), I'm thinking it's much cheaper to do this yourself.
The email is another story altogether. It would be nice to have a reliable lifetime email address, even if it means paying a small amount every year to keep it. While there are some niche players out there, and some nice freebies, I think an email account with solid corporate backing would make a lot of users happy.
It doesn't look like the new mac.com will offer a good email solution, given that it includes a lot of extra overpriced baggage already.
I have to agree; while there's a lot you can do with X the average Windows user simply won't be willing to accomidate the complexities of this.
We need a lot more than a X config wizard, we need something to replace X; IMHO. And it needs to be consistent, with a clipboard that makes sense, AA fonts, a UI that follows conventions between applications, high-performance video support (maybe even via a kernel module), comprehensive multimedia support, and a visual development tool to tie everything together.
KDE and Gnome are band-aids, they cover-up X with a fancy UI. If we had a new windowing environment, legacy X apps would be unable to run and new apps would be forced to conform to the UI standards.
Slashdot had a posting on X11 alternatives not too long ago; unfortunately, in that list nothing really came to light. There are a few projects out there, but none of them are trul mature yet.
Maybe we could latch on to the Mac OS X display layer somehow? I don't know much about that, other than the fact that it is possible to shoe-horn X apps onto Mac OS X... I'm not entirely convinced that is a desirable feature.
That's pretty funny; somebody give this a few points.
Seriously, as others have already alluded to, I think advertisements in open source software are a lost cause. With the source code open, developers can simply go in and remove the ads -- then before you know it there is a new distribution without ads... and we're right back where we started from.
RedHat has the right approach, IMHO (as does IBM, Microsoft, Sun, HP...) You charge for premium support services, customers will take a free product as long as they know they can pay someone for support. There's no reason we can't have successful companies supporting Linux while contributing to the open source development of the kernel.
While you can certainly change to a metric timescale for the sake of consistancy, it can't be consistant with itself so it really makes no sense.
Unlike other non-metric measurement systems, time has scientific meaning. We are really more concerned with when the sun comes up, when it is the middle of the day and when the sun goes down again. Metric time could certainly be used to fill in the intervals inbetween, but it will never replace the yearly cycle of the earth in its orbit around the sun.
A better unit of time would simply be the day. Forget hours in the day. Obviously this doesn't have enough granularity for most people, so we need to slice the day up a little.
Well, this slicing is based on where the sun is in the sky -- which makes total sense to me. A metric scale would only confuse things.
That almost makes sense; too bad the SFU is yet another example of a great set of utilities that Microsoft didn't write.
At any rate, if this really is a showcase for XP embedded there better be some kind of Linux tie-in. Maybe show how you can develop XP-embedded apps on Linux? If they don't demonstrate some kind of capability on Linux, they have no place at the show.
Aside from the fact that the Palladium endeavor is monopolistic, it does suggest the possibility that the os could be hardened to protect users further from unauthorized code.
What about digitally signed apps? When the kernel is built it embeds a private key into itself that is used to unlock apps that are compiled for this kernel. Compiles can optionally require another password in order to futher secure the process.
A program must then be built for this kernel on the same system it will be used, and the public key is embedded into the binary. Legacy apps could still run, but they will prompt before execution indicating they may damage your system.
While this won't stop a Perl script from wrecking havoc, it will limit the kinds of trojans that can be dropped onto your system. Maybe scripted code requires a password to execute?
Hm. Just a random thought; Gentoo would be ideal for this. OTOH, maybe something like this already exists?
FWIW, I've never really liked the MS Media Player since it was overhauled (was that v7?). It's big & bloated now, it's impossible to figure out how to just "Play" a CD without making a copy of it, and as far as using your DVD's it's faster to reformat, install Linux and fire-up Ogle.
The UI has definitely been "dumbed-down", there are very few options that can be configured. For neophite users this will make Gnome much more approachable, but more seasoned users will quickly grow frustrated by the limited configuration options.
It's not stable yet, either. Inside of my first 15 minutes with it this version of Gnome has crashed 3 times. I'll need to use it more to understand what it was that might have made it become unresponsive.
The default config is very Mac-like, but ironically this is after the Mac has already moved on to a new UI. Sun will probably be retisent to adopting anything that remotely resembles OpenWin, so this notalgic Mac look for Gnome 2 is likely to persist.
Everything looks much more professional. Obviously the anti-aliased fonts are a win-win, and the Gnome Panel has been reworked significantly.
I'm not at all happy with Nautilus. In contrast with Konqueror, there are too many basic file operations missing. For example, right-click on any file in Konqueror and you can pick anywhere on your computer to copy / move the file. In Nautilus, you're forced to use the sluggish copy/paste methodology. While certainly much more impressive, I think the Konqueror solution is faster in most cases.
While I enjoy by bubble CPU monitor, I'm not sure this is going to be enough to keep me with Gnome. Version 2.0 isn't looking significantly better, so I may find myself back in KDE soon.
Remember when railroad companies used different sized track to eliminate competition? They also eliminated any semblance of travel convenience for the consumer.
The commercial software establishment is largely like these now non-existant railroad franchises. People have discovered that it's just software, and they are happier to enjoy a level of compatibility accross a variety of systems.
Of course, once the railroad industries agreed on standards it became possible for mass production of standards-based railway hardware. This eliminated much of the guesswork, tracks were wide enough to support trains of various sizes and shapes. Without these standards, the golden age of travel would have been unachievable.
Software needs to adopt standards, and the open source community has been vital to that process.
Given Microsoft's track record in this area, I think the best outcome now is for the judge to force MS to abide by standards for all present and future networking protocols. If a networked feature of MS software does not employ a documented RFC, W3C recommendation, etc. it must be fixed.
And there should be a federal committee responsible for reviewing and enforcing this. It is not acceptable that standards can be implemented along with a proprietary MS protocol (eg: MS Exchange).
Ultimately, all commercial software should be made to follow these rules, only the open source community will be allowed to innovate networking protocols. Most of the significant protocols came from open source / public domain anyway, let's mandate that tradition and stop companies like Microsoft from meddling with a good thing.
Agreed. That was my first impression, this legacy JVM lacks many of the API's that are part of the Java 2.0 specification. To make matters worse, MS included a variety of native calls to enable COM/COM+ integration with Java so non-MS users are bound to experience errors.
The result? Windows users will grow frustrated by the MS legacy JVM incompatibilities and will migrate towards the MS ActiveX "standard". This is the worst thing that could happen to Java in the browser, we were better off when MS had stopped supporting the format.
Java applications will suffer as well. Users with the braindead MS implementation will find themselves unable to run most modern Java code. In most cases, when it doesn't work users assume a format is broken.
Even with the monthly TiVo membership fee, once you add the cost of VCR cleanings, tapes, missed episodes, time required to program the device, etc. you're likely to be under the cost of a VCR in the long run.
Slashdot Mirror
To some extent, it is logical why higher contract bids are preferred. This is generally taken as an indication that the contractor is going to be able to offer a certain level of service that a low-bidder may not be able to offer. I liked the example of the $1 coffee -vs- the $3 coffee.
Unfortunately, software development organizations also undertake similiar endeavors. Rather than leverage existing technologies, a development organization is generally content over-developing (and thus over-spending) to meet its goals. For some reason, the more LOC's written the greater the "quality" of the product.
So because it has more features that means it's "better"? That's a pretty lousy definition of better, IMHO.
My real complaint is the characterization that Vim is popular. There is nothing about Vim that anyone here uses that isn't part of vi. If it is so popular why are we scrambling to deconfigure this thing to make it behave like vi? We obviously don't want these features, so let's not call vim popular when it really is not.
You can even put that into your .vimrc.
OTOH, all of this wrecks havoc when you go to a box with a vim that's not compatible.
For my money, the original posters suggestion to remove system settings for VIM is the money-shot. Whenever possible I try to install a true-blue vi myself, can't stand any of the vim features; I've got nedit, jEdit and Emacs if I really need anything fancy.
TBPH, I don't think vim is "popular" at all. An informal survey of linux users here at work indicates that no one likes the "features" of vim and would prefer a vanilla vi.
Imagine if universities got into deals with book publishers that said they were going to use only books they published for educating students. Granted that some professors like to push books they have written, but a deal with a publisher would result in an extremely spotty curriculum and would ultimately produce a rather narrow education.
.NET and C# the students (and therefore ultimately faculty and staff) will be forced to gravitate towards this platform. While companies like Sun are entrenched in academia, at least a C/C++ compiler can run on NT, Solaris, Irix, HP/UX, Linux, etc. etc. etc.
It's not so bad that C# is being taught, what is more devistating here is students will end up having to purchase Microsoft software to do their work. Academia has traditionally tended towards heterogenious computing environments (mainframe, UNIX, MS Windows, Mac, etc.), but with deals that force programming to happen with
Universities should wait until C# proves itself commercially and academically. There are plenty of language implementations out there, schools should be focused on those languages that teach programming principals and that have demonstrated themselves as historically significant.
You know what, when you reboot {n} times, that's {n} times that something can go wrong. Think about it from a users perspective; your machine is now rebooting with very little real feedback other than a notice that this is something Windows needs to do.
... but you opted to skip that step.
Now what? Do you remove the CD from the drive? Should you insert another CD? Maybe Windows needs a floppy to get the system started again
It's silly. There's no reason for upteen zillion reboots when a modern operating system like Linux can perform the same task in a single step.
Another interesting quandry for Windows users is what to do when their media stops working. My Win98 CD no longer functions properly, can I just send this to Redmond and get a new copy in the mail? Seriously, what happens when MS stops supports W2K and your original media has a defect. You're not left with many options; once again, open source proves itself.
When your RedHat CD stops working, you can actually download and burn a new copy to get your install going again as quickly as possible. Try that with a W2K coaster. TBPH, I haven't yet gotten a RH coaster from any copy purchased; my only coasters have resulted from a flaky CDRW drive.
Incidentally, the same can be said for any new Windows CD. While my older discs are getting flaky, new copies of Win98, WinNT, W2K and XP have all been flawless - I've only ever seen people have problems with pirated copies of Windows.
I can't agree more, SSL is really only practical for preventing intermediate parties from sniffing the wire and accumulating data in cleartext.
With that said, how is this attack any different than any other man-in-the-middle attack?
Given the volatility of the ISP market, a national email infrastructure would have been a wonderful thing. You could maintain a permanent email with the US Post Office and not have to worry about what might happen to your address if your service provider should change.
Imagine not having to worry about @mediaone.com suddenly not working for you. Just about every major provider has undergone a substantial shift in how they process emails, resulting in everything from new domain names to new mail accounts. I can't tell you how many people I can no longer find @compuserve.com.
I think there is a natural rift between CLI users and GUI users.
The GUI user tends to be an end-user who is looking for simplicity. This is typically someone who has a workstation and wants to run a few programs, like a wordprocessor and an email client, and doesn't want to be hassled with learning the commands needed to make these run. The CLI user, on the other hand, tends to want to glue various commands together to perform operations that would normally be tedious in a GUI.
These modes of operation do not compliment each other. It's like the difference between riding a motorcycle and riding a sports car, while they both accomplish the same thing you won't ever be able to have a sports car that's also a motorcycle.
I think the original author was trying to say something more significant than CLI -vs- GUI. Perhaps there is a more natural interface paradigm for the personal computer, using voice or maybe even thought to manage tasks on your PC. Imagine not typing at all, but strapping on a helmet and thinking your way through all of your tasks.
Ironically, computer programmers would be least-likely to enjoy any of these newfangled interfaces.
At any rate, once any product reaches a commodity status it is very difficult (read: nearly impossible) to provide consumers with any improvements on that design. There's a lot more we can do with the automobile, for example, but most consumers don't really care because they have a car that works exactly the way they have become accustom to it working.
Computers are the same way. As another poster pointed out, his grandmother switched WinXP back to the old Windows95 look. Why? Because she has become accustom to that interface.
For better or for worse, most people have seen Windows and expect a graphical computer to look that way. Breaking that with something as significant as a paradigm shift will be almost impossible. Complimenting that with a very simliar UI (aka: KDE) is much more plausible.
I think the open source community is right on target with mimicing existing windowing environments. Everybody copies from everybody else anyway, while we are in need of some UI guidelines to help maintain consistancy between applications on Linux it is at least some solace that Microsoft doesn't abide by its own UI standards anyway.
Somehow I doubt the commercial software establishment will be eager to adopt this kind of sales strategy. For that matter, why would academic / non-profit organizations have any insentive to buy MS software to begin with.
To some extent, the real problem is the version of Word that arrived on one machine is incompatible with the version that arrived on another. For that matter, the entire OS is incompatible between systems and these institutions are being forced to upgrade for even the most basic computing needs. The commercial software establishment has created this problem, I don't think piracy is the solution but there needs to be some leaniency with how licenses get managed by schools or non-profit organizations.
Perhaps a more realistic solution is to provide these user groups with a truly "open license" that presents them with a fixed cost to maintain all software in their organization. The granularity should not be at a per-seat level, but rather more like small / medium and large schools. Private schools should be treated differently than state or public institutions, and non-profit organizations should get a fixed-rate. This would get sent to the US government, which in turn would provide commercial software companies with tax incentives and/or cash infusions.
Of course, given the complexity of implementing any reasonable licensing scheme with commercial software, it's ultimately going to be easier to adopt open-source solutions like Linux.
Read the next sentence: "[it] would not include a pre-negotiated debt-for-equity swap as some bondholders had hoped".
Not good. This could spell the end of WorldCom; start shopping for your next service provider or you're a WorldCom customer!
I'm not convinced the user really gets a choice here.
....) that all did a better job than any contemporary MS app.
My gaming system got nuked by lightning last night. It was a cheap box with a cheap UPS and Win98 (it wasn't even turned on!). At any rate, while shopping for a computer the popular vendors (Dell & GW2K) offer pre-installed XP with either MS Works or a varient of MS Office XP.
No option. Can't opt-out of the Office or the XP install. What kind of *choice* is that? I'm not surprised that a Microsoft os would be mandetory, but I am surprised that Office XP is pretty much a pre-requisit as well.
I've already bought XP and don't want it, my copy of Win98 is perfect for this system.
To make matters worse, the only option seems to be XP or W2K Pro and I *know* that my games won't work with either of this. Why would I want to pay for something that I know won't work for me?
The fact is, most consumers end up doing this and this perpetuates the "popularity" of Microsoft applications. While they certainly write some neat looking programs, I can think of quite a few apps offhand (WordPerfect, 123, Harvard Graphics, Ami Pro, Lotus Notes,
Let's face it, MS won by cheating. There's nothing wrong with that, but let's not say that consumer had a choice in the matter.
Rasterman is right. Windows has won the desktop on the Intel i386 platform. It will be interesting to see if they can also win the IA64 platform, they are losing on PowerPC and the IA32 is showing its age. I believe MS will find it impossible to get beyond the i386, consumers don't want to repurchase software again and closed-source solutions make this a requirement.
So it'll cost me $350 for 1GB of online storage? Last time I checked, 1GB was selling for about $2. Even after factoring in the cost of putting this online (say $50 a month for high speed Internet access), I'm thinking it's much cheaper to do this yourself.
The email is another story altogether. It would be nice to have a reliable lifetime email address, even if it means paying a small amount every year to keep it. While there are some niche players out there, and some nice freebies, I think an email account with solid corporate backing would make a lot of users happy.
It doesn't look like the new mac.com will offer a good email solution, given that it includes a lot of extra overpriced baggage already.
I have to agree; while there's a lot you can do with X the average Windows user simply won't be willing to accomidate the complexities of this.
We need a lot more than a X config wizard, we need something to replace X; IMHO. And it needs to be consistent, with a clipboard that makes sense, AA fonts, a UI that follows conventions between applications, high-performance video support (maybe even via a kernel module), comprehensive multimedia support, and a visual development tool to tie everything together.
KDE and Gnome are band-aids, they cover-up X with a fancy UI. If we had a new windowing environment, legacy X apps would be unable to run and new apps would be forced to conform to the UI standards.
Slashdot had a posting on X11 alternatives not too long ago; unfortunately, in that list nothing really came to light. There are a few projects out there, but none of them are trul mature yet.
Maybe we could latch on to the Mac OS X display layer somehow? I don't know much about that, other than the fact that it is possible to shoe-horn X apps onto Mac OS X... I'm not entirely convinced that is a desirable feature.
did anyone else notice the recently-cracked USA Today site switched from Solaris this year?
Hmm, I wonder if they're kicking themselves now.
That's pretty funny; somebody give this a few points.
Seriously, as others have already alluded to, I think advertisements in open source software are a lost cause. With the source code open, developers can simply go in and remove the ads -- then before you know it there is a new distribution without ads... and we're right back where we started from.
RedHat has the right approach, IMHO (as does IBM, Microsoft, Sun, HP...) You charge for premium support services, customers will take a free product as long as they know they can pay someone for support. There's no reason we can't have successful companies supporting Linux while contributing to the open source development of the kernel.
While you can certainly change to a metric timescale for the sake of consistancy, it can't be consistant with itself so it really makes no sense.
Unlike other non-metric measurement systems, time has scientific meaning. We are really more concerned with when the sun comes up, when it is the middle of the day and when the sun goes down again. Metric time could certainly be used to fill in the intervals inbetween, but it will never replace the yearly cycle of the earth in its orbit around the sun.
A better unit of time would simply be the day. Forget hours in the day. Obviously this doesn't have enough granularity for most people, so we need to slice the day up a little.
Well, this slicing is based on where the sun is in the sky -- which makes total sense to me. A metric scale would only confuse things.
That almost makes sense; too bad the SFU is yet another example of a great set of utilities that Microsoft didn't write.
At any rate, if this really is a showcase for XP embedded there better be some kind of Linux tie-in. Maybe show how you can develop XP-embedded apps on Linux? If they don't demonstrate some kind of capability on Linux, they have no place at the show.
Hilarious! Somebody +1 this guy for humour value.
Aside from the fact that the Palladium endeavor is monopolistic, it does suggest the possibility that the os could be hardened to protect users further from unauthorized code.
What about digitally signed apps? When the kernel is built it embeds a private key into itself that is used to unlock apps that are compiled for this kernel. Compiles can optionally require another password in order to futher secure the process.
A program must then be built for this kernel on the same system it will be used, and the public key is embedded into the binary. Legacy apps could still run, but they will prompt before execution indicating they may damage your system.
While this won't stop a Perl script from wrecking havoc, it will limit the kinds of trojans that can be dropped onto your system. Maybe scripted code requires a password to execute?
Hm. Just a random thought; Gentoo would be ideal for this. OTOH, maybe something like this already exists?
FWIW, I've never really liked the MS Media Player since it was overhauled (was that v7?). It's big & bloated now, it's impossible to figure out how to just "Play" a CD without making a copy of it, and as far as using your DVD's it's faster to reformat, install Linux and fire-up Ogle.
The UI has definitely been "dumbed-down", there are very few options that can be configured. For neophite users this will make Gnome much more approachable, but more seasoned users will quickly grow frustrated by the limited configuration options.
It's not stable yet, either. Inside of my first 15 minutes with it this version of Gnome has crashed 3 times. I'll need to use it more to understand what it was that might have made it become unresponsive.
The default config is very Mac-like, but ironically this is after the Mac has already moved on to a new UI. Sun will probably be retisent to adopting anything that remotely resembles OpenWin, so this notalgic Mac look for Gnome 2 is likely to persist.
Everything looks much more professional. Obviously the anti-aliased fonts are a win-win, and the Gnome Panel has been reworked significantly.
I'm not at all happy with Nautilus. In contrast with Konqueror, there are too many basic file operations missing. For example, right-click on any file in Konqueror and you can pick anywhere on your computer to copy / move the file. In Nautilus, you're forced to use the sluggish copy/paste methodology. While certainly much more impressive, I think the Konqueror solution is faster in most cases.
While I enjoy by bubble CPU monitor, I'm not sure this is going to be enough to keep me with Gnome. Version 2.0 isn't looking significantly better, so I may find myself back in KDE soon.
It was part of my high-school history. Granted, that was 20 years ago - but somehow it went TOS.
Remember when railroad companies used different sized track to eliminate competition? They also eliminated any semblance of travel convenience for the consumer.
The commercial software establishment is largely like these now non-existant railroad franchises. People have discovered that it's just software, and they are happier to enjoy a level of compatibility accross a variety of systems.
Of course, once the railroad industries agreed on standards it became possible for mass production of standards-based railway hardware. This eliminated much of the guesswork, tracks were wide enough to support trains of various sizes and shapes. Without these standards, the golden age of travel would have been unachievable.
Software needs to adopt standards, and the open source community has been vital to that process.
Given Microsoft's track record in this area, I think the best outcome now is for the judge to force MS to abide by standards for all present and future networking protocols. If a networked feature of MS software does not employ a documented RFC, W3C recommendation, etc. it must be fixed.
And there should be a federal committee responsible for reviewing and enforcing this. It is not acceptable that standards can be implemented along with a proprietary MS protocol (eg: MS Exchange).
Ultimately, all commercial software should be made to follow these rules, only the open source community will be allowed to innovate networking protocols. Most of the significant protocols came from open source / public domain anyway, let's mandate that tradition and stop companies like Microsoft from meddling with a good thing.
Agreed. That was my first impression, this legacy JVM lacks many of the API's that are part of the Java 2.0 specification. To make matters worse, MS included a variety of native calls to enable COM/COM+ integration with Java so non-MS users are bound to experience errors.
The result? Windows users will grow frustrated by the MS legacy JVM incompatibilities and will migrate towards the MS ActiveX "standard". This is the worst thing that could happen to Java in the browser, we were better off when MS had stopped supporting the format.
Java applications will suffer as well. Users with the braindead MS implementation will find themselves unable to run most modern Java code. In most cases, when it doesn't work users assume a format is broken.
Well, there's always TiVo
http://www.tivo.com
Even with the monthly TiVo membership fee, once you add the cost of VCR cleanings, tapes, missed episodes, time required to program the device, etc. you're likely to be under the cost of a VCR in the long run.