Windows NT/2000 might be a 32-bit OS, but Windows programmers are still 16-bit. Someday far into the future, windows programmers might realize they are no longer coding for DOS, and might even join Linux, mac, and all other programmers, in the 21st century. But until that day arrives, we have to put up with these idiots who can only babble in eight letter words.
If IT workers are smart and their employers are stupid, then the IT workers should start their own companies and, using their superior knowlege, wipe the floor with their former employers. If you can't work with'em, beat'em.
Packaging (and shared code) inevitably causes more problems than it solves. Maybe a few geeks like us would appreciate the better use of resources that shared code gives. But most users (and probably a few geeks as well) would be better served making each program its own seperate entity--One program, one set of code. Some will no doubt decry my advocacy for static linking (I'll make an exception for things like glibc), but take into account that Linux is starting to get into the regular consumer market. The regular joe, home consumer market has a totally different set of rules and goals than the
workplace IT market. And currently, I see the linux movement making a very large mistake by approaching the desktop market with the same exact strategies and objective as the server market. Success and failure in the two markets is very different. On the server, a crash is failure. Some script kiddie rooting the box is failure. On the destkop, a user who, on Christmas morning, getting messages that Barbie Magic Funhouse can't be installed because it conflicts with sendmail and will break dependencies for
Evolution is complete, utter and total, unadulterated failure. And up to this point, the linux community has done nothing but bury its head in the sand and try to rely on the internet to solve problems that static linking could easily solve.
The funny thing about the word "dependency"--it usually comes after the words "co" and "drug".
Seriously, ask 1000 people outside a CompUSA what they would rather have: An OS that uses their memory and disk space more efficiently, or an OS that lets them install as much software as they want without breaking anything and which will never preclude the installation of any other software. 999 of them will go with the latter.
Most people want computers that simply just work. This is The Reality That Is The Desktop.
>What's to stop somebody from developing a Linux >Distro that mirrors OS X? We've been strugling >for years to find the "Linux your mother could >use", and Cupertino has shown it to us.
Nothing, which is why I'm currently working on a linux distribution that incorporates essential ideas of mac interface design. I'm currently rewriting parts of gnome to make the stuff that looks like it was written by the folks in redmond look like the stuff written by the folks in cupertino. Then I'll throw it all into a nice distribution your grandmother can use
>In computing, the contemporary GUI sets aside >the idea of mastery as not only unnecessary but >unattainable.
This is exactly what I believe. The sign of a good GUI is not that a user never exposed to it can immediately master the fundamental basics, rather that mastery of the fundamental basics can be easily applied to the mastery of new programs.
>But the specifics of the spatial metaphor vary >widely. Consider the icons on the Windows >desktop: the Network Neighborhood, My Computer, >Recycle Bin, and any shortcut and file icons >that may be present all have the same basic >appearance -- but they function very >differently. In some applications, dialog boxes >that accomplish file management functions use a >left-to-right hierarchy; others may choose top->to-bottom; still others will make use of both.
The problem with windows is that it makes a lot of exceptions to the desktop metaphor. Move a folder to somewhere else, that's all well and good. Try moving the "control panels" or "dial-up-networking" folder, you're told you can't do it.
Drag a a file to the desktop--no problem. Drag an program to the destop, windows makes a shortcut instead.
>Worse yet, when there is a system problem, and >the structures which supply ease are replaced >with "Unexpected exception in module >Vx00f323.dll" or "Oops. Dumping core . ..," >the "have nots" may find their computers >unusable [13].
Again, another great M$ inconsistency. Objects like "My Computer" are given Plain English names that make sense, but then we take a look at the system files and they're all DOS 8.3 naming convention gobbledygook. Windows is a 32 bit OS, unfortunately windows programmers are still 16-bit. It is intersting to note that the MacOS has a filename limit of 31 characters, but still has system files with far more intuitive names. The example the author gave also illustrates another M$ inconsistency: half the time they try to do everything for the user with their wizards and talking paperclips--and wind up getting in the user's way. The other half of the time they default to their old DOS heritage--and shamelessly and completely abandon the user to a system so arcane, confusing, and erratic that it makes kernel hacking look intuitive.
People say that the desktop metaphor has failed. I say that it is not the desktop metaphor that has failed, but rather the failure of non-mac programmers to adequately understand this metaphor and implement it consistently. Good GUI programmers don't try to do everything for the user, they allow the user to make better, more informed choices in a highly productive manner.
A cool way of hiding forbidden code places where the corporate machine could never get to it would be in mp3s (or other files commonly downloaded and shared). Imagine embedding DeCSS in a Britney Spears MP3 (which would actually give the mp3 some *real* value;) and then putting that in a Napster directory on your machine. Thousands upon thousands of ten year olds would download DeCSS along with the mp3 and put it in their napster directory (And a Britney spears MP3 would finally contain something of value!)Thus, DeCSS would be spread to people who never even knew they had it. While the MPAA might have a few good spiders here and there, for them to try to grab every mp3 on Napster (or Gnutella) and decode it to search for DeCSS would be futile. Especially if napster were absolutely chock full of mp3's containing DeCSS. You could even move beyond mp3s, embedding DeCSS in porn (surely to get distributed). One could write something like an apache module that would embed the forbidden code in a graphic file every time the file is served. Thus, you could embed the forbidden code right into the fabric of the internet and, again, the ____________ (fill in your facist corporate entity) wouldn't even begin to have the computer resources to check everyone's gifs and PNG's for DeCSS.
I've got a novel idea. Linux is open source, which means it's possible to tinker with or modify the OS and many of its apps as one sees fit. The real strengths and beauty of MacOS has been its consistent and well thought out interface, its robust way of dealing with metadata and configuration files, and the fact that most mac programmers put a value on UI design and easy of use that most programmers simply don't have. Technically, there's really nothing preventing anyone from taking all these OS and UI design ideals from MacOS and modifying GNOME and linux code to the extent that it copies these ideals. Or in layman's terms, porting MacOS to Linux. For example, if there's were some hypothetical source file for some hypothetical GNOME feature called Gnome App Helper, which provides a hypothetical default framework for creating really crappy ripoffs of the crappy windows UI, one could hypothetically modify that framework to create default UI's resembling MacOS'. Any GNOME program that linked against this new version of the GNOME library could hypothetically produce programs with mac-like interfaces. To duplicate the highly robust mac metadata model, one could hypothetically create some modified version of ext2 that generates resource and data forks. Certainly all these things seem hypothetically possible.
The million dollar question is--is this all just hypothetical?
I have a novel idea. Since the MPAA wants DeCSS, why don't we give them more copies than they ever want. Okay, it's not really a denial of service attack, but the jist of the idea is that everyone with DeCSS should send one (and exactly one) email of complaint (to make the whole thing legitimate) to the movie studios sponsoring DeCSS. Included attached to this e-mail would be a copy of DeCSS. There would be so many legitimate grievances flowing into the studios that their mail servers would get bogged down. And because each person would be sending exactly one message with one email attachment (as opposed to flooding the site) prosecuting the individual for a denial of service attack would be difficult.
Why don't we simply follow apple's guidelines, and then we'll already have the book written for us? Seriously, I feel that the only thing app helper does is help enforce a shoddier version of already Microsoft's shoddy UI design. F3 is not an acceptable shortcut for open, and accelerators needlessly steal the most rational choice for the shortcut modifier key while cluttering up the entire menu system with crap that only 2% of users will use anyhow. I believe in standards, but not badly chosen standards copied from people who never knew what the hell they were doing. Maybe it's time for 2 GNOMEs.
It is actually quite easy to make gnome programs *fairly* maclike (alt keyboard shortcut, ripping out accelerator key systems, etc). I'm working on trying to create a linux distribution that adopts macintosh interface ideas into GNOME (which means having to create my own version of GNOME. oh well). If I can make every single signification GNOME application mac-like and consistent, I will win the linux desktop. The only programmers who have ever shown any talent, ability, desire, or tradition of making consistent interfaces have been macintosh programmers. Therefore, it is only natural to take their lead and build the future of the linux desktop on the foundation of the mac. Its time to stop the Redmondification of GNOME!
Long Live Clarux the Penguincow!!!
The same exact thing can be said (and has been said) of Microsoft Visual DOS, err, I mean, Windows when compared to MacOS. MacOS is vastly easier to use and less confusing and far less ambiguous than windows. MacOS actually has consistency in UI between applications, something windows (or rather, windows programmers) still can't grasp. MacOS actually uses Plain English for it's system file names. No MFC42.DLL DOS crap. Look in a mac's system folder, then look in a windows machine's "windows" folder and see which platform users more words that can be found in Websters Dictionary.
Windows might beat Linux on ease of use, but MacOS slams the bejeezus out of windows in that very same category. It's amazing just how hypocritical the windows world really is.
The mandrake install is pretty awful. The checkboxes are very ambiguous and to the untrained eye it is hard to tell if they are checked or not. A check box is supposed to be a big, white box with a clearly visiable black worder that has a nice, fat black checkmark (or X) going through it. Apparently, Mandrake doesn't understand this, opting for some extrememly ambiguous checkbox that resembles a motif widget.
And there are a few checkboxes in even more confusing star shapes. And then they further screw up the UI by taking the colored balls they used in the last version as interactive bullets/radio buttons and replacing those with the confusing stars are well. Arrgh. There are so many people in the computer industry (especially in the linux and windows areas) who think they know everything about ease of use (such as GNOME) but in reality they do not have the slightest clue about making *anything* easy. But in the end this doesn't really matter, 'cause I've got the code.
One of the biggest problems with gnutella is that almost all the interfaces I have seem for gnutella, both win32 and linux, totally suck. For example, in the linux Gnubile client, for every file that you want to add to your share list, you have to open up a file browser that (I believe) opens up to the users home directory, and then navigate the path to wherever the file in question in. There is no "keep file dialog open and allow for multiple selections" option. While geeks like me can write perl scripts to "batch add" resources to the gnubile share config file, many people would simply say "screw it" and simply be content to download. In the win32 side of things, all the gnutella clients I have ever tried (even the official one) did not work properly and performed voodoo on the registry I still can't get rid of. The one bright and shinning example was gnucleus, which doesn't deal in any way with the registry and gives several good default hosts for getting client IP's from. If there is going to be no committment to making it easy for the average joe to set up his gnutella client and easily and efficiently setting up shared file, then people shouldn't complain about the system not working. Napster got as popular as they did because they made their client easy to set up and to use. If gnutella wants Napsters success, they must do the same.
It's called binary, which is the only true non-metaphoric way of presenting computer data. Ascii letters are dumbed down version of binary. Hex is a dumbed down version of binary. Anything that is not displayed as 10101010101010101 is inherently a dumbed down version of binary. Unless you have 1010101010010101 flashing across your screen at all times, you've got no right to complain about metaphors, human.
can't create a language worth a damn. They are incompetant in all areas, and language creation is really no different. Anyone who creates a language that delimits blocks of codes via new lines and has parameter passing by reference as the default cannot be taken seriously.
In The Reality That Is The Desktop, the most important thing is for ordinary (aka non-geek) users is the ability to install/deinstall absolutely any software without penalty or pain. This concern overrides security, it overrides performance, it overrides more efficient usage of resources, and overrides just about anything else. Period. In the discipline of engineering, it is understood that you trade off one thing for another to achieve the necessary goals. In The Reality That Is The Desktop, you trade off the more efficient usage of resources that dynamic linking provides with the robustness of installation that static linking gives and end users requires. Right now, there is such intense fear among windows users that anything they do, any upgrade they make, any new software they install will, in their words, "screw up my computer". So, your average joe computer user will not think very well of Linux when, on Christmas morning, RPM reports that Barbie Magic Funhouse refuses to install because it conflicts with Evolution. Once again, this is The Reality That Is The Desktop. In order to succeed in The Reality That Is The Desktop, one must statically link, as it is the only thing that insures the software will actually install without destroying existing software or precluding the installation of any future software. Dynamic linking simply does not fit in with a healthy model of average joe, consumer desktop computing. Ask 1000 people outside of a CompUSA if,given a choice, whether they would rather have a computer that uses less resources (RAM, disk space, etc) or a computer that will let add or delete absolutely any piece of software without destroying previous software or precluding the installation of any future sofware. I guarentee you that at least 95% of the people will choose the second option. They have all had their share of windows doing unacceptable things when adding or removing software, and they will not accept the unacceptable from an operating system touted as superior to windows. Anyone who argues otherwise does not understand the desktop market and probably has no business whatsoever trying to bring linux to the desktop. Static linking not only provides robustness of installation, it also adds an element of ease of use, too. To deinstall a statically linked program, a user only has to click on the program folder (which can be placed in any damn directory the user chooses) and drag to trash. Empty trash, and you deinstall program. Ridiculously simple and keeps consistency with the UI concept of getting rid of something by dragging to it trash. To install program, the user copies the program folder from CD-ROM to wherever they want it. That's it. To install from internet, the user only has to download and unzip the program folder, then place the program folder where you want it. Simple and quite effective. These are the steps that must be taken for linux to succeed in The Reality That Is The Desktop.
Oh, and whether you agree or disagree with me doesn't matter.
'Cause I've got the code.
Cracking will be made obselete
on
Biotransistors
·
· Score: 1
Now you can bring down someone's webserver with penecillin
The point is that from an ease-of-use perspective, windows doesn't exactly come out smelling like a rose and usually get it butt kicked by MacOS in any matter of ease of use. Windows is to linux as MacOS is to windows. On the stability side of things windows is problematic, but what people don't realize is that on the user interface side of things, windows is also very problematic. That's why I am against duplication of the windows interface--it duplicates all of window's UI mistakes.
Here's a quote from tom piwowar, the mac guy of computer guys radio show(http://www.wamu.org/computerguys/index.html). I like linux but this quote makes a very important point regarding OpenWindows " I made a very important discovery. There is a very important function for Linux. Linux dispells ignorance for windows users. All these years, windows users have been mystified at the reaction they get from mac users when they come over and talk about windows. The mac user looks at you like you just pooped on the floor! You go [imitating windows user] 'Hey, what did I just..Bill Gates is god..what do you mean, did I say something wrong?' [laughing] Well, now what windows users need to do is go over and look at the Linux X-windows, the GUI. And they've got office applications that run under that. And they need to use it for a while and see how they feel. And what they're going to feel like is that this is a very impoverished user interface, that everything is much slower, that there are more steps, and that it's much more complicated. They're going to feel just like a macintosh user feels when they see windows! So all you windows users, go out there and spend some time with Linux. And then you'll understand what where talking about!"--Tom Piwowar
Slashdot Mirror
Windows NT/2000 might be a 32-bit OS, but Windows programmers are still 16-bit. Someday far into the future, windows programmers might realize they are no longer coding for DOS, and might even join Linux, mac, and all other programmers, in the 21st century. But until that day arrives, we have to put up with these idiots who can only babble in eight letter words.
Open source porn would rock. But Richard Stallman GNUde would be a scary sight.
Windows Me is a 32 bit simulator. Get your byte alignment straight!
If IT workers are smart and their employers are stupid, then the IT workers should start their own companies and, using their superior knowlege, wipe the floor with their former employers. If you can't work with'em, beat'em.
So, how much did Tipper go for?
Packaging (and shared code) inevitably causes more problems than it solves. Maybe a few geeks like us would appreciate the better use of resources that shared code gives. But most users (and probably a few geeks as well) would be better served making each program its own seperate entity--One program, one set of code. Some will no doubt decry my advocacy for static linking (I'll make an exception for things like glibc), but take into account that Linux is starting to get into the regular consumer market. The regular joe, home consumer market has a totally different set of rules and goals than the workplace IT market. And currently, I see the linux movement making a very large mistake by approaching the desktop market with the same exact strategies and objective as the server market. Success and failure in the two markets is very different. On the server, a crash is failure. Some script kiddie rooting the box is failure. On the destkop, a user who, on Christmas morning, getting messages that Barbie Magic Funhouse can't be installed because it conflicts with sendmail and will break dependencies for Evolution is complete, utter and total, unadulterated failure. And up to this point, the linux community has done nothing but bury its head in the sand and try to rely on the internet to solve problems that static linking could easily solve. The funny thing about the word "dependency"--it usually comes after the words "co" and "drug". Seriously, ask 1000 people outside a CompUSA what they would rather have: An OS that uses their memory and disk space more efficiently, or an OS that lets them install as much software as they want without breaking anything and which will never preclude the installation of any other software. 999 of them will go with the latter. Most people want computers that simply just work. This is The Reality That Is The Desktop.
>What's to stop somebody from developing a Linux >Distro that mirrors OS X? We've been strugling >for years to find the "Linux your mother could >use", and Cupertino has shown it to us.
Nothing, which is why I'm currently working on a linux distribution that incorporates essential ideas of mac interface design. I'm currently rewriting parts of gnome to make the stuff that looks like it was written by the folks in redmond look like the stuff written by the folks in cupertino. Then I'll throw it all into a nice distribution your grandmother can use
This is outright heresy ;)
>In computing, the contemporary GUI sets aside >the idea of mastery as not only unnecessary but >unattainable.
This is exactly what I believe. The sign of a good GUI is not that a user never exposed to it can immediately master the fundamental basics, rather that mastery of the fundamental basics can be easily applied to the mastery of new programs.
>But the specifics of the spatial metaphor vary >widely. Consider the icons on the Windows >desktop: the Network Neighborhood, My Computer, >Recycle Bin, and any shortcut and file icons >that may be present all have the same basic >appearance -- but they function very >differently. In some applications, dialog boxes >that accomplish file management functions use a >left-to-right hierarchy; others may choose top->to-bottom; still others will make use of both.
The problem with windows is that it makes a lot of exceptions to the desktop metaphor. Move a folder to somewhere else, that's all well and good. Try moving the "control panels" or "dial-up-networking" folder, you're told you can't do it.
Drag a a file to the desktop--no problem. Drag an program to the destop, windows makes a shortcut instead.
>Worse yet, when there is a system problem, and >the structures which supply ease are replaced >with "Unexpected exception in module >Vx00f323.dll" or "Oops. Dumping core . .
Again, another great M$ inconsistency. Objects like "My Computer" are given Plain English names that make sense, but then we take a look at the system files and they're all DOS 8.3 naming convention gobbledygook. Windows is a 32 bit OS, unfortunately windows programmers are still 16-bit. It is intersting to note that the MacOS has a filename limit of 31 characters, but still has system files with far more intuitive names. The example the author gave also illustrates another M$ inconsistency: half the time they try to do everything for the user with their wizards and talking paperclips--and wind up getting in the user's way. The other half of the time they default to their old DOS heritage--and shamelessly and completely abandon the user to a system so arcane, confusing, and erratic that it makes kernel hacking look intuitive.
People say that the desktop metaphor has failed. I say that it is not the desktop metaphor that has failed, but rather the failure of non-mac programmers to adequately understand this metaphor and implement it consistently. Good GUI programmers don't try to do everything for the user, they allow the user to make better, more informed choices in a highly productive manner.
but you still want it in plain sight on that rare occasion you need it. Menus items are no different.
A cool way of hiding forbidden code places where the corporate machine could never get to it would be in mp3s (or other files commonly downloaded and shared). Imagine embedding DeCSS in a Britney Spears MP3 (which would actually give the mp3 some *real* value ;) and then putting that in a Napster directory on your machine. Thousands upon thousands of ten year olds would download DeCSS along with the mp3 and put it in their napster directory (And a Britney spears MP3 would finally contain something of value!)Thus, DeCSS would be spread to people who never even knew they had it. While the MPAA might have a few good spiders here and there, for them to try to grab every mp3 on Napster (or Gnutella) and decode it to search for DeCSS would be futile. Especially if napster were absolutely chock full of mp3's containing DeCSS. You could even move beyond mp3s, embedding DeCSS in porn (surely to get distributed). One could write something like an apache module that would embed the forbidden code in a graphic file every time the file is served. Thus, you could embed the forbidden code right into the fabric of the internet and, again, the ____________ (fill in your facist corporate entity) wouldn't even begin to have the computer resources to check everyone's gifs and PNG's for DeCSS.
I've got a novel idea. Linux is open source, which means it's possible to tinker with or modify the OS and many of its apps as one sees fit. The real strengths and beauty of MacOS has been its consistent and well thought out interface, its robust way of dealing with metadata and configuration files, and the fact that most mac programmers put a value on UI design and easy of use that most programmers simply don't have. Technically, there's really nothing preventing anyone from taking all these OS and UI design ideals from MacOS and modifying GNOME and linux code to the extent that it copies these ideals. Or in layman's terms, porting MacOS to Linux. For example, if there's were some hypothetical source file for some hypothetical GNOME feature called Gnome App Helper, which provides a hypothetical default framework for creating really crappy ripoffs of the crappy windows UI, one could hypothetically modify that framework to create default UI's resembling MacOS'. Any GNOME program that linked against this new version of the GNOME library could hypothetically produce programs with mac-like interfaces. To duplicate the highly robust mac metadata model, one could hypothetically create some modified version of ext2 that generates resource and data forks. Certainly all these things seem hypothetically possible.
The million dollar question is--is this all just hypothetical?
Long live Clarux the PengiunCow!!!
I have a novel idea. Since the MPAA wants DeCSS, why don't we give them more copies than they ever want. Okay, it's not really a denial of service attack, but the jist of the idea is that everyone with DeCSS should send one (and exactly one) email of complaint (to make the whole thing legitimate) to the movie studios sponsoring DeCSS. Included attached to this e-mail would be a copy of DeCSS. There would be so many legitimate grievances flowing into the studios that their mail servers would get bogged down. And because each person would be sending exactly one message with one email attachment (as opposed to flooding the site) prosecuting the individual for a denial of service attack would be difficult.
Why don't we simply follow apple's guidelines, and then we'll already have the book written for us? Seriously, I feel that the only thing app helper does is help enforce a shoddier version of already Microsoft's shoddy UI design. F3 is not an acceptable shortcut for open, and accelerators needlessly steal the most rational choice for the shortcut modifier key while cluttering up the entire menu system with crap that only 2% of users will use anyhow. I believe in standards, but not badly chosen standards copied from people who never knew what the hell they were doing. Maybe it's time for 2 GNOMEs.
It is actually quite easy to make gnome programs *fairly* maclike (alt keyboard shortcut, ripping out accelerator key systems, etc). I'm working on trying to create a linux distribution that adopts macintosh interface ideas into GNOME (which means having to create my own version of GNOME. oh well). If I can make every single signification GNOME application mac-like and consistent, I will win the linux desktop. The only programmers who have ever shown any talent, ability, desire, or tradition of making consistent interfaces have been macintosh programmers. Therefore, it is only natural to take their lead and build the future of the linux desktop on the foundation of the mac. Its time to stop the Redmondification of GNOME! Long Live Clarux the Penguincow!!!
The same exact thing can be said (and has been said) of Microsoft Visual DOS, err, I mean, Windows when compared to MacOS. MacOS is vastly easier to use and less confusing and far less ambiguous than windows. MacOS actually has consistency in UI between applications, something windows (or rather, windows programmers) still can't grasp. MacOS actually uses Plain English for it's system file names. No MFC42.DLL DOS crap. Look in a mac's system folder, then look in a windows machine's "windows" folder and see which platform users more words that can be found in Websters Dictionary.
Windows might beat Linux on ease of use, but MacOS slams the bejeezus out of windows in that very same category. It's amazing just how hypocritical the windows world really is.
The mandrake install is pretty awful. The checkboxes are very ambiguous and to the untrained eye it is hard to tell if they are checked or not. A check box is supposed to be a big, white box with a clearly visiable black worder that has a nice, fat black checkmark (or X) going through it. Apparently, Mandrake doesn't understand this, opting for some extrememly ambiguous checkbox that resembles a motif widget.
And there are a few checkboxes in even more confusing star shapes. And then they further screw up the UI by taking the colored balls they used in the last version as interactive bullets/radio buttons and replacing those with the confusing stars are well. Arrgh. There are so many people in the computer industry (especially in the linux and windows areas) who think they know everything about ease of use (such as GNOME) but in reality they do not have the slightest clue about making *anything* easy. But in the end this doesn't really matter, 'cause I've got the code.
One of the biggest problems with gnutella is that almost all the interfaces I have seem for gnutella, both win32 and linux, totally suck. For example, in the linux Gnubile client, for every file that you want to add to your share list, you have to open up a file browser that (I believe) opens up to the users home directory, and then navigate the path to wherever the file in question in. There is no "keep file dialog open and allow for multiple selections" option. While geeks like me can write perl scripts to "batch add" resources to the gnubile share config file, many people would simply say "screw it" and simply be content to download. In the win32 side of things, all the gnutella clients I have ever tried (even the official one) did not work properly and performed voodoo on the registry I still can't get rid of. The one bright and shinning example was gnucleus, which doesn't deal in any way with the registry and gives several good default hosts for getting client IP's from. If there is going to be no committment to making it easy for the average joe to set up his gnutella client and easily and efficiently setting up shared file, then people shouldn't complain about the system not working. Napster got as popular as they did because they made their client easy to set up and to use. If gnutella wants Napsters success, they must do the same.
It's called binary, which is the only true non-metaphoric way of presenting computer data. Ascii letters are dumbed down version of binary. Hex is a dumbed down version of binary. Anything that is not displayed as 10101010101010101 is inherently a dumbed down version of binary. Unless you have 1010101010010101 flashing across your screen at all times, you've got no right to complain about metaphors, human.
Those who think statically link. For user panic, link dynamic.
can't create a language worth a damn. They are incompetant in all areas, and language creation is really no different. Anyone who creates a language that delimits blocks of codes via new lines and has parameter passing by reference as the default cannot be taken seriously.
In The Reality That Is The Desktop, the most important thing is for ordinary (aka non-geek) users is the ability to install/deinstall absolutely any software without penalty or pain. This concern overrides security, it overrides performance, it overrides more efficient usage of resources, and overrides just about anything else. Period. In the discipline of engineering, it is understood that you trade off one thing for another to achieve the necessary goals. In The Reality That Is The Desktop, you trade off the more efficient usage of resources that dynamic linking provides with the robustness of installation that static linking gives and end users requires. Right now, there is such intense fear among windows users that anything they do, any upgrade they make, any new software they install will, in their words, "screw up my computer". So, your average joe computer user will not think very well of Linux when, on Christmas morning, RPM reports that Barbie Magic Funhouse refuses to install because it conflicts with Evolution. Once again, this is The Reality That Is The Desktop. In order to succeed in The Reality That Is The Desktop, one must statically link, as it is the only thing that insures the software will actually install without destroying existing software or precluding the installation of any future software. Dynamic linking simply does not fit in with a healthy model of average joe, consumer desktop computing. Ask 1000 people outside of a CompUSA if,given a choice, whether they would rather have a computer that uses less resources (RAM, disk space, etc) or a computer that will let add or delete absolutely any piece of software without destroying previous software or precluding the installation of any future sofware. I guarentee you that at least 95% of the people will choose the second option. They have all had their share of windows doing unacceptable things when adding or removing software, and they will not accept the unacceptable from an operating system touted as superior to windows. Anyone who argues otherwise does not understand the desktop market and probably has no business whatsoever trying to bring linux to the desktop. Static linking not only provides robustness of installation, it also adds an element of ease of use, too. To deinstall a statically linked program, a user only has to click on the program folder (which can be placed in any damn directory the user chooses) and drag to trash. Empty trash, and you deinstall program. Ridiculously simple and keeps consistency with the UI concept of getting rid of something by dragging to it trash. To install program, the user copies the program folder from CD-ROM to wherever they want it. That's it. To install from internet, the user only has to download and unzip the program folder, then place the program folder where you want it. Simple and quite effective. These are the steps that must be taken for linux to succeed in The Reality That Is The Desktop. Oh, and whether you agree or disagree with me doesn't matter. 'Cause I've got the code.
Now you can bring down someone's webserver with penecillin
The point is that from an ease-of-use perspective, windows doesn't exactly come out smelling like a rose and usually get it butt kicked by MacOS in any matter of ease of use. Windows is to linux as MacOS is to windows. On the stability side of things windows is problematic, but what people don't realize is that on the user interface side of things, windows is also very problematic. That's why I am against duplication of the windows interface--it duplicates all of window's UI mistakes.
Here's a quote from tom piwowar, the mac guy of computer guys radio show(http://www.wamu.org/computerguys/index.html). I like linux but this quote makes a very important point regarding OpenWindows " I made a very important discovery. There is a very important function for Linux. Linux dispells ignorance for windows users. All these years, windows users have been mystified at the reaction they get from mac users when they come over and talk about windows. The mac user looks at you like you just pooped on the floor! You go [imitating windows user] 'Hey, what did I just..Bill Gates is god..what do you mean, did I say something wrong?' [laughing] Well, now what windows users need to do is go over and look at the Linux X-windows, the GUI. And they've got office applications that run under that. And they need to use it for a while and see how they feel. And what they're going to feel like is that this is a very impoverished user interface, that everything is much slower, that there are more steps, and that it's much more complicated. They're going to feel just like a macintosh user feels when they see windows! So all you windows users, go out there and spend some time with Linux. And then you'll understand what where talking about!"--Tom Piwowar