Companies not caring about the usability of their products is a problem that will not be solved no matter how many new technological advances are thrown at it. While open source/free software allows people who give a damn to make changes, the people who are currently doing open source/free software are just as apathetic towards usability as most companies.
It's not a tech problem; it's a people problem, and those are always a lot harder to solve.
Gandalf kicks some booty in Isengard while Johnny Cash's "ring of fire" plays in the background. It annoys the purists, but I think the scene kicks ass.
Yup. It's packed with lots more people and destructive goodies than the Enterprise C. And there's and Enterprise E that's going to be ever more larger and deadly than the D class.
Enterprises are like bra sizes; when you go up a letter they get bigger and more impressive.
If you look a C:\WINDOWS, you'll notice that most files still have 8 letters or less. It's like in the new Harry Potter movie, technically the evil is dead but it's memory is enough to still give you the willies.
I was wondering why the digitally remastered "Easy Rider" DVD I bought from Amazon had Dennis Hopper and Peter Fonda riding around on scooters. I had figured it was just the director's cut.
Microsoft has billions of dollars in the bank and with all that money they must be spending hundreds of millions of those dollars on security research, therefore windows must be more secure than linux, right? There's no way that some weird Finnish guy and a couple hundred of non-rich hobbyists submitting patches could come up with an OS core more secure than something made by a large corporation like Microsoft, right?
Contrary to popular belief in the linux community, Microsoft is actually one of the software companies most frequently criticized by usability professionals. They are the most frequent inductee in the Interface Hall of ShameJust because they've got more money than some industrialized nations doesn't mean they aren't capable of cranking out some horrendously bad designs. If Microsoft had effective usability, they never would have come out with Window-in-Window MDI, multi-row tabs, or any of the other atrocities they've released over the years. Unfortunately, Open Source Software has incorporated more than their fair share of these stupid designs in the mistaken belief that microsoft knew what they were doing.
A developer community is only successful in areas where they have very strong beliefs and values that are advantageous. Linux has succeeded so well on the server because the linux development community had very strong values regarding security and stability, and these sorts of values were advantageous on the server. Unfortunately, linux people are unix people, and unix people have had a long standing tradition of calling end-users stupid, telling them to go RTFM, and decrying the field of usability as BS and usability folks as "whiners".
Who'd want to do usability for free for people who say things like:
"Don't whine about what you're getting for free"
"Free Software does not entitle you to a usable interface"
"I can't believe some people get paid to criticize the work of others"
"Usability is in the eye of the beholder. Don't listen to any of these 'Usability Experts'"
"If you want to improve the interface, learn how to code and submit a patch."
Open Source doesn't need money to improve usability. It needs an attitude adjustment.
Besides, I have recently begin doubting the existance of libraries. We have plenty of space these days, why not make everything statistically linked?
Because this idea is extremely distasteful to many geeks who value 4MB of RAM and 100MB of space on an 80G disk more than they value a good, quick, painless install that fits in with the interface paradigm and is damn near impervious to dependency shennanigans and upgrade landmines.
I myself would favor a system where an application comes with all the dynamic libraries it needs to run (excepting a few monsters like libc and X) which are stored in its application folder. At first the application would try to link to the comptuers centralized repository of dynamic libraries (e.g./usr/lib). Assuming it would successfully link to those libraries, it would have the advantage of traditional shared-library systems (save resources and use upgraded stuff). But if the app couldn't link to a centralized dynamic library, it would use the version of that dynamic library that resides in its folder. Yes, this entire system might end up costing several GB for on a 80GB disk. So what?
Volow's First Law of Computing: The cost of something not working far exceeds any other cost
Here the author obviously hasn't used a PocketPC. With the PPC its very very easy not to close applications. What happens? The system slows down to a crawl as it tried to run 5 or 6 different applications. Again, this is the user being in control of the computer. I want the ability to close applications when I'm not using them. That is my decision, not the computers. It's the desktop analogy. Once i've finished with a book, I put it away because otherwise my desk gets cluttered. I don't leave it out because otherwise my desk gets full and working becomes a problem. Sure, we could get around this by having the PC unload or suspend applications that aren't used in a while - but how does it decide? Just because I've not typed something into Word for the past 30 minutes doesn't mean that I'm not using it. You'd get to the point where the cleverness of the OS/Application was causing me more hassle as it tried to be helpful and suspend stuff for me.
On an all-in-ram storage device, if you are competant (i.e. not Microsoft or Sharp) you want to take advantage of a stateless interface model, where there is no quitting of applications or saving. You go to a different work area and the process is suspended in ram as you left it, and your data is just as you left it. It is still you making the decision; you are making the decision to put aside your work and go back to it at a later time. When you're not using an application, it's last operating state (maybe aside from the stack) is saved in ram. Of course, this design has traditionally only worked with devices that are mostly nothing but memory and it is generally understood by developers that applications are to be written to use as little memory as possible because RAM is the only storage you have. Devices like the Palm come to mind.
On a Palm, you don't quit the datebook, you write some stuff into it, go use another app, and when you go back to the datebook, you're back at the last thing your wrote, down to the last character and the last cursor position. No multiple processes necessary, because instead of having several multiple processes running at once and switching between them, one process is suspended while the user goes to another process.
PocketPC, on the other hand, is a total interface design failure and one of the shining examples of why the linux community looks like such idiots when they claim "Microsoft must spend billions on UI research, it's smart to copy what they do". PocketPC was basically a desktop UI grafted to a PDA with no consideration whatsoever as to the difference in design demands between a sit-down desktop with unlimited resources and where the user has hours to get work done and a mobile device with scarce resources where the user has 20 seconds to get down a phone number. Microsoft thought the mini-windows UI would work better than something like the Palm because for users coming from a Windows Desktop PC "it would be familiar". Bad UI decisions being made because some techie thought "it would be familiar". Where have I seen this before *cough*Xandros*cough*Lindows*cough*KDE*cough*
Perhaps what mpt was suggesting was that with today's fast, unreasonably large hard drives and copious amounts of RAM and CPU cycles, we now have enough resources to emulate the document-oriented, stateless, all-in-ram feel of something like a Palm. Mpt's not a hard-core techie, so he might not have considered all the process manipulation voodoo required to give the feeling of statelessness. But with some modifications to the kernel and perhaps the linker it might just now be possible with the technology we have.
They are an eye-opener for any one who has wondered why linux is still not ready for the desktop despite the prescense of so many talented programmers in the Free Software Community
I second the parent post. One really neat thing about the Canon Cat was that you type in the word processor:
5 6
highlight them and (I think) hit the 'add' key and on the next line you would get the number
11
No "Calculator" application necessary. If you were a 1337 Cat user, you'd type in something like "forth", highlight it, hit some weird key, and you'd have access to the forth interpreter. The Cat software was written in Forth, so it kind of makes sense when you think about it. From what I've heard (never tried it myself), you could type forth commands into the word processor and execute them.
The entire state of Cat memory (as mentioned by the poster) was stored on a floppy disk. You'd put the floppy in the Cat, turn it on, and it would boot in less than 10 seconds. During the bootup, you would actually see a screenshot bitmap of the last state of the computer before it was turned off. You'd turn the damn thing on and immediately you'd see the bitmap of what you were last working on (as opposed to the splashscreens/diagnostic messages of today's os'es), and in 7-8 seconds you could start working. Sort of like Xsession except it doesn't suck.
Jef Raskin has started an open source project that aims to take the best ideas of the Cat and re-write them in python for a variety of platforms. You can find out more about this project at Jef's Humane Environment Sourceforge page.
The proper response to RTFM would be for a group of people of who genuinely care about end-users having high quality documentation creating a new type of documentation license that mandates a minimum level of quality for modification of their documentation and bars distribution or linking to of bad documentation. For example, a license like this would ban any modification of the documentation that reduced the number of diagrams to less than three (i.e. converting the document to all-text is strictly prohibted). In addition to this, the license would bar distribution or linking to of HOW-TO's. If someone like Debian distributed all-text documents, they would be barred from linking to or distributing the high-quality documentation.
In essence, the license would be saying "We will place the RTFM'ing bastards on permanent lockout."
I call this license the "Anti User-Hostility Documentation License".
In a recent move to reduce the confusing differences in pronounciations of the word "potato", Red Hat pissed off both the "po-tay-to" and "po-tah-to" factions by decreeing that the word "potato" will now be pronounced "starchy self-reproducing tuber with eyes".
It's called a double standard. Kernel hackers' field of study is considered sacred, yet when usability people of equal expertise in their field try to explain how to correct some very bad usability problems with free software, they are told by technically-inclined linux zealots that their field of study, put between a sarcastic pair of quotes, is BS. A prominant kernel hacker once told me that he couldn't believe that "[usability] people get paid to criticize the work of others". I think of this quote everytime someone poses the quote "why linux has had such extraordinary success on the server yet such a hard time on the desktop?".
Technical expertise is lauded and user interface expertise is devalued and mocked. There's a reason why there is such a dearth of usability people in the free software community. It's attitudes like these, people.
I often say that Bill Gates doesn't have to lift a finger to crush desktop linux because so many people in the linux community are doing his job *for* him.
The name of the game is deterrance. It's like cold war.
Apple works on a partial x86 port of OS X (Darwin x86). Not that they are ever going to deploy a full x86 OS X, but they want to let microsoft know they can do so at any moment.
Microsoft keeps "re-evaluating" whether they should be releasing Mac versions of Office. Not that they are ever going to cancel Office for Macintosh, but they want to let Apple know they can do so at any moment.
Slashdot Mirror
If I had any mod points, I'd give'em to you.
Companies not caring about the usability of their products is a problem that will not be solved no matter how many new technological advances are thrown at it. While open source/free software allows people who give a damn to make changes, the people who are currently doing open source/free software are just as apathetic towards usability as most companies.
It's not a tech problem; it's a people problem, and those are always a lot harder to solve.
It begs the question of what else could have been done with the same money and far superior management."
.
A moon sized space station capabable of destroying rebel bases.
Assuming, of course, there isn't some OSHA regulation against telepathically strangling incompetant middle-level management
Gandalf kicks some booty in Isengard while Johnny Cash's "ring of fire" plays in the background. It annoys the purists, but I think the scene kicks ass.
According to the legacy computer systems, whose calendar runs out in the year 2000, the world has already ended.
You can use the parabolic dishes, but you have to compensate for a weakened structural integrity forcefield.
If you just stayed in Starfleet Academy another year they would have taught you about the virtues of model airplace glue.
Yup. It's packed with lots more people and destructive goodies than the Enterprise C. And there's and Enterprise E that's going to be ever more larger and deadly than the D class.
Enterprises are like bra sizes; when you go up a letter they get bigger and more impressive.
I'd rather have my own dark tower. It impresses the neighbors and strikes fear into the hearts of travelling mormons.
If you look a C:\WINDOWS, you'll notice that most files still have 8 letters or less. It's like in the new Harry Potter movie, technically the evil is dead but it's memory is enough to still give you the willies.
I've sworn at computers for years. Now I can stare them down, too.
I was wondering why the digitally remastered "Easy Rider" DVD I bought from Amazon had Dennis Hopper and Peter Fonda riding around on scooters. I had figured it was just the director's cut.
Not really.
Ximian decided not to support VI because it isn't a prime number. You'll have to wait for VII to come out.
Contrary to popular belief in the linux community, Microsoft is actually one of the software companies most frequently criticized by usability professionals. They are the most frequent inductee in the Interface Hall of ShameJust because they've got more money than some industrialized nations doesn't mean they aren't capable of cranking out some horrendously bad designs. If Microsoft had effective usability, they never would have come out with Window-in-Window MDI, multi-row tabs, or any of the other atrocities they've released over the years. Unfortunately, Open Source Software has incorporated more than their fair share of these stupid designs in the mistaken belief that microsoft knew what they were doing.
A developer community is only successful in areas where they have very strong beliefs and values that are advantageous. Linux has succeeded so well on the server because the linux development community had very strong values regarding security and stability, and these sorts of values were advantageous on the server. Unfortunately, linux people are unix people, and unix people have had a long standing tradition of calling end-users stupid, telling them to go RTFM, and decrying the field of usability as BS and usability folks as "whiners".
Who'd want to do usability for free for people who say things like:
Open Source doesn't need money to improve usability. It needs an attitude adjustment.
The apps would only load their own copy iff they have trouble linking to the centralized library.
Besides, I have recently begin doubting the existance of libraries. We have plenty of space these days, why not make everything statistically linked?
/usr/lib). Assuming it would successfully link to those libraries, it would have the advantage of traditional shared-library systems (save resources and use upgraded stuff). But if the app couldn't link to a centralized dynamic library, it would use the version of that dynamic library that resides in its folder. Yes, this entire system might end up costing several GB for on a 80GB disk. So what?
Because this idea is extremely distasteful to many geeks who value 4MB of RAM and 100MB of space on an 80G disk more than they value a good, quick, painless install that fits in with the interface paradigm and is damn near impervious to dependency shennanigans and upgrade landmines.
I myself would favor a system where an application comes with all the dynamic libraries it needs to run (excepting a few monsters like libc and X) which are stored in its application folder. At first the application would try to link to the comptuers centralized repository of dynamic libraries (e.g.
Volow's First Law of Computing: The cost of something not working far exceeds any other cost
Here the author obviously hasn't used a PocketPC. With the PPC its very very easy not to close applications. What happens? The system slows down to a crawl as it tried to run 5 or 6 different applications. Again, this is the user being in control of the computer. I want the ability to close applications when I'm not using them. That is my decision, not the computers. It's the desktop analogy. Once i've finished with a book, I put it away because otherwise my desk gets cluttered. I don't leave it out because otherwise my desk gets full and working becomes a problem. Sure, we could get around this by having the PC unload or suspend applications that aren't used in a while - but how does it decide? Just because I've not typed something into Word for the past 30 minutes doesn't mean that I'm not using it. You'd get to the point where the cleverness of the OS/Application was causing me more hassle as it tried to be helpful and suspend stuff for me.
On an all-in-ram storage device, if you are competant (i.e. not Microsoft or Sharp) you want to take advantage of a stateless interface model, where there is no quitting of applications or saving. You go to a different work area and the process is suspended in ram as you left it, and your data is just as you left it. It is still you making the decision; you are making the decision to put aside your work and go back to it at a later time. When you're not using an application, it's last operating state (maybe aside from the stack) is saved in ram. Of course, this design has traditionally only worked with devices that are mostly nothing but memory and it is generally understood by developers that applications are to be written to use as little memory as possible because RAM is the only storage you have. Devices like the Palm come to mind.
On a Palm, you don't quit the datebook, you write some stuff into it, go use another app, and when you go back to the datebook, you're back at the last thing your wrote, down to the last character and the last cursor position. No multiple processes necessary, because instead of having several multiple processes running at once and switching between them, one process is suspended while the user goes to another process.
PocketPC, on the other hand, is a total interface design failure and one of the shining examples of why the linux community looks like such idiots when they claim "Microsoft must spend billions on UI research, it's smart to copy what they do". PocketPC was basically a desktop UI grafted to a PDA with no consideration whatsoever as to the difference in design demands between a sit-down desktop with unlimited resources and where the user has hours to get work done and a mobile device with scarce resources where the user has 20 seconds to get down a phone number. Microsoft thought the mini-windows UI would work better than something like the Palm because for users coming from a Windows Desktop PC "it would be familiar". Bad UI decisions being made because some techie thought "it would be familiar". Where have I seen this before *cough*Xandros*cough*Lindows*cough*KDE*cough*
Perhaps what mpt was suggesting was that with today's fast, unreasonably large hard drives and copious amounts of RAM and CPU cycles, we now have enough resources to emulate the document-oriented, stateless, all-in-ram feel of something like a Palm. Mpt's not a hard-core techie, so he might not have considered all the process manipulation voodoo required to give the feeling of statelessness. But with some modifications to the kernel and perhaps the linker it might just now be possible with the technology we have.
The author of the article, Matthew Thomas, also wrote two very good pieces
"Why Free Software Usability Tends To Suck"
"Why Free Software Usability Tends To Suck Even More".
They are an eye-opener for any one who has wondered why linux is still not ready for the desktop despite the prescense of so many talented programmers in the Free Software Community
I second the parent post. One really neat thing about the Canon Cat was that you type in the word processor:
5
6
highlight them and (I think) hit the 'add' key and on the next line you would get the number
11
No "Calculator" application necessary. If you were a 1337 Cat user, you'd type in something like "forth", highlight it, hit some weird key, and you'd have access to the forth interpreter. The Cat software was written in Forth, so it kind of makes sense when you think about it. From what I've heard (never tried it myself), you could type forth commands into the word processor and execute them.
The entire state of Cat memory (as mentioned by the poster) was stored on a floppy disk. You'd put the floppy in the Cat, turn it on, and it would boot in less than 10 seconds. During the bootup, you would actually see a screenshot bitmap of the last state of the computer before it was turned off. You'd turn the damn thing on and immediately you'd see the bitmap of what you were last working on (as opposed to the splashscreens/diagnostic messages of today's os'es), and in 7-8 seconds you could start working. Sort of like Xsession except it doesn't suck.
Jef Raskin has started an open source project that aims to take the best ideas of the Cat and re-write them in python for a variety of platforms. You can find out more about this project at Jef's Humane Environment Sourceforge page.
The proper response to RTFM would be for a group of people of who genuinely care about end-users having high quality documentation creating a new type of documentation license that mandates a minimum level of quality for modification of their documentation and bars distribution or linking to of bad documentation. For example, a license like this would ban any modification of the documentation that reduced the number of diagrams to less than three (i.e. converting the document to all-text is strictly prohibted). In addition to this, the license would bar distribution or linking to of HOW-TO's. If someone like Debian distributed all-text documents, they would be barred from linking to or distributing the high-quality documentation.
In essence, the license would be saying "We will place the RTFM'ing bastards on permanent lockout."
I call this license the "Anti User-Hostility Documentation License".
In a recent move to reduce the confusing differences in pronounciations of the word "potato", Red Hat pissed off both the "po-tay-to" and "po-tah-to" factions by decreeing that the word "potato" will now be pronounced "starchy self-reproducing tuber with eyes".
It's called a double standard. Kernel hackers' field of study is considered sacred, yet when usability people of equal expertise in their field try to explain how to correct some very bad usability problems with free software, they are told by technically-inclined linux zealots that their field of study, put between a sarcastic pair of quotes, is BS. A prominant kernel hacker once told me that he couldn't believe that "[usability] people get paid to criticize the work of others". I think of this quote everytime someone poses the quote "why linux has had such extraordinary success on the server yet such a hard time on the desktop?".
Technical expertise is lauded and user interface expertise is devalued and mocked. There's a reason why there is such a dearth of usability people in the free software community. It's attitudes like these, people.
I often say that Bill Gates doesn't have to lift a finger to crush desktop linux because so many people in the linux community are doing his job *for* him.
Cartoon Jackie Chan vs. Cartoon Mr. T: who would win?
That's no AI. That's Clippy. He's been running in spite of human intervention for years.