When he visited my school to receive an honorary doctorate, he handed out to everyone who came to the party two $2 bills joined together and told us the story about the casino.
For those of us who consider usability and ease-of-use essential software freedoms, Apple tends to be an organization that has supported software freedom since 1984, while GNU has been an organization that has stripped end users of this essential software freedom since 1984.
While I disagree with the statement about Linux in general having better documentation (Apple has no stigma against putting pictures in their documentation; can't say the same about how-tos and most linux books), I do agree that Apple developer documentation is exceptionally bad in some areas. Often, those areas where they have sucky documentation are for C-based API's. It really sucks when you're looking for documentation on QuickTime and you have to refer to a 1997 edition of Inside Macintosh or you need to look up using AppleEvents and all the code examples are in Pascal. Even some new C-based technologies SearchKit and the audio system have terrible documentation, which basically killed my ideas for an e-mail list searching program and a number of applications where I wanted to give the user the ability to record sound. How a company that prides itself on being built on multimedia can have such bad documentation on building multimedia technologies is beyond me.
After having worked with XGrid for a few months, I would say because Apple recognizes that the time and difficulty to set the thing up and the dedicated staff you have to pay are also costs, and they make their stuff 5 times as easier to set up as their competitors.
Hmmm. Where is FairPlay/AAC on that list? Speaking of a free press, click here to see the latest Apple stories on Slashdot. 2 or 3 on the first page alone are about Apple censoring users and closing communications. Definitely more Kim Jong Il than Johnny Appleseed here. This just does not happen nearly as much in the OSS world.
OSS seems to have it's own brand of censorship. People who have legitimate grievances OSS usability and with how damn hard OSS is to used are attacked by the OSS community and are called whiners and their posts are often removed from forums. Reminds me sort of what Stalin did when people complained about food shortages.
Or lets talk about people who fork OSS projects because they feel the parent project wasn't paying adequate attention to usability issues. I forked KDE because they were pretty damn negligent in the user-friendliness area. I changed UI stuff and I made my damn code GPL. I try posting to Freshmeat, the central OSS announcment site. They wouldn't post the announcement for my project, because they considered changing a UI "merely a patch" and they told me to go talk to the same people (the KDE developers) who wouldn't listen or make the needed changes in the first place. Total censorship. So much for this freedom thing I keep hearing OSS people talking about.
At least in the mac community, users are free to criticize software for being hard to use. It's considered a popular past time, and the brutal and honest criticism mac developers receive makes the software constantly improve.
Usability is an essential software freedom. Freedom to criticize software is an essential software freedom. The mac community recognizes both, and the OSS community recognizes neither.
When end-users with justified complaints are called whiners and told to stop criticizing gifts, especially when talking about usability issues, it becomes enormously apparent why Open Source and Free Software are still getting their ass kicked in the end-user desktop arena.
I've also noticed that many of the "it's just a gift" people, when you turn your back, are start advocating for their software to be pushed into areas like businesses and governments where the users won't have any choice in what they run and will be forced to use the software. In my opinion, when something is forced on someone, it stops being "just a gift" and the person who made it stops being "merely a volunteer".
Most of Linux's problems are cultural, not technical.
Linux was born of a culture (the unix culture) with a hardcore dislike of anything graphical. This parent culture has had a longstanding practice of revelling in things that are cryptic and difficult to use and labelling end-users who are thwarted by difficult and confusing things as stupid.
The unix culture is also a parts-based culture, which tries to break things down modularly into their smallest parts; they cringe at anything that has permanent, wide-scale integration. It's impossible to try to explain to a dyed-in-the-wool unix hacker that a successful environment for end users is one where the whole of the interface and code that makes it has to be greater than the sum of it's parts. All they see is a bunch of supposedly interchangeable parts, so naturally having ten billion different GUI toolkits seems perfectly justified to them.
In addition to the past 30 years of unix holding Linux back from the desktop, the current Linux culture also holds itself back. People are yelled for asking questions on IRC and told to RTFM. Then, when they get frustrated and (correctly) vent that the user interface for the software they were using really blew, they are called "whiners", they are told they should "go code it themselves", they are told they should drop whatever they are doing and become programmers, they are told they demean "the gift of volunteers" by criticizing it, or they are informed that the UI is not confusing, it's just that they're "used to Windows". But it gets better. Once the angry and frustrated user decides to stay with his proprietary alternative that only slaps him or her in the face three or four times a day and developers pretend to be nice and to give a damn because they're being paid money, the developers of the Linux application the user had tried to use blame the proprietary developers (read Microsoft) for the user not switching and hold themselves blameless. And since they have destroyed virtually every legitimate route to getting their software installed on end-users machines, they must then try to lobby to get their software installed in governments and in workplaces, where the user has no choice in what they run and must take whatever crap the Linux developers give them.
How could any effort built on such cultural foundations ever succeed?
MDI is a horrible design, and any piece of software that gets rid of it does the world a favor.
That being said, I'm tired of seeing users with legitimate grievances with Linux usability being called "whiners" and then seeing the Linux community whine about how end-users aren't adopting their software and how "it's all Microsoft's fault".
The reason why there are so many complaints about something not being done the MS way is because the FOSS people were dumb enough to try to make their stuff look like Windows in some foolish attempt to make their software less intimidating. Guess what--if people see something that has been made to look just like Windows, they expect it to act just like Windows. So when your software doesn't act like Windows, the users expectations were getting violated. The reason why there are fewer complaints about OS X is that Apple didn't try to make their stuff look like Microsoft's stuff down to the last usability flaw.
Nothing but Office will ever feel like Office. Every time anyone ever tries to emulate a user experience, they always get something out of place or botch it on the small things that everyone takes for granted.
The only thing worse than a totally alien interface is one that tricks the users into thinking that what they are now using is just like what they were using before, and then pulls the rug out from under them at the last minute when they try to rely on old habits. Contrary to popular OSS belief, a lot of problems users coming from windows have with the Linux UI's are not because "they are addicted to microsoft" but rather that the dumb linux programmers thought they could make things better by trying to make the UI like Windows and they got stuff wrong, leading to lots of nasty surprises for the people using their software. Hence the common "it's not like windows complaint"--you make it look just like Microsoft, people expect it to act just like Microsoft.
Any new piece of software will feel strange and unfamiliar. It simply can't be helped. The only thing that can be changed is whether it has a radically better user experience than what came before it.
If I had a dime for every time I've seen an OSS zealot play down the usability problems of OS X Star/Open Office and tell us it's a perfect alternative to Office, I could probably buy them a clue about why linux still isn't ready for the desktop.
When I first saw the headline I read it as "space probe uses jupiter moon to help reverse-engineer Apple networking technology".
I probably made such an interpretation because I've spent three months trying to create an XGrid interoperability layer in python only to be continually thwarted by strange undocumented stuff. At this point, misusing heavenly bodies for personal gain doesn't sound like such a bad idea.
In all these discussions about electronic voting, the entire issue of the usability of the electronic voting process has been ignored, irrespective of disability. There's plenty of discussion about whether these systems will be secure from hackers trying to influence elections, but I've really heard very little about what will be done to make sure that voters (with disability or otherwise) will be able to accurate choose the candidate they wanted to choose.
We've already had problems with the butterfly ballots in Florida, where the bad usability problems of *paper* ballots caused enough people to cast the wrong vote that the entire election process was screwed up. A confusing user interface for a digital voting system could potentially wreak as much havoc as the most malicious intruder.
Re:There are other examples...
on
IT Literacy Test
·
· Score: 2, Insightful
(several years before in the bowels of Hewlett-Packard)
Interaction Designer: we really need to start designing the UI of this printing system from the very beginning.
Programmer: Don't be silly. That's the purty GUI front-end part. We usually save that for last. Why would you want to do that first?
Interaction Designer: So we can do things like build into the networked printing system stuff that will keep the user aware that their job is still being processed, so that they won't keep hitting Ctrl+P and wondering why they're getting no response.
Programmer: Go away. Star Trek starts in thirty minutes.
Here's a letter to Linux magazine I wrote in response to the above linked article, which was printed in the Augest edition of Linux magazine. As Linux magazine doesn't publish letters in the archive, I'll have to republish my letter in the post.
Our Way Is The Hard Way
The analysis of the Sharp Zaurus' failure in May's "The Hard Way" column is a textbook example of how we in the Open Source community blow it big time on usability issues and then blame the lack of mass adoption on "evil" proprietary companies like Sharp.
A successful PDA for the masses should minimize the number of taps required to perform a task. However, for every tap on my Palm, I had to do 2-3 more taps in the Qtopia applications on my Zaurus. Qtopia buttons and menus also took up needless screen real estate--real estate the designer of the Palm user interface was smart enough to conserve.
A successful PDA for the masses has a developer community that understands how to minimize taps and screen real-estate use. Palm developers get it. Zaurus developers, on the other hand, would constantly tell me "It's not a 'usability problem', you're just familiar with the Palm UI."
A successful PDA for the masses has a user community that rants and raves about how they can organize stuff and plan out their day with a few taps of their styli. The Palm had this. The Zaurus had a user community that ranted and raved about how they could run a terminal on their PDA to ssh into servers.
We in Open Source often hurt mainstream adoption more than any proprietary company ever could. It's time we realize this and stop blaming others for the problems of our own making.
While the Open Source community did significantly hurt the user experience of the Zaurus, in all fairness some of that blame also belongs to Sharp. For example, they left the power button exposed and uncovered by the flip-down visor, which often results in the Zaurus getting turned on while it's still in your pocket and the battery being kaput the second you whip it out.
I think that whether future handhelds will be successful depends on whether they are "Linux Handhelds" guided primarily by Unix design values or "Handhelds That Just Happen To Run Linux" that are guided primary by PDA/Ergonomic values.
What the heck, though, is someone going to do with 78 NES decks?
Actually, if you got 78 Nintendo ROB's, you could could control them all and have yourself a fairly respectable army of robots that could destroy your enemies by stacking up little piles of discs.
I understand the unix philosophy perfectly, and you've described exactly what I'm describing as the problem. The unix philosophy of first making a black box backend with a documented programmatic interface and trying to keep that completely seperate in design from GUI front-end that's sprayed on at a later point.
User interaction takes place across the entire computing process, not just in the "pretty little front-end" that most programmers think it does. Your choice of algorithms, the names you give files, what engineering tradeoffs are made--everything is going to have some affect on the user experience.
No matter how elegant the programmers think they will have made something, if they try to make the technical part in total isolation from the usability part, they will leave out something critical, and one or more of three bad things are going to happen: the programmers will have to go back and rewrite zillions of lines of code and totally rework the architecture and/or the usability people designing the UI will be massively constrained in what they can do because of the shortsightedness of the programmers and/or there will be lots of legacy UI that confuses users and provides inefficient interaction but is kept in for some or other technical reason (usability guru Alan Cooper refers to this as "scar tissue").
While the MVC design pattern, which has been successfully used to create usable software, does make some attempt at seperation of UI (View) from code (Model), to exploit this pattern to make the software usable and feel truly integrated, you have to design the view first. If you design the model first and the view with the optimal usability breaks the model or the model can't be accessed fast enough, you're screwed. To be able to access your model fast enough or in the right way with your view, you have to make design decisions and tradeoffs regarding how the model is built. How these design decisions are going to be made will depend on how the user accesses the view. You can't simply build a model then try to attach a view to it.
In most open source projects I'm aware of, all power flows from the output of a compiler--Peter Trudelle
In the workplace, usability folk deal day in and day out with political issues involving changes to UI's. There are zillions of people we have to deal with who do not respect our field of expertise, who thwart our every attempt to make the interface usable, who refuse to empathize with people who aren't them, and who refuse to understand that we can only really be effective if brought in at the beginning of the project before any code is written. And all the while, we're getting paid good money to put up with this crap.
Why the hell would we want to put with this kind of crap ten times over in our spare time without any kind of pay at all?
Until the OSS world is willing (just to list a few)
To renounce the unix methodology and culture.
To renounce the command line.
To accept the need to design the user experience first and then design the technical stuff.
To make engineering tradeoffs that favor end users instead of programmers' sense of elegence.
To see our work as just valuable and as just important as that of kernel hackers.
To hire as many usability experts as they hire kernel hackers.
To spend 10% of each year's development budget on usability reserach.
To buy out usability companies like Nielsen Norman group for the same ludicrous sums they bought out companies like Cygnus (650M), Ximian (Lotta M), and Suse(200M).
To stop trying to bring us in on projects in a half-assed and half-hearted way to do damage control on a UI we never had any input in to begin with.
To accept that some of the most important and needed contributions can come from people who don't know how to code.
To stop regarding legitimate criticisms of the UI as Microsoft propoganda.
To accept responsbility for usability problems and not try to melt into some anonymous crowd of "volunteers" who we're accused of attacking every time we point these problems out.
To stop equating suffering through a bad user experience with "learning".
To stop attributing end-user problems to lack of intelligence on the part of those users.
Most usability people will not want to have anything to do with OSS. Usable OSS requires a major cultural restructing, not some additional amount of technical progress.
One final note: it is interesting that the products Maddog Hall mentioned such as the Palm (originally modelled after a block of wood and a chopstick) were produced by designing the user experience first and then designing the technical stuff--a clear violation of the unix philosophy that OSS is built on.
a) SourceForge a the hosting site. Freshmeat is an announcing site. The two serve entirely different needs.
b) The same people who run SourceForge (OSDN) also run Freshmeat. I couldn't trust OSDN to not delete announcements for my project. Why the hell should I give them the power to delete the disk space which contains all my work and any CVS commits from contributors?
2. Broader Issues
The interesting thing about the Open Source community is that they present themselves as a single entity when they talk about how they're united and taking over the world and how damned superior Open Source is. But the instant someone has a problem, whether that problem is something like the way they've been treated by the OSS community, usability problems, etc, the Open Source community dissolves into a random collection of people who accuse the person with the problem of criticizing a random collection of people, none of whom represent Open Source. And once they're done chewing that person out, they go back to being a "single entity" that considers itself united in its struggle to take over the world through its "better" method of creating software.
Guerrillas and paramilitary death squads have similar "attack as one->disperse into a large anonymous crowd that can't be attacked->regroup->attack" MO's. And so do proprietary companies who try to deflect criticism by breaking the product up into a million vendors each of whom doesn't accept responsibility and blames the other vendor.
It's ironic how more and more the Open Source community resembles the people they preach against.
While I completeley disagree with your feelings about where the GNOME project has taken things (I think they should have gone much further and totally flipped off the unix geeks and shouldn't have blindly copied so many of microsoft's mistakes), I do respect you for your decision to fork, as the GNOME guys have been complete and utter jackasses about many things (such as usability, or lack thereof). I have had the same idea as you, albeit to fork GNOME in a completely opposite direction with the Clarux project and making GNOME far more mac-like. While I totally disagree with what you're doing, I'm glad at least someone had the same idea, even if it does run counter to mine.
One piece of advice to the opposition: the Free Software community says they promote freedom, but often, that's not the case. A while back, I created a fork of KDE that removed some really stupid usability problems the project had refused to deal with for years. I provided all my changes as source code people could download, I complied with the GPL, but Freshmeat refused to post the project because they considered it "only a patch". If you do something considered "significant" like modify someone else's code, it can be considered a distribution. But if you modify something that the Free Software community considers "insignificant", like the user experience, it's only considered "a patch". People in the Free Software development community might tell you "if you think you can do better, make your own version"; the thing is, they don't really mean it. So I'm warning you now, if you are really planning on forking a major desktop environment, you won't be able to rely on traditional community outlets for promoting it.
Last piece of advice--post as yourself. Stop this silly oGaLaxYo/Anonymous Coward crap. Post as Ali Agaa, be proud of your opinion, and be proud of what you're trying to stand up for and accomplish (even if it is rather silly).
Your post is definately the best one in regards to the article. Most anyone who has some sort of usability background and who has had some involvement in Open Source can relate to the whole "second-class citizen" feeling.
Slashdot Mirror
When he visited my school to receive an honorary doctorate, he handed out to everyone who came to the party two $2 bills joined together and told us the story about the casino.
For those of us who consider usability and ease-of-use essential software freedoms, Apple tends to be an organization that has supported software freedom since 1984, while GNU has been an organization that has stripped end users of this essential software freedom since 1984.
While I disagree with the statement about Linux in general having better documentation (Apple has no stigma against putting pictures in their documentation; can't say the same about how-tos and most linux books), I do agree that Apple developer documentation is exceptionally bad in some areas. Often, those areas where they have sucky documentation are for C-based API's. It really sucks when you're looking for documentation on QuickTime and you have to refer to a 1997 edition of Inside Macintosh or you need to look up using AppleEvents and all the code examples are in Pascal. Even some new C-based technologies SearchKit and the audio system have terrible documentation, which basically killed my ideas for an e-mail list searching program and a number of applications where I wanted to give the user the ability to record sound. How a company that prides itself on being built on multimedia can have such bad documentation on building multimedia technologies is beyond me.
After having worked with XGrid for a few months, I would say because Apple recognizes that the time and difficulty to set the thing up and the dedicated staff you have to pay are also costs, and they make their stuff 5 times as easier to set up as their competitors.
Hmmm. Where is FairPlay/AAC on that list? Speaking of a free press, click here to see the latest Apple stories on Slashdot. 2 or 3 on the first page alone are about Apple censoring users and closing communications. Definitely more Kim Jong Il than Johnny Appleseed here. This just does not happen nearly as much in the OSS world.
OSS seems to have it's own brand of censorship. People who have legitimate grievances OSS usability and with how damn hard OSS is to used are attacked by the OSS community and are called whiners and their posts are often removed from forums. Reminds me sort of what Stalin did when people complained about food shortages.
Or lets talk about people who fork OSS projects because they feel the parent project wasn't paying adequate attention to usability issues. I forked KDE because they were pretty damn negligent in the user-friendliness area. I changed UI stuff and I made my damn code GPL. I try posting to Freshmeat, the central OSS announcment site. They wouldn't post the announcement for my project, because they considered changing a UI "merely a patch" and they told me to go talk to the same people (the KDE developers) who wouldn't listen or make the needed changes in the first place. Total censorship. So much for this freedom thing I keep hearing OSS people talking about.
At least in the mac community, users are free to criticize software for being hard to use. It's considered a popular past time, and the brutal and honest criticism mac developers receive makes the software constantly improve.
Usability is an essential software freedom. Freedom to criticize software is an essential software freedom. The mac community recognizes both, and the OSS community recognizes neither.
When end-users with justified complaints are called whiners and told to stop criticizing gifts, especially when talking about usability issues, it becomes enormously apparent why Open Source and Free Software are still getting their ass kicked in the end-user desktop arena.
I've also noticed that many of the "it's just a gift" people, when you turn your back, are start advocating for their software to be pushed into areas like businesses and governments where the users won't have any choice in what they run and will be forced to use the software. In my opinion, when something is forced on someone, it stops being "just a gift" and the person who made it stops being "merely a volunteer".
A heroin-like addiction to emacs and vi.
Most of Linux's problems are cultural, not technical.
Linux was born of a culture (the unix culture) with a hardcore dislike of anything graphical. This parent culture has had a longstanding practice of revelling in things that are cryptic and difficult to use and labelling end-users who are thwarted by difficult and confusing things as stupid.
The unix culture is also a parts-based culture, which tries to break things down modularly into their smallest parts; they cringe at anything that has permanent, wide-scale integration. It's impossible to try to explain to a dyed-in-the-wool unix hacker that a successful environment for end users is one where the whole of the interface and code that makes it has to be greater than the sum of it's parts. All they see is a bunch of supposedly interchangeable parts, so naturally having ten billion different GUI toolkits seems perfectly justified to them.
In addition to the past 30 years of unix holding Linux back from the desktop, the current Linux culture also holds itself back. People are yelled for asking questions on IRC and told to RTFM. Then, when they get frustrated and (correctly) vent that the user interface for the software they were using really blew, they are called "whiners", they are told they should "go code it themselves", they are told they should drop whatever they are doing and become programmers, they are told they demean "the gift of volunteers" by criticizing it, or they are informed that the UI is not confusing, it's just that they're "used to Windows". But it gets better. Once the angry and frustrated user decides to stay with his proprietary alternative that only slaps him or her in the face three or four times a day and developers pretend to be nice and to give a damn because they're being paid money, the developers of the Linux application the user had tried to use blame the proprietary developers (read Microsoft) for the user not switching and hold themselves blameless. And since they have destroyed virtually every legitimate route to getting their software installed on end-users machines, they must then try to lobby to get their software installed in governments and in workplaces, where the user has no choice in what they run and must take whatever crap the Linux developers give them.
How could any effort built on such cultural foundations ever succeed?
MDI is a horrible design, and any piece of software that gets rid of it does the world a favor.
That being said, I'm tired of seeing users with legitimate grievances with Linux usability being called "whiners" and then seeing the Linux community whine about how end-users aren't adopting their software and how "it's all Microsoft's fault".
The reason why there are so many complaints about something not being done the MS way is because the FOSS people were dumb enough to try to make their stuff look like Windows in some foolish attempt to make their software less intimidating. Guess what--if people see something that has been made to look just like Windows, they expect it to act just like Windows. So when your software doesn't act like Windows, the users expectations were getting violated. The reason why there are fewer complaints about OS X is that Apple didn't try to make their stuff look like Microsoft's stuff down to the last usability flaw.
Nothing but Office will ever feel like Office. Every time anyone ever tries to emulate a user experience, they always get something out of place or botch it on the small things that everyone takes for granted.
The only thing worse than a totally alien interface is one that tricks the users into thinking that what they are now using is just like what they were using before, and then pulls the rug out from under them at the last minute when they try to rely on old habits. Contrary to popular OSS belief, a lot of problems users coming from windows have with the Linux UI's are not because "they are addicted to microsoft" but rather that the dumb linux programmers thought they could make things better by trying to make the UI like Windows and they got stuff wrong, leading to lots of nasty surprises for the people using their software. Hence the common "it's not like windows complaint"--you make it look just like Microsoft, people expect it to act just like Microsoft.
Any new piece of software will feel strange and unfamiliar. It simply can't be helped. The only thing that can be changed is whether it has a radically better user experience than what came before it.
If I had a dime for every time I've seen an OSS zealot play down the usability problems of OS X Star/Open Office and tell us it's a perfect alternative to Office, I could probably buy them a clue about why linux still isn't ready for the desktop.
They've already had that sort of thing for quite a while. In fact, you could call it "the world's oldest time sharing".
This fine service is available in most big cities, but you can only write it off as a business expense in Nevada.
Joel Spolsky has probably done the best job I've seen of explaining What's Wrong with Unix.
When I first saw the headline I read it as "space probe uses jupiter moon to help reverse-engineer Apple networking technology".
I probably made such an interpretation because I've spent three months trying to create an XGrid interoperability layer in python only to be continually thwarted by strange undocumented stuff. At this point, misusing heavenly bodies for personal gain doesn't sound like such a bad idea.
In all these discussions about electronic voting, the entire issue of the usability of the electronic voting process has been ignored, irrespective of disability. There's plenty of discussion about whether these systems will be secure from hackers trying to influence elections, but I've really heard very little about what will be done to make sure that voters (with disability or otherwise) will be able to accurate choose the candidate they wanted to choose.
We've already had problems with the butterfly ballots in Florida, where the bad usability problems of *paper* ballots caused enough people to cast the wrong vote that the entire election process was screwed up. A confusing user interface for a digital voting system could potentially wreak as much havoc as the most malicious intruder.
(several years before in the bowels of Hewlett-Packard)
Interaction Designer: we really need to start designing the UI of this printing system from the very beginning.
Programmer: Don't be silly. That's the purty GUI front-end part. We usually save that for last. Why would you want to do that first?
Interaction Designer: So we can do things like build into the networked printing system stuff that will keep the user aware that their job is still being processed, so that they won't keep hitting Ctrl+P and wondering why they're getting no response.
Programmer: Go away. Star Trek starts in thirty minutes.
While the Open Source community did significantly hurt the user experience of the Zaurus, in all fairness some of that blame also belongs to Sharp. For example, they left the power button exposed and uncovered by the flip-down visor, which often results in the Zaurus getting turned on while it's still in your pocket and the battery being kaput the second you whip it out.
I think that whether future handhelds will be successful depends on whether they are "Linux Handhelds" guided primarily by Unix design values or "Handhelds That Just Happen To Run Linux" that are guided primary by PDA/Ergonomic values.
What the heck, though, is someone going to do with 78 NES decks?
Actually, if you got 78 Nintendo ROB's, you could could control them all and have yourself a fairly respectable army of robots that could destroy your enemies by stacking up little piles of discs.
I understand the unix philosophy perfectly, and you've described exactly what I'm describing as the problem. The unix philosophy of first making a black box backend with a documented programmatic interface and trying to keep that completely seperate in design from GUI front-end that's sprayed on at a later point.
User interaction takes place across the entire computing process, not just in the "pretty little front-end" that most programmers think it does. Your choice of algorithms, the names you give files, what engineering tradeoffs are made--everything is going to have some affect on the user experience.
No matter how elegant the programmers think they will have made something, if they try to make the technical part in total isolation from the usability part, they will leave out something critical, and one or more of three bad things are going to happen: the programmers will have to go back and rewrite zillions of lines of code and totally rework the architecture and/or the usability people designing the UI will be massively constrained in what they can do because of the shortsightedness of the programmers and/or there will be lots of legacy UI that confuses users and provides inefficient interaction but is kept in for some or other technical reason (usability guru Alan Cooper refers to this as "scar tissue").
While the MVC design pattern, which has been successfully used to create usable software, does make some attempt at seperation of UI (View) from code (Model), to exploit this pattern to make the software usable and feel truly integrated, you have to design the view first. If you design the model first and the view with the optimal usability breaks the model or the model can't be accessed fast enough, you're screwed. To be able to access your model fast enough or in the right way with your view, you have to make design decisions and tradeoffs regarding how the model is built. How these design decisions are going to be made will depend on how the user accesses the view. You can't simply build a model then try to attach a view to it.
I hope I've clarified my point of view a little.
In the workplace, usability folk deal day in and day out with political issues involving changes to UI's. There are zillions of people we have to deal with who do not respect our field of expertise, who thwart our every attempt to make the interface usable, who refuse to empathize with people who aren't them, and who refuse to understand that we can only really be effective if brought in at the beginning of the project before any code is written. And all the while, we're getting paid good money to put up with this crap.
Why the hell would we want to put with this kind of crap ten times over in our spare time without any kind of pay at all?
Until the OSS world is willing (just to list a few)
Most usability people will not want to have anything to do with OSS. Usable OSS requires a major cultural restructing, not some additional amount of technical progress.
One final note: it is interesting that the products Maddog Hall mentioned such as the Palm (originally modelled after a block of wood and a chopstick) were produced by designing the user experience first and then designing the technical stuff--a clear violation of the unix philosophy that OSS is built on.
Given the nature of the magazine, I have to wonder whether interviewer will bring up the inevitable Booble controversy.
1. Pragmatic Issues
a) SourceForge a the hosting site. Freshmeat is an announcing site. The two serve entirely different needs.
b) The same people who run SourceForge (OSDN) also run Freshmeat. I couldn't trust OSDN to not delete announcements for my project. Why the hell should I give them the power to delete the disk space which contains all my work and any CVS commits from contributors?
2. Broader Issues
The interesting thing about the Open Source community is that they present themselves as a single entity when they talk about how they're united and taking over the world and how damned superior Open Source is. But the instant someone has a problem, whether that problem is something like the way they've been treated by the OSS community, usability problems, etc, the Open Source community dissolves into a random collection of people who accuse the person with the problem of criticizing a random collection of people, none of whom represent Open Source. And once they're done chewing that person out, they go back to being a "single entity" that considers itself united in its struggle to take over the world through its "better" method of creating software.
Guerrillas and paramilitary death squads have similar "attack as one->disperse into a large anonymous crowd that can't be attacked->regroup->attack" MO's. And so do proprietary companies who try to deflect criticism by breaking the product up into a million vendors each of whom doesn't accept responsibility and blames the other vendor.
It's ironic how more and more the Open Source community resembles the people they preach against.
Ali,
While I completeley disagree with your feelings about where the GNOME project has taken things (I think they should have gone much further and totally flipped off the unix geeks and shouldn't have blindly copied so many of microsoft's mistakes), I do respect you for your decision to fork, as the GNOME guys have been complete and utter jackasses about many things (such as usability, or lack thereof). I have had the same idea as you, albeit to fork GNOME in a completely opposite direction with the Clarux project and making GNOME far more mac-like. While I totally disagree with what you're doing, I'm glad at least someone had the same idea, even if it does run counter to mine.
One piece of advice to the opposition: the Free Software community says they promote freedom, but often, that's not the case. A while back, I created a fork of KDE that removed some really stupid usability problems the project had refused to deal with for years. I provided all my changes as source code people could download, I complied with the GPL, but Freshmeat refused to post the project because they considered it "only a patch". If you do something considered "significant" like modify someone else's code, it can be considered a distribution. But if you modify something that the Free Software community considers "insignificant", like the user experience, it's only considered "a patch". People in the Free Software development community might tell you "if you think you can do better, make your own version"; the thing is, they don't really mean it. So I'm warning you now, if you are really planning on forking a major desktop environment, you won't be able to rely on traditional community outlets for promoting it.
Last piece of advice--post as yourself. Stop this silly oGaLaxYo/Anonymous Coward crap. Post as Ali Agaa, be proud of your opinion, and be proud of what you're trying to stand up for and accomplish (even if it is rather silly).
If I had any mod points, I'd give them to you.
Your post is definately the best one in regards to the article. Most anyone who has some sort of usability background and who has had some involvement in Open Source can relate to the whole "second-class citizen" feeling.