"You should be employing more than one manager. One of them should be YOU. "
And if you've read/watched "A Scanner Darkly", you should know that even YOU might not be fully trustworthy, and you should consider diversifying into multiple personalities to keep an eye on that sneaky...
Paranoia. It's the sign of a healthy, prosperous society!
"You tell me college football, if you have possession of the ball and your knee touches the ground, the ball is down, whether or not the player was tackled. I give you a college football game, and the first time you try to kick a field goal, the ball is downed 7 yards behind the line of scrimmage because the holder's knee is touching the ground."
Can you translate that for those of us who only speak Rugby?
"At the moment, finding the few billions it might take to get us off-world just seems expensive, but at some point finding those billions may actually require taking the decision to stop feeding some people and that will be a tough decision for anyone to make! "
And how exactly is getting a few dozen people into space going to help feed anything?
Plants don't grow on vacuum alone. Here on Earth we have a whole biosphere for free. If we can't preserve that, we sure won't be able to grow a whole new one from scratch.
Conversely, if we have the technology to grow new biospheres from scratch, we can do it on Earth much more cheaply than doing it in space.
You seem to be under the impression that 'trillions of cubic kilometers of vacuum' is a resource, a 'new frontier' like America was in the 1500s. But in reality space seems more like a hole full of nothing. There's no tobacco, no tomatoes, no corn, no coffee, no beavers, no buffalo out there. Nothing to bring back in the whole solar system except dry dust, rock and sunlight, and we already have those. And it's thousands of years to the next star at plausible acceleration rates unless you can beat Einstein.
"Spending all the money fixing this world does nothing to get all of our eggs out of the basket, and if anything harms that basket, then we are screwed."
But if, for example, a virus wipes out the Earth... if you've got a space infrastructure, how is that virus not going to spread through your network of ships and stations as well?
And what tiny percentage of the human population is ever going to be able to inhabit space stations? Currently we're doing, what, three to eight people at once? How many could we boost that up to given that in space, water and oxygen are rare, and we don't even know if edible plantsf really can grow in lunar soil, let alone places further out like Mars. Then there's radiation, genetic fragility...
It seems to me that unless we discover warp drive (so we can get Earthlike planets for free), 'expanding into space' will be slow, inherently self-limiting, and only ever available to a tiny scientific/industrial elite who are supported by a huge Earth-based resource pyramid.
At best we might learn how to maintain self-supporting ecologies, which isn't nothing, but it's not like space is a 'solution' for anything; when it comes down to it, there just aren't any useful resources out there except rock. So barring a major overthrow of what scientific data we've collected so far, it seems like just a very expensive and ultimately pointless hobby for the ultra-rich or ultra-lucky.
"I restart everything in X. It's quick, it's painless and way quicker than rebooting the whole computer. "
That's an illusion though. It's a little faster, sure, but in terms of the potential for losing user data, restarting X is just as invasive as restarting the whole box. It is most certainly NOT 'painless' unless all your documents are already saved and you've configured GNOME to reload all open apps on startup - and even then it's not quick.
One thing I've never understood. X contains within itself the built-in capability for the X server to run as a completely different process - heck, even on a different computer - than the X apps. In which case, if the X server crashed it wouldn't take out the applications and could be safely restarted. And yet, on all our Linux builds, we rig it so that gdm launches X and then launches the applications, so restarting X kills all the applications. Why do we do this? Wouldn't it be more stable if we separated the two? We'd actually be USING the neat distributed features of the X Protocol.
Case in point: Bug 332949, the update-notifier 'upgrade' in Jaunty which most users agree is functionality they want which was working, but Mark Shuttleworth thought he'd change because he could.
"One thing that Microsoft has done well is to maintain continuity with the past. The desktop of Windows 95 is still available on all consumer versions of Windows up to Windows 7."
They did, until Office 2007. Getting rid of the menu bar was a huge break with backwards compatibility, and IMO not at all a good one.
"Honestly, I don't think that kind of UI design is all that critical. If it'd been a few steps higher up like workflow design, then I'm all with you. "
And here's my question to you:
Why can't the user create their *own* workflow on a modern desktop?
Why aren't there tools available to allow the user to script and remix modes and functions of applications into their own new applications?
Why can't I take a GUI application that annoys me because the buttons are laid out wrong, and edit the window so the buttons are 35% bigger and slightly to the left, then post just that change somewhere safely on the Web so others can critique it and use it?
If I see a spelling error in a dialog box in a free application, why can't I *instantly* click somewhere, fix that error, and repost it, just like I can on Wikipedia?
How much do I need to know about thread-safe signal-handling GUI event loops in order to change a badly-drawn icon or resize a scroll bar?
We're not leveraging the full power of the Free Software mentality unless we can enable small, safe, incremental fixes like this, all across the user base.
"What if what happens to developers is that they don't give a damn about what "the users" want or need?"
Precisely, and this I think is why we need to erase the developer/user distinction by creating languages and tools which allow USERS to create their own solutions. Then the users BECOME the developers, and make the system they want.
This doesn't happen at the moment because it's too hard - there's a big gulf between 'user' and 'developer' level tools. The conventional default way of creating GUI apps - C/C++ plus huge unwieldy libraries - is not a good fit with how users want to evolve their own interface experience. This separation between user/developer has emerged from the commercial history of software development. But it doesn't have to be like that. We could and should do much, much better.
Don't you think it's insane that you get a program like, say, Evolution or F-Spot, and you get it as a big monolithic whole - windows, widgets, control logic, database, storage formats, all rolled together, with 'no user-servicable parts'? And no way of configuring it between ticking the few GNOME HIG-compliant checkboxes (the fewer the better) or learning Mono and Gtk and hacking the source directly - then fighting with version control and package management and create my own rogue repository somewhere.
If we had a way where users could make tweaks to things like how GUI windows were laid out - in such a way that control logic and data storage and transfer logic were cleanly separated from interface concerns, so people could easy share and modify and create whole new 'skins' for products, much like how the Firefox extension ecosystem works, or how spreadsheet templates can be created, wouldn't it help bridge that gap?
Why, for example, can't I use Gconf, make a bunch of configuration settings, then save them as a template file and export and share them to users as a 'configuration mode' or something for a program I like? Why do I have to make all configuration changes from scratch, by hand, by myself, with no modularity?
And isn't this exactly what the whole Object-Oriented and Component Programming revolution way back in the 80s was supposed to be about? Reusable code? Why didn't it happen? Why did OOP stay stuck in the 'use objects to build big applications, then ship them as entire systems plus a few shared libraries which are so fragile they have to be centrally managed at the OS level' mode?
"A library which removes 'noise' is also removing signal that I want. "
Yes, exactly. Especially if you're interesting in anomaly research, fringe science, and the paranormal, which I am. A lot of valuable research and case histories get censored from official channels by overzealous self-appointed guardians of "rationality"; you can only locate this information via the Net.
Yes, there might be false information and lots of teeny-bopper emo drama angst. But I'm the one who wants to apply my own filter, because I've discovered that the official filters are worse than useless for the kind of research I'm doing.
The Net lets us bypass the gatekeepers and get straight to primary sources. That can be good and bad. Like C, it lets you shoot yourself in the foot. But sometimes you need the raw power of unfiltered data, and it can be actively harmful to aggressively filter content in service of a "truth maintenance" agenda.
"There is so much crap on the internet that it undermines all the information that is out there. Conversely, if you go to the 500 and 600 sections of the library, you can be somewhat assured that you are getting at least -something- that is accurate."
In other words, a library is better because it is censored by central knowledge-gatekeepers, while the Internet is worse because it is decentralised and democratic?
While that may be true, it seems like it's a viewpoint that has interesting political consequences, if followed through rigorously.
Exactly. Also, if you're interested in somewhat off-the-beaten-tracks topics, like say ufology, you're much more likely to find interesting resources on the Internet than at your library. For example, the Journal of Scientific Exploration isn't likely to be carried at your local.
I was a library rat since I was a preteen, but my knowledge base only really started to expand once the Web - and then Wikipedia - came online. Most of the topics I wanted to know about were very, very esoteric and specialised and most of the published books I could find were at the popular digest level, third- or fourth-hand summaries.
And far worse, most of the teen science fiction books I grew up with and loved got purged mercilessly in the 1990s. Great young adult writers like Nicholas Fisk and John Christopher and Douglas Hill have been erased from the stacks, just because... I don't understand why, really, just they weren't considered sufficiently 'relevant' I suppose. So libraries don't always keep the important works around.
Having said that, I do love libraries, the feel and the smell of them, although I switched to bookshops in my thirties and started specifically ordering books I'd find online. I can walk into any given library and instantly get sucked into a book; it's a bit of a hazard. I think the best way is a symbiosis between the Internet for cataloguing and exotic/up-to-date material, and well-managed libraries for finding the rest.
I read online obsessively, but I also buy print books.
Increasingly, I'm finding Google Books a big temptation. Through a random search somewhere I discovered Strategic Computing, and after reading the first couple of chapters online decided heck with it and placed an order at my local bookstore for the real thing.
Without Google Books and the ability to read text I would never have done this.
One data point maybe, but for me it's pretty obvious: digitising books in the way that Google is doing doesn't replace print books, it promotes them.
Ray Bradbury's welcome to miss out on this if he wants. He's always been anti-technology, but I didn't think he was anti *reading*. Guess I was wrong.
So then his drones will fight the government drones, and the TV newsdrones will be recording it, and then the SWAT drones will turn up, and then the blogger drones, and before you know it some smart-alec kid's wearing an EVA-01 suit and that's when things get *really* out of hand.
"As our economy becomes increasingly information-based and decreasingly manufacturing-based, old manufacturing centers are crumbling. But a new economy is rising, one based on information and information technology."
Slashdot Mirror
"You should be employing more than one manager. One of them should be YOU. "
And if you've read/watched "A Scanner Darkly", you should know that even YOU might not be fully trustworthy, and you should consider diversifying into multiple personalities to keep an eye on that sneaky...
Paranoia. It's the sign of a healthy, prosperous society!
"You tell me college football, if you have possession of the ball and your knee touches the ground, the ball is down, whether or not the player was tackled. I give you a college football game, and the first time you try to kick a field goal, the ball is downed 7 yards behind the line of scrimmage because the holder's knee is touching the ground."
Can you translate that for those of us who only speak Rugby?
"We tried opening it and some guy's face melted."
WONTFIX. This behaviour is by design. RTFM.
Amadeus are you okay, Amadeus are you okay, are you okay Amadeus...
In Advanced Facebook the final level boss is *yourself*.
Oh smeg is right, matey.
"At the moment, finding the few billions it might take to get us off-world just seems expensive, but at some point finding those billions may actually require taking the decision to stop feeding some people and that will be a tough decision for anyone to make! "
And how exactly is getting a few dozen people into space going to help feed anything?
Plants don't grow on vacuum alone. Here on Earth we have a whole biosphere for free. If we can't preserve that, we sure won't be able to grow a whole new one from scratch.
Conversely, if we have the technology to grow new biospheres from scratch, we can do it on Earth much more cheaply than doing it in space.
You seem to be under the impression that 'trillions of cubic kilometers of vacuum' is a resource, a 'new frontier' like America was in the 1500s. But in reality space seems more like a hole full of nothing. There's no tobacco, no tomatoes, no corn, no coffee, no beavers, no buffalo out there. Nothing to bring back in the whole solar system except dry dust, rock and sunlight, and we already have those. And it's thousands of years to the next star at plausible acceleration rates unless you can beat Einstein.
"Spending all the money fixing this world does nothing to get all of our eggs out of the basket, and if anything harms that basket, then we are screwed."
But if, for example, a virus wipes out the Earth... if you've got a space infrastructure, how is that virus not going to spread through your network of ships and stations as well?
And what tiny percentage of the human population is ever going to be able to inhabit space stations? Currently we're doing, what, three to eight people at once? How many could we boost that up to given that in space, water and oxygen are rare, and we don't even know if edible plantsf really can grow in lunar soil, let alone places further out like Mars. Then there's radiation, genetic fragility...
It seems to me that unless we discover warp drive (so we can get Earthlike planets for free), 'expanding into space' will be slow, inherently self-limiting, and only ever available to a tiny scientific/industrial elite who are supported by a huge Earth-based resource pyramid.
At best we might learn how to maintain self-supporting ecologies, which isn't nothing, but it's not like space is a 'solution' for anything; when it comes down to it, there just aren't any useful resources out there except rock. So barring a major overthrow of what scientific data we've collected so far, it seems like just a very expensive and ultimately pointless hobby for the ultra-rich or ultra-lucky.
"Wolfram Alpha and Bing have recently shown that there is room even in search for serious innovation"
Bing innovates? How?
"I restart everything in X. It's quick, it's painless and way quicker than rebooting the whole computer. "
That's an illusion though. It's a little faster, sure, but in terms of the potential for losing user data, restarting X is just as invasive as restarting the whole box. It is most certainly NOT 'painless' unless all your documents are already saved and you've configured GNOME to reload all open apps on startup - and even then it's not quick.
One thing I've never understood. X contains within itself the built-in capability for the X server to run as a completely different process - heck, even on a different computer - than the X apps. In which case, if the X server crashed it wouldn't take out the applications and could be safely restarted. And yet, on all our Linux builds, we rig it so that gdm launches X and then launches the applications, so restarting X kills all the applications. Why do we do this? Wouldn't it be more stable if we separated the two? We'd actually be USING the neat distributed features of the X Protocol.
Case in point: Bug 332949, the update-notifier 'upgrade' in Jaunty which most users agree is functionality they want which was working, but Mark Shuttleworth thought he'd change because he could.
"One thing that Microsoft has done well is to maintain continuity with the past. The desktop of Windows 95 is still available on all consumer versions of Windows up to Windows 7."
They did, until Office 2007. Getting rid of the menu bar was a huge break with backwards compatibility, and IMO not at all a good one.
And here's a follow-up question set:
What might be the GUI software equivalent of Wikipedia's 'page' and 'edit history' and 'locking'?
Can our current GUI programming languages support safe, massively distributed user-led editing like Wikipedia?
If not, why not, and how can we evolve them towards such a massively distributed, integrated, language / protocol / environment?
Do C/C++/Java/Mono allow us to even think about this kind of paradigm? What about Ruby or Python?
If neither, can we envisage a suitable language which might?
"Honestly, I don't think that kind of UI design is all that critical. If it'd been a few steps higher up like workflow design, then I'm all with you. "
And here's my question to you:
Why can't the user create their *own* workflow on a modern desktop?
Why aren't there tools available to allow the user to script and remix modes and functions of applications into their own new applications?
Why can't I take a GUI application that annoys me because the buttons are laid out wrong, and edit the window so the buttons are 35% bigger and slightly to the left, then post just that change somewhere safely on the Web so others can critique it and use it?
If I see a spelling error in a dialog box in a free application, why can't I *instantly* click somewhere, fix that error, and repost it, just like I can on Wikipedia?
How much do I need to know about thread-safe signal-handling GUI event loops in order to change a badly-drawn icon or resize a scroll bar?
We're not leveraging the full power of the Free Software mentality unless we can enable small, safe, incremental fixes like this, all across the user base.
"What if what happens to developers is that they don't give a damn about what "the users" want or need?"
Precisely, and this I think is why we need to erase the developer/user distinction by creating languages and tools which allow USERS to create their own solutions. Then the users BECOME the developers, and make the system they want.
This doesn't happen at the moment because it's too hard - there's a big gulf between 'user' and 'developer' level tools. The conventional default way of creating GUI apps - C/C++ plus huge unwieldy libraries - is not a good fit with how users want to evolve their own interface experience. This separation between user/developer has emerged from the commercial history of software development. But it doesn't have to be like that. We could and should do much, much better.
Don't you think it's insane that you get a program like, say, Evolution or F-Spot, and you get it as a big monolithic whole - windows, widgets, control logic, database, storage formats, all rolled together, with 'no user-servicable parts'? And no way of configuring it between ticking the few GNOME HIG-compliant checkboxes (the fewer the better) or learning Mono and Gtk and hacking the source directly - then fighting with version control and package management and create my own rogue repository somewhere.
If we had a way where users could make tweaks to things like how GUI windows were laid out - in such a way that control logic and data storage and transfer logic were cleanly separated from interface concerns, so people could easy share and modify and create whole new 'skins' for products, much like how the Firefox extension ecosystem works, or how spreadsheet templates can be created, wouldn't it help bridge that gap?
Why, for example, can't I use Gconf, make a bunch of configuration settings, then save them as a template file and export and share them to users as a 'configuration mode' or something for a program I like? Why do I have to make all configuration changes from scratch, by hand, by myself, with no modularity?
And isn't this exactly what the whole Object-Oriented and Component Programming revolution way back in the 80s was supposed to be about? Reusable code? Why didn't it happen? Why did OOP stay stuck in the 'use objects to build big applications, then ship them as entire systems plus a few shared libraries which are so fragile they have to be centrally managed at the OS level' mode?
"So the Internet is a series of tubes in the air somewhere...?"
Pneumatic tubes.
"A library which removes 'noise' is also removing signal that I want. "
Yes, exactly. Especially if you're interesting in anomaly research, fringe science, and the paranormal, which I am. A lot of valuable research and case histories get censored from official channels by overzealous self-appointed guardians of "rationality"; you can only locate this information via the Net.
Yes, there might be false information and lots of teeny-bopper emo drama angst. But I'm the one who wants to apply my own filter, because I've discovered that the official filters are worse than useless for the kind of research I'm doing.
The Net lets us bypass the gatekeepers and get straight to primary sources. That can be good and bad. Like C, it lets you shoot yourself in the foot. But sometimes you need the raw power of unfiltered data, and it can be actively harmful to aggressively filter content in service of a "truth maintenance" agenda.
"There is so much crap on the internet that it undermines all the information that is out there. Conversely, if you go to the 500 and 600 sections of the library, you can be somewhat assured that you are getting at least -something- that is accurate."
In other words, a library is better because it is censored by central knowledge-gatekeepers, while the Internet is worse because it is decentralised and democratic?
While that may be true, it seems like it's a viewpoint that has interesting political consequences, if followed through rigorously.
Exactly. Also, if you're interested in somewhat off-the-beaten-tracks topics, like say ufology, you're much more likely to find interesting resources on the Internet than at your library. For example, the Journal of Scientific Exploration isn't likely to be carried at your local.
I was a library rat since I was a preteen, but my knowledge base only really started to expand once the Web - and then Wikipedia - came online. Most of the topics I wanted to know about were very, very esoteric and specialised and most of the published books I could find were at the popular digest level, third- or fourth-hand summaries.
And far worse, most of the teen science fiction books I grew up with and loved got purged mercilessly in the 1990s. Great young adult writers like Nicholas Fisk and John Christopher and Douglas Hill have been erased from the stacks, just because... I don't understand why, really, just they weren't considered sufficiently 'relevant' I suppose. So libraries don't always keep the important works around.
Having said that, I do love libraries, the feel and the smell of them, although I switched to bookshops in my thirties and started specifically ordering books I'd find online. I can walk into any given library and instantly get sucked into a book; it's a bit of a hazard. I think the best way is a symbiosis between the Internet for cataloguing and exotic/up-to-date material, and well-managed libraries for finding the rest.
I read online obsessively, but I also buy print books.
Increasingly, I'm finding Google Books a big temptation. Through a random search somewhere I discovered Strategic Computing, and after reading the first couple of chapters online decided heck with it and placed an order at my local bookstore for the real thing.
Without Google Books and the ability to read text I would never have done this.
One data point maybe, but for me it's pretty obvious: digitising books in the way that Google is doing doesn't replace print books, it promotes them.
Ray Bradbury's welcome to miss out on this if he wants. He's always been anti-technology, but I didn't think he was anti *reading*. Guess I was wrong.
"Laser fusion, on the other hand, is just there so the Dept. of Energy can make better simulation of nukes in their supercomputers."
And that's what pays the bills in the military-industrial complex, as it always has.
It's like Slashdot, but on hash.
News for stoners. Stuff that... wow.
The world's #1 rouge nation is France.
Makes me blush to think of it.
So then his drones will fight the government drones, and the TV newsdrones will be recording it, and then the SWAT drones will turn up, and then the blogger drones, and before you know it some smart-alec kid's wearing an EVA-01 suit and that's when things get *really* out of hand.
"As our economy becomes increasingly information-based and decreasingly manufacturing-based, old manufacturing centers are crumbling. But a new economy is rising, one based on information and information technology."
And hunter-killer robots, right?
"3. We build the crap out of modern safe designs for fission plants and let that hold us until fusion finally gets into production."
1. Two words: Peak uranium
2. Never mind "production", when is fusion even projected to demonstrate theoretical breakeven?