Rudy still makes the same noises in Nintendo's Pinball Hall of Fame Williams Collection. Taxi makes the sound of a cat getting run over, Pinbots... you name it. I bought it new - 10 games for $20 - a few years ago - everyone likes pinball games because they're so easy to play, and so hard to beat.
Xandros - a friend of mine was using it for years to go online to protect his computer. It did everything he wanted, and he would have paid for updates, same as Apple charges.
On "How will you write "for" loops with a GUI" - Computer Associates dbFast for Windows 3x FTW. It was a cute product, but I quickly stopped using the GUI development tools and just wrote code, because it was easier to modify.
I wouldn't be too harsh on Robert - we've all been there at one time or another, either with F/LOSS or with something else in our lives.
Now here's a question - is there room for a PAY-FOR proprietary (as in "it can actually generate sales and has to be responsive to user demands") OS that doesn't have the problems of updates b0rking systems (no dynamic linking, etc), simpified installation and management (none of this UUID= garbage in grub, for example*, scripts that are dead simple and GUIs for each one), and a common-sense desktop?
* people who make excuses for that sort of stuff don't realize that it ultimately causes more problems than it solves, and treating anyone who has to move hard drives around as an idiot is a poor policy. It's their box - let them mess around with fstab as they want, w/o this uuid crap.
Cut workforce, moved a lot of stuff to other countries, and most important - it's because they're hugely profitable and pay a regular dividend.
Investors like profits, but they LOVE dividends, especially in this economy.
So when it comes to OS vendors, we have Apple being the most profitable, Microsoft at #2, Google at #3, and RedHat at #4.
Any distro except RedHat running at a profit? Canonical? Nope, never did, never will. Mandriva? Struggling to stay out of bankruptcy court. Suse? Dependent on another $100 million of "license purchases" over the next 4 years since it was purchased by Microsoft's new BFF Attachmate. Fedora? Supported by RedHat. Debian? Survives on donations, not really enough to pay for staff. Mint? Thanks to Canonical dropping the ball, Mint actually now has two paid full-time staff!!!.
None of the distros, with the exception of RedHat, is something that an investor should put serious money into. And that's unfortunate, but it should tell you that fragmentation is a disease, and maybe Canonical closing up shop sometime in the next couple of years will wake people up.
Well, @$50 a copy, people would buy it, AND it would get rid of the legacy copies of XP running after it's been EOL'd (because people will continue to run them, and when they get too messed up, just restore from a disk image or re-install).
I don't see the linux desktop usage numbers ever going above where they are today - they haven't really moved in a decade. Even Novell, with its decades of experience, couldn't get people to renew their SLES licneses (the ones Microsoft sold customers) w/o giving discounts of 80% to 90% - and Novell never made a profit with linux. I wonder how much they would have lost w/o Microsoft sending hundreds of millions their way... but with numbers like that, a "deal with the devil" was their only alternative.
I was thinking exclusively corporate support for mixed environments. Marketing to consumers is a losing proposition, as you pointed out. Consumers rightly take the need for an extended warranty as an indicator that the product is crap.
I personally think it's going to be neck-and-neck which is no longer "officially supported" - Ubuntu or XP... it's going to be a tight race, for sure - and when XP is finally EOL, it will probably still have 20x the user base Ubuntu has (though at the current rate of attrition, it may be 100x in 2 years), so much for "fixing bug #1".
Basically, the only ways to monetize the desktop in meaningful numbers are to create a proprietary desktop, or create a proprietary application for an existing desktop. Free just doesn't work any more than free cars works.
Take transportation as an example. Free cars won't work - it just costs to much to give everyone a free car, and increases pollution, traffic, and road wear and tear. Free buses, on the other hand, could work because they reduce the need for capital outlays for wider roads, bridges, etc., as well as ongoing maintenance, resulting in enough savings to offset the cost of free buses. So buses are an indirect investment in infrastructure, but cars aren't. Linux is infrastructure. It's worth it for companies to invest in infrastructure, because they share in the benefits. It's not worth it for them to give everyone a free car.
Please remember, we're talking Unity here - something that was supposed to be for mobile devices. Typing is a PITA in a touch-screen environment. You have to launch the keyboard, then "type" on it (nothing nearly as good as a real keyboard), then "disappear" the keyboard again. So the HUD is slower.
Second, if you have 10 apps that you want to launch quickly, what's the problem? Most DEs support a place to dock launchers for your frequently-used apps. Or you could just stick them in your rc file.
Third, Alt-F2 works on most *nix DEs no need for anything "special".
But using Blobs for storing email isn't usually a good idea in a db. More easier to save it as a separate file referenced by the record on the db.
Not really - each file uses up an inode, so there's 4k gone per file. A better solution would be to store an arbitrary number of emails in each file, compressed and then concatenated, and just store the filename:offset:length of each one in the db. Each individual email is quickly recovered, way fewer inodes used.
There *is* one way... but people will howl and scream like crazy. There are work-arounds that let you use GPL'd code in closed-source products, completely within the restrictions of current copyright law - and without resorting to TIVOization.
The question is, is it even worth it? Probably not, the market just isn't there. That's the real Bug #1, and it's not going to change, because TANSTAAFL. Linux is great for infrastructure, and for people who know what they're doing (for example, I'm currently running Fedora 16, but I certainly wouldn't just give an install disk to someone and expect them to be able to use it - most of their programs won't run, which is the root cause of the real-life Bug #1).
Bug #0, on the other hand, is social. It's the GPL itself, which has turned software into a closed-minded quasi-pseudo-religion for too many people. It's why more projects are switching to free/non-restrictive ABM licenses (Apache, BSD, MIT). They still get code contributions back, but they get them because the contributors find it's more effective in the long term, not because they "have to." Cooperation, not coercion.
An awful lot of what you say sounds like something I wrote over the weekend (to be seen elsewhere in a week).
In the meantime, remember this? Doesn't the Google Heads Up Display stuff sound an awful lot like the first step? So let them mod down all they want - I get it right more often than not, and there's only one way to salvage Canonical
0. Drop the Ubuntu brand - it's tarnished beyond repair.
1. On the support side - support ALL distros, ALL OSes. One-stop tech support for small and medium businesses and local governments. Or close it. Nothing in between will work - it's too niche.
2. On the projects side - cancel Unity, cancel UbuntuTV (it's just samygo.tv anyway), stop pretending that this "Ubuntu on Android" is anything new (and don't try to distribute the Debian hack from last summer that it's based on - it's not worth getting sued by Googlerola), cancel Ubuntu 1, cancel Ubuntu cloud.
3. As a show of good faith, immediately reinstate Kubuntu as a supported spin. Ditto for Xubuntu.
4. Cancel Bug #1 - it's childish in a world where interoperability is more important.
5. Add value by actually adding value, not surfing the net for ideas other people have developed (like UbuntuTV and Ubuntu on Android or whatever they call it) and slapping your name on it.
6. Fire all the disloyal people (the ones who were so busy being "yes-men" and "yes-women" rather than saying "hey, this is ****ed up, we need to get back to basics and stop throwing our users under the nearest bus").
Of course, the real question is, will the Self-Appointed Benevolent Dictator For Life do that, or just pull the plug?
Canonical had a goal of "fixing bug #1 - Windows majority market share" back in 2004. Here we are, 2012, and creaky old XP has 20x the market share.
Last year Shuttleworth set a goal of 200 million Ubuntu users by 2015. Since then, he's thrown all the Kubuntu users under the bus, same as he did to the Gnome users.
This latest announcement is old news - others have been running Debian on the same hardware combo since last summer. But seriously, turning a smartphone into a netbook? And one that can't even run Android apps when in "netbook mode"?????!!!!???? No OEM is going to bother with this. Netbooks are dead. And with the higher-res screens coming to tablets, Unity will be sub-optimal. Just like the idea of having to type stuff to run a program (their stupid "Heads-Up Display") is a step backwards for users.
That's okay - Mint was happy to take most of those disaffected Ubuntu users, along with their mindshare, and Fedora, Arch, and a few other distros are also seeing significant upticks. Choice is good, right?
Employees trying to justify being yes-persons to Shuttleworth's "Self-Appointed Benevolent Dictator for Life" so they can continue to collect a paycheck, duh!
Nobody's going to tell him his ideas are seriously out of date, that Canonical is a failure from the venture capital perspective, and that they should never have abandoned the goal of making the best desktop distro, period, instead of turning around and throwing their user base under the bus.
That's okay - Canonical's mistakes are encouraging others to step in to fill the void (Mint, for example), so in the end, it's all good, right?
If you really consider this as "Ubuntu for Android" (it's not, it's just that the Atrix originally came with a cut-down linux for when it was docked, and Android for when it wasn't - you're just hibernating one and waking up the other, and sharing the data files and some hardware), and Canonical doesn't want to share it, use last year's Debian on Atrix - same hardware, 7 months earlier.
Just like last months' UbuntuTV was samygo.tv running Ubuntu instead of whatever linux distro you wanted to run.
Who knows what "new" product Canonical will hype next month that someone already demoed last year? Even Matt Asay, their former COO, thinks they're basically scr*wed.
There wasn't much "work" to do - the Atrix already ran a cusomized linux when it was docked - all Canonical did was swap it out for Ubuntu. Debian on the Atrix - Last Summer.
Turning your smartphone into a netbook - why? Netbooks are dead, and nobody is going to carry around a docking station. The Atrix is a marketing failure, and those stupid commercials with the TSA guys didn't help.
All they did, same as Canonical, was replace the customized linux os that was running when it's docked with Ubuntu, same as the Debian guy swapped it out for Debian.
And let's be honest, nobody wants to haul around a docking station that turns your smartphone into a netbook. Even Matt Asay - Canonical's former COO - says Canonical lacks focus and that netbooks are dead - two facts most of us have been aware of for a while... so be happy you didn't blow millions trying to make this turkey fly.
You can't run them "side-by-side" in the conventional sense. The hardware activates one OS image or the other. And none of this is new - here's a video of Debian running on the same device, back in August of last year http://www.youtube.com/watch?v=G8-92J9hfkA
Canonical announced their "Android Execution Environment" 3 years ago, then abandoned it 2 years ago because they couldn't do it.
This is as "innovative" as UbuntuTV was - which was just Canonical customizing the freely-available samygo.tv software to run Ubuntu instead of another distro http://www.samygo.tv/ You too can have your own brand TV distro running right inside your TV - Slackware, RedHat, Debian, even MythTV...
Of course, the other question is why anyone would want to run a half-baked "mobile UI" on a desktop display like they propose...
Sorry, you're right. I should have done better - both OS images are stored in the phone, it's just the display, kb, etc., that are external. btw, here's a video of Debian on the Atrix from August 2nd, 2011.
It's because it's not actually "running in the smartphone" - all they did was replace the linux os in the Lapdock100 (which is a separate accessory for the smartphone) with Ubuntu.
In other words, nothing to see here. Ubuntu still can't run Android, or Android apps, almost 3 years after Canonical announced their "Android Execution Environment."
Actually, you have to run them side-by-side. Canonical failed to deliver their "Android Execution Environment" that they announced withi such a big splash in 2009.
It's simple, really - you use the existing hypervisor in the Atrix, and just replace the desktop os with the os of your choice.
Freetards - alive and well, and now with twice the stupidity.
Re: TV - over-modulated ads haven't been much of a problem since BEFORE the regulations requiring normalization of sound - unless you bought a cheap set that didn't have auto volume control. As for the length of time of TV ads, I prefer a 3-minute block of ads - it lets me put away the dishes or fold some laundry or make the bed or any one of a number of things that need to be done.
Re: Radio - if you don't like the ads on it, again, the technical solution is to carry your own music, audiobooks, etc.
Re: Streaming video: so pay to stream stuff ad-free (oh wait, you want it for nothing)
Re: Websites: So go elsewhere. Start your own. Go into the big blue room and take a break. Relax.
Re: Electronic billboards: Get a better-shielded radio, duh! Or some duct tape and tin foil...
Re: Email: so what's the problem again, since you don't use webmail or html mail? Oh right, there isn't one, you just want to complain.
Re: Search engines: So instead of using them, just bookmark the damn site once and be done with it. Is it *that* hard? Or create your own. It's not that hard to write a crawler. It's even easier to pay someone else to, if you can't. Just put your money where your mouth is.
Say Windows determines Photoshop is a frequently used program, after it's loaded for the first time, if the user closes it, the entire program should stay resident in memory unless that space is needed by something else. So if Photoshop is loaded again, it should be able to know if it's already in memory, and skip all of the splash screen garbage and just get down to business.
One reason to close the program between uses is to release leaked memory. Memory might be cheap, but it's not infinite.
There are a few technical problems with your proposal.
1. Programs that save their state when closing - none of them will have the same screen layout next time around.
2. Changed screen dimensions / colour depths / dpi between runs.
3. The MRU list (most recently used files list) will still have to be rebuilt from scratch anyway. So will any containers and parent objects/data structures
4. Updates to the software will also trigger a rebuild
5. Updates to system libraries will cause it to um... behave erratically, since the blob maps to old library entry points.
6. Any memory leaks become cumulative... not a good idea.
People don't mind the advertising so much - it's the tracking that bugs us the most. The all-prevasive, obsessive-cumpulsive, privacy-invading stalking behavioural tracking...
Just like we don't mind ads on TV - we either watch them, channel surf, or go do a couple minutes worth of housework, but if TV ads watched us back, we'd find that unacceptable as well.
Cookies to preserve state in client-server web apps == good. Cookies to track peoples behaviour == bad. Cross-site cookies == bad by definition.
How would you suggest that they implement the "+1" feature on other web sites without doing this?
Short answer - don't! I'm sick and tired of all this bogus tracking crap. If you absolutely insist on having a +1, then also have a -1, so we can tell you how we really feel! Same as if you're going to have a Like button, also have a Hate button.
Slashdot Mirror
Rudy still makes the same noises in Nintendo's Pinball Hall of Fame Williams Collection. Taxi makes the sound of a cat getting run over, Pinbots ... you name it. I bought it new - 10 games for $20 - a few years ago - everyone likes pinball games because they're so easy to play, and so hard to beat.
Xandros - a friend of mine was using it for years to go online to protect his computer. It did everything he wanted, and he would have paid for updates, same as Apple charges.
On "How will you write "for" loops with a GUI" - Computer Associates dbFast for Windows 3x FTW. It was a cute product, but I quickly stopped using the GUI development tools and just wrote code, because it was easier to modify.
I wouldn't be too harsh on Robert - we've all been there at one time or another, either with F/LOSS or with something else in our lives.
Now here's a question - is there room for a PAY-FOR proprietary (as in "it can actually generate sales and has to be responsive to user demands") OS that doesn't have the problems of updates b0rking systems (no dynamic linking, etc), simpified installation and management (none of this UUID= garbage in grub, for example*, scripts that are dead simple and GUIs for each one), and a common-sense desktop?
* people who make excuses for that sort of stuff don't realize that it ultimately causes more problems than it solves, and treating anyone who has to move hard drives around as an idiot is a poor policy. It's their box - let them mess around with fstab as they want, w/o this uuid crap.
Cut workforce, moved a lot of stuff to other countries, and most important - it's because they're hugely profitable and pay a regular dividend.
Investors like profits, but they LOVE dividends, especially in this economy.
So when it comes to OS vendors, we have Apple being the most profitable, Microsoft at #2, Google at #3, and RedHat at #4.
Any distro except RedHat running at a profit? Canonical? Nope, never did, never will. Mandriva? Struggling to stay out of bankruptcy court. Suse? Dependent on another $100 million of "license purchases" over the next 4 years since it was purchased by Microsoft's new BFF Attachmate. Fedora? Supported by RedHat. Debian? Survives on donations, not really enough to pay for staff. Mint? Thanks to Canonical dropping the ball, Mint actually now has two paid full-time staff!!!.
None of the distros, with the exception of RedHat, is something that an investor should put serious money into. And that's unfortunate, but it should tell you that fragmentation is a disease, and maybe Canonical closing up shop sometime in the next couple of years will wake people up.
Well, @$50 a copy, people would buy it, AND it would get rid of the legacy copies of XP running after it's been EOL'd (because people will continue to run them, and when they get too messed up, just restore from a disk image or re-install).
I don't see the linux desktop usage numbers ever going above where they are today - they haven't really moved in a decade. Even Novell, with its decades of experience, couldn't get people to renew their SLES licneses (the ones Microsoft sold customers) w/o giving discounts of 80% to 90% - and Novell never made a profit with linux. I wonder how much they would have lost w/o Microsoft sending hundreds of millions their way ... but with numbers like that, a "deal with the devil" was their only alternative.
I was thinking exclusively corporate support for mixed environments. Marketing to consumers is a losing proposition, as you pointed out. Consumers rightly take the need for an extended warranty as an indicator that the product is crap.
I personally think it's going to be neck-and-neck which is no longer "officially supported" - Ubuntu or XP ... it's going to be a tight race, for sure - and when XP is finally EOL, it will probably still have 20x the user base Ubuntu has (though at the current rate of attrition, it may be 100x in 2 years), so much for "fixing bug #1".
Basically, the only ways to monetize the desktop in meaningful numbers are to create a proprietary desktop, or create a proprietary application for an existing desktop. Free just doesn't work any more than free cars works.
Take transportation as an example. Free cars won't work - it just costs to much to give everyone a free car, and increases pollution, traffic, and road wear and tear. Free buses, on the other hand, could work because they reduce the need for capital outlays for wider roads, bridges, etc., as well as ongoing maintenance, resulting in enough savings to offset the cost of free buses. So buses are an indirect investment in infrastructure, but cars aren't. Linux is infrastructure. It's worth it for companies to invest in infrastructure, because they share in the benefits. It's not worth it for them to give everyone a free car.
Please remember, we're talking Unity here - something that was supposed to be for mobile devices. Typing is a PITA in a touch-screen environment. You have to launch the keyboard, then "type" on it (nothing nearly as good as a real keyboard), then "disappear" the keyboard again. So the HUD is slower.
Second, if you have 10 apps that you want to launch quickly, what's the problem? Most DEs support a place to dock launchers for your frequently-used apps. Or you could just stick them in your rc file.
Third, Alt-F2 works on most *nix DEs no need for anything "special".
Next thing you know, they'll be snorting coke!
Not really - each file uses up an inode, so there's 4k gone per file. A better solution would be to store an arbitrary number of emails in each file, compressed and then concatenated, and just store the filename:offset:length of each one in the db. Each individual email is quickly recovered, way fewer inodes used.
There *is* one way ... but people will howl and scream like crazy. There are work-arounds that let you use GPL'd code in closed-source products, completely within the restrictions of current copyright law - and without resorting to TIVOization.
The question is, is it even worth it? Probably not, the market just isn't there. That's the real Bug #1, and it's not going to change, because TANSTAAFL. Linux is great for infrastructure, and for people who know what they're doing (for example, I'm currently running Fedora 16, but I certainly wouldn't just give an install disk to someone and expect them to be able to use it - most of their programs won't run, which is the root cause of the real-life Bug #1).
Bug #0, on the other hand, is social. It's the GPL itself, which has turned software into a closed-minded quasi-pseudo-religion for too many people. It's why more projects are switching to free/non-restrictive ABM licenses (Apache, BSD, MIT). They still get code contributions back, but they get them because the contributors find it's more effective in the long term, not because they "have to." Cooperation, not coercion.
In the meantime, remember this? Doesn't the Google Heads Up Display stuff sound an awful lot like the first step? So let them mod down all they want - I get it right more often than not, and there's only one way to salvage Canonical
0. Drop the Ubuntu brand - it's tarnished beyond repair.
1. On the support side - support ALL distros, ALL OSes. One-stop tech support for small and medium businesses and local governments. Or close it. Nothing in between will work - it's too niche.
2. On the projects side - cancel Unity, cancel UbuntuTV (it's just samygo.tv anyway), stop pretending that this "Ubuntu on Android" is anything new (and don't try to distribute the Debian hack from last summer that it's based on - it's not worth getting sued by Googlerola), cancel Ubuntu 1, cancel Ubuntu cloud.
3. As a show of good faith, immediately reinstate Kubuntu as a supported spin. Ditto for Xubuntu.
4. Cancel Bug #1 - it's childish in a world where interoperability is more important.
5. Add value by actually adding value, not surfing the net for ideas other people have developed (like UbuntuTV and Ubuntu on Android or whatever they call it) and slapping your name on it.
6. Fire all the disloyal people (the ones who were so busy being "yes-men" and "yes-women" rather than saying "hey, this is ****ed up, we need to get back to basics and stop throwing our users under the nearest bus").
Of course, the real question is, will the Self-Appointed Benevolent Dictator For Life do that, or just pull the plug?
Canonical had a goal of "fixing bug #1 - Windows majority market share" back in 2004. Here we are, 2012, and creaky old XP has 20x the market share.
Last year Shuttleworth set a goal of 200 million Ubuntu users by 2015. Since then, he's thrown all the Kubuntu users under the bus, same as he did to the Gnome users.
This latest announcement is old news - others have been running Debian on the same hardware combo since last summer. But seriously, turning a smartphone into a netbook? And one that can't even run Android apps when in "netbook mode"?????!!!!???? No OEM is going to bother with this. Netbooks are dead. And with the higher-res screens coming to tablets, Unity will be sub-optimal. Just like the idea of having to type stuff to run a program (their stupid "Heads-Up Display") is a step backwards for users.
That's okay - Mint was happy to take most of those disaffected Ubuntu users, along with their mindshare, and Fedora, Arch, and a few other distros are also seeing significant upticks. Choice is good, right?
Employees trying to justify being yes-persons to Shuttleworth's "Self-Appointed Benevolent Dictator for Life" so they can continue to collect a paycheck, duh!
Nobody's going to tell him his ideas are seriously out of date, that Canonical is a failure from the venture capital perspective, and that they should never have abandoned the goal of making the best desktop distro, period, instead of turning around and throwing their user base under the bus.
That's okay - Canonical's mistakes are encouraging others to step in to fill the void (Mint, for example), so in the end, it's all good, right?
Just like last months' UbuntuTV was samygo.tv running Ubuntu instead of whatever linux distro you wanted to run.
Who knows what "new" product Canonical will hype next month that someone already demoed last year? Even Matt Asay, their former COO, thinks they're basically scr*wed.
Turning your smartphone into a netbook - why? Netbooks are dead, and nobody is going to carry around a docking station. The Atrix is a marketing failure, and those stupid commercials with the TSA guys didn't help.
Not going to happen.
This "latest innovation" is old news - some guy got Debian to run on the exact same phone last summer. So tell us, why would any manufacturer go to Canonical for this?
All they did, same as Canonical, was replace the customized linux os that was running when it's docked with Ubuntu, same as the Debian guy swapped it out for Debian.
And let's be honest, nobody wants to haul around a docking station that turns your smartphone into a netbook. Even Matt Asay - Canonical's former COO - says Canonical lacks focus and that netbooks are dead - two facts most of us have been aware of for a while ... so be happy you didn't blow millions trying to make this turkey fly.
You can't run them "side-by-side" in the conventional sense. The hardware activates one OS image or the other. And none of this is new - here's a video of Debian running on the same device, back in August of last year http://www.youtube.com/watch?v=G8-92J9hfkA
Canonical announced their "Android Execution Environment" 3 years ago, then abandoned it 2 years ago because they couldn't do it.
This is as "innovative" as UbuntuTV was - which was just Canonical customizing the freely-available samygo.tv software to run Ubuntu instead of another distro http://www.samygo.tv/ You too can have your own brand TV distro running right inside your TV - Slackware, RedHat, Debian, even MythTV ...
Of course, the other question is why anyone would want to run a half-baked "mobile UI" on a desktop display like they propose ...
http://www.youtube.com/watch?v=G8-92J9hfkA
It's really funny how their "touch UI" (disUnity) only runs on the non-touch device :-)
It's because it's not actually "running in the smartphone" - all they did was replace the linux os in the Lapdock100 (which is a separate accessory for the smartphone) with Ubuntu.
In other words, nothing to see here. Ubuntu still can't run Android, or Android apps, almost 3 years after Canonical announced their "Android Execution Environment."
Both. It's not an either-or thing. False dichotomy and all that ...
It's simple, really - you use the existing hypervisor in the Atrix, and just replace the desktop os with the os of your choice.
Freetards - alive and well, and now with twice the stupidity.
Re: TV - over-modulated ads haven't been much of a problem since BEFORE the regulations requiring normalization of sound - unless you bought a cheap set that didn't have auto volume control. As for the length of time of TV ads, I prefer a 3-minute block of ads - it lets me put away the dishes or fold some laundry or make the bed or any one of a number of things that need to be done.
Re: Radio - if you don't like the ads on it, again, the technical solution is to carry your own music, audiobooks, etc.
Re: Streaming video: so pay to stream stuff ad-free (oh wait, you want it for nothing)
Re: Websites: So go elsewhere. Start your own. Go into the big blue room and take a break. Relax.
Re: Electronic billboards: Get a better-shielded radio, duh! Or some duct tape and tin foil ...
Re: Email: so what's the problem again, since you don't use webmail or html mail? Oh right, there isn't one, you just want to complain.
Re: Search engines: So instead of using them, just bookmark the damn site once and be done with it. Is it *that* hard? Or create your own. It's not that hard to write a crawler. It's even easier to pay someone else to, if you can't. Just put your money where your mouth is.
One reason to close the program between uses is to release leaked memory. Memory might be cheap, but it's not infinite.
1. Programs that save their state when closing - none of them will have the same screen layout next time around. ... not a good idea.
2. Changed screen dimensions / colour depths / dpi between runs.
3. The MRU list (most recently used files list) will still have to be rebuilt from scratch anyway. So will any containers and parent objects/data structures
4. Updates to the software will also trigger a rebuild
5. Updates to system libraries will cause it to um... behave erratically, since the blob maps to old library entry points.
6. Any memory leaks become cumulative
People don't mind the advertising so much - it's the tracking that bugs us the most. The all-prevasive, obsessive-cumpulsive, privacy-invading stalking behavioural tracking ...
Just like we don't mind ads on TV - we either watch them, channel surf, or go do a couple minutes worth of housework, but if TV ads watched us back, we'd find that unacceptable as well.
Cookies to preserve state in client-server web apps == good. Cookies to track peoples behaviour == bad. Cross-site cookies == bad by definition.
Short answer - don't! I'm sick and tired of all this bogus tracking crap. If you absolutely insist on having a +1, then also have a -1, so we can tell you how we really feel! Same as if you're going to have a Like button, also have a Hate button.