The Bush Administration hosted private federal government emails on Republican National Committee servers. When that was discovered, the data was subpoenaed in connection with the firing of US Attorneys in 2007. And the RNC wiped the data rather than turn it over.
How come every conservative has forgotten about this? They're all out for Hillary's blood without noting that she's following precedent by President George Bush and Vice President Dick Cheney. I will gladly support her prosecution for this - just as long as they're prosecuted first, since they broke the same fucking laws first but nobody conservative cares.
That's a fair point. In any event, I think if Nix or Guix ever get wide adoption it will be a long road because the way they store and manage packages is so fundamentally different from traditional Unixes and Unix package managers. Interesting idea, though.
People like you and me want to own the whole stack from top to bottom for reasons of long term security of our business model.
But people willing to ride the tidal wave and risk getting the floor yanked out from underneath them (to mix metaphors) build Candy Crush and Words with Friends and so forth.
For it to be useful you would need for it to work seamlessly with both, right? You want local copies constantly synchronized so you can keep working during a loss of connectivity but you want remote copies constantly synchronized so that a problem with your local machine doesn't lose hours (or more) of work.
I concede your point on the TFQ because of your arguments elsewhere. (No sarcasm.)
But with respect to emacs or Vim, if you're only using it like Windows Notepad, then you are nowhere near as productive as someone in an IDE. You can get work done, and in your free time you can use Vim only knowing 'i', 'Esc:q!', and 'Esc:wq' (or 'Esc:ZZ') and just typing away in insert mode. But if you want to work anywhere near as quickly as someone with an IDE, you'll need to learn navigating by word ending and beginning, jumping to the beginning and ending of a line, regex search, regex replace, multi-line indent and un-indent, visual select and then copy or delete, paging up and paging down without taking your fingers off the normal position in the keyboard, editing undo and redo, copying and pasting, etc... and that learning curve is much rougher - and in my opinion, the expertise is much harder to pass along to someone else - than working with an IDE.
A separate factor in the IDE vs. CLI question is the language in use. I happen to be working in Java professionally. I neither love nor hate the language, but it's a high ceremony language and the autocomplete and jump-to-declaration features of the IDE can be incredibly handy. A user would get much less utility from an IDE when writing in Python, Ruby, Perl, PHP, Scheme, etc... and on the all-too-rare but happy occasions my employer asks me to work on Python or Javascript, I close Eclipse and work in Vim.
Well, there's grounds for discussion, right? If you met someone that only works in nano and C and refuses to learn emacs or Vim or sed or Perl, Python, Ruby, Lisp, Scheme, C++, Smalltalk, is that "sticking with proficient tools" or plain stubbornness?
So if a potential new employer asks you to try something out, the sensible action is to give it a try and account for the fact that some of your initial slowness is unfamiliarity. Then you have to evaluate whether you can ever be as efficient as you were with your previous preferred setup. If the answer is no, and you did your best to be objective, then refuse with a clear conscience.
Good point on sed vs. IDE for text replacements and the speed of preview.
The audio advice for the IDE is relatively rare. "It would be faster if you did Ctrl-Alt-____" every few minutes because most of the navigation is visual. Mouse wheel to scroll. Right click for options. etc... etc... A novice can still work pretty quickly with only occasional prompting. With Vim, everything except typing actual text is a funky command, so unless you want to watch someone navigate one keystroke at a time with hjkl you're going to be barking directions constantly.
With my regex example, the person I was working with was familiar with ":% s/foo/bar/g" but not ":(start line),(end line)" or Vim's handling of capture groups. I can't remember the syntactic differences between Vim regex and Perl regex, but he stumbled across one of them.
With 'rock climbing', I just meant that the learning curve for the useful features is almost a vertical wall. I'm sorry I didn't make the metaphor clearer. I am confident you're an emacs wizard and can move quickly, and that's great. But for two mostly equal new developers, the person with emacs plus sed will need months - frustrating months - to get to the point where they can pace the person who started with an IDE.
I've used Eclipse and Netbeans most recently, and they do have previews for text changes.
The problem with emacs and Vim isn't what they can do, it's the learning curve. IntelliJ, Visual Studio, KDevelop, Eclipse, Netbeans, etc... etc... have a learning curve that looks like a slope. Barebones Emacs and Vim usage is an easy slope, but if you want to be highly productive it's like rock-climbing.
Or to put it another way, once in a while at work we pair together over a screen-sharing application. When my colleagues and I use Eclipse, the other person will helpfully suggest menus and keyboard shortcuts to accelerate some tasks. When I'm watching someone use Vim, dictating, "escape colon one four two comma one nine five space ess forward slash caret space open square bracket...." to do a regex replacement on certain lines or create and run a macro or navigate with anything fancier than hjkl will give you a stroke. You'll either go insane watching the novice move slowly or drive them insane bombarding them with incomprehensible key combinations.
HTML5 supports local storage, right? So in theory all of your local storage requirements can be met with a web-based development environment. Whether any of the existing web-based IDEs actually have that fully and seamlessly implemented, I have no idea.
But to me, there is no killer feature here. The closest thing to killer feature is nearly instant developer environment replacement. If my kids spill a drink on my work laptop or it's stolen, if I have a recent full disk image backup it will take me a few days to get back up and running. I have to buy the same hardware, and then put the backup image on the storage, and then I'm back to work. If I don't have a recent full disk image backup then I have to buy some kind of replacement and spend another day getting it ready. With a fully web-based IDE the loss of hardware is a ten minute annoyance - just boot up any other machine with a web browser, log back in, get back to work.
If you're a professional software developer using a $200 Chrome OS laptop with a 12 inch, 1366x768 screen and 2GB of RAM, you have a problem. But the Chromebook Pixel would be a fine developer device.
An "intending team member" that won't use the tool because he or she is stubborn isn't worth having on your team, and is thus not a lost. To be fair, though, an "intending team member" that won't use the tool because the tool cuts their productivity down by 10% or 30% or 80% even after climbing a fair bit of the learning curve is a sign the tool is not ready for serious use.
I don't think the problem there is the free software community and homebrew, the problem is sucky PC wireless cards and poor Linux drivers. Fixing that isn't trivial.
It would be interesting to try, though. In my area, Comcast's pricing for 150 Mbps down is only $14 per month more than 25 Mbps down. I upgraded my service, then realized my router only has a 100 Mbps port on the WAN side. That's not a huge problem, I lived with 25 Mbps down for ten years. But now I'm looking at upgrades, and I would prefer to do something like this author - but with 802.11ac wireless, too.
"buying profitable companies at their peak is not a sound business strategy" - Microsoft has over ten billion dollars in profit per year. It is a completely valid business strategy for them to just buy and add Microsoft branding to any innovators that might someday eat their lunch. 1.2 billion dollars was a big deal to Notch, but peanuts to Redmond and now Redmond owns a game my kids spent a thousand hours playing in the past three years.
Right. I understand why Minecraft took the world by storm. But Minetest ( http://www.minetest.net/ ) and Terasology ( http://terasology.org/ ) are open source, just as extensible, just as fun. Writing mods for Minecraft is a royal pain because you have to deal with the fact that the core program is proprietary. Minetest and Terasology don't have that.
I make a good living serving evil corporate masters myself. We all have to eat.
But that doesn't mean it's good for the future of humanity for proprietary software to dominate everything. If you want freedom, you need open source software on everything. The only way to keep that open source software from being forked and then replaced by proprietary software is copyleft licensing.
But consider, the power generation for a city is already massive. I'm trying to find a useful example and it's hard, but look at 2013 Rhode Island electricity use in 2013, http://www.eia.gov/electricity... - 7.8 million MWh purchased (if I'm reading the page correctly). That's for about a million state residents.
So to pick a nuclear power plant at relatively random, the Braidwood nuclear plant in Illinois generates about 20 million MWh per year at a construction cost of $5.2 billion. To generate the electricity needs for Rhode Island, you need about 40% of its production capacity. That would cost 40% of the $5.2 billion, or about $2 billion. To pick a coal power plant at random, the colossal W.H. Sammis power plant also generates about 20 million MWh per year, and I couldn't find a construction cost but a recent retrofit cost $1.5 billion.
$1.5 billion or $2 billion buys an awful lot of kites, and an awful lot of land to fly them over.
Since the anti-environmental types tend to be the pro-capitalism types, I ask the parent post writer to look at it this way: the biggest expense for a fossil fuel power generation plant is the fuel itself. So they have the best reason in the world to pursue maximum efficiency: profitability.
So unless someone presents credible evidence the fossil fuel power industry has been intentionally wasting billions of their own dollars, we have to assume modern energy plant efficiency is pretty close to optimal for the technology being used.
Satya Nadella hasn't stopped Microsoft's offensive use of patents, which to me indicates he's just as ruthless as Gates or Ballmer. IE11 and Edge browser are standards compliant only because that currently makes the most money for Microsoft. As soon as it's in their financial interest to break compatibility with the standard, all bets are off.
Open source is a million miles better than proprietary, so if you're not willing to use GPL/EPL/MPL or other copyleft but you'll use MIT/BSD/Apache/ISC, that's still fine with me and we can be allies.
But to beat this dead horse - copyleft keeps the code free for everyone. MIT/BSD/Apache/ISC is primarily free as in labor - Microsoft, Google, Apple, Amazon, Oracle, VMWare, Cisco, Juniper, Sony, and others contribute and take liberal license code and then they use it to build their proprietary software empires that own the world. There has never been more liberal license free software code available on the internet than there is now - and user freedom to use purchased devices any (legal) way they see fit, user freedom from DRM, and user freedom from privacy violations are all being eaten away worse than they were in the recent past. Maybe this doesn't bother you, because you're satisfied to be a wealthy corporate drone. Me, I don't like it.
I hate Microsoft's business tactics throughout its history. I hate its FUD campaigns against competitors, and the way it intentionally broke APIs and standards compliance to shut other businesses down. I hate how it wields its patents like a weapon when it's crystal clear the original intent of the US patent system has been twisted to protect giant corporations against new innovators instead of the other way around.
But that doesn't mean all of the code they released as open source is automatically garbage or a trap. All or almost all of the code released by Microsoft on Github is under an MIT license, and also carries an additional free patent use grant from Microsoft. So there is nothing wrong with adopting the.NET Framework or Chakra Core or anything else Microsoft put on Github for your own personal or business uses.
I'm a member of the FSF. I think proprietary operating systems are fundamentally compatible with end user rights and privacy. So while Microsoft's actions might be friendly to my goals here and there, their core business model cannot ever align with my goals. The whole company would need to be reinvented. As far as I'm concerned, the ideal software companies are ones like Red Hat - there is no proprietary software in the mix.
But further, I think it's crystal clear that Microsoft is playing nice with open source licenses, the HTML standard, the Javascript/ECMAScript standard, HTTP 2.0, etc... etc... because it doesn't have a sufficient monopoly to force the world into its proprietary technologies. If the world computing market started to look like it did in the 1990s, they would go back to their dirty tricks (breaking APIs in DOS to kill competitors to Excel, later intentionally making Internet Explorer differ from open standards to push users and developers away from supporting competing browsers, launching Visual J+ as Windows-only Java, etc...). And if they had a monopoly position again, it would again be in their "enlightened self interest" to use those exact tactics.
"Enlightened self interest" is just clever marketing gloss for advocating a strategy of waiting until you can get away with it before slitting the other guy's throat. Their goal is still a return to monopoly, they just aren't making progress as well as they want.
This would need a huge market share before incorporating Windows-specific Chakra code did anything other than ruin 90% of your visitors' experiences with your site, and I suspect if Chakra Core did see widespread adoption there would be more work on this specific problem or possibly forks of the project that block it.
Their fundamental business model is built upon getting people to pay licensing fees for proprietary software. I am very grateful and appreciative for all of the excellent code they have released as open source. But the company will never be a complete friend to the open source software community.
Slashdot Mirror
The Bush Administration hosted private federal government emails on Republican National Committee servers. When that was discovered, the data was subpoenaed in connection with the firing of US Attorneys in 2007. And the RNC wiped the data rather than turn it over.
How come every conservative has forgotten about this? They're all out for Hillary's blood without noting that she's following precedent by President George Bush and Vice President Dick Cheney. I will gladly support her prosecution for this - just as long as they're prosecuted first, since they broke the same fucking laws first but nobody conservative cares.
That's a fair point. In any event, I think if Nix or Guix ever get wide adoption it will be a long road because the way they store and manage packages is so fundamentally different from traditional Unixes and Unix package managers. Interesting idea, though.
People like you and me want to own the whole stack from top to bottom for reasons of long term security of our business model.
But people willing to ride the tidal wave and risk getting the floor yanked out from underneath them (to mix metaphors) build Candy Crush and Words with Friends and so forth.
For it to be useful you would need for it to work seamlessly with both, right? You want local copies constantly synchronized so you can keep working during a loss of connectivity but you want remote copies constantly synchronized so that a problem with your local machine doesn't lose hours (or more) of work.
I concede your point on the TFQ because of your arguments elsewhere. (No sarcasm.)
But with respect to emacs or Vim, if you're only using it like Windows Notepad, then you are nowhere near as productive as someone in an IDE. You can get work done, and in your free time you can use Vim only knowing 'i', 'Esc:q!', and 'Esc:wq' (or 'Esc:ZZ') and just typing away in insert mode. But if you want to work anywhere near as quickly as someone with an IDE, you'll need to learn navigating by word ending and beginning, jumping to the beginning and ending of a line, regex search, regex replace, multi-line indent and un-indent, visual select and then copy or delete, paging up and paging down without taking your fingers off the normal position in the keyboard, editing undo and redo, copying and pasting, etc... and that learning curve is much rougher - and in my opinion, the expertise is much harder to pass along to someone else - than working with an IDE.
A separate factor in the IDE vs. CLI question is the language in use. I happen to be working in Java professionally. I neither love nor hate the language, but it's a high ceremony language and the autocomplete and jump-to-declaration features of the IDE can be incredibly handy. A user would get much less utility from an IDE when writing in Python, Ruby, Perl, PHP, Scheme, etc... and on the all-too-rare but happy occasions my employer asks me to work on Python or Javascript, I close Eclipse and work in Vim.
Well, there's grounds for discussion, right? If you met someone that only works in nano and C and refuses to learn emacs or Vim or sed or Perl, Python, Ruby, Lisp, Scheme, C++, Smalltalk, is that "sticking with proficient tools" or plain stubbornness?
So if a potential new employer asks you to try something out, the sensible action is to give it a try and account for the fact that some of your initial slowness is unfamiliarity. Then you have to evaluate whether you can ever be as efficient as you were with your previous preferred setup. If the answer is no, and you did your best to be objective, then refuse with a clear conscience.
Good point on sed vs. IDE for text replacements and the speed of preview.
The audio advice for the IDE is relatively rare. "It would be faster if you did Ctrl-Alt-____" every few minutes because most of the navigation is visual. Mouse wheel to scroll. Right click for options. etc... etc... A novice can still work pretty quickly with only occasional prompting. With Vim, everything except typing actual text is a funky command, so unless you want to watch someone navigate one keystroke at a time with hjkl you're going to be barking directions constantly.
With my regex example, the person I was working with was familiar with ":% s/foo/bar/g" but not ":(start line),(end line)" or Vim's handling of capture groups. I can't remember the syntactic differences between Vim regex and Perl regex, but he stumbled across one of them.
With 'rock climbing', I just meant that the learning curve for the useful features is almost a vertical wall. I'm sorry I didn't make the metaphor clearer. I am confident you're an emacs wizard and can move quickly, and that's great. But for two mostly equal new developers, the person with emacs plus sed will need months - frustrating months - to get to the point where they can pace the person who started with an IDE.
Right. The only Chrome OS developer environment that might make sense with the right web IDE is one of Google's Chromebook Pixel machines.
I've used Eclipse and Netbeans most recently, and they do have previews for text changes.
The problem with emacs and Vim isn't what they can do, it's the learning curve. IntelliJ, Visual Studio, KDevelop, Eclipse, Netbeans, etc... etc... have a learning curve that looks like a slope. Barebones Emacs and Vim usage is an easy slope, but if you want to be highly productive it's like rock-climbing.
Or to put it another way, once in a while at work we pair together over a screen-sharing application. When my colleagues and I use Eclipse, the other person will helpfully suggest menus and keyboard shortcuts to accelerate some tasks. When I'm watching someone use Vim, dictating, "escape colon one four two comma one nine five space ess forward slash caret space open square bracket...." to do a regex replacement on certain lines or create and run a macro or navigate with anything fancier than hjkl will give you a stroke. You'll either go insane watching the novice move slowly or drive them insane bombarding them with incomprehensible key combinations.
HTML5 supports local storage, right? So in theory all of your local storage requirements can be met with a web-based development environment. Whether any of the existing web-based IDEs actually have that fully and seamlessly implemented, I have no idea.
But to me, there is no killer feature here. The closest thing to killer feature is nearly instant developer environment replacement. If my kids spill a drink on my work laptop or it's stolen, if I have a recent full disk image backup it will take me a few days to get back up and running. I have to buy the same hardware, and then put the backup image on the storage, and then I'm back to work. If I don't have a recent full disk image backup then I have to buy some kind of replacement and spend another day getting it ready. With a fully web-based IDE the loss of hardware is a ten minute annoyance - just boot up any other machine with a web browser, log back in, get back to work.
If you're a professional software developer using a $200 Chrome OS laptop with a 12 inch, 1366x768 screen and 2GB of RAM, you have a problem. But the Chromebook Pixel would be a fine developer device.
An "intending team member" that won't use the tool because he or she is stubborn isn't worth having on your team, and is thus not a lost. To be fair, though, an "intending team member" that won't use the tool because the tool cuts their productivity down by 10% or 30% or 80% even after climbing a fair bit of the learning curve is a sign the tool is not ready for serious use.
Thanks for the info, I appreciate it. In what way is master (AP) mode shitty? No snark intended - what went wrong?
I don't think the problem there is the free software community and homebrew, the problem is sucky PC wireless cards and poor Linux drivers. Fixing that isn't trivial.
It would be interesting to try, though. In my area, Comcast's pricing for 150 Mbps down is only $14 per month more than 25 Mbps down. I upgraded my service, then realized my router only has a 100 Mbps port on the WAN side. That's not a huge problem, I lived with 25 Mbps down for ten years. But now I'm looking at upgrades, and I would prefer to do something like this author - but with 802.11ac wireless, too.
yeah, me too
"buying profitable companies at their peak is not a sound business strategy" - Microsoft has over ten billion dollars in profit per year. It is a completely valid business strategy for them to just buy and add Microsoft branding to any innovators that might someday eat their lunch. 1.2 billion dollars was a big deal to Notch, but peanuts to Redmond and now Redmond owns a game my kids spent a thousand hours playing in the past three years.
Right. I understand why Minecraft took the world by storm. But Minetest ( http://www.minetest.net/ ) and Terasology ( http://terasology.org/ ) are open source, just as extensible, just as fun. Writing mods for Minecraft is a royal pain because you have to deal with the fact that the core program is proprietary. Minetest and Terasology don't have that.
I make a good living serving evil corporate masters myself. We all have to eat.
But that doesn't mean it's good for the future of humanity for proprietary software to dominate everything. If you want freedom, you need open source software on everything. The only way to keep that open source software from being forked and then replaced by proprietary software is copyleft licensing.
But consider, the power generation for a city is already massive. I'm trying to find a useful example and it's hard, but look at 2013 Rhode Island electricity use in 2013, http://www.eia.gov/electricity... - 7.8 million MWh purchased (if I'm reading the page correctly). That's for about a million state residents.
So to pick a nuclear power plant at relatively random, the Braidwood nuclear plant in Illinois generates about 20 million MWh per year at a construction cost of $5.2 billion. To generate the electricity needs for Rhode Island, you need about 40% of its production capacity. That would cost 40% of the $5.2 billion, or about $2 billion. To pick a coal power plant at random, the colossal W.H. Sammis power plant also generates about 20 million MWh per year, and I couldn't find a construction cost but a recent retrofit cost $1.5 billion.
$1.5 billion or $2 billion buys an awful lot of kites, and an awful lot of land to fly them over.
Since the anti-environmental types tend to be the pro-capitalism types, I ask the parent post writer to look at it this way: the biggest expense for a fossil fuel power generation plant is the fuel itself. So they have the best reason in the world to pursue maximum efficiency: profitability.
So unless someone presents credible evidence the fossil fuel power industry has been intentionally wasting billions of their own dollars, we have to assume modern energy plant efficiency is pretty close to optimal for the technology being used.
Satya Nadella hasn't stopped Microsoft's offensive use of patents, which to me indicates he's just as ruthless as Gates or Ballmer. IE11 and Edge browser are standards compliant only because that currently makes the most money for Microsoft. As soon as it's in their financial interest to break compatibility with the standard, all bets are off.
Open source is a million miles better than proprietary, so if you're not willing to use GPL/EPL/MPL or other copyleft but you'll use MIT/BSD/Apache/ISC, that's still fine with me and we can be allies.
But to beat this dead horse - copyleft keeps the code free for everyone. MIT/BSD/Apache/ISC is primarily free as in labor - Microsoft, Google, Apple, Amazon, Oracle, VMWare, Cisco, Juniper, Sony, and others contribute and take liberal license code and then they use it to build their proprietary software empires that own the world. There has never been more liberal license free software code available on the internet than there is now - and user freedom to use purchased devices any (legal) way they see fit, user freedom from DRM, and user freedom from privacy violations are all being eaten away worse than they were in the recent past. Maybe this doesn't bother you, because you're satisfied to be a wealthy corporate drone. Me, I don't like it.
I hate Microsoft's business tactics throughout its history. I hate its FUD campaigns against competitors, and the way it intentionally broke APIs and standards compliance to shut other businesses down. I hate how it wields its patents like a weapon when it's crystal clear the original intent of the US patent system has been twisted to protect giant corporations against new innovators instead of the other way around.
.NET Framework or Chakra Core or anything else Microsoft put on Github for your own personal or business uses.
But that doesn't mean all of the code they released as open source is automatically garbage or a trap. All or almost all of the code released by Microsoft on Github is under an MIT license, and also carries an additional free patent use grant from Microsoft. So there is nothing wrong with adopting the
I'm a member of the FSF. I think proprietary operating systems are fundamentally compatible with end user rights and privacy. So while Microsoft's actions might be friendly to my goals here and there, their core business model cannot ever align with my goals. The whole company would need to be reinvented. As far as I'm concerned, the ideal software companies are ones like Red Hat - there is no proprietary software in the mix.
But further, I think it's crystal clear that Microsoft is playing nice with open source licenses, the HTML standard, the Javascript/ECMAScript standard, HTTP 2.0, etc... etc... because it doesn't have a sufficient monopoly to force the world into its proprietary technologies. If the world computing market started to look like it did in the 1990s, they would go back to their dirty tricks (breaking APIs in DOS to kill competitors to Excel, later intentionally making Internet Explorer differ from open standards to push users and developers away from supporting competing browsers, launching Visual J+ as Windows-only Java, etc...). And if they had a monopoly position again, it would again be in their "enlightened self interest" to use those exact tactics.
"Enlightened self interest" is just clever marketing gloss for advocating a strategy of waiting until you can get away with it before slitting the other guy's throat. Their goal is still a return to monopoly, they just aren't making progress as well as they want.
This would need a huge market share before incorporating Windows-specific Chakra code did anything other than ruin 90% of your visitors' experiences with your site, and I suspect if Chakra Core did see widespread adoption there would be more work on this specific problem or possibly forks of the project that block it.
Nonsense. Microsoft was already free to take (or if you prefer, steal) anything it liked from the open source V8 Javascript project.
Their fundamental business model is built upon getting people to pay licensing fees for proprietary software. I am very grateful and appreciative for all of the excellent code they have released as open source. But the company will never be a complete friend to the open source software community.