Okay, stop right there - he made the "announcement" in the form of an email directly to the guy who wrote an "open" email to him, and who then published that answer without asking Jobs first because his email was open.
And if Guinnes rated mastery of getting free PR via leaks, Apple would probably be the current top dog and certainly in the top handfull.
Why should PR by leak be limited to just product announcements? The principle applies in general and there are lots of other memes a company benefits by promulgating.
Apple's products are hardware. There aren't any "open-source alternatives" to those, so this argument makes little sense.
But Apple's products are distinguished from other similar products by two things: Appearance and software.
For instance: The iPad is Apple's all-or-nothing bet to stay viable thorugh the next decade. But it's just a general purpose portable computer (with a typical compliment of storage and peripherals for a portable) in a big touch tablet, with Apple's software and service infrastructure behind it. Other vendors produce such devices (and the more successful iPad is the more will appear on the market). If they can run open source software that provides all the things Apple provides but for far less money and/or with significantly improved features and/or with additional functionality Apple doesn't supply and/or unlocked and controllable by the user, Apple loses market share. Each of those improvements, if large enough, will pull a segment of their customer base and combinations of them will pull additional segments.
Drive vendors away from using Open Source software on their tablet devices or providing content they can play in open formats via nebulous fear of patent suits and Apple protects a "purely hardware" product's market share.
Apple doesn't have anything in particular to gain from pushing h.264 on others. They don't own it, and they own a single patent out of hundreds and hundreds in the patent pool.
They don't need to own the patents to h.264. If they can create the perception that their products will remain on the market and supported (because they have or can get licenses to any component that is open to a patent suit) while the open source alternatives might expose their users to lawsuits, they can drive users away from those alternatives.
Target isn't just the end users of the products. It's also the content providers.
If you were putting out program material to sell to users for pennies or single-digit dollars or serve on the web for some monitization scheme that pays pennies per view, would you use a format that might open you to a patent lawsuit claim that might cost you hundreds of dollars per unit of "content" sold or served? Doesn't matter if the risk is real: It's enough to create the perception of risk, and that's what FUD does.
VHS beat Betamax, despite Betamax's higher technical quality, because most prerecorded movies were available on VHS and few on Beta. Now we have another format war...
Meanwhile, open source tools can form the basis for competitors to iTunes that would provide better customer service, content quality, etc. Why not spike them to keep the gravy train rolling?
Looks to me, Steve Jobs just knows there are people looking into suing Theora. Not Steve Jobs (or Apple) is going to sue Theora.
Even if that's the case he made the announcement in the form of a FUD attack on Theora and the other open source CODECs.
Now lots of potential adopters will instead be waiting for the other shoe to drop before considering an open source solution - and paying for proprietary stuff meanwhile. And if the shoe never drops they'll wait, and pay, a very long time. This is the magic of FUD.
I work for a major hospital in the Northeast. Recently the hospital has taken it upon itself to increase its general level of computer security. As a result they now require full-disk encryption on any computer connected to their network on site.
So encrypt your disk already.
Do they require you use some particular software or do they let you use one of your choice? Are you running Linux or something else?
(If the answers are "your choice" and "Linux" I know there's a solution - because I use it on a me-configured laptop with MY gargantuan employer, which requires full disk encryption and up-to-date software on any offsite or portable machine where you view or store company data or access its network.)
The Ubuntu distribution, for instance, has a full-disk encryption option. Encrypts everything but the boot partition (gotta have SOMETHING in the clear to get started) as one big loop-mounted device where the rest of the partitions and swap area are built.
Don't know if it's available on the Live CD on Karmic or Lucid. But for Hardy and Jaunty it was only available on the "alternate" installation disk. (Do a default install from the live CD to see how the distribution wants to be partitioned first, then clone that mapping when partitioning the encrypted hard drive when installing from the alternate CD.)
No conversion on-the-fly that I know of. But save your data offline, reinstall, and load it back onto the new install. Then you're back on the air with full disk encryption.
DON'T forget the passphrase: Unlike commercial systems with administrator backdoors the passphrase IS the key and if you lose it you're hosed.
... but I also think the employer is going to have to pay somebody to go through the entire configuration because they had a guy working there who they didn't trust, so redoing the router configuration may turn out to be a small part of that.
Yep.
But if I were that guy I'd want the password so I could download the actual configuration of the routers, etc. Otherwise, even with a lot of sniffing and probing before lobotomizing the routers, I'd probably still end up killing the net, maybe for weeks, maybe multiply. This would not look good on my resume or in the news articles with my name attached. So I wouldn't take the job. B-)
If you have access to the hardware you can set the password to anything you want. You don't need the old password.
But in doing so you also cause some boxes to erase the current configuration - a security "feature" to prevent its leakage to anyone with physical access to the box. You need the current password to dump the configuration.
The network was running. Forced password reset would shut it down until a replacement configuration was loaded. The remaining employees didn't have a reliable copy of the configuration information and couldn't get one from the box without the password they didn't have. So forced password reset would take down the running network with no reliable way to bring it back up any time soon. Like until somebody else redid all of Childs' work...
After he was arrested and placed in custody is when he stated that he would only give the password to the mayor, not becuase it was a rule or directive but becuase Mayor Newsom was "the only person he felt he could trust".
I don't understand his logic here. It didn't occur to him that the mayor would simply hand that password off to the very same people asking Mr. Childs for it directly?
I understand it fine:
Policy says don't give it to your boss, but doesn't specify who is boss enough to override it. By definition (in the absence of ordinances to the contrary) the Mayor is the head of the executive branch in question and his formal order to hand over the passwords amends the policy or grants an exception. (But in absence of an explicit order to ignore the crowd of eavesdroppers the sysadmin may still be bound by the policy to not disclose it in an insecure setting.)
Once the mayor turns around and hands it to the pointy-haired boss it's the Mayor's problem, not the sysadmin's.
Doesn't mean I agree with all of that. But it is a plausible and understandable interpretation.
I can barely type well with an onscreen keyboard, and soon they'll want me to air-type? No thanks.
How about parsing ASL (American Sign Language)?
Then again, finger spelling might be easer - since that only involves one hand, rather than both and the body, letting you hold the phone in the other and reducing the camera requiremens. Also a much more limited "vocabulary" consisting mainly of the letters and numbers.
So we want cars to steer towards what we are looking at? Seriously? You want to have all the cute women in the world run over?
While the comment WAS funny there is a problem with something like that already.
It's been known for decades that drunk drivers tend to fixate on flashing yellow lights and then steer toward them. This makes using flashing yellow lights as a warning counter-productive.
Oregon, for instance, long ago switched away from blinky-yellow lights to the rear on police cars to use as warning lights when they have people pulled over - with a significant reduction in car-hits-cop-at-traffic-stop incidents.
California, of course, has standardized on big yellow blinky-lights for cop car pullover warnings. (I recall a few years back when San Jose was lamenting how many of their new fleet of cruisers had been smashed by drunk drivers that year...)
Saw an article about some recent work with animal models that suggests cellphone usage MIGHT cause a significant (40%ish?) REDUCTION in probability of developing clinical alzheimers.
I wonder if, should that turn out to be true, this study would pick it up?
If only C++ had been been consistent during the construction (and destruction) of derived class member variables of object types...
What do you mean? All objects in C++ have the same construction sequence rules, whether they're auto or static variables, created on heap via "new", or are fields of another object.
The issue is what happens if a base class constructor exports the "this" pointer and the constructor of a member variable of the derived class (or a function called from the argument to an initializer) does something that causes that pointer to be used to call a virtual function that is overridden in the derived class. Does it get the base or derived version of the member function? (Similarly going down the destruction hierarchy when a member object's destructor causes a virtual function of the partially-destroyed instance to be called.)
I maintain that during the initialization of the member variables, like that of the base class, the result should be the BASE class version of the overridden function. Member variables and the base class(es) are peers for this purpose. Before the constructor is entered, attempting to use the derived class version means using functionality whose support has not yet been initialized - leading to error. Once you're in the constructor the derived class version of the function and the constructor can interact to handle any issues about what's initialized and what's not. So the switch of what version of the virtual functions that would be called should occur as the step IMMEDIATELY BEFORE the execution of the first user-written statement of the constructor body. (For instance, in a virtual-function-table implementation of the code generator, such as cfront, the storage of the virtual function pointer(s) should occur as the last step before executing the first constructor body statement.)
The same issue occurs in reverse on the destructor: Immediately after the last statement of the destructor and before the destruction of the members, the class should shift to the base class versions of the virtual member functions.
This was not an academic issue. Back before the first ANSI standard (when "catch" and "throw" were merely reserved words) a company I worked with attempted to write a hypertext database server using C++. We did our own exception handling and garbage collection for memory management, both of which were badly hosed by the lack of this virtual-function timing semantics. For example: The garbage collector used "smart pointer" member variables, with virtual functions enumerating them and a preprocessor to fill in the virtual function guts for a given level of each class. Much of the work of a class instance occurred during its construction, and a given class might have many members that were also classes - and might allocate and construct a lot of other objects, provoking garbage collection to salvage memory. If the garbage collector got the derived-class version of the member funciton it would follow uninitialized pointers full of heap junk, resulting in total havoc. (Similar havoc could occur with exceptions thrown from constructors, resulting in destructors on unconstructed objects being called.) To avoid this we had to refrain from using member objects with any significant construction/destruction behavior: Instead we used ONLY smartpointers (which could be safely initialized to NULL without provoking garbage collection or exceptions), then allocate the "members" on the heap and point the smart pointers to them during construction. This turned classes that should have been a single heap object into a Smalltalk-style web of separately allocated heap objects strung together by pointers - vastly bloating the memory allocation and garbage collection overhead.
With a "right" and "wrong" way to do the virtual function selection on both construction and destruction, there are four ways total to do it - one right and three "wrong". I examined the availabl
Fedora claims about twice as many active users as Canonical does, and most people consider it to be a desktop Linux distro.... many of us can at least agree that Ubuntu is undoubtedly the most hyped Linux distro ever.
I dropped Red Hat for my next upgrade when Red Hat dropped its non-corporate customers. Hurrah for the community picking up support when the company bailed - but my die was cast.
When I next did an install (Gutsy came out two days after I got serious about migrating to my new work laptop) my experience convinced me that Ubuntu had made it to prime time - especially for me, as somebody who COULD dig into the guts to admin it but now had other things to do with that time.
At this point I see no other distribution with significant advantages for me, let alone a big enough edge to pay for the additional effort of switching. And now that my wife's vertigo is under control and she's looking at making the move from Windows to "a real OS" and possibly digging into its guts later, I have no problem recommending it as a starter platform for her as well.
I just picked up a couple laptops - one each for me and the wife. I would prefer to put the next LTS on them rather than an older release followed by an upgrade within weeks. B-(
(But if we have to wait a bit for the LTS release to be a GOOD one, or have a backrev of a major component that needs an upgrade in a month or so, that's the breaks.)
The point is that they get found and then get fixed fast.
Ubuntu's problem occurred because they have a shipping deadline and a really bad bug got inserted late and detected about a week from the scheduled release. So there's not much time left for testing a fix, another if the first fails, rinse-and-repeat...
Deadlines: "The light at the end of the tunnel is an oncoming locomotive."
It's a kinda goofy dialect of C with objects tacked on. They borrow some Smalltalk idioms and pepper the mixture with a generous helping of unnecessary brackets and parentheses.
That's fair.
It's one of the two attempts to do an object-oriented programming extension to C that succeed well enough to go mainstream. I think it actually predates C++ slightly.
It's heavily influenced by Smalltalk and gets wrong, the Smalltalk way, some stuff that C++ gets right. (In particular: derived class methods override the base class versions of member functions even during construction - blowing up your debugging of the base class constructors. In C++ you don't get the derived-class version until you're up to the derived class constructor. If only C++ had been been consistent during the construction (and destruction) of derived class member variables of object types... B-( ) It also has some useful features C++ does not.
Jobs used Objective C for the Next and its NextStep OS. These days it's mainly used with Mac OSX (particularly: the Cocoa API) and the iPhone.
And once it IS signed it's still legal if you're in one of a number of other countries when you do it. (I wonder if the EU laws on personal information apply to the caller-ID info retrieval step if it's done there?)
As far as I can see (IANAL) the only step that's currently illegal in the US is cracking past the voicemail password. That's illegal under the Computer Fraud and Abuse act (accessing a protected computer) and occurs at the server location even if it's initiated from outside the US so there's jurisdiction.
* Talking about a password in front of others. (i.e. the Mayor apparently first asked him to tell the password with many others clearly listening in.)
Bingo. Thanks.
And when the mayor asked him in a secure setting he gave him the password. His work here is done.
I hope that the jury finds for him - and that he then gets a big win on a follow-on suit for whatever a good lawyer can identify as legally improper action by the city, its personnel, and/or the prosecutor.
= = =
If I were in a position to hire him (for something where his long-past criminal record didn't disqualify him), and he was willing to work with and/or train up others to also be competent to work on the network (rather than keep it to himself), I'd do it in a New York minute.
Then I'd explain to his new colleagues, with him present, that he's hired explicitly BECAUSE he has proven he is both skilled and willing to adhere to the company security policy even at great cost to himself - and that at our company we're going to have more than one person AUTHORIZED to admin the network so he'll never be put in such jeopardy with us.
Originally, the term meant to someone who creates furniture with an axe.
Then, by extension, someone who can substitute skill and persistence for advanced tooling to successfully construct something of high quality and function.
Though hosted on a San Francisco government site, that document self-identifies as being the product of a trade organization composed of County sysadmins (and it does not list the "City and County of San Francisco" as one of the Counties whose members contributed.) Indeed, "San Francisco" doesn't appear in the document at all.
Can you also post a link to a place on the site where the city says they adopted this document as their policy?
(Also the quoted text doesn't support the allegation that the password was only to be "disclosed to the mayor in a secure setting". "Mayor" doesn't appear in the document, and "chief" only appears as part of "chief information security officer", not "chief executive".)
Slashdot Mirror
Okay, stop right there - he made the "announcement" in the form of an email directly to the guy who wrote an "open" email to him, and who then published that answer without asking Jobs first because his email was open.
And if Guinnes rated mastery of getting free PR via leaks, Apple would probably be the current top dog and certainly in the top handfull.
Why should PR by leak be limited to just product announcements? The principle applies in general and there are lots of other memes a company benefits by promulgating.
Apple's products are hardware. There aren't any "open-source alternatives" to those, so this argument makes little sense.
But Apple's products are distinguished from other similar products by two things: Appearance and software.
For instance: The iPad is Apple's all-or-nothing bet to stay viable thorugh the next decade. But it's just a general purpose portable computer (with a typical compliment of storage and peripherals for a portable) in a big touch tablet, with Apple's software and service infrastructure behind it. Other vendors produce such devices (and the more successful iPad is the more will appear on the market). If they can run open source software that provides all the things Apple provides but for far less money and/or with significantly improved features and/or with additional functionality Apple doesn't supply and/or unlocked and controllable by the user, Apple loses market share. Each of those improvements, if large enough, will pull a segment of their customer base and combinations of them will pull additional segments.
Drive vendors away from using Open Source software on their tablet devices or providing content they can play in open formats via nebulous fear of patent suits and Apple protects a "purely hardware" product's market share.
Apple doesn't have anything in particular to gain from pushing h.264 on others. They don't own it, and they own a single patent out of hundreds and hundreds in the patent pool.
They don't need to own the patents to h.264. If they can create the perception that their products will remain on the market and supported (because they have or can get licenses to any component that is open to a patent suit) while the open source alternatives might expose their users to lawsuits, they can drive users away from those alternatives.
Target isn't just the end users of the products. It's also the content providers.
If you were putting out program material to sell to users for pennies or single-digit dollars or serve on the web for some monitization scheme that pays pennies per view, would you use a format that might open you to a patent lawsuit claim that might cost you hundreds of dollars per unit of "content" sold or served? Doesn't matter if the risk is real: It's enough to create the perception of risk, and that's what FUD does.
VHS beat Betamax, despite Betamax's higher technical quality, because most prerecorded movies were available on VHS and few on Beta. Now we have another format war...
Meanwhile, open source tools can form the basis for competitors to iTunes that would provide better customer service, content quality, etc. Why not spike them to keep the gravy train rolling?
Looks to me, Steve Jobs just knows there are people looking into suing Theora. Not Steve Jobs (or Apple) is going to sue Theora.
Even if that's the case he made the announcement in the form of a FUD attack on Theora and the other open source CODECs.
Now lots of potential adopters will instead be waiting for the other shoe to drop before considering an open source solution - and paying for proprietary stuff meanwhile. And if the shoe never drops they'll wait, and pay, a very long time. This is the magic of FUD.
I work for a major hospital in the Northeast. Recently the hospital has taken it upon itself to increase its general level of computer security. As a result they now require full-disk encryption on any computer connected to their network on site.
So encrypt your disk already.
Do they require you use some particular software or do they let you use one of your choice? Are you running Linux or something else?
(If the answers are "your choice" and "Linux" I know there's a solution - because I use it on a me-configured laptop with MY gargantuan employer, which requires full disk encryption and up-to-date software on any offsite or portable machine where you view or store company data or access its network.)
The Ubuntu distribution, for instance, has a full-disk encryption option. Encrypts everything but the boot partition (gotta have SOMETHING in the clear to get started) as one big loop-mounted device where the rest of the partitions and swap area are built.
Don't know if it's available on the Live CD on Karmic or Lucid. But for Hardy and Jaunty it was only available on the "alternate" installation disk. (Do a default install from the live CD to see how the distribution wants to be partitioned first, then clone that mapping when partitioning the encrypted hard drive when installing from the alternate CD.)
No conversion on-the-fly that I know of. But save your data offline, reinstall, and load it back onto the new install. Then you're back on the air with full disk encryption.
DON'T forget the passphrase: Unlike commercial systems with administrator backdoors the passphrase IS the key and if you lose it you're hosed.
... but I also think the employer is going to have to pay somebody to go through the entire configuration because they had a guy working there who they didn't trust, so redoing the router configuration may turn out to be a small part of that.
Yep.
But if I were that guy I'd want the password so I could download the actual configuration of the routers, etc. Otherwise, even with a lot of sniffing and probing before lobotomizing the routers, I'd probably still end up killing the net, maybe for weeks, maybe multiply. This would not look good on my resume or in the news articles with my name attached. So I wouldn't take the job. B-)
If you have access to the hardware you can set the password to anything you want. You don't need the old password.
But in doing so you also cause some boxes to erase the current configuration - a security "feature" to prevent its leakage to anyone with physical access to the box. You need the current password to dump the configuration.
The network was running. Forced password reset would shut it down until a replacement configuration was loaded. The remaining employees didn't have a reliable copy of the configuration information and couldn't get one from the box without the password they didn't have. So forced password reset would take down the running network with no reliable way to bring it back up any time soon. Like until somebody else redid all of Childs' work...
I understand it fine:
Policy says don't give it to your boss, but doesn't specify who is boss enough to override it. By definition (in the absence of ordinances to the contrary) the Mayor is the head of the executive branch in question and his formal order to hand over the passwords amends the policy or grants an exception. (But in absence of an explicit order to ignore the crowd of eavesdroppers the sysadmin may still be bound by the policy to not disclose it in an insecure setting.)
Once the mayor turns around and hands it to the pointy-haired boss it's the Mayor's problem, not the sysadmin's.
Doesn't mean I agree with all of that. But it is a plausible and understandable interpretation.
I can barely type well with an onscreen keyboard, and soon they'll want me to air-type? No thanks.
How about parsing ASL (American Sign Language)?
Then again, finger spelling might be easer - since that only involves one hand, rather than both and the body, letting you hold the phone in the other and reducing the camera requiremens. Also a much more limited "vocabulary" consisting mainly of the letters and numbers.
Since I heard it word-of-mouth I don't have any handy. But I'll see if I can trace it down.
Meanwhile, if you're really interested, you might try doing that yourself. My day job involves herding packets, not cars. B-)
So we want cars to steer towards what we are looking at? Seriously? You want to have all the cute women in the world run over?
While the comment WAS funny there is a problem with something like that already.
It's been known for decades that drunk drivers tend to fixate on flashing yellow lights and then steer toward them. This makes using flashing yellow lights as a warning counter-productive.
Oregon, for instance, long ago switched away from blinky-yellow lights to the rear on police cars to use as warning lights when they have people pulled over - with a significant reduction in car-hits-cop-at-traffic-stop incidents.
California, of course, has standardized on big yellow blinky-lights for cop car pullover warnings. (I recall a few years back when San Jose was lamenting how many of their new fleet of cruisers had been smashed by drunk drivers that year...)
Saw an article about some recent work with animal models that suggests cellphone usage MIGHT cause a significant (40%ish?) REDUCTION in probability of developing clinical alzheimers.
I wonder if, should that turn out to be true, this study would pick it up?
The issue is what happens if a base class constructor exports the "this" pointer and the constructor of a member variable of the derived class (or a function called from the argument to an initializer) does something that causes that pointer to be used to call a virtual function that is overridden in the derived class. Does it get the base or derived version of the member function? (Similarly going down the destruction hierarchy when a member object's destructor causes a virtual function of the partially-destroyed instance to be called.)
I maintain that during the initialization of the member variables, like that of the base class, the result should be the BASE class version of the overridden function. Member variables and the base class(es) are peers for this purpose. Before the constructor is entered, attempting to use the derived class version means using functionality whose support has not yet been initialized - leading to error. Once you're in the constructor the derived class version of the function and the constructor can interact to handle any issues about what's initialized and what's not. So the switch of what version of the virtual functions that would be called should occur as the step IMMEDIATELY BEFORE the execution of the first user-written statement of the constructor body. (For instance, in a virtual-function-table implementation of the code generator, such as cfront, the storage of the virtual function pointer(s) should occur as the last step before executing the first constructor body statement.)
The same issue occurs in reverse on the destructor: Immediately after the last statement of the destructor and before the destruction of the members, the class should shift to the base class versions of the virtual member functions.
This was not an academic issue. Back before the first ANSI standard (when "catch" and "throw" were merely reserved words) a company I worked with attempted to write a hypertext database server using C++. We did our own exception handling and garbage collection for memory management, both of which were badly hosed by the lack of this virtual-function timing semantics. For example: The garbage collector used "smart pointer" member variables, with virtual functions enumerating them and a preprocessor to fill in the virtual function guts for a given level of each class. Much of the work of a class instance occurred during its construction, and a given class might have many members that were also classes - and might allocate and construct a lot of other objects, provoking garbage collection to salvage memory. If the garbage collector got the derived-class version of the member funciton it would follow uninitialized pointers full of heap junk, resulting in total havoc. (Similar havoc could occur with exceptions thrown from constructors, resulting in destructors on unconstructed objects being called.) To avoid this we had to refrain from using member objects with any significant construction/destruction behavior: Instead we used ONLY smartpointers (which could be safely initialized to NULL without provoking garbage collection or exceptions), then allocate the "members" on the heap and point the smart pointers to them during construction. This turned classes that should have been a single heap object into a Smalltalk-style web of separately allocated heap objects strung together by pointers - vastly bloating the memory allocation and garbage collection overhead.
With a "right" and "wrong" way to do the virtual function selection on both construction and destruction, there are four ways total to do it - one right and three "wrong". I examined the availabl
Sounded like at least one voice missed the note a couple postings back. B-)
Fedora claims about twice as many active users as Canonical does, and most people consider it to be a desktop Linux distro. ... many of us can at least agree that Ubuntu is undoubtedly the most hyped Linux distro ever.
I dropped Red Hat for my next upgrade when Red Hat dropped its non-corporate customers. Hurrah for the community picking up support when the company bailed - but my die was cast.
When I next did an install (Gutsy came out two days after I got serious about migrating to my new work laptop) my experience convinced me that Ubuntu had made it to prime time - especially for me, as somebody who COULD dig into the guts to admin it but now had other things to do with that time.
At this point I see no other distribution with significant advantages for me, let alone a big enough edge to pay for the additional effort of switching. And now that my wife's vertigo is under control and she's looking at making the move from Windows to "a real OS" and possibly digging into its guts later, I have no problem recommending it as a starter platform for her as well.
I just picked up a couple laptops - one each for me and the wife. I would prefer to put the next LTS on them rather than an older release followed by an upgrade within weeks. B-(
(But if we have to wait a bit for the LTS release to be a GOOD one, or have a backrev of a major component that needs an upgrade in a month or so, that's the breaks.)
but... Linux doesn't have bugs!
Sure it does.
The point is that they get found and then get fixed fast.
Ubuntu's problem occurred because they have a shipping deadline and a really bad bug got inserted late and detected about a week from the scheduled release. So there's not much time left for testing a fix, another if the first fails, rinse-and-repeat...
Deadlines: "The light at the end of the tunnel is an oncoming locomotive."
It's a kinda goofy dialect of C with objects tacked on. They borrow some Smalltalk idioms and pepper the mixture with a generous helping of unnecessary brackets and parentheses.
That's fair.
It's one of the two attempts to do an object-oriented programming extension to C that succeed well enough to go mainstream. I think it actually predates C++ slightly.
It's heavily influenced by Smalltalk and gets wrong, the Smalltalk way, some stuff that C++ gets right. (In particular: derived class methods override the base class versions of member functions even during construction - blowing up your debugging of the base class constructors. In C++ you don't get the derived-class version until you're up to the derived class constructor. If only C++ had been been consistent during the construction (and destruction) of derived class member variables of object types... B-( ) It also has some useful features C++ does not.
Jobs used Objective C for the Next and its NextStep OS. These days it's mainly used with Mac OSX (particularly: the Cocoa API) and the iPhone.
You can still spoof as long as you aren't doing so to deceive or defraud.
Seems to me that spoofing caller ID in order to trick the database into delivering information on some other phone user constitutes intent to defraud.
And once it IS signed it's still legal if you're in one of a number of other countries when you do it. (I wonder if the EU laws on personal information apply to the caller-ID info retrieval step if it's done there?)
As far as I can see (IANAL) the only step that's currently illegal in the US is cracking past the voicemail password. That's illegal under the Computer Fraud and Abuse act (accessing a protected computer) and occurs at the server location even if it's initiated from outside the US so there's jurisdiction.
I don't see any indication of when this first went out.
(My wife runs McAfee and launched an update around 3 AM PDT before hitting the sack...)
* Talking about a password in front of others. (i.e. the Mayor apparently first asked him to tell the password with many others clearly listening in.)
Bingo. Thanks.
And when the mayor asked him in a secure setting he gave him the password. His work here is done.
I hope that the jury finds for him - and that he then gets a big win on a follow-on suit for whatever a good lawyer can identify as legally improper action by the city, its personnel, and/or the prosecutor.
= = =
If I were in a position to hire him (for something where his long-past criminal record didn't disqualify him), and he was willing to work with and/or train up others to also be competent to work on the network (rather than keep it to himself), I'd do it in a New York minute.
Then I'd explain to his new colleagues, with him present, that he's hired explicitly BECAUSE he has proven he is both skilled and willing to adhere to the company security policy even at great cost to himself - and that at our company we're going to have more than one person AUTHORIZED to admin the network so he'll never be put in such jeopardy with us.
OK, I see where it says not to give your password to your boss.
I don't say where it says to give it, on termination, only to the chief executive of the government and only in a secure setting.
Can you help me with that, too?
(If it's there, or in some other policy, and if that was brought up in court, it seems to me this should be an open-and-shut not-guilty.)
My usage:
Originally, the term meant to someone who creates furniture with an axe.
Then, by extension, someone who can substitute skill and persistence for advanced tooling to successfully construct something of high quality and function.
Though hosted on a San Francisco government site, that document self-identifies as being the product of a trade organization composed of County sysadmins (and it does not list the "City and County of San Francisco" as one of the Counties whose members contributed.) Indeed, "San Francisco" doesn't appear in the document at all.
Can you also post a link to a place on the site where the city says they adopted this document as their policy?
(Also the quoted text doesn't support the allegation that the password was only to be "disclosed to the mayor in a secure setting". "Mayor" doesn't appear in the document, and "chief" only appears as part of "chief information security officer", not "chief executive".)