Thanks to self-interested politicians like Utah's Orrin Hatch and others who'd rather fatten up on pork, China has a viable space program, while the US just has a money sink that keeps corporations flush in fat lobbying budgets.
What is it that determines whether a jerk sticks around long enough to succeed and become someone at the top that others excuse and follow or is fired with a "do not rehire" mark on his HR record?
"Sometimes people don't understand the responsibilities that CEOs have, so sometimes they'll take that as, 'Why is he being such an asshole?'" Wolff did allow that Pincus sometimes uses language devoid of "soothing qualities," and could be challenging to work for, depending on how you adapted to his management style. "He's moving at 100 miles per hour. You've either got to get on the bus, or you're not on the bus," he says. "Most people have a buffer.... Mark's not like that. He thinks it, and he says it."
Perhaps the gameplay of FarmVille offers a lesson: you'll put up with shit if you get a little candy now and then, with a promise of a big payout later. Guys like Pincus have a knack for showing just enough promising results (cash profit) to whet the appetite while exploiting the "compulsion loop" of the people they work with. Nice work, if you can stand yourself.
I figured any software developer worthy of the title had heard the story and understood the details already, so I merely linked to the Wikipedia article for reference. Yes, the full filesystem was the root cause, but the repeated reboots and failure of commands to the rover to go into night-time shutdown arose from a race condition within a critical sequence. If you want a more technical analysis, you'll find many, but one is http://www.cs.princeton.edu/courses/archive/fall09/cos109/mars.rover.pdf
While no doubt the atomicity of rename() will prop up the code of naive programmers, there's a great deal more to the operation in question than just changing the file name. Once again, oversimplifying the solution ensures that subtle bugs remain. As for the comment "relatively safe", that's like saying a 39-story fall from the roof of a 40-story building is "relatively healthy".
And of course, if your world is only POSIX and Windows, then portability is still an unrealized goal.
Every two-bit peachfuzz-face programmer in the open source world thinks he has the solution to lots of long-standing bugs. Being able to read what a spec says and actually understanding how it works in actual implementations. For example, what's the behavior of rename() if the file is actually a symbolic link? Patches or GTFO.
The fix is trivial: simply write the new config file out-of-place, and then replace the original with it once it has been fully written.
You left out, "and do it in a way that avoids race conditions". You must also do it in a way that is portable, so it can't depend on OS-specific crutches, and is secure and doesn't interact badly with locking mechanisms that may be transparently in place on the underlying filesystem (NFS, for example). Programming 101 may teach you some things, but most assuredly if that's all you know, a "trivial" write-new-and-replace implementation will be worse. If you are so 100% sure you know how to avoid all these problems, there's some programmers who troubleshot a similar race-condition error from millions of miles away who might have even more possible failure modes to throw at you.
OK smart boy. The mozilla project is open source. If this bug is so all-fired important to you, get to coding, or because you admit you don't have the skills, hire someone who can code it. Don't go whining "THEY won't fix a bug I care about". This 'bug' isn't going to get fixed simply because the folks who could fix it have far better things to spend time on than working around an arbitrary limitation in a not-too-far-in-the-future legacy 32-bit platform. I would be willing to wager that your assertion that web pages have lots of GIFs any more is false anyway. The PNG and JPG formats have pretty much completely replaced that 80s-era 256-color format.
As a matter of fact, with the introduction of things like infallible malloc, you're likely to see more badly-engineered platforms crashing on crappy web pages.
But again I say, if YOU don't like the bug, then YOU fix it, and submit a patch. If they don't accept your patch, then you can build your own FF from source and patch it however crazy way you want.
Comparing 3D tv to color is completely missing the problem. The move to color made sense and was a natural progression that mirrored film. The move to 3d is just another to way to get consumers to consider their old equipment obsolete and force them into buying new TVs. Like a lot of the consumer electronic gadgetry out there, the benefits are questionable at best and the upgrade treadmill only benefits the vendors. Now that they've sold digital to everyone and forced the decades-old standard into obsolescence the electronics makers smell blood. What better way to ensure profits than following the personal computer model of convincing everyone that the perfectly good equipment they bought two years ago is now "outdated" and needs replacing.
Anyone remember the 1981 movie "Comin' At Ya!"? Noted at the time for all the action shots of exaggerated movement towards the camera (arrows, knives, boobies), it was nothing but a gimmick. Sounds like almost 30 years later the salespeople are still pushing the same tricks.
There was a time, in the early days of "horseless carriages" when the law required that a person on foot carrying a flag precede the presumably scary, steam-huffing monstrosity, to warn the pedestrians and horses that the oncoming fuming and chuffing machine of the devil was approaching.
There a many corporate-sponsored contests like this photography, mostly geared at amateurs. Back in the 80s I learned to look at the terms carefully, and if anywhere in them was a clause giving up rights to the photographs entered to the contest-holder, to run far away. Prestigious contests always make it clear that all rights remain with the photographer, although they may legitimately request a time-limited right to display entries for promotional purposes only, not for resale ever.
Stock agencies used to use these contests to pick up vast swaths of decent, if unremarkable, photographs for almost nothing, and with no pesky trouble like having to keep track of who took the photo for credit and payment. I imagine now with Flickr and the flood of digital images, they don't really have work even that hard.
All this remains true for programming contests, and really any contest where the creative work of an individual is made available to another party.
The article starts from a false assumption: that the postal service must be profitable, or at least break even.
Framing the issue this way has nothing to do with what the USPS should or should not carry, or how much they should charge.
Why is that so for the postal service but not for the military, department of transportation, or most any other government agency that provides a service? Universal free mail delivery is something that the citizens of the US want -- or at least did at one time. As a government service, it's something taxpayers agree to pay for.
Now clearly the two authors of this article, management consultants, have a different view of that need. Perhaps they are ideologically inclined to expect that government services should break even or better, in which case, they ought to take on a real challenge and explain to the Pentagon how they can "save" the armed forces. Or perhaps they have a financial interest in private delivery services like FedEX and UPS, who knows? It's clear from early in the article, "Should the federal government continue to compete against the private sector?" that the authors have a sense that somehow there's money to made for UPS, FedEx, and other private delivery services if the postal service was forced to compete on the same level as them. I'm sure they wouldn't advocate for reforming USPS if they thought it would take money away from the private sector.
In any case, before people go trying to reform USPS, let's first decide if we want to continue to support the current expectation of free (for the recipient) door-to-door mail service for everyone in the country everywhere. If citizens clearly want that, then budget (and tax) for it, and shut up about billion dollar "losses" that pale compared to the "losses" racked up by other services we expect as a modern nation. On the other hand, if the country decides that hey, we don't need to deliver everywhere any more, then go ahead, revamp the postal service to be just another profit-motivated competitor.
If I were SpaceX founder Elon Musk, I'd be hopping mad right now. After developing Falcon9 and Dragon on the basis of a truly competitive commercial space program, the porkbarrel senators for aerospace/defense contractor states wrote a new NASA budget to basically hand money over to Boeing and the rest of the usual cast of trough-feeders to continue but with changes and more delays the Ares/Orion program. This craft will see about as much reality as the Orion did before Boeing is behind schedule and over budget and requests yet more money.
The whole goal is to crowd out the smaller guys while maintaining the jobs programs in states like Washington, Utah, and Florida.
Did anyone notice that they don't say where they are going in this capsule? Where are the senators who called Obama's proposed budget a mission to nowhere? This new NASA program doesn't have a destination, either, but at least the dollars keep flowing to the same interests.
I haven't looked at the guts of any of Apple's NS* classes, but I don't doubt there's some places where it's a mess -- it's ancient code, those, where the Android code is new. But just because the feature is part of the SpringBoard application doesn't mean it's badly coded. What classes in the NS SDK specifically?
You're right about one thing: You've reminded once again that I made the right choice in quitting the industry after holding a variety of lucrative sysadmin, software development, IT, and technical lead positions from 1983 to 2009. Too many projects where getting it done mattered more than getting it right, ending up in the software equivalent of a Deepwater Horizon rig explosion. I'm so glad to be done with that.
Holy hell the code for the Android OS StatusBarPolicy in the StatusBarPolicy.java file is a stinking mess. So much for Google having the best programmers in the world. A single public method -- installIcons() at the class level, and a pile of private methods doing all sorts of things. Hundreds of lines of different private variables and worst of all the slew of private anonymous classes.
This sort of mess make single responsibility principle weep.
Slashdot Mirror
Thanks to self-interested politicians like Utah's Orrin Hatch and others who'd rather fatten up on pork, China has a viable space program, while the US just has a money sink that keeps corporations flush in fat lobbying budgets.
... and wwwwrrrrrrrrrrrriggling
What is it that determines whether a jerk sticks around long enough to succeed and become someone at the top that others excuse and follow or is fired with a "do not rehire" mark on his HR record?
Perhaps the gameplay of FarmVille offers a lesson: you'll put up with shit if you get a little candy now and then, with a promise of a big payout later. Guys like Pincus have a knack for showing just enough promising results (cash profit) to whet the appetite while exploiting the "compulsion loop" of the people they work with. Nice work, if you can stand yourself.
http://www.angryflower.com/aposter3.jpg
Makes total sense, as Zynga's 'games' are far more like casino slots and other sorts of gambling than real skill- or puzzle-based games.
Bjeat mje tjøø ijt.
I figured any software developer worthy of the title had heard the story and understood the details already, so I merely linked to the Wikipedia article for reference. Yes, the full filesystem was the root cause, but the repeated reboots and failure of commands to the rover to go into night-time shutdown arose from a race condition within a critical sequence. If you want a more technical analysis, you'll find many, but one is http://www.cs.princeton.edu/courses/archive/fall09/cos109/mars.rover.pdf
While no doubt the atomicity of rename() will prop up the code of naive programmers, there's a great deal more to the operation in question than just changing the file name. Once again, oversimplifying the solution ensures that subtle bugs remain. As for the comment "relatively safe", that's like saying a 39-story fall from the roof of a 40-story building is "relatively healthy".
And of course, if your world is only POSIX and Windows, then portability is still an unrealized goal.
Every two-bit peachfuzz-face programmer in the open source world thinks he has the solution to lots of long-standing bugs. Being able to read what a spec says and actually understanding how it works in actual implementations. For example, what's the behavior of rename() if the file is actually a symbolic link? Patches or GTFO.
The fix is trivial: simply write the new config file out-of-place, and then replace the original with it once it has been fully written.
You left out, "and do it in a way that avoids race conditions". You must also do it in a way that is portable, so it can't depend on OS-specific crutches, and is secure and doesn't interact badly with locking mechanisms that may be transparently in place on the underlying filesystem (NFS, for example). Programming 101 may teach you some things, but most assuredly if that's all you know, a "trivial" write-new-and-replace implementation will be worse. If you are so 100% sure you know how to avoid all these problems, there's some programmers who troubleshot a similar race-condition error from millions of miles away who might have even more possible failure modes to throw at you.
September is Cyborg month, and the 18th is Talk Like a Pirate Day. What does that mean?
http://zenisstupid.com/wp-content/uploads/2010/01/cyborg-pirate-ninja-jesus.jpg
OK smart boy. The mozilla project is open source. If this bug is so all-fired important to you, get to coding, or because you admit you don't have the skills, hire someone who can code it. Don't go whining "THEY won't fix a bug I care about". This 'bug' isn't going to get fixed simply because the folks who could fix it have far better things to spend time on than working around an arbitrary limitation in a not-too-far-in-the-future legacy 32-bit platform. I would be willing to wager that your assertion that web pages have lots of GIFs any more is false anyway. The PNG and JPG formats have pretty much completely replaced that 80s-era 256-color format.
As a matter of fact, with the introduction of things like infallible malloc, you're likely to see more badly-engineered platforms crashing on crappy web pages.
But again I say, if YOU don't like the bug, then YOU fix it, and submit a patch. If they don't accept your patch, then you can build your own FF from source and patch it however crazy way you want.
And by the way, the wikipedia page on GIF, http://en.wikipedia.org/wiki/Graphics_Interchange_Format, doesn't crash FF 3.6.10 running on OS X Snow Leopard in 32-bit mode. As they say in the business, WORKSFORME.
Comparing 3D tv to color is completely missing the problem. The move to color made sense and was a natural progression that mirrored film. The move to 3d is just another to way to get consumers to consider their old equipment obsolete and force them into buying new TVs. Like a lot of the consumer electronic gadgetry out there, the benefits are questionable at best and the upgrade treadmill only benefits the vendors. Now that they've sold digital to everyone and forced the decades-old standard into obsolescence the electronics makers smell blood. What better way to ensure profits than following the personal computer model of convincing everyone that the perfectly good equipment they bought two years ago is now "outdated" and needs replacing.
Anyone remember the 1981 movie "Comin' At Ya!"? Noted at the time for all the action shots of exaggerated movement towards the camera (arrows, knives, boobies), it was nothing but a gimmick. Sounds like almost 30 years later the salespeople are still pushing the same tricks.
There was a time, in the early days of "horseless carriages" when the law required that a person on foot carrying a flag precede the presumably scary, steam-huffing monstrosity, to warn the pedestrians and horses that the oncoming fuming and chuffing machine of the devil was approaching.
Wear a different hat each day.
I put on my robe and wizard hat.
There a many corporate-sponsored contests like this photography, mostly geared at amateurs. Back in the 80s I learned to look at the terms carefully, and if anywhere in them was a clause giving up rights to the photographs entered to the contest-holder, to run far away. Prestigious contests always make it clear that all rights remain with the photographer, although they may legitimately request a time-limited right to display entries for promotional purposes only, not for resale ever.
Stock agencies used to use these contests to pick up vast swaths of decent, if unremarkable, photographs for almost nothing, and with no pesky trouble like having to keep track of who took the photo for credit and payment. I imagine now with Flickr and the flood of digital images, they don't really have work even that hard.
All this remains true for programming contests, and really any contest where the creative work of an individual is made available to another party.
Yes but what you linked isn't ready for prime time yet. Good specs to be shooting for, though.
How about 50kW?
The article starts from a false assumption: that the postal service must be profitable, or at least break even.
Framing the issue this way has nothing to do with what the USPS should or should not carry, or how much they should charge.
Why is that so for the postal service but not for the military, department of transportation, or most any other government agency that provides a service? Universal free mail delivery is something that the citizens of the US want -- or at least did at one time. As a government service, it's something taxpayers agree to pay for.
Now clearly the two authors of this article, management consultants, have a different view of that need. Perhaps they are ideologically inclined to expect that government services should break even or better, in which case, they ought to take on a real challenge and explain to the Pentagon how they can "save" the armed forces. Or perhaps they have a financial interest in private delivery services like FedEX and UPS, who knows? It's clear from early in the article, "Should the federal government continue to compete against the private sector?" that the authors have a sense that somehow there's money to made for UPS, FedEx, and other private delivery services if the postal service was forced to compete on the same level as them. I'm sure they wouldn't advocate for reforming USPS if they thought it would take money away from the private sector.
In any case, before people go trying to reform USPS, let's first decide if we want to continue to support the current expectation of free (for the recipient) door-to-door mail service for everyone in the country everywhere. If citizens clearly want that, then budget (and tax) for it, and shut up about billion dollar "losses" that pale compared to the "losses" racked up by other services we expect as a modern nation. On the other hand, if the country decides that hey, we don't need to deliver everywhere any more, then go ahead, revamp the postal service to be just another profit-motivated competitor.
You mean toeing the line.
If I were SpaceX founder Elon Musk, I'd be hopping mad right now. After developing Falcon9 and Dragon on the basis of a truly competitive commercial space program, the porkbarrel senators for aerospace/defense contractor states wrote a new NASA budget to basically hand money over to Boeing and the rest of the usual cast of trough-feeders to continue but with changes and more delays the Ares/Orion program. This craft will see about as much reality as the Orion did before Boeing is behind schedule and over budget and requests yet more money.
The whole goal is to crowd out the smaller guys while maintaining the jobs programs in states like Washington, Utah, and Florida.
Did anyone notice that they don't say where they are going in this capsule? Where are the senators who called Obama's proposed budget a mission to nowhere? This new NASA program doesn't have a destination, either, but at least the dollars keep flowing to the same interests.
I haven't looked at the guts of any of Apple's NS* classes, but I don't doubt there's some places where it's a mess -- it's ancient code, those, where the Android code is new. But just because the feature is part of the SpringBoard application doesn't mean it's badly coded. What classes in the NS SDK specifically?
You're right about one thing: You've reminded once again that I made the right choice in quitting the industry after holding a variety of lucrative sysadmin, software development, IT, and technical lead positions from 1983 to 2009. Too many projects where getting it done mattered more than getting it right, ending up in the software equivalent of a Deepwater Horizon rig explosion. I'm so glad to be done with that.
Yes, a C programmer with just enough familiarity with higher-order function and lexical closure constructs to be dangerous.
Holy hell the code for the Android OS StatusBarPolicy in the StatusBarPolicy.java file is a stinking mess. So much for Google having the best programmers in the world. A single public method -- installIcons() at the class level, and a pile of private methods doing all sorts of things. Hundreds of lines of different private variables and worst of all the slew of private anonymous classes.
This sort of mess make single responsibility principle weep.
http://www.angryflower.com/bobsqu.gif