In Descent II, there's a level towards the end with a very large room just past the start room, with a door directly across from the door to the start room, and lots of crossbarred windows to rooms all over the level that require keys for the ship to reach, but which missiles can fly between. If you start the level with a full load of guided missiles, and you get really, really good, you can take out more than half the bots in the level without even moving from the start spot. I probably got more replay value out of my saved game at the beginning of that level than the rest of the game combined.
...to suppress the output of a particular gdb feature. With a particular combination of kernel and gdb versions, single-stepping broke control flow, but enabling this feature (which produced a line of output per instruction) fixed it. The customer apparently had very strict change control, so rather than update the kernel, they preferred to have hundreds of their developers use a debug switch they didn't really want, and then use a second switch, found nowhere else in the world, to hide the debug data they had just requested, so it wouldn't clutter the output and hide what they really cared about.
I think the hallucinations began around day 6. I was coding for a class project, and keeping notes. The notes got very weird around then. Coding is a bit less stressful than combat though.
More than anything else, an undiagnosed anxiety disorder. There was also caffeine and adderall, but not in any higher doses than I'd used regularly in the past. I was spending most of the time coding, until that somehow transitioned to studying topology on wikipedia so I could describe the performance of my networking protocol in a 4-dimensional toroidal universe. Based on my notes, I think I went psychotic around day 6.
I didn't say I was functional! It started out as a coding binge for a class project. Based on my notes, I later concluded that some time around day 6 I began reading wikipedia articles on topology, to formally describe the performance of my networking protocol in a network built in a 4-dimensional toroidal universe.
Adderall, caffeine, and an undiagnosed anxiety disorder were also involved.
I once went 9 days without sleep. After 22 hours of sleep I woke up in severe pain, as an injury I had suffered halfway through, which seemed very mild in my sensory-depressed state, was in fact something that required medical attention. If it had been only a tiny bit worse, I could have developed life-threatening complications after several days of ignoring and aggravating it. Impaired motor control, pain sensitivity, awareness, and judgment, all at the same time, is a dangerous combination.
The CEO of the studio that released "The Pink Panther 2" is in no position to lecture anyone about a sense of entitlement. The Onion's commentary on this is barely even satire at this point:
Just write a one-liner that replaces all calls to memcpy with a call to memcpy_s, duplicating the size parameter.
I'm only half-joking. This is exactly how people will (mis)use memcpy_s. If you want safe memory access, you need to ban the entire C language. For those cases where you need C, you'll just have to make sure your programmers know what they're doing.
Most of them won't go into detail, but Wall Street firms have immense server farms. Some of them are limited in size by the amount of electricity the New York City power grid can supply them. They also have huge data centers in less prime real estate, but microseconds are dollars in the financial markets, so they try to keep as many of their systems as close to the action as possible. There are entire floors of NYC skyscrapers full of racks modeling the financial markets in real time, conducting transactions, and crunching numbers for human analysts.
...is that if you don't sell, they'll just make it themselves. They have vastly more resources than you do, so they could swamp you unless you get some serious venture capital immediately. But it's cheaper for them to buy you, so they'd rather do that.
The money you could make going it alone is by no means certain, but the money you make by selling is. As long as the terms are such that you're allowed to leave, with the money, and start another company doing something similar (they'll probably have a non-compete for something identical), you get to keep the freedom and start over with more resources.
Many entrepreneurs go through this cycle several times, building their experience and their resources until they reach the point where they can raise the kind of venture capital that allows them to become the megacorp themselves. Sometimes they enjoy being on top, and sometimes they avail themselves of the freedom to do whatever they want, and go on to start more successful companies.
Personally, I'd sell out, as long as the terms gave me plenty of freedom. You might even enjoy working at the megacorp. Some of them have very healthy cultures.
...is worth a pound of cure, as the saying goes. Most of the demand for antivirus software is in compensating for the afterthought-hacked-on security model of Windows. For a tiny fraction of the cost of treating this problem, the US government is sponsoring work on SELinux and OpenBSD to create more robust security models that thwart novel attacks with negligible maintenance and overhead compared to constantly scanning a system for known malware using an always-out-of-date database.
The glaring hole in this economic argument is that subsidizing antivirus software would decrease the incentive for Microsoft to make its OS more secure in order to maintain a competitive TCO with other operating systems, and decrease the incentive for people to migrate to more secure operating system.
In the US, HD cable providers are required by the FCC to give you a receiver with firewire output if you ask. You may have all the hardware you need for mythTV, without even buying a tuner card.
I'd be curious to see a comparison of kidney and thyroid diseases in these areas, as well as other psychiatric indicators. Lithium doesn't just reduce suicide risk.
It's not the bandwidth, it's the latency. In a quiet house, in the dead of night, I get 2 ms pings to my 802.11g router, in the best case. The average is around 3 ms, and if I so much as sneeze it'll spike into the tens of milliseconds, due to retransmits. Those are tiny packets, and there's negligible interference in the area. When you pack an office full of users, 2.4 GHz bluetooth cell phones, microwave ovens, etc., you'll get averages in the tens of milliseconds with frequent spikes in the hundreds. Some packets will get dropped completely, causing TCP stalls and UDP DNS timeouts.
Honestly, the average web/popmail user *still* won't notice this, but the instant you start opening files on network shares and large IMAP mailboxes, demand paging data as you scroll through anything large, everything will slow to a crawl.
At my old job, IMAP was faster when I was logged in from home (wired or wirelessly, since my home was relatively RF-quiet) than when I was in the office and unplugged. Large attachments still went faster at work, because I had plenty of bandwidth, but the latency made many tasks painfully slow.
For many small businesses, the article's assertion may be accurate, but don't go ripping the cables out of large cube farms just yet.
If you're doing a 100 meter run, your boss is absolutely right. If you're going a few feet, hand-made is fine. The cut-off point depends on your skills and equipment, and varies widely.
Seriously, I thought it was all about the social stature, earnings potential, open culture, plentiful recreational substances, and sea of prospective sex partners. Classes are when you sleep.
a) where the bomb threat came from. b) which building the suicidal student needs to get talked down from. c) who impersonated the professor to cancel an assignment. d) how a lab router ended up sniffing for passwords.
All of these things happened while I was in campus IT, but I never heard about an RIAA/MPAA complaint about something that happened less than two weeks prior, so this really doesn't look like undue outside influence to invade student privacy. It's just responsible network management.
...with naive freshmen, students of mine I dared not touch, desperate for friendship in a new place and approval from authority in the midst of their newfound freedom, hitting on me.
I'm sure it would have some effect, but it certainly wouldn't make them get into the CPU/GPU/chipset business by buying AMD. They're a software company, aside from the Zune and Xbox, which are made with other companies' components.
When Red Hat bought JBoss, Oracle retaliated with Unbreakable Linux, but that doesn't change the fact that Oracle servers are Red Hat's bread and butter, and that Oracle needs RHEL to keep Microsoft at bay. It's called coopetition. Sun and IBM both have their own architecture, OS, and database, yet they make a fortune selling x86 Linux servers that run Oracle apps.
My experience with memtest is that it will improperly attempt to test MMIO regions, which will fail because of side effects of reads and writes, giving false positives that are usually repeatable and at low addresses.
I've also seen it trigger thermal problems that you might encounter during high-load activities like gaming, but never have a problem with during light use. On one such machine, using the powersave CPU governor kept peak power consumption down enough during heavy gaming that it remained usable, albeit somewhat slower. On another machine that didn't support frequency scaling, it was sufficient to place an old floppy disk between the CPU heatsink and the RAM to keep the RAM from overheating.
Memtest will tell you if you have a potential memory problem, but it won't tell you why, and it has no idea if the problem will affect you during normal use or if there's an easy workaround. I value its results, but I never make a decision based on that information alone.
Slashdot Mirror
In Descent II, there's a level towards the end with a very large room just past the start room, with a door directly across from the door to the start room, and lots of crossbarred windows to rooms all over the level that require keys for the ship to reach, but which missiles can fly between. If you start the level with a full load of guided missiles, and you get really, really good, you can take out more than half the bots in the level without even moving from the start spot. I probably got more replay value out of my saved game at the beginning of that level than the rest of the game combined.
...to suppress the output of a particular gdb feature. With a particular combination of kernel and gdb versions, single-stepping broke control flow, but enabling this feature (which produced a line of output per instruction) fixed it. The customer apparently had very strict change control, so rather than update the kernel, they preferred to have hundreds of their developers use a debug switch they didn't really want, and then use a second switch, found nowhere else in the world, to hide the debug data they had just requested, so it wouldn't clutter the output and hide what they really cared about.
...embedded.
Nothing to see here. Move along.
I was in grad school. I don't remember it very well, but my notes got really weird around day 6.
I think the hallucinations began around day 6. I was coding for a class project, and keeping notes. The notes got very weird around then. Coding is a bit less stressful than combat though.
More than anything else, an undiagnosed anxiety disorder. There was also caffeine and adderall, but not in any higher doses than I'd used regularly in the past. I was spending most of the time coding, until that somehow transitioned to studying topology on wikipedia so I could describe the performance of my networking protocol in a 4-dimensional toroidal universe. Based on my notes, I think I went psychotic around day 6.
I didn't say I was functional! It started out as a coding binge for a class project. Based on my notes, I later concluded that some time around day 6 I began reading wikipedia articles on topology, to formally describe the performance of my networking protocol in a network built in a 4-dimensional toroidal universe.
Adderall, caffeine, and an undiagnosed anxiety disorder were also involved.
I once went 9 days without sleep. After 22 hours of sleep I woke up in severe pain, as an injury I had suffered halfway through, which seemed very mild in my sensory-depressed state, was in fact something that required medical attention. If it had been only a tiny bit worse, I could have developed life-threatening complications after several days of ignoring and aggravating it. Impaired motor control, pain sensitivity, awareness, and judgment, all at the same time, is a dangerous combination.
The CEO of the studio that released "The Pink Panther 2" is in no position to lecture anyone about a sense of entitlement. The Onion's commentary on this is barely even satire at this point:
http://www.theonion.com/content/news/vindictive_movie_studio_threatens
...will be podcast.
Just write a one-liner that replaces all calls to memcpy with a call to memcpy_s, duplicating the size parameter.
I'm only half-joking. This is exactly how people will (mis)use memcpy_s. If you want safe memory access, you need to ban the entire C language. For those cases where you need C, you'll just have to make sure your programmers know what they're doing.
Most of them won't go into detail, but Wall Street firms have immense server farms. Some of them are limited in size by the amount of electricity the New York City power grid can supply them. They also have huge data centers in less prime real estate, but microseconds are dollars in the financial markets, so they try to keep as many of their systems as close to the action as possible. There are entire floors of NYC skyscrapers full of racks modeling the financial markets in real time, conducting transactions, and crunching numbers for human analysts.
...is that if you don't sell, they'll just make it themselves. They have vastly more resources than you do, so they could swamp you unless you get some serious venture capital immediately. But it's cheaper for them to buy you, so they'd rather do that.
The money you could make going it alone is by no means certain, but the money you make by selling is. As long as the terms are such that you're allowed to leave, with the money, and start another company doing something similar (they'll probably have a non-compete for something identical), you get to keep the freedom and start over with more resources.
Many entrepreneurs go through this cycle several times, building their experience and their resources until they reach the point where they can raise the kind of venture capital that allows them to become the megacorp themselves. Sometimes they enjoy being on top, and sometimes they avail themselves of the freedom to do whatever they want, and go on to start more successful companies.
Personally, I'd sell out, as long as the terms gave me plenty of freedom. You might even enjoy working at the megacorp. Some of them have very healthy cultures.
...that Trent could use a warm blankie.
Of the various things about raising children, finding a job closer to home (or moving home closer to work) is not one of the harder ones.
Clearly you haven't been job-hunting lately.
...is worth a pound of cure, as the saying goes. Most of the demand for antivirus software is in compensating for the afterthought-hacked-on security model of Windows. For a tiny fraction of the cost of treating this problem, the US government is sponsoring work on SELinux and OpenBSD to create more robust security models that thwart novel attacks with negligible maintenance and overhead compared to constantly scanning a system for known malware using an always-out-of-date database.
The glaring hole in this economic argument is that subsidizing antivirus software would decrease the incentive for Microsoft to make its OS more secure in order to maintain a competitive TCO with other operating systems, and decrease the incentive for people to migrate to more secure operating system.
In the US, HD cable providers are required by the FCC to give you a receiver with firewire output if you ask. You may have all the hardware you need for mythTV, without even buying a tuner card.
I'd be curious to see a comparison of kidney and thyroid diseases in these areas, as well as other psychiatric indicators. Lithium doesn't just reduce suicide risk.
It's not the bandwidth, it's the latency. In a quiet house, in the dead of night, I get 2 ms pings to my 802.11g router, in the best case. The average is around 3 ms, and if I so much as sneeze it'll spike into the tens of milliseconds, due to retransmits. Those are tiny packets, and there's negligible interference in the area. When you pack an office full of users, 2.4 GHz bluetooth cell phones, microwave ovens, etc., you'll get averages in the tens of milliseconds with frequent spikes in the hundreds. Some packets will get dropped completely, causing TCP stalls and UDP DNS timeouts.
Honestly, the average web/popmail user *still* won't notice this, but the instant you start opening files on network shares and large IMAP mailboxes, demand paging data as you scroll through anything large, everything will slow to a crawl.
At my old job, IMAP was faster when I was logged in from home (wired or wirelessly, since my home was relatively RF-quiet) than when I was in the office and unplugged. Large attachments still went faster at work, because I had plenty of bandwidth, but the latency made many tasks painfully slow.
For many small businesses, the article's assertion may be accurate, but don't go ripping the cables out of large cube farms just yet.
If you're doing a 100 meter run, your boss is absolutely right. If you're going a few feet, hand-made is fine. The cut-off point depends on your skills and equipment, and varies widely.
Seriously, I thought it was all about the social stature, earnings potential, open culture, plentiful recreational substances, and sea of prospective sex partners. Classes are when you sleep.
...is just enough time to figure out:
a) where the bomb threat came from.
b) which building the suicidal student needs to get talked down from.
c) who impersonated the professor to cancel an assignment.
d) how a lab router ended up sniffing for passwords.
All of these things happened while I was in campus IT, but I never heard about an RIAA/MPAA complaint about something that happened less than two weeks prior, so this really doesn't look like undue outside influence to invade student privacy. It's just responsible network management.
...with naive freshmen, students of mine I dared not touch, desperate for friendship in a new place and approval from authority in the midst of their newfound freedom, hitting on me.
It was torture.
I'm sure it would have some effect, but it certainly wouldn't make them get into the CPU/GPU/chipset business by buying AMD. They're a software company, aside from the Zune and Xbox, which are made with other companies' components.
When Red Hat bought JBoss, Oracle retaliated with Unbreakable Linux, but that doesn't change the fact that Oracle servers are Red Hat's bread and butter, and that Oracle needs RHEL to keep Microsoft at bay. It's called coopetition. Sun and IBM both have their own architecture, OS, and database, yet they make a fortune selling x86 Linux servers that run Oracle apps.
My experience with memtest is that it will improperly attempt to test MMIO regions, which will fail because of side effects of reads and writes, giving false positives that are usually repeatable and at low addresses.
I've also seen it trigger thermal problems that you might encounter during high-load activities like gaming, but never have a problem with during light use. On one such machine, using the powersave CPU governor kept peak power consumption down enough during heavy gaming that it remained usable, albeit somewhat slower. On another machine that didn't support frequency scaling, it was sufficient to place an old floppy disk between the CPU heatsink and the RAM to keep the RAM from overheating.
Memtest will tell you if you have a potential memory problem, but it won't tell you why, and it has no idea if the problem will affect you during normal use or if there's an easy workaround. I value its results, but I never make a decision based on that information alone.