I was going to use the last of my mod points on some other posts in this story that gave some really good tips. At least until I saw this post.
Nothing personal, because I'm sure your comments were made with the best of intentions, and in the PC/non-G5 computing world they'd be great suggestions. However, I would not do any of the "standard tricks" with a G5. Here's why:
Apple has done a serious amount of engineering with the cooling in the G5s. Nine fans, 21 temperature sensors and four "thermal zones". Not to mention those gigantor heatsinks. Replacing their fans, or going so far as to run them at different voltages (and thus speeds), may well cause the opposite effect you're looking for. Mac OS X may end up running the fans much more often to keep the ambient case temperature in the zones you've "played with" at their target temperature(s).
You can get a brief overview of their cooling design here: http://www.apple.com/powermac/design.html
Are you still in the original warranty, or did you add AppleCare? If either of those is true, you may not want to start mucking around with the cooling system components. If I were you, I'd check the warranty/AppleCare legalese very closely first to make sure that you're not voiding those agreements -- especially if you end up damaging your system.
Here's what I would suggest instead:
- Close all unnecessary background programs (fewer programs == less CPU usage == less waste heat == less fan noise) when you are recording your voice overs.
- The type of recording discussed in the parent article won't need remotely near the power of the G5. Because of this, it may also help to go into the Energy Saver system preference panel and switch to the "reduced performance" power profile. This should reduce the need for the fans. When you're done recording, switch back to Automatic or Highest.
- Keep discs out of your superdrive when recording. That drive, when it has a disc in there spinning around, can be loud.
- Get a good directional mic and don't have your G5 tower in front of the mic, where you'll be sitting to do the recording.
Having said all that, you may have something else going on that's causing the noise in your machine. I vaguely recall something about a batch of power supplies in earlier G5s that (I think) Apple will replace under warranty/AppleCare because of excess noise/hum. Maybe you could look into that, as well.
And just to put all my comments in more context, I also have a G5 (dual 2GHz), an Mbox and have used ProTools for voice overs. My G5 is dead silent, and always has been. When I'm sitting at my desk, I literally hear more fan noise from other people's machines down the hallway than I have ever heard from something inside my office, including my G5.
I can't recall, since getting my G5 last October, ever hearing the fans kick into a high speed (and this includes the times I've had FCP4, ProTools, Cinema4D, DVD Studio Pro, Photoshop and LiveType all running at once). And on the recording side, for reference, I've been using a Shure KSM27 mic.
With one weird, but possibly-not-related, exception.
I did the 10.2.8 update on my G4 PowerMac. After the initial reboot and before I had a chance to actually do anything, my keyboard and mouse stopped responding. But a power cycle later and things seem to be working fine. I've since run a couple Classic apps (mainly the scanning software for a Canon), plus OS X Photoshop 7 and connected to a couple Samba shares on a Linux server. The only network connection is wired Ethernet.
I should have thought to try and ssh in before cycling the machine, but I didn't. So, for all I know it could have either been just the kb/mouse, or it could have been the entire machine, that froze up.
What about game saves? Sure, it could possibly include drivers for the current filesystems (of course, NTFS r/w was/is still pretty sketchy under Linux last time I checked), and the boot process could seek out the current drives. Of course, they'd also need to be able to support all the ways people connect drives to their systems. If I'm saving my progress in a PC game right now, it doesn't matter whether I'm saving it to an IDE drive, SCSI, USB, Firewire, network share (Samba, NFS, Netware, etc.), or even battery-backed RAM disk.
But, even assuming they could manage to handle all the currently supported filesystem types and all the ways of connecting them that already exist, what happens when new FS types come along? "Sorry, sir, but your machine is too new for our game to be able to provide you the ability to save your progress." I don't think that will cut it.
I was going to mod this down (overrated), but decided I'd rather reply.
No one said that honeytokens are superior in every way to honeypots and should be used in place of the latter. That you pulled out of your hindquarters. Basically, what you said could be expressed similarly in this example: "Seat belts are not absolutely superior in every way to the steel frame of a car, so what's the point in buckling up?"
I would hope that makes it clear how faulty your logic is. Like using seat belts in addition to a protective steel frame, to provide added protection, honeytokens could be used in addition to honeypots. Their ultimate goals are the same: protect your life (frame/seat belts) or your data (honey[pot|token]). If your life/data is that important, why not provide all the layers of security you can?
One advantage that honeytokens do have is in who they can help protect against. Honeypots are typically deployed to detect and help figure out how to protect against external threats. Anyone with a shred of sense about security knows, however, that you also need to protect against internal threats. Deploying honeytokens can help in that vein, by posssibly detecting internal abuse of your systems.
Just because honeytokens won't protect against everything, solve global hunger, and bring about world peace, doesn't mean they shouldn't or can't be used effectively.
I've been playing VJ for a couple days now (got mine Tuesday -- imported it through videogamedepot.com) and love it. I've always been a platformer at heart, and this game definitely makes me happy so far. For any US cuber's who want to get this *now* but don't speak Japanese and are worried about not knowing what's going on, stop worrying and just get it. Much of it is in English anyway.
My only beef with it was when I first popped it in and just wanted to get straight to the action. You have to sit through a drawn out intro that can't be skipped (it seemed to go on forever the first time around). If you do import it (and if this is your first import), make sure you either have a modded GCN or the "Freeloader" so that you can actually play it, and also make sure to have a separate memory card (you cannot save US and JP games on the same card).
And some people seem to forget that just doing something "over the Internet" does not exclude you from the laws of your country/state/province/city/etc. Slashdot is owned by a company registered and operating in the U.S. There are a crapload of laws regulating contests, sweepstakes, and other things of that nature in the U.S. (which vary by state, many of them placing all sorts of restrictions when there are prizes, monetary or otherwise, involved). Slashdot, as with all companies in the U.S., is required to abide by those laws (or restrict entry from residents of states where they cannot or will not abide by the laws).
And before you decide to then stop blaming Slashdot and blame "the U.S.", pick any other industrialized country in the world and know that it will be the same way there. If Slashdot was a Belgian company, they'd almost certainly be restricting the contest to Belgium (or at most EU countries), which would still leave out a significant majority of the world.
I have a couple of those and the quality is very solid (held up through many washes, as well). Those kinds of designs probably aren't done more often because of the higher costs of production compared to a shirt with two simple, solid colors in a basic design. Not to mention you have to spend a lot more time actually creating the design, checking samples for registration misalignments, etc.
Err... I did fail to mention that the Japanese CDs are more expensive than the Korean/Chinese ones. K/HK music generally goes for $7-$9/CD, but the Japanese CDs go for the usual "import prices" ($22-$30). Forgot about that when I first posted, since I listen to a lot more K/HK music.
If those are your thing, check out YesAsia. I've been shopping with them for years now. Once a month or more, I place a huge order with them. Music, movies, anime, etc. Most of the CDs (yes, legit) are $7-$9, which I actually consider to be a fair price (opposed to most american pop albums at $18-$20).
I also buy most of my movies from them. A lot of the movies can be bought three at a time for $10/each (if you don't buy them in 3s to get the discount, they're usually $12-$15). VCDs are even cheaper if you want to go that route. You can get region-free DVD players from them as well (the Shinsonic is not a bad set, although there is a weird "vertical line ghosting" issue on large blocks of static, dark, solid colors).
I very highly recommend them. One note of warning is that they take a few days to process and ship orders, since they often have to ship some stuff from their HK (I assume) warehouse to California before it will ship out to a US customer. They do seem to have been trimming this turnaround down a little (average delivery time, from time of order, using overnight is now one week), so either they're getting a little more efficient or they know me well enough that I'll keep on coming back spending hundreds each month, as long as they keep me happy.;-)
Anyway, if you're into Korean, Japanese or Chinese music/movies/etc, definitely give them a look. And, no, I do not work for them or have any relationship with them other than being a very satisfied customer.
In his article he actually does address this very question. He even gives, what I feel at least, is an interesting answer.
So, you don't run an open relay. You're not going to slow down the spammer directly, but you will slow down all the connections that come from that open relay to your mail server. For a particularly abused open relay, that could lead to such problems that the admin of that open relay will finally get a clue and look in to configuring their server properly.
Hence, a cascading effect that will eventually harm the spammers. Admins of open relays that get a clue will tighten their servers, thus depriving the spammers of one more relay they can abuse.
Without much hassle, you can use XS to allow your Perl scripts to interface with C/C++ libraries in a natural way. In fact, many Perl modules are written in C/C++ and wrapped with XS. This allows you to get the best of both worlds in some situations -- streamlined C code to offload intensive operations into separate libraries, then rapidly prototypable Perl front ends to those libraries (whether it be mod_perl for web apps, Perl/GTK for GUI, or any other type of front end you need).
It's not always the perfect solution, but your post makes it seems as if you don't think there's any way to get C and Perl to talk. Few things are further from the truth. And this isn't just some nifty Perl thing -- many of the "scripting" languages (I can't speak for PHP) offer similar ways to accessing C/C++/etc. I'm just most familiar with Perl's way.
Very poor analogy, actually. Microsoft is in a very special class of companies: monopolies. Those companies have to play by a whole new set of laws that don't apply to other companies. Especially when said monopoly infringes those laws with regards to competing companies. Not to mention Microsoft is a publicly-traded company, whereas Google is privately held, though that fact has less bearing on pointing out the fallacy of your comment than the monopoly issue.
Actually, "in" would be inappropriate for the use you describe. Using "in" would imply that "Linux" or "Unix" are programming languages. It would be better to use "under" or possibly "on". "Programming under Linux" or "Programming on Unix in C".
I agree with your basic point, however, that simply saying "Programming Linux" implies development work on the core components of a Linux system. If you really wanted to be pedantic about it, it would imply simply the Linux kernel.
Try this version instead, it follows the incredible verbosity of the original question better:
I've been wrestling with this problem for some time now. I'm looking to shoot myself in the foot using some sort of firearm. I will probably use a pistol or revolver, as I believe that will be easier to handle as I assume I will be pointing directly downward (unless I decide to shoot myself in the foot while in bed). Unfortunately, the pain that I hear may result could prevent me from possibly being able to fire more than a couple rounds. My question is this: Is there a real benefit from using four smaller bullets as opposed to using two larger bullets when I shoot myself in the foot? I've found some people seem to think that a 9mm round will be more effective for this use... But, purchasing two.38 rounds appears to be cheaper than four 9mm rounds. Will the effectiveness of shooting myself in the foot suffer if the two.38 rounds are used, even if they are larger? I've searched high and low using google and have yet to find someone else who has collected any data on this matter. Take these firearm/ammunition combinations as an example: A revolver with two.38 rounds versus a pistol with four 9mm rounds. The revolver would be more expensive, because I am limiting myself to a model which is not used very frequently and is slightly more difficult to find readily available. I would also have to purchase a full box of ammunition even though I am only considering using two rounds (I don't know how to ask the clerks for only two rounds -- but that's another Ask Slashdot), thus inflating the price/use ratio. And I know a.38 round wouldn't have the spread of a 5 gauge buckshot round (if it existed) for a shotgun, although this comment has almost no relevance to my question (and makes a comparison to a product that doesn't even exist, nor will ever), but yet I still want to think it would be more effective. The firearms I am looking at are both manufactured by Colt and have the same basic features. Maybe some Slashdot readers, who have shot their own feet, could shed some light on this topic?
In raw disk storage, maybe. But you're forgetting actually putting those drives into a useable state with disaster recovery plans.
In other words, someone dealing with 50TB and who wants backups of that data will be spending many, many times the amount it would cost to just purchase enough hard drives to get the bragging rights of 50TB. And a backup located in the same room/floor/rackspace/whatever as the source data will be pointless in the event of fire, floods, nuclear fallout, etc. So, they would also need a way to transfer all that data to offsite backups in a timely manner (waiting five weeks for a full backup to transfer over a 100Mb/s pipe would probably not be acceptable).
Aside from backups, how would the drives be accessible? Even as JBOD, you're talking 40 IDE/ATA controllers (assuming 320GB drives and 4 ports per controller), or 20 SCSI channels (assuming 160GB per drive and 15 non-host devices per channel) to support that many disks. You could also use Fibre Channel and get away with only a couple arbitrated loops. Physically, you're talking about hundreds of disks that need to be mounted somewhere, so you would also need dozens of chassis to hold the drives.
But, hundreds of disks in a JBOD configuration means you'll have hundreds of partitions, each separate from the others. Hell, if the clients are Windows machines, they won't even be able to access more than a couple dozen at a time. And even for operating systems with better partition/mount-point addressing, it would be unmanageable.
So, now you get in to needing a RAID solution that can tie hundreds of disks together. If you're talking about hooking these up to standard servers through PCI RAID cards, you'll need several of those machines to be able to host all the controllers necessary (especially if all the disks are not 160GB or larger each).
The only realistic solution for this much storage, at least until we have 5TB hard drives, is a SAN-like setup. Specialized hardware designed to house hundreds of disks in stand-alone cabinets and provide advanced RAID and partitioning features. SANs don't come cheap.
Add to the SAN the various service plans, installation, freight, configuration, management and the occasional drive swapping as individual disks fail and you've already multiplied that $50K several times, as a bare minimum (and you still haven't priced out the backup solution).
There's a lot more to it than just having a pile of hard drives on the floor. I wouldn't even be surprised if the drives are the cheapest component.
I generally have two schedules throughout the year; a "big project" schedule and a "maintenance" schedule. The first involves 14 to 18 hours a day, six days a week (regardless of holidays), and tends to last a month or two at a time. The other one involves generally between 9 to 12 hours a day, five days a week (and actually staying home on holidays).
I typically do not have the big project schedule more than twice a year. I work for a small company (well, technically, a small division of a medium sized company) and am solely responsible for the development of our software. There's no COTS or open source software out there that meets our needs for functionality, ease-of-use and performance (at least not any that is less expensive, including whatever it would cost for 24/7 onsite service and support from the developers who wrote the software, than my salary and benefits), so it falls on me to write virtually all the software that runs our division (minus the operating systems, office suites and database servers, of course).
But, there are two main reasons why the difference in schedules and the very heavy loads that happen a couple times a year don't bother me.
The first reason is that project planning in our division is managed well, for the most part. I can think of a couple times where things did not go as planned, but they're in the minority. We also generally plan well in advance. I'm currently in the middle of a big project schedule right now, but I have known this was coming since the beginning of the year. Also, because we are a small group (less than a dozen people total for our division), I have a strong say in the project plans. If I mark something as unreasonable, we generally either don't do it, or we come up with an alternate version that is reasonable.
The second reason is that I *really* enjoy my work. If I'm not at work programming, chances are that I'm at home programming (usually for work) or sleeping. I do very little other than code. I enjoy having to solve difficult problems. You could certainly take the perspective that I simply have no personal life, and I would not argue one bit with that. I don't have any illusions of having a personal life -- but that doesn't bother me at all. I am a very dull person outside of work, anyway, because I tend to only think about work.
Supporting reasons for why I don't mind include the fact that I feel I am compensated well for my work and that I am valued by the other people in the company. I am also constantly learning -- especially when I am confronted with the more complex problems. Solving the really difficult problems may be painful while I'm doing it, but at the same time I consider it to be extremely rewarding because it gives me an excellent opportunity to expand my knowledge.
Like the saying goes, your finest teachers are your worst adversaries (in my case, the most complex problems). If everything is "point and click" easy, you are never challenged to think and confront problems head on and find their solutions. You have significantly reduced chances to grow or have the opportunity to expand yourself if you aren't challenged in some way.
Possibly just as important, psychologically, for my acceptance of, and willingness to commit to long work hours and tight deadlines is that I have a significant emotional/personal investment in the company, having been one of the people that started it. I still hold on to a lot of the "founder" and "startup" work ethic mentality from our company's early days, even though we have grown beyond those stages.
Will I burn out one day because of the long hours I work? Probably, but I hope that day is in the very distant future. And just to tie this post in with the original question, I do not believe that the long hours I put in negatively impact my productivity. I actually put in the long hours mainly because I work better that way.
However, were I not working for a company I cared deeply about, or were I not being challenged intellectually by my work, I would most likely burn out quickly and my productivity would suffer. I consider myself lucky to be in a position where I actually have some very significant and positive motivating factors.
It happens every once in a while. Places like Register.com, Verisign, et. al. that put up those "site coming soon," "buy this domain for $x," and other messages switch their web server software every so often. Don't ask me why they do it exactly, but when it happens it can create a very sudden change in the numbers.
This report, I believe, had a note that one of those domain name parkers switched to Apache from IIS, which probably accounted for at least a noticable chunk of the 6% swing.
Thanks for the info on that. I had not been aware that the Orbs were so poorly thought of. And, considering I'm not a physicist (just a mere programmer), I would never have thought of heat transfer issues with Al-Cu.
Tell that to an Athlon 1.2GHz I had for a while. No overclocking, whatsoever. I had an Orb on the Athlon, rounded cables for better airflow, and 7 case fans (all 80mm or larger -- three in front, three in back, one on the side), plus the two fans on the power supply, plus an Antec "PCI Bay" fan.
I *never* saw the Athlon in this box under 65C, and under stress (games mainly) it would hit 70+C. After months of trying everything I could, short of water cooling, this _non_overclocked chip finally killed itself because of the heat.
I replaced it with a P4 2.53GHz, in a different case with only three fans, plus the stock Intel heatsink. I've never seen the P4 above 44C sans overclocking, even under stress. Even when I overclocked the chip, the core temp never went over 55C.
Both of these machines were in the same location of my apartment, which is kept at an ambient temp of 60-65F.
Some areas can get BBC World News broadcasts. I only get an hour of it once or twice a day on the local "public" station, but that's enough to cover the major news topics for the day. Course, I signed up with XM Radio last January and BBC News has an entire 24/7 station on XM.
You can also stream the news (audio and video) from the BBC website.
Too much trouble to read their press release? It's not that long. If you had read it, your first two questions would have been immediately addressed by the following:
"Clients which connect to our peer-to-peer clients, and then afterwards attempt to illegally access the network will be immediately blacklisted from Information Wave's network."
It does not directly address your third question about spoofed IP addresses, but they could address that at the routers.
PCI cards still make sense for some of us, even if we've never stepped foot inside a Walmart (or outside the office, for that matter). If you can show me a motherboard that has four AGP slots for all my video cards (I run a multihead setup), then I'll no longer need PCI video cards. Alternately, show me an AGP card with at least four high resolution outputs that works as solidly under Linux as four individual nVidia based cards.
Agreed. However, regarding the comment about the costs of moving from 12 to 13 being the same as moving from 12 to 18, I'd like to add something. I agree that the costs will probably be exactly the same (or such a miniscule difference you'd be a fool to make the same mistake by only going to 13).
You might be surprised by how many people would not make that connection. I have personally dealt with another company that, when faced with a limitation in their software that was truncating object identifiers from our software (the two communicate various information back and forth on a regular basis through XML streams), they "solved" the problem by adding one extra byte to the variable causing the problem. It was currently 8 bytes and they made it 9.
What made it worse is that these identifiers increment rapidly, and the "solution" they came up with would only hold out for a few months and would have to be changed again. I tried to explain it to them, and let them know that our software expects these to be able to expand up to 48 bytes -- which would hold us for decades at minimum.
So, they changed it to be 10 bytes on their end. They obviously didn't get it. I can't really say that I think they would be the exception to the rule, either. I wouldn't be surprised if a lot of other less-than-forward-thinking people would make the same mistake and only increase the upper limit to the bare minimum needed to fix the immediate problem -- without ensuring further expansion in the future.
Slashdot Mirror
I was going to use the last of my mod points on some other posts in this story that gave some really good tips. At least until I saw this post.
Nothing personal, because I'm sure your comments were made with the best of intentions, and in the PC/non-G5 computing world they'd be great suggestions. However, I would not do any of the "standard tricks" with a G5. Here's why:
Apple has done a serious amount of engineering with the cooling in the G5s. Nine fans, 21 temperature sensors and four "thermal zones". Not to mention those gigantor heatsinks. Replacing their fans, or going so far as to run them at different voltages (and thus speeds), may well cause the opposite effect you're looking for. Mac OS X may end up running the fans much more often to keep the ambient case temperature in the zones you've "played with" at their target temperature(s).
You can get a brief overview of their cooling design here: http://www.apple.com/powermac/design.html
Are you still in the original warranty, or did you add AppleCare? If either of those is true, you may not want to start mucking around with the cooling system components. If I were you, I'd check the warranty/AppleCare legalese very closely first to make sure that you're not voiding those agreements -- especially if you end up damaging your system.
Here's what I would suggest instead:
- Close all unnecessary background programs (fewer programs == less CPU usage == less waste heat == less fan noise) when you are recording your voice overs.
- The type of recording discussed in the parent article won't need remotely near the power of the G5. Because of this, it may also help to go into the Energy Saver system preference panel and switch to the "reduced performance" power profile. This should reduce the need for the fans. When you're done recording, switch back to Automatic or Highest.
- Keep discs out of your superdrive when recording. That drive, when it has a disc in there spinning around, can be loud.
- Get a good directional mic and don't have your G5 tower in front of the mic, where you'll be sitting to do the recording.
Having said all that, you may have something else going on that's causing the noise in your machine. I vaguely recall something about a batch of power supplies in earlier G5s that (I think) Apple will replace under warranty/AppleCare because of excess noise/hum. Maybe you could look into that, as well.
And just to put all my comments in more context, I also have a G5 (dual 2GHz), an Mbox and have used ProTools for voice overs. My G5 is dead silent, and always has been. When I'm sitting at my desk, I literally hear more fan noise from other people's machines down the hallway than I have ever heard from something inside my office, including my G5.
I can't recall, since getting my G5 last October, ever hearing the fans kick into a high speed (and this includes the times I've had FCP4, ProTools, Cinema4D, DVD Studio Pro, Photoshop and LiveType all running at once). And on the recording side, for reference, I've been using a Shure KSM27 mic.
With one weird, but possibly-not-related, exception.
I did the 10.2.8 update on my G4 PowerMac. After the initial reboot and before I had a chance to actually do anything, my keyboard and mouse stopped responding. But a power cycle later and things seem to be working fine. I've since run a couple Classic apps (mainly the scanning software for a Canon), plus OS X Photoshop 7 and connected to a couple Samba shares on a Linux server. The only network connection is wired Ethernet.
I should have thought to try and ssh in before cycling the machine, but I didn't. So, for all I know it could have either been just the kb/mouse, or it could have been the entire machine, that froze up.
What about game saves? Sure, it could possibly include drivers for the current filesystems (of course, NTFS r/w was/is still pretty sketchy under Linux last time I checked), and the boot process could seek out the current drives. Of course, they'd also need to be able to support all the ways people connect drives to their systems. If I'm saving my progress in a PC game right now, it doesn't matter whether I'm saving it to an IDE drive, SCSI, USB, Firewire, network share (Samba, NFS, Netware, etc.), or even battery-backed RAM disk.
But, even assuming they could manage to handle all the currently supported filesystem types and all the ways of connecting them that already exist, what happens when new FS types come along? "Sorry, sir, but your machine is too new for our game to be able to provide you the ability to save your progress." I don't think that will cut it.
I was going to mod this down (overrated), but decided I'd rather reply.
No one said that honeytokens are superior in every way to honeypots and should be used in place of the latter. That you pulled out of your hindquarters. Basically, what you said could be expressed similarly in this example: "Seat belts are not absolutely superior in every way to the steel frame of a car, so what's the point in buckling up?"
I would hope that makes it clear how faulty your logic is. Like using seat belts in addition to a protective steel frame, to provide added protection, honeytokens could be used in addition to honeypots. Their ultimate goals are the same: protect your life (frame/seat belts) or your data (honey[pot|token]). If your life/data is that important, why not provide all the layers of security you can?
One advantage that honeytokens do have is in who they can help protect against. Honeypots are typically deployed to detect and help figure out how to protect against external threats. Anyone with a shred of sense about security knows, however, that you also need to protect against internal threats. Deploying honeytokens can help in that vein, by posssibly detecting internal abuse of your systems.
Just because honeytokens won't protect against everything, solve global hunger, and bring about world peace, doesn't mean they shouldn't or can't be used effectively.
I've been playing VJ for a couple days now (got mine Tuesday -- imported it through videogamedepot.com) and love it. I've always been a platformer at heart, and this game definitely makes me happy so far. For any US cuber's who want to get this *now* but don't speak Japanese and are worried about not knowing what's going on, stop worrying and just get it. Much of it is in English anyway.
My only beef with it was when I first popped it in and just wanted to get straight to the action. You have to sit through a drawn out intro that can't be skipped (it seemed to go on forever the first time around). If you do import it (and if this is your first import), make sure you either have a modded GCN or the "Freeloader" so that you can actually play it, and also make sure to have a separate memory card (you cannot save US and JP games on the same card).
And some people seem to forget that just doing something "over the Internet" does not exclude you from the laws of your country/state/province/city/etc. Slashdot is owned by a company registered and operating in the U.S. There are a crapload of laws regulating contests, sweepstakes, and other things of that nature in the U.S. (which vary by state, many of them placing all sorts of restrictions when there are prizes, monetary or otherwise, involved). Slashdot, as with all companies in the U.S., is required to abide by those laws (or restrict entry from residents of states where they cannot or will not abide by the laws).
And before you decide to then stop blaming Slashdot and blame "the U.S.", pick any other industrialized country in the world and know that it will be the same way there. If Slashdot was a Belgian company, they'd almost certainly be restricting the contest to Belgium (or at most EU countries), which would still leave out a significant majority of the world.
Capture the Bear tshirt (for Megatokyo).
I have a couple of those and the quality is very solid (held up through many washes, as well). Those kinds of designs probably aren't done more often because of the higher costs of production compared to a shirt with two simple, solid colors in a basic design. Not to mention you have to spend a lot more time actually creating the design, checking samples for registration misalignments, etc.
I have two of these, bought through ThinkGeek, and they're great. Both of mine are still on the first set of batteries.
Nice (Flash) cartoon site. Very appropriate for kids, especially the very young, impressionable ones.
Err... I did fail to mention that the Japanese CDs are more expensive than the Korean/Chinese ones. K/HK music generally goes for $7-$9/CD, but the Japanese CDs go for the usual "import prices" ($22-$30). Forgot about that when I first posted, since I listen to a lot more K/HK music.
I also buy most of my movies from them. A lot of the movies can be bought three at a time for $10/each (if you don't buy them in 3s to get the discount, they're usually $12-$15). VCDs are even cheaper if you want to go that route. You can get region-free DVD players from them as well (the Shinsonic is not a bad set, although there is a weird "vertical line ghosting" issue on large blocks of static, dark, solid colors).
I very highly recommend them. One note of warning is that they take a few days to process and ship orders, since they often have to ship some stuff from their HK (I assume) warehouse to California before it will ship out to a US customer. They do seem to have been trimming this turnaround down a little (average delivery time, from time of order, using overnight is now one week), so either they're getting a little more efficient or they know me well enough that I'll keep on coming back spending hundreds each month, as long as they keep me happy. ;-)
Anyway, if you're into Korean, Japanese or Chinese music/movies/etc, definitely give them a look. And, no, I do not work for them or have any relationship with them other than being a very satisfied customer.
In his article he actually does address this very question. He even gives, what I feel at least, is an interesting answer.
So, you don't run an open relay. You're not going to slow down the spammer directly, but you will slow down all the connections that come from that open relay to your mail server. For a particularly abused open relay, that could lead to such problems that the admin of that open relay will finally get a clue and look in to configuring their server properly.
Hence, a cascading effect that will eventually harm the spammers. Admins of open relays that get a clue will tighten their servers, thus depriving the spammers of one more relay they can abuse.
Without much hassle, you can use XS to allow your Perl scripts to interface with C/C++ libraries in a natural way. In fact, many Perl modules are written in C/C++ and wrapped with XS. This allows you to get the best of both worlds in some situations -- streamlined C code to offload intensive operations into separate libraries, then rapidly prototypable Perl front ends to those libraries (whether it be mod_perl for web apps, Perl/GTK for GUI, or any other type of front end you need).
It's not always the perfect solution, but your post makes it seems as if you don't think there's any way to get C and Perl to talk. Few things are further from the truth. And this isn't just some nifty Perl thing -- many of the "scripting" languages (I can't speak for PHP) offer similar ways to accessing C/C++/etc. I'm just most familiar with Perl's way.
Very poor analogy, actually. Microsoft is in a very special class of companies: monopolies. Those companies have to play by a whole new set of laws that don't apply to other companies. Especially when said monopoly infringes those laws with regards to competing companies. Not to mention Microsoft is a publicly-traded company, whereas Google is privately held, though that fact has less bearing on pointing out the fallacy of your comment than the monopoly issue.
Actually, "in" would be inappropriate for the use you describe. Using "in" would imply that "Linux" or "Unix" are programming languages. It would be better to use "under" or possibly "on". "Programming under Linux" or "Programming on Unix in C".
I agree with your basic point, however, that simply saying "Programming Linux" implies development work on the core components of a Linux system. If you really wanted to be pedantic about it, it would imply simply the Linux kernel.
I've been wrestling with this problem for some time now. I'm looking to shoot myself in the foot using some sort of firearm. I will probably use a pistol or revolver, as I believe that will be easier to handle as I assume I will be pointing directly downward (unless I decide to shoot myself in the foot while in bed). Unfortunately, the pain that I hear may result could prevent me from possibly being able to fire more than a couple rounds. My question is this: Is there a real benefit from using four smaller bullets as opposed to using two larger bullets when I shoot myself in the foot? I've found some people seem to think that a 9mm round will be more effective for this use... But, purchasing two .38 rounds appears to be cheaper than four 9mm rounds. Will the effectiveness of shooting myself in the foot suffer if the two .38 rounds are used, even if they are larger? I've searched high and low using google and have yet to find someone else who has collected any data on this matter. Take these firearm/ammunition combinations as an example: A revolver with two .38 rounds versus a pistol with four 9mm rounds. The revolver would be more expensive, because I am limiting myself to a model which is not used very frequently and is slightly more difficult to find readily available. I would also have to purchase a full box of ammunition even though I am only considering using two rounds (I don't know how to ask the clerks for only two rounds -- but that's another Ask Slashdot), thus inflating the price/use ratio. And I know a .38 round wouldn't have the spread of a 5 gauge buckshot round (if it existed) for a shotgun, although this comment has almost no relevance to my question (and makes a comparison to a product that doesn't even exist, nor will ever), but yet I still want to think it would be more effective. The firearms I am looking at are both manufactured by Colt and have the same basic features. Maybe some Slashdot readers, who have shot their own feet, could shed some light on this topic?
In raw disk storage, maybe. But you're forgetting actually putting those drives into a useable state with disaster recovery plans.
In other words, someone dealing with 50TB and who wants backups of that data will be spending many, many times the amount it would cost to just purchase enough hard drives to get the bragging rights of 50TB. And a backup located in the same room/floor/rackspace/whatever as the source data will be pointless in the event of fire, floods, nuclear fallout, etc. So, they would also need a way to transfer all that data to offsite backups in a timely manner (waiting five weeks for a full backup to transfer over a 100Mb/s pipe would probably not be acceptable).
Aside from backups, how would the drives be accessible? Even as JBOD, you're talking 40 IDE/ATA controllers (assuming 320GB drives and 4 ports per controller), or 20 SCSI channels (assuming 160GB per drive and 15 non-host devices per channel) to support that many disks. You could also use Fibre Channel and get away with only a couple arbitrated loops. Physically, you're talking about hundreds of disks that need to be mounted somewhere, so you would also need dozens of chassis to hold the drives.
But, hundreds of disks in a JBOD configuration means you'll have hundreds of partitions, each separate from the others. Hell, if the clients are Windows machines, they won't even be able to access more than a couple dozen at a time. And even for operating systems with better partition/mount-point addressing, it would be unmanageable.
So, now you get in to needing a RAID solution that can tie hundreds of disks together. If you're talking about hooking these up to standard servers through PCI RAID cards, you'll need several of those machines to be able to host all the controllers necessary (especially if all the disks are not 160GB or larger each).
The only realistic solution for this much storage, at least until we have 5TB hard drives, is a SAN-like setup. Specialized hardware designed to house hundreds of disks in stand-alone cabinets and provide advanced RAID and partitioning features. SANs don't come cheap.
Add to the SAN the various service plans, installation, freight, configuration, management and the occasional drive swapping as individual disks fail and you've already multiplied that $50K several times, as a bare minimum (and you still haven't priced out the backup solution).
There's a lot more to it than just having a pile of hard drives on the floor. I wouldn't even be surprised if the drives are the cheapest component.
I generally have two schedules throughout the year; a "big project" schedule and a "maintenance" schedule. The first involves 14 to 18 hours a day, six days a week (regardless of holidays), and tends to last a month or two at a time. The other one involves generally between 9 to 12 hours a day, five days a week (and actually staying home on holidays).
I typically do not have the big project schedule more than twice a year. I work for a small company (well, technically, a small division of a medium sized company) and am solely responsible for the development of our software. There's no COTS or open source software out there that meets our needs for functionality, ease-of-use and performance (at least not any that is less expensive, including whatever it would cost for 24/7 onsite service and support from the developers who wrote the software, than my salary and benefits), so it falls on me to write virtually all the software that runs our division (minus the operating systems, office suites and database servers, of course).
But, there are two main reasons why the difference in schedules and the very heavy loads that happen a couple times a year don't bother me.
The first reason is that project planning in our division is managed well, for the most part. I can think of a couple times where things did not go as planned, but they're in the minority. We also generally plan well in advance. I'm currently in the middle of a big project schedule right now, but I have known this was coming since the beginning of the year. Also, because we are a small group (less than a dozen people total for our division), I have a strong say in the project plans. If I mark something as unreasonable, we generally either don't do it, or we come up with an alternate version that is reasonable.
The second reason is that I *really* enjoy my work. If I'm not at work programming, chances are that I'm at home programming (usually for work) or sleeping. I do very little other than code. I enjoy having to solve difficult problems. You could certainly take the perspective that I simply have no personal life, and I would not argue one bit with that. I don't have any illusions of having a personal life -- but that doesn't bother me at all. I am a very dull person outside of work, anyway, because I tend to only think about work.
Supporting reasons for why I don't mind include the fact that I feel I am compensated well for my work and that I am valued by the other people in the company. I am also constantly learning -- especially when I am confronted with the more complex problems. Solving the really difficult problems may be painful while I'm doing it, but at the same time I consider it to be extremely rewarding because it gives me an excellent opportunity to expand my knowledge.
Like the saying goes, your finest teachers are your worst adversaries (in my case, the most complex problems). If everything is "point and click" easy, you are never challenged to think and confront problems head on and find their solutions. You have significantly reduced chances to grow or have the opportunity to expand yourself if you aren't challenged in some way.
Possibly just as important, psychologically, for my acceptance of, and willingness to commit to long work hours and tight deadlines is that I have a significant emotional/personal investment in the company, having been one of the people that started it. I still hold on to a lot of the "founder" and "startup" work ethic mentality from our company's early days, even though we have grown beyond those stages.
Will I burn out one day because of the long hours I work? Probably, but I hope that day is in the very distant future. And just to tie this post in with the original question, I do not believe that the long hours I put in negatively impact my productivity. I actually put in the long hours mainly because I work better that way.
However, were I not working for a company I cared deeply about, or were I not being challenged intellectually by my work, I would most likely burn out quickly and my productivity would suffer. I consider myself lucky to be in a position where I actually have some very significant and positive motivating factors.
Domain name parking.
It happens every once in a while. Places like Register.com, Verisign, et. al. that put up those "site coming soon," "buy this domain for $x," and other messages switch their web server software every so often. Don't ask me why they do it exactly, but when it happens it can create a very sudden change in the numbers.
This report, I believe, had a note that one of those domain name parkers switched to Apache from IIS, which probably accounted for at least a noticable chunk of the 6% swing.
Thanks for the info on that. I had not been aware that the Orbs were so poorly thought of. And, considering I'm not a physicist (just a mere programmer), I would never have thought of heat transfer issues with Al-Cu.
Tell that to an Athlon 1.2GHz I had for a while. No overclocking, whatsoever. I had an Orb on the Athlon, rounded cables for better airflow, and 7 case fans (all 80mm or larger -- three in front, three in back, one on the side), plus the two fans on the power supply, plus an Antec "PCI Bay" fan.
I *never* saw the Athlon in this box under 65C, and under stress (games mainly) it would hit 70+C. After months of trying everything I could, short of water cooling, this _non_overclocked chip finally killed itself because of the heat.
I replaced it with a P4 2.53GHz, in a different case with only three fans, plus the stock Intel heatsink. I've never seen the P4 above 44C sans overclocking, even under stress. Even when I overclocked the chip, the core temp never went over 55C.
Both of these machines were in the same location of my apartment, which is kept at an ambient temp of 60-65F.
You can also stream the news (audio and video) from the BBC website.
Too much trouble to read their press release? It's not that long. If you had read it, your first two questions would have been immediately addressed by the following:
"Clients which connect to our peer-to-peer clients, and then afterwards attempt to illegally access the network will be immediately blacklisted from Information Wave's network."
It does not directly address your third question about spoofed IP addresses, but they could address that at the routers.
PCI cards still make sense for some of us, even if we've never stepped foot inside a Walmart (or outside the office, for that matter). If you can show me a motherboard that has four AGP slots for all my video cards (I run a multihead setup), then I'll no longer need PCI video cards. Alternately, show me an AGP card with at least four high resolution outputs that works as solidly under Linux as four individual nVidia based cards.
Agreed. However, regarding the comment about the costs of moving from 12 to 13 being the same as moving from 12 to 18, I'd like to add something. I agree that the costs will probably be exactly the same (or such a miniscule difference you'd be a fool to make the same mistake by only going to 13).
You might be surprised by how many people would not make that connection. I have personally dealt with another company that, when faced with a limitation in their software that was truncating object identifiers from our software (the two communicate various information back and forth on a regular basis through XML streams), they "solved" the problem by adding one extra byte to the variable causing the problem. It was currently 8 bytes and they made it 9.
What made it worse is that these identifiers increment rapidly, and the "solution" they came up with would only hold out for a few months and would have to be changed again. I tried to explain it to them, and let them know that our software expects these to be able to expand up to 48 bytes -- which would hold us for decades at minimum.
So, they changed it to be 10 bytes on their end. They obviously didn't get it. I can't really say that I think they would be the exception to the rule, either. I wouldn't be surprised if a lot of other less-than-forward-thinking people would make the same mistake and only increase the upper limit to the bare minimum needed to fix the immediate problem -- without ensuring further expansion in the future.