Re:What Edison would say if alive today...
on
Engineer in a Box?
·
· Score: 1
Actually,
I think Edison would disagree with the above question. There is currently no substitute for the artistic side of the human brain. Pure genius is usually knowledge applied in a new and creative way. Without the mathmatical reference point for the brain to view through, engineering wonders may not even be possible.
One should note that while there are many stories of child prodigies in art and music such as Mozart, nowhere in history has a small child made leap-through discoveries in math and science. This is not because they don't have the mental capacity, but because they don't have the math/science reference level to work with.
Which raises an interesting question.. when will countries start claiming territory on the moon?
The U.N. has specifically declared space to be "the province of all Mankind". Since all of the space capable nations are members of the U.N., my answer would be not anytime soon.
Sorry to pop your bubble, but E-1 to E-4 pay ranges from $1022.70 to $1752.30 per month depending on time in service (dfas.mil). This is on top of the fact that all your food, housing, electricity, water, trash, medical, dental, and training are provided to you free of charge. In other words, the only financial obligations you have are the ones you create yourself. Even if you take a loan out on a car($300/mo. + $100/mo. for expensive insurance + $50/mo. for gas/upkeep), get a cell phone ($50/mo with a ton of minutes), and get cable plus wide-band access (up to $100/mo) that still leaves over $400 pure spending cash for even a boot E-1. Average promotion times are 6 mo. to E-2, 8 more mo. to E-3 and an average of 18 months for E-4 depending on how good of a worker you are. Most folks straight out of high school with little to no work experience usually don't have it nearly that good. Also, starting Oct 1st, the Military pays 100% tuition assistance for undergrad education, so free college to boot!
Now, that being said, the military is still a tad behind the civilian world in overall fiscal compensation. But it's a bit too far to say that enlisted make "almost nothing". As far as the IT field goes, most military people suck up the free training and bail after the 4 year stint for better paying jobs. This rapid attrition rate allows those that stay in to reach the ranks of E-5 and E-6 faster. Benefits begin to increase in the form of priviledges which makes staying in just about as worth while as getting out.
FYI, the daily unclassified, non-critical networks that the E-1 through E-4's usually administer have terrible up-time rates and is usually directly attributed to the lack of experience and education. Most of these self-proclaimed IT wizards couldn't manage a Nintendo without their roomate's assistance.
Before the blasting starts, I enlisted 9 years ago and have worked my way to the officer ranks. I think I made plenty of money then and make plenty of money now with the increase in pay directly related to the increase in my education level and responsibilities.
<Interresting(Insightful)?...The moderators don't know what they're moderating, you think?> Rough translation as both my Kanji dictionary and excite.co.jp had trouble with JI and AN being used together.
Never seen someone bitch about moderation in Japanese. Wonder why they didn't use English if they understood enough to disagree with the moderators?
The article is quite definitely outdated. The system they are referring to is known as an 'Expert System'. These systems are developed to learn facts about a given subject based on a set of predefined rules. The rules also react to facts and can initiate actions when a certain fact becomes true or known. More advenced systems are even capable of creating new rules, or modifying old ones based on the facts in their knowledge base. The medical community is probably in the lead in this field as they struggle to provide a reliable system that will accurately diagnose a patient, freeing up the doctor's valuable time for the actual treatment. One of the key requirements for an expert system is that it should be able to explain in detail how it reached a certain conclusion or action. IBM is simply trying to build something that will become an expert on troubleshooting. It should be noted that NASA has been working on this for years in order to provide more reliable satelites that are capable of conducting simple repairs and reconfiguration to react to the many mishaps that occur 50+ miles above the techs.
Unfortunately, as long as their are humans involved, corruption will always be there. From the guys paid to write the software, to the DB admins, to our friends at M$ who will undoubtably provide a security-lacking OS to run the system on, voting will always be called into question when it gets as close as it did between Gore and Bush.
Intelectual property is a modern concept, invented by greedy bastards that ignore that ideas belongs to everyone.
As much as I disagree with the way the MPAA and the RIAA have handled this brave new world of technology, they are fundamentally right. You have apparently never had something that you put serious effort into creating taken without your consent. Intellectual property is a way of quantifying that a song, or video, or picture, or peice of software exists through the expenditure of someones resources (time, money, etc..) and that expenditure should be justly compensated.
Who are you to claim that a song should belong to you by right. What the hell did you do to help Eminem write the lyrics? How much money did you give Metallica to help them by their instruments? How much recording equipment did you donate to Sony? How many hours in the factory did you volunteer to help press the albums? Who did you hire to teach Dave Matthews to play the Guitar?
Again, I'm not saying that the current actions of the RIAA are morally sound. But having had a peice of software stolen from me in College and billed as somebody elses work woke me up to the fact that people pour a bit of their soul into the things they create and they at the very least deserve to be recognized and if they so desire, compensated for their effort. Recognize that the RIAA and MPAA are products of the industry itself. They are trying to work in an arena they had no hand in building and really don't have the option to just sit back and say, "fuck it, nobody get's paid anymore." Artists and studios make a ton of money because we the people are willing to pay it for the enjoyment of their product. No matter how "Robin Hood" altruistic the intent may have been, stealing the music as a response to the high prices is not the answer. Unless someone can come up with a way to restructure these industries to where the artists are compensated for their work and the consumers are granted that work at reasonable prices to make it accesible to everyone, the RIAA will continue to raid houses of 'innocent' teenagers who were just downloading their favorite NSYNC song.
FYI, I used Napster. The difference is, when I found something I liked, I went out and bought the album. Many thanks to the hard working people who spend years perfecting a talent so that I can sit lazily on my couch and enjoy their work.
In just the US, yes....big city is a geographic minority. I presently work in Japan where big city is the majority. However, my point was simply that most people spend most of their time in a position where it is either impossible, or dangerous to use the extra power provided by "sports/muscle" cars. Whether this be the time spent sleeping, working, eating, socializing, or driving down roads where the speed limit is 45 and you know there are cops watching; spending high $$$ on a vehicle with some kick might not be worth it. i.e. if you pay an extra $5000 for the power and ten times per year over 5 years you get to really open it up and go, you are paying $100 per few seconds/minutes of rush plus any associated costs(lower fuel efficiency, speeding ticket, higher insurance, medical and deductible costs if you wreck while going fast). I do appreciate the rush of going fast and occasionally do it out here on the express way, but I don't see it as being so valuable as to be worth all the associated costs both fiscally and environmentally.
I still want a car with a lot of horsepower and low end torque. When I can get one like that, I might be interested
Why is it that people desire extra super-charged power stuff that they can't use due to regulations or other inhibitors? You probably also are chomping at the bit for a cool new 3GHz P4 which runs just about as slow as the 800MHz PIII due to the frontside bus bottleneck.
Does anybody around here think about using the money they waste on overpowered, inefficient toys for more practical uses like a bigger pr0n collection???
Maybe it's because I'm used to big city traffic; but be it EV, Hybrid, or a turbo-charged V12, they all perform exactly the same in a traffic jam which is unfortunately what most people deal with at least once a day. It's a shame the EV concept didn't take hold. Smog is pretty damned depressing.
That's odd...I distinctly remember always being the SERVER for Q2 and Q3 at school since I had the highest end machine. Everyone else connected to my game via the seach if they were on my segment, or via an IP I provided if they weren't.
As long as there are configurable parameters in the game, I don't think they will ever seperate from a server based concept. One machine has got to be the athority on which map is being used and how much health the players have, etc... Now, perhaps all machines will advertize the games they are connected to so there doesn't have to be a need for GameSpy. It will be interresting to see how they implement this.
Cost is directly related to possiblity. For example, the Maximum hardware performance available to me in a computer would be an Athlon 2600+ system. Sure, ASCII White is available, but not to me due to cost. Moore's law was more geared towards the 'affordability' of the power. Cheaper technology leads to higher demand...higher demand makes higher production....higher production leads to competition and increased affordability and AVAILABILITY of the tech.
If I were just to judge off their LINUX Mags over here I would say turbolinux just got hooked up. An average issue weighs in at nearly 180 pages. That's 50% larger than this month's Journal not to mention the extra density of text in the Japanese Language....
I'm not affiliated with Internet2, but reading the docs indicates that it is routing both IPv4 and IPv6. The former done over BGP and the latter over IS-IS. It goes into how the BGP is encrypted via MD5 and that the IS-IS only shares info via a plain text authentication. I imagine it will eventually migrate to just IPv6, but prolly is waiting on support from M$....
While we're on the subject of cartels...the American flower industry is robbing everyone blind on any day of the year. I'm stationed over in Japan and was stunned to find I could get a dozen roses, wrapped with baby's breath and ribbons and cute paper designs and the rest of the nine yards for a paltry 1000 yen, which is about $9 at todays rate. Considering the effort the florist put into the boquette(sp?), I feel I got the roses for free and damn near gave into the urge to tip the guy. (tipping is considered rude here)...
Good grief! Did you go to Bovine University? Hash tables easily fit within the ten most commonly used data structures. They take all of 30 seconds to explain, and in my experience are usually introduced in the same breath as lists and trees.
I went to the United States Naval Academy where INTRO to CS was a course designed to introduce people to computer science. It covered basic programming practices such as pseudocode, flowcharting, top-down design, and intoduced the basics of the C language by covering data types, input/output, flow control, basic structures and introduced pointers. I have also spoken to a few folks who work for me and their intro courses were similar at Florida State, University of Maryland and University of Delaware. The typical next course was data structures and covered all forms of lists and trees as well as the classical structures such as stacks and queues. Perhaps you had a background in CS as I did and were not required to take the INTRO course. I also tested out of the structures course as you may have as well. My point was that a BEGINNING CS student who just learned how to make a for loop and is still getting dinged for his flow charts will have significant difficulty understanding a hash table at the code level. Could you explain the concept of hashing in 30 seconds?... of course. Could the INTRO students go home that night and write a scalable hash table for homework, or even for a final exam?...possible, but not very likely...especially not the students who decide in the INTRO course that they weren't meant to program computers.
Also, you've never heard of sequential hashing (no linked lists)? Or perfect hashing (hideously, monstrously expensive to add keys, but one-step look-up every time)?
I have heard of both and your knowlege of machine language is lacking. The offset in the lookup table must be calculated everytime unless you plan on the table occupying a sick amount of memory and you don't allow duplicate keys. In that case, you can simply use the bytes of the key as the relative offset and and use the indirect addressing mode available on every processor I've ever used to read the data. However, a filesystem which was the point of discussion, has to be allowed duplicate keys due to the pysical memory limitations...i.e., the byte value of the filename cannot be used as the offset because most filenames exceed the length in bytes that can be physically addressed in memory. This means that steps must be taken to calculate the offset and that there must also be a list of some size existing at that offset due to similar filenames..i.e. setihome.ini and setihome.scr. You can rehash all you want, but at somepoint you will run into a physical limitation on your offset table and will then be forced to allow the lists to grow. Also, depending on how your offset is calculated you also may not be able to avoid lists of varying size i.e. if I create 200 files that all start with myapplicationdataset and then have a 3 digit number attached, most hash designs will result in a single list of 200 file locations and a call to open myapplicationdataset199.dat will take longer than a call to myapplicationdataset000.
Big O is completely appropriate any time you are discussing time delays. No one cares about best case or average case, they want to know they will have adequate performance under worst case.
I would be curious to know your educational background in relation to why you found it suprising that I don't think beginning students should be covering hash tables. As the lead software engineer for my office, I often find myself filling in gaps in my programmers education and if there is a better way to build a computer scientist, please let me know. (I couldn't think of a way to state that that didn't sound somewhat sarcastic, but no offense is intended...I really do respect your views on this and have been around long enough to know I will always have more to learn)...
I normally don't reply to inflamatory posts, but I also don't want a beginning CS student to read this and be taken astray.
The concept behind a hash table is to break up large lists into a group of smaller lists based on some data eleement in the list. With a file system, that data element could likely be the first character of a file. If we decide that our file names can only begin with a letter or a number, then that would result in 36 small lists as opposed to one giant list.
A well hashed list will result in equal sized smaller lists thus resulting in the near "constant average time" for a search as mentioned by the Anonymous Coward. Unfortunately, at any point you add another item to the list, everything must be rehashed to keep it that efficient. Add enough items to the list and then the average size of the hashed lists will begin to grow. He was mostly correct with his guess of O(log N) except he was missing one peice of information. The correct notation would be O(log_a_ N - 'log base a of N') where a is the number of seperate hash lists. The bottom line is that the poster proved his first statement wrong with this very post. The function is still with respect to N and always will be until someone finds a way to make a processor always know where everything is in its storage, or at least be able to find anything in a fixed amount of time.
Note to the readers: Advanced search algorithms such as hashing are rarely covered in an intro course. (I've never heard of it) And if they were, I would have to question the quality of the educational institution. Most beginning CS students are still strugling with the concept of pointers at the end of an intro course. I really can't see them effectively absorbing something that requires a good grasp on dynamically allocated structures in some form of a linked list....
I think it's a bit premature for us linux advocates to say it is immune to this. While a package manager certainly does help reduce dependancy issues, I don't think it addresses the physical issues.
First and foremost to consider is that there is no such think as a O(1) search algorithm("read 'Big Oh of one' for the non computer scientists in the crowd: notation used to measure the efficiency of an algorithm). The ammount of time required to search a list will always be a function of its size.
As the user base of Linux grows, so will the demand and supply of software. slowly, but surely,/usr/* will start to get quite fat with binaries and their required libraries. Everytime one of those apps are called, it will take additional time to find the binary itself, then tack on the additional latency produced by the libraries having to be located and loaded into memory and the rusult will unavoidably be the visual degrade in performance.
Plus as more and more non computer literate people start using Linux, we will have to ensure that the software to support installing and upgrading packages on the system is user-proof, or other problems will result.
Unless someone can win a Nobell prize or Fields medal for finding a O(1) search, I'm afraid the above article is correct....
I must concur with the other posts on this subject. Any reference to the way wireless is handled in America is like trying to teach modern history with a 20 year old book. Before the flamers start up, I'm from America, but have spent a majority of my Military career living abroad. I am now happy to see American cellular technology finally catching up to the rest of the world. (although that Verizon commercial about the text messaging on the calculator style black and white diplay screen is still quite depressing)
In Europe and Asia, land lines are damned expensive and local calls are billed on top of your monthly rate to just have the line. Here in Japan it runs about 3 cents a minute (3 yen) to call local land lines and 7 cents (9 yen) to call cell phones. If you could pay a flat rate of ~$35 (5000 yen) for unlimitted wireless to your computer and your cell phone (256,000 color display, email, pictures, full color video games, and a variety of PDA like functions) runs you about the same rate, why would you ever get a land line? Even is you only spent 120 hours online, which I do on a regular basis, that's 21,600 yen (~$168 at today's rate) plus the ISP cost plus the basic land line charge.
Needless to say, I have yet to meet anyone here in Okinawa who uses dial up and very few who use DSL. Most rely entirely on the cell phone for calls and either cable or wireless for the net.
Even for those of us on US bases who do have free local calling, there is no broadband solution offered on base. So wireless is our only solution.
Pick up the books out of the trash, and look at their binding. They're published by either TSR or Wizards of the Coast--not a publishing house of any credulity, but rather a niche publisher that does an amazing job catering to the low-brow fantasy market.
Or rather, a credible publishing house that owns the rights to the Dungeons and Dragons universe. Hence why no one else publishes the stories. They aren't allowed to. TSR was purchased by Wizards of the Coast and so they are inherrently the same entity when it comes to publishing. I'm not sure what you deem as credible, or what you deem as quality, but the simple fact is that D&D books, including Salvatore's, have been around for years. And the popular ones, some of Salvatore's included, continue to be reprinted because people continue to read, and re-read them.
As far as the movie quality sucking, Two things:
#1 We've built our hopes up for years over these films and quite possibly built them too high.
#2 The whole reason why the films are being made is because someone with a rudimentary grasp of economics realized the baby-boomer generation would line up like little kids to pay for the hope of more childhood magic. (which they were when the original movies came out, and as we have seen so far for both movies, they have) The sad truth is that even if the second movie sucks as bad as everyone is saying, the same lines will be formed for Episode 3. Sure, they might be a little shorter, but lines they will be with cash in hand.
Just a note on the verbose fight scenes in the book. The reviewer is clearly unfamiliar with Salvatore's writing style. In depth fight scenes are what made Drizzt a legend in the Forgotten Realms world and it's not suprise to see Salvatore sticking with what worked in the past.
There are dozens of amateur OS's out there that have functioning GUI's. With a little bit of research, MS could come back into court with proof that they're not the only ones who integrated the browser and that it must be a good idea.
Note: I don't know of any of those OS's that did integrate the browser, but I have not downloaded and played with every single one either.
Several years ago, NTT Docomo released a phone in Japan that could store and play mp3's. The problem with the concept was a combination of price and target audience. The average person who desired to have their phone double as a walkman was college age and younger. The phone was priced at a modest $600 which most college students couldn't afford.
The idea of the camera in the phone is quite useful and has been available here in Japan for quite some time and is very reasonably priced. Almost all the phones over here have a great deal of PDA functionality and connect to the internet which is very useful.
But we return again to the target audience. The novelty of the phone being able to play my favorite mp3's and mpg's is going to wear off about two seconds after I read the price tag. The reason being is that like most salary workers, my day is fairly evenly divided up between work, commuting, eating, sleeping, and trying desparately to have something called a social life. Aside from amazing my friends during the "social life" part of my day, I have no time to play with gadgets, nor do I have any professional use for playing mp3's or mpg's. I didn't even get the camera phone over here because the cost didn't justify the very few times I would actually use that feature. IMO the P800 looks really cool for a technology demo, but I don't think they will be able to produce it at a sufficiently low cost for it to succeed. It's simple economics....when you have to pay rent/bills, support a girlfriend/wife/kids, and still manage to feed and entertain yourself, you find yourself seriously asking, "How often would I REALLY use this?"
This is remeniscent of the P4 vs Athlon video that showed the P4 surviving while the athlon cooked. Camera angle wasn't wide enough to see if a high powered fan was blowing on the P4 case and you have no idea if the room temps were the same for both, so you have to put your faith in the person who created it.
For the record, my girlfriends computer is completely toasted right now. But if I had a digital camera, I could give you some great shots of it POSTing. It's really easy considering that's where it hangs.
Slashdot Mirror
Actually,
I think Edison would disagree with the above question. There is currently no substitute for the artistic side of the human brain. Pure genius is usually knowledge applied in a new and creative way. Without the mathmatical reference point for the brain to view through, engineering wonders may not even be possible.
One should note that while there are many stories of child prodigies in art and music such as Mozart, nowhere in history has a small child made leap-through discoveries in math and science. This is not because they don't have the mental capacity, but because they don't have the math/science reference level to work with.
Which raises an interesting question.. when will countries start claiming territory on the moon?
The U.N. has specifically declared space to be "the province of all Mankind". Since all of the space capable nations are members of the U.N., my answer would be not anytime soon.
Sorry to pop your bubble, but E-1 to E-4 pay ranges from $1022.70 to $1752.30 per month depending on time in service (dfas.mil). This is on top of the fact that all your food, housing, electricity, water, trash, medical, dental, and training are provided to you free of charge. In other words, the only financial obligations you have are the ones you create yourself. Even if you take a loan out on a car($300/mo. + $100/mo. for expensive insurance + $50/mo. for gas/upkeep), get a cell phone ($50/mo with a ton of minutes), and get cable plus wide-band access (up to $100/mo) that still leaves over $400 pure spending cash for even a boot E-1. Average promotion times are 6 mo. to E-2, 8 more mo. to E-3 and an average of 18 months for E-4 depending on how good of a worker you are. Most folks straight out of high school with little to no work experience usually don't have it nearly that good. Also, starting Oct 1st, the Military pays 100% tuition assistance for undergrad education, so free college to boot!
Now, that being said, the military is still a tad behind the civilian world in overall fiscal compensation. But it's a bit too far to say that enlisted make "almost nothing". As far as the IT field goes, most military people suck up the free training and bail after the 4 year stint for better paying jobs. This rapid attrition rate allows those that stay in to reach the ranks of E-5 and E-6 faster. Benefits begin to increase in the form of priviledges which makes staying in just about as worth while as getting out.
FYI, the daily unclassified, non-critical networks that the E-1 through E-4's usually administer have terrible up-time rates and is usually directly attributed to the lack of experience and education. Most of these self-proclaimed IT wizards couldn't manage a Nintendo without their roomate's assistance.
Before the blasting starts, I enlisted 9 years ago and have worked my way to the officer ranks. I think I made plenty of money then and make plenty of money now with the increase in pay directly related to the increase in my education level and responsibilities.
every 4 years over 250 Million taxpayers get together to beg the government to work as a team and look how far that's gotten!
<Interresting(Insightful)?...The moderators don't know what they're moderating, you think?>
Rough translation as both my Kanji dictionary and excite.co.jp had trouble with JI and AN being used together.
Never seen someone bitch about moderation in Japanese. Wonder why they didn't use English if they understood enough to disagree with the moderators?
The article is quite definitely outdated. The system they are referring to is known as an 'Expert System'. These systems are developed to learn facts about a given subject based on a set of predefined rules. The rules also react to facts and can initiate actions when a certain fact becomes true or known. More advenced systems are even capable of creating new rules, or modifying old ones based on the facts in their knowledge base. The medical community is probably in the lead in this field as they struggle to provide a reliable system that will accurately diagnose a patient, freeing up the doctor's valuable time for the actual treatment. One of the key requirements for an expert system is that it should be able to explain in detail how it reached a certain conclusion or action. IBM is simply trying to build something that will become an expert on troubleshooting. It should be noted that NASA has been working on this for years in order to provide more reliable satelites that are capable of conducting simple repairs and reconfiguration to react to the many mishaps that occur 50+ miles above the techs.
My 2 cents
Unfortunately, as long as their are humans involved, corruption will always be there. From the guys paid to write the software, to the DB admins, to our friends at M$ who will undoubtably provide a security-lacking OS to run the system on, voting will always be called into question when it gets as close as it did between Gore and Bush.
Intelectual property is a modern concept, invented by greedy bastards that ignore that ideas belongs to everyone.
As much as I disagree with the way the MPAA and the RIAA have handled this brave new world of technology, they are fundamentally right. You have apparently never had something that you put serious effort into creating taken without your consent. Intellectual property is a way of quantifying that a song, or video, or picture, or peice of software exists through the expenditure of someones resources (time, money, etc..) and that expenditure should be justly compensated.
Who are you to claim that a song should belong to you by right. What the hell did you do to help Eminem write the lyrics? How much money did you give Metallica to help them by their instruments? How much recording equipment did you donate to Sony? How many hours in the factory did you volunteer to help press the albums? Who did you hire to teach Dave Matthews to play the Guitar?
Again, I'm not saying that the current actions of the RIAA are morally sound. But having had a peice of software stolen from me in College and billed as somebody elses work woke me up to the fact that people pour a bit of their soul into the things they create and they at the very least deserve to be recognized and if they so desire, compensated for their effort. Recognize that the RIAA and MPAA are products of the industry itself. They are trying to work in an arena they had no hand in building and really don't have the option to just sit back and say, "fuck it, nobody get's paid anymore." Artists and studios make a ton of money because we the people are willing to pay it for the enjoyment of their product. No matter how "Robin Hood" altruistic the intent may have been, stealing the music as a response to the high prices is not the answer. Unless someone can come up with a way to restructure these industries to where the artists are compensated for their work and the consumers are granted that work at reasonable prices to make it accesible to everyone, the RIAA will continue to raid houses of 'innocent' teenagers who were just downloading their favorite NSYNC song.
FYI, I used Napster. The difference is, when I found something I liked, I went out and bought the album. Many thanks to the hard working people who spend years perfecting a talent so that I can sit lazily on my couch and enjoy their work.
In just the US, yes....big city is a geographic minority. I presently work in Japan where big city is the majority. However, my point was simply that most people spend most of their time in a position where it is either impossible, or dangerous to use the extra power provided by "sports/muscle" cars. Whether this be the time spent sleeping, working, eating, socializing, or driving down roads where the speed limit is 45 and you know there are cops watching; spending high $$$ on a vehicle with some kick might not be worth it. i.e. if you pay an extra $5000 for the power and ten times per year over 5 years you get to really open it up and go, you are paying $100 per few seconds/minutes of rush plus any associated costs(lower fuel efficiency, speeding ticket, higher insurance, medical and deductible costs if you wreck while going fast). I do appreciate the rush of going fast and occasionally do it out here on the express way, but I don't see it as being so valuable as to be worth all the associated costs both fiscally and environmentally.
I still want a car with a lot of horsepower and low end torque. When I can get one like that, I might be interested
Why is it that people desire extra super-charged power stuff that they can't use due to regulations or other inhibitors? You probably also are chomping at the bit for a cool new 3GHz P4 which runs just about as slow as the 800MHz PIII due to the frontside bus bottleneck.
Does anybody around here think about using the money they waste on overpowered, inefficient toys for more practical uses like a bigger pr0n collection???
Maybe it's because I'm used to big city traffic; but be it EV, Hybrid, or a turbo-charged V12, they all perform exactly the same in a traffic jam which is unfortunately what most people deal with at least once a day. It's a shame the EV concept didn't take hold. Smog is pretty damned depressing.
That's odd...I distinctly remember always being the SERVER for Q2 and Q3 at school since I had the highest end machine. Everyone else connected to my game via the seach if they were on my segment, or via an IP I provided if they weren't.
As long as there are configurable parameters in the game, I don't think they will ever seperate from a server based concept. One machine has got to be the athority on which map is being used and how much health the players have, etc... Now, perhaps all machines will advertize the games they are connected to so there doesn't have to be a need for GameSpy. It will be interresting to see how they implement this.
Cost is directly related to possiblity. For example, the Maximum hardware performance available to me in a computer would be an Athlon 2600+ system. Sure, ASCII White is available, but not to me due to cost. Moore's law was more geared towards the 'affordability' of the power. Cheaper technology leads to higher demand...higher demand makes higher production....higher production leads to competition and increased affordability and AVAILABILITY of the tech.
If I were just to judge off their LINUX Mags over here I would say turbolinux just got hooked up. An average issue weighs in at nearly 180 pages. That's 50% larger than this month's Journal not to mention the extra density of text in the Japanese Language. ...
I'm not affiliated with Internet2, but reading the docs indicates that it is routing both IPv4 and IPv6. The former done over BGP and the latter over IS-IS. It goes into how the BGP is encrypted via MD5 and that the IS-IS only shares info via a plain text authentication. I imagine it will eventually migrate to just IPv6, but prolly is waiting on support from M$. ...
While we're on the subject of cartels...the American flower industry is robbing everyone blind on any day of the year. I'm stationed over in Japan and was stunned to find I could get a dozen roses, wrapped with baby's breath and ribbons and cute paper designs and the rest of the nine yards for a paltry 1000 yen, which is about $9 at todays rate. Considering the effort the florist put into the boquette(sp?), I feel I got the roses for free and damn near gave into the urge to tip the guy. (tipping is considered rude here) ...
Good grief! Did you go to Bovine University? Hash tables easily fit within the ten most commonly used data structures. They take all of 30 seconds to explain, and in my experience are usually introduced in the same breath as lists and trees.
...
I went to the United States Naval Academy where INTRO to CS was a course designed to introduce people to computer science. It covered basic programming practices such as pseudocode, flowcharting, top-down design, and intoduced the basics of the C language by covering data types, input/output, flow control, basic structures and introduced pointers. I have also spoken to a few folks who work for me and their intro courses were similar at Florida State, University of Maryland and University of Delaware. The typical next course was data structures and covered all forms of lists and trees as well as the classical structures such as stacks and queues. Perhaps you had a background in CS as I did and were not required to take the INTRO course. I also tested out of the structures course as you may have as well. My point was that a BEGINNING CS student who just learned how to make a for loop and is still getting dinged for his flow charts will have significant difficulty understanding a hash table at the code level. Could you explain the concept of hashing in 30 seconds?... of course. Could the INTRO students go home that night and write a scalable hash table for homework, or even for a final exam?...possible, but not very likely...especially not the students who decide in the INTRO course that they weren't meant to program computers.
Also, you've never heard of sequential hashing (no linked lists)? Or perfect hashing (hideously, monstrously expensive to add keys, but one-step look-up every time)?
I have heard of both and your knowlege of machine language is lacking. The offset in the lookup table must be calculated everytime unless you plan on the table occupying a sick amount of memory and you don't allow duplicate keys. In that case, you can simply use the bytes of the key as the relative offset and and use the indirect addressing mode available on every processor I've ever used to read the data. However, a filesystem which was the point of discussion, has to be allowed duplicate keys due to the pysical memory limitations...i.e., the byte value of the filename cannot be used as the offset because most filenames exceed the length in bytes that can be physically addressed in memory. This means that steps must be taken to calculate the offset and that there must also be a list of some size existing at that offset due to similar filenames..i.e. setihome.ini and setihome.scr. You can rehash all you want, but at somepoint you will run into a physical limitation on your offset table and will then be forced to allow the lists to grow. Also, depending on how your offset is calculated you also may not be able to avoid lists of varying size i.e. if I create 200 files that all start with myapplicationdataset and then have a 3 digit number attached, most hash designs will result in a single list of 200 file locations and a call to open myapplicationdataset199.dat will take longer than a call to myapplicationdataset000.
Big O is completely appropriate any time you are discussing time delays. No one cares about best case or average case, they want to know they will have adequate performance under worst case.
I would be curious to know your educational background in relation to why you found it suprising that I don't think beginning students should be covering hash tables. As the lead software engineer for my office, I often find myself filling in gaps in my programmers education and if there is a better way to build a computer scientist, please let me know. (I couldn't think of a way to state that that didn't sound somewhat sarcastic, but no offense is intended...I really do respect your views on this and have been around long enough to know I will always have more to learn)
I normally don't reply to inflamatory posts, but I also don't want a beginning CS student to read this and be taken astray.
...
The concept behind a hash table is to break up large lists into a group of smaller lists based on some data eleement in the list. With a file system, that data element could likely be the first character of a file. If we decide that our file names can only begin with a letter or a number, then that would result in 36 small lists as opposed to one giant list.
A well hashed list will result in equal sized smaller lists thus resulting in the near "constant average time" for a search as mentioned by the Anonymous Coward. Unfortunately, at any point you add another item to the list, everything must be rehashed to keep it that efficient. Add enough items to the list and then the average size of the hashed lists will begin to grow. He was mostly correct with his guess of O(log N) except he was missing one peice of information. The correct notation would be O(log_a_ N - 'log base a of N') where a is the number of seperate hash lists. The bottom line is that the poster proved his first statement wrong with this very post. The function is still with respect to N and always will be until someone finds a way to make a processor always know where everything is in its storage, or at least be able to find anything in a fixed amount of time.
Note to the readers: Advanced search algorithms such as hashing are rarely covered in an intro course. (I've never heard of it) And if they were, I would have to question the quality of the educational institution. Most beginning CS students are still strugling with the concept of pointers at the end of an intro course. I really can't see them effectively absorbing something that requires a good grasp on dynamically allocated structures in some form of a linked list.
I think it's a bit premature for us linux advocates to say it is immune to this. While a package manager certainly does help reduce dependancy issues, I don't think it addresses the physical issues.
/usr/* will start to get quite fat with binaries and their required libraries. Everytime one of those apps are called, it will take additional time to find the binary itself, then tack on the additional latency produced by the libraries having to be located and loaded into memory and the rusult will unavoidably be the visual degrade in performance.
...
First and foremost to consider is that there is no such think as a O(1) search algorithm("read 'Big Oh of one' for the non computer scientists in the crowd: notation used to measure the efficiency of an algorithm). The ammount of time required to search a list will always be a function of its size.
As the user base of Linux grows, so will the demand and supply of software. slowly, but surely,
Plus as more and more non computer literate people start using Linux, we will have to ensure that the software to support installing and upgrading packages on the system is user-proof, or other problems will result.
Unless someone can win a Nobell prize or Fields medal for finding a O(1) search, I'm afraid the above article is correct.
No wonder the whole world hates us...
-----
I must concur with the other posts on this subject. Any reference to the way wireless is handled in America is like trying to teach modern history with a 20 year old book. Before the flamers start up, I'm from America, but have spent a majority of my Military career living abroad. I am now happy to see American cellular technology finally catching up to the rest of the world. (although that Verizon commercial about the text messaging on the calculator style black and white diplay screen is still quite depressing)
In Europe and Asia, land lines are damned expensive and local calls are billed on top of your monthly rate to just have the line. Here in Japan it runs about 3 cents a minute (3 yen) to call local land lines and 7 cents (9 yen) to call cell phones. If you could pay a flat rate of ~$35 (5000 yen) for unlimitted wireless to your computer and your cell phone (256,000 color display, email, pictures, full color video games, and a variety of PDA like functions) runs you about the same rate, why would you ever get a land line? Even is you only spent 120 hours online, which I do on a regular basis, that's 21,600 yen (~$168 at today's rate) plus the ISP cost plus the basic land line charge.
Needless to say, I have yet to meet anyone here in Okinawa who uses dial up and very few who use DSL. Most rely entirely on the cell phone for calls and either cable or wireless for the net.
Even for those of us on US bases who do have free local calling, there is no broadband solution offered on base. So wireless is our only solution.
-------------
Pick up the books out of the trash, and look at their binding. They're published by either TSR or Wizards of the Coast--not a publishing house of any credulity, but rather a niche publisher that does an amazing job catering to the low-brow fantasy market.
Or rather, a credible publishing house that owns the rights to the Dungeons and Dragons universe. Hence why no one else publishes the stories. They aren't allowed to. TSR was purchased by Wizards of the Coast and so they are inherrently the same entity when it comes to publishing. I'm not sure what you deem as credible, or what you deem as quality, but the simple fact is that D&D books, including Salvatore's, have been around for years. And the popular ones, some of Salvatore's included, continue to be reprinted because people continue to read, and re-read them.
As far as the movie quality sucking, Two things:
#1 We've built our hopes up for years over these films and quite possibly built them too high.
#2 The whole reason why the films are being made is because someone with a rudimentary grasp of economics realized the baby-boomer generation would line up like little kids to pay for the hope of more childhood magic. (which they were when the original movies came out, and as we have seen so far for both movies, they have) The sad truth is that even if the second movie sucks as bad as everyone is saying, the same lines will be formed for Episode 3. Sure, they might be a little shorter, but lines they will be with cash in hand.
Just a note on the verbose fight scenes in the book. The reviewer is clearly unfamiliar with Salvatore's writing style. In depth fight scenes are what made Drizzt a legend in the Forgotten Realms world and it's not suprise to see Salvatore sticking with what worked in the past.
There are dozens of amateur OS's out there that have functioning GUI's. With a little bit of research, MS could come back into court with proof that they're not the only ones who integrated the browser and that it must be a good idea.
Note: I don't know of any of those OS's that did integrate the browser, but I have not downloaded and played with every single one either.
-----------------
Since I had to break out the calculator to visualize these specs in English units...here's what it comes out to.
H: 4.6in
W: 2.3in
D: 1.1in
Weight: 5.6oz (.35lbs)
----------
Several years ago, NTT Docomo released a phone in Japan that could store and play mp3's. The problem with the concept was a combination of price and target audience. The average person who desired to have their phone double as a walkman was college age and younger. The phone was priced at a modest $600 which most college students couldn't afford.
The idea of the camera in the phone is quite useful and has been available here in Japan for quite some time and is very reasonably priced. Almost all the phones over here have a great deal of PDA functionality and connect to the internet which is very useful.
But we return again to the target audience. The novelty of the phone being able to play my favorite mp3's and mpg's is going to wear off about two seconds after I read the price tag. The reason being is that like most salary workers, my day is fairly evenly divided up between work, commuting, eating, sleeping, and trying desparately to have something called a social life. Aside from amazing my friends during the "social life" part of my day, I have no time to play with gadgets, nor do I have any professional use for playing mp3's or mpg's. I didn't even get the camera phone over here because the cost didn't justify the very few times I would actually use that feature. IMO the P800 looks really cool for a technology demo, but I don't think they will be able to produce it at a sufficiently low cost for it to succeed. It's simple economics....when you have to pay rent/bills, support a girlfriend/wife/kids, and still manage to feed and entertain yourself, you find yourself seriously asking, "How often would I REALLY use this?"
----------
Here Here!!
This is remeniscent of the P4 vs Athlon video that showed the P4 surviving while the athlon cooked. Camera angle wasn't wide enough to see if a high powered fan was blowing on the P4 case and you have no idea if the room temps were the same for both, so you have to put your faith in the person who created it.
For the record, my girlfriends computer is completely toasted right now. But if I had a digital camera, I could give you some great shots of it POSTing. It's really easy considering that's where it hangs.