I actually have a friend who knows a great deal about the x86 ISA and does prefer the x86 architecture. I don't know how well he knows the PPC ISA, but would guess that he knows enough to form his opinion.
My preference for the PPC ISA has to do with the abundance of registers more than anything else (ok, decoding x86 instructions is kind of a pain). My preference for the G5 chip has more to do with power consumption than with the ISA. But, there are other systems that satisfy that need. Transmeta has a great position in the performance per watt and an acceptable position in the performance per dollar categories. I think those are the only thing transmeta has going for their chips (not that it isn't enough). It's the performance per watt and not the ISA or the performance per dollar that have made me consider getting the orion multisystem 12 node box.
I'm of the same opinion as the grandparent, so I will respond to your question. I would most likely buy a P5 OSX over a G6 OSX. Here's my reasoning:
1. Right now I favor my dual G5 over my dual Xeon because: a. my dual G5 is plugged into a 23" display and my dual Xeon is plugged into a poor 19" display. If I upgraded to one of the new displays, ADC support is dropped and a P5 box will likely have hardware that supports the nice DVI display. b. Most of my commercial software licenses are for my G5. This is actually a strike against switching to the P5 over the G6. c. The dual G5 box is significantly quieter than the dual Xeon box. Again, this is a strike against switching.
2. Cost of hardware: Umm... my dual Xeon with SCSI hard drives cost 30% less than my dual G5 for approximately the same processor performance (better drive performance). I'll assume the same cost difference applies in the hypothetical scenario.
3. There exist very good compilers for x86. I ran some benchmarks against both of my boxes and the Xeon outperformed the G5 in areas where the Xeon should have lost hands down. Inspecting the assembly code I realized that gcc (for C++ especially, but even for C) generates much better x86 code than PPC code. In fact, the C++ PPC code generation was terrible (particularly for large array/martix math). Intel's compiler is even better for C++ than g++, and the lack of an alternative on PPC makes the gap even larger.
This really irritates me. The PPC architecture *does* have advantages over x86. However, they are almost completely unrealized due to the lack of a decent compiler (metrowerks on PPC is decent, and fares well againt g++ on x86, but doesn't compare to intel's x86 compiler).
If (3) was fixed, I may be willing to accept (2) and then change my answer, particularly because my commercial software licenses more than make up the difference.
I really like OS X. As far as driver support, it works with my wacom tablet, my microsoft mouse, my sandisk usb drive, my iRiver MP3 player, my Canon camera, and my Epson scanner. It failed to work with my Labtec usb headphones/microphone (which doesn't matter since I can't play CounterStrike on it anyway), my Canon scanner (which happened to be one of only two Canoscan models not supported--that irritated me), and my USB wireless adapter (which I tried just to see if it worked. I already have a wireless card. Besides, the USB fucker bluescreened my windows laptop on multiple occasions, so I wouldn't call its windows support great either). I would recommend OSX iMacs to people like my grandmother as and OSX powerbooks to fellow CS students.
Well, it most likely would have roughly double the transistors. Most of the difference is in the model it presents to the system. The system would need to be able to identify the single chip as two processors, kinda like hyperthreading. Unlike hyperthreading, this would really be two processors: double the number of registers, double the number of ALUs, double the cache, etc.
An OS is supposed to MANAGE programs, not be AFFECTED by them.
True unless the program is not segregated from the OS memory footprint (like a driver, or a debugger that allows poking all memory locations). These programs generally require administrator priveledges to install (and setuid patented equivalent to run). Unfortunately, 90% of programs on windows require admin install, and for no good reason. They spew dll's in system directories instead of application local directories, write to HKEY_LOCAL_MACHINE instead of HKEY_CURRENT_USER, or simply use restricted API functions instead of less capable alternatives.
I'm not absolving Microsoft of all of their sins, but most crashes are caused because most windows applications do not follow the microsoft guidelines, and users still install unsafe programs. As an analogy: if all Linux software required kernel.h and required setuid to run, it too could take down your linux box if it crashed. However, most linux programs do not require such priveledges, and if one does, the user generally tries to find an alternative program. So some of the blame should be shouldered by the developers of the applications and the users who install them.
Of course, on the flip side, any linux development guide will tell you that you do *not* put kernel.h in userland programs. Microsoft makes it very developer friendly to do the equivalent (by providing better documentation on all of the naughty stuff I stated above). Well, they've changed on that recently, but they can't erase all of the old articles and books that teach the bad form.
heh. You still have the problem of implementing the "feasible" algorithm. Take triangulation of simple polygons. It was shown that a linear time algorithm exists -- in the decades that have passed nobody has ever implemented it. People typically use a randomized algorithm that is linear expected time with nlogn worst case.
Yes, I know that nlogn is still "feasible", but the point is that even the discovery of an algorithm doesn't imply that it has an easy implementation.
Ok, then maybe my "really boost the odds in your favor" was an overstatement. The parent is right, some people are just ignorant, and there is nothing you can do about that. I think even if it "doesn't count," it should still show something about the character of the person: they actually found work while in school.
I found that it helped *me* in my job search. Also, and when *I* was hiring people, it mattered. In fact, on more than one occasion I hired a current student instead of hiring a recent graduate with no experience. Three reasons (maybe not great, but it was my thinking at the time):
1. The student is making an effort to get work experience. 2. I could pay them less for approximately the same skills (which I estimated at "Zero, but trainable" for both). 3. I could contact professors I trust at the school and get their personal opinions of the current students. Only one of the recent grads was even recognized by the professors I knew.
The downside was having to work around a student's school schedule, but that only bit me in the ass twice in 2 years (and both times I was able to put in weekend time to compensate).
That said, the way I found the students I hired was through the school directly, instead of passively receiving resumes. So, more advice to the students: find which professors have industry ties and let them know you are interested in work positions. Professors are frequently approached by alumni looking to poach talent.
I worked for 3 years after graduating from college before heading back to grad school. My toys generally trump those of undergrads. As an undergrad, I was also happy enough to afford school without building debt.
I didn't like the old iMac myself. Let me rephrase: I hated it initially, but the design grew on me with the release of the 20", when the proportions actually worked. Then I liked it until I played with one in CompUSA. When I tilted the monitor down the screen went black! The connectors were shoddy, and tilting it back fixed it. Maybe it was a particularly bad display model, but I usually expect the premium price on Macs to buy higher quality.
I have had zero problems with my dual G5 or my G4 powerbook. I was looking at the G4 iMac as an office computer, and decided against it because of the display model. Instead, I bought an LCD projector (hey, cheaper than a 23" display), mouse, etc. to dock my laptop at the office.
I have to buy a computer for a coworker, who needs a mac set up by the end of September. I was hoping (more than a little) that G5 powerbooks would be coming in time for the new school year, as that would be ideal for my coworker. So now he and I get to decide between a G5 iMac, a G4 powerbook, or a bigger budget for a G5 desktop (which isn't likely). Of course, if I go out and buy the 30" display, I could loan the office my 23" display. Hmmm... I'll need to check with the boss (at home) on that one.
On a more personal note, I was sort of hoping that the new iMacs would be pretty aluminum instead of white plastic.
Thanks for posting that. She almost went with that option. Actually, after having a couple credit card applications rejected, she just went to Washington Mutual, where she has had a savings and a checking account for a few years (and just started a ROTH IRA). Washington Mutual was willing to give her a card with no credit history because she had an account history.
Bank of America required the secure card, because BofA outsourced their credit application process (no, not to India, just to another company). I find this part funny, my girlfriend used to work at BofA and had accounts there longer than at WaMu.
Another thing to check when leaving college (not highschool) is that there exist many loan programs with special rates for new graduates. I got my first car with a 3% loan because of one of those programs.
Oh, and more advice for people in college is to get summer internships or jobs. The job market may be tough when you get out of school, and having experience in the workforce will really boost the odds in your favor.
Heh. Earlier this year I was stuck in an undergraduate course (until I passed out of it). A buddy of mine was taking the course but ditching regularly. I was taking notes on my laptop (12" Powerbook -- the battery lasts forever). Unfortunately the class was the computer architecture class, and diagrams were sometimes necessary. Enter cell phone camera! I drew the diagrams on some scratch paper and took pictures of them. Bluetoothed them to the laptop and placed them in the notes. I had better lecture notes than the professor. Gratuitous use of technology.
As an added bonus, I was able to email my buddy the notes when he was out of town without having to scan in my terrible handwriting.
The follow up to this advice is to get a credit card anyway, and make sure you buy your books using it (and then pay them off). You will still have a credit history this way. With no credit card, you have no credit history. It makes it harder to buy a car years later (as my girlfriend is finding out at age 24). You won't have a great credit score--like having paid off a 10,000 dollar debt, making minimal payments, but never missing the payment--but it is better than nothing (and a lot better than bad credit).
Visio was better before Microsoft bought it and rolled it back one version (by cutting a lot of functionality that was in the last version). It's great again--even better than before the purchase--but the version rollback was sad and pissed off a lot of Visio users. The only thing they added was the MS logo.
Also, I spent 399.99 for Office:mac 2004 standard edition (no upgrade). No project, no access available. Unfortunately, I needed it for work (otherwise I just accept the few problems I have with OOo). The kicker was when OOo was able to open a file that Office:mac reported errors on.
Why do you think Project is great? I will grant that it is cheap, but have you ever used any other project management software? The *only* thing project does well is produce Gantt charts.
Powerpoint on the mac rocks. It's all about the rotating cube transition. Seriously. Also, have you ever used Powerpoint with a projector connected as a second monitor instead of as a mirror of your primary display? Doing so will allow you to see your speaker notes and outline while the audience just sees your slides. AFAIK, OOo doesn't have this capability (somebody please correct me).
When was the last time you used Access? I just trained a Psychologist to use it rather effectively with about an hour of training. I think Access is extremely usable, for what it does. Also, recent versions of Access allow you to use MSDE instead of Jet as the database engine. MSDE is SQL Server crippled, but it is still extremely capable.
Your post makes me happy and sad at the same time. I am happy that you learned to enjoy the theory aspects of your education. I am sad that the applied curriculum was so outdated. I used to think that Computer Science schools had all become code monkey factories. That was my (ill informed) opinion 10 years ago.
Now I have gone back to school at UCSB in the PhD program. I have been quite impressed with how current the applied curriculum is, both at the graduate and at the undergraduate levels (there are a couple excellent theory courses for grad students. I know nothing about the undergraduate theory).
At the graduate level, a course was offered on language implementation. Topics included bytecode verification of java and CIL, garbage collection techniques, virtual machine implementation, JIT compiling optimizations, etc. Quite interesting, and with a fantastic instructor.
I was the TA for the Operating Systems class. The first assignment was to write a shell using readline. The next assignment was a basic "let's learn pthreads" simulation. Those two assignments were cute and had theoretical importance, but mostly dated as far as practicality goes. The third assignment was to modify the proc system in linux (using User Mode Linux), and write a loadable kernel module. I'm not saying that linux is modern *wink*, just that a "modify what you need in these thousands of files that you didn't write to add feature X" assignment is quite a practical experience.
I am enjoying almost every aspect of being back in school. The only problem I have is balancing my time between being a full time student and working full time.
Oh, and VB has got to be one of the greatest languages ever designed. The exception handling is amazing: "If an error happens, just pick up where you left off (On Error Resume Next)" or "Hmm, the code hit an error. Let's reset all the values to default and retry the code from before the error happened". That's just awesome! (And to figure out if I am serious or not, you should know that I hate VB.Net)
I would say no. If there is no copyright, then the GPL could not be enforced (*ignores arguments about enforcement of GPL*). Therefore people could still take otherwise free software, modify it, and release only the binaries. You would no longer have the freedom to inspect and modify the modified code (except in any form you could derive from the binaries).
Sure, you could freely distribute the binaries to others, but that's only one of the freedoms you get from the GPL. Of course, you now have the additional freedom of not being bound by the GPL, but I digress...
Have you ever tried to program a traffic simulation? There are few simulations more trivial, so don't be impressed.
Second, traffic simulations based on human behavior are always post hoc analysis. Twiddle the parameters until it looks right, then make up the behavior that fits the parameters. If you've ever had a chance to play with one, they are a lot of fun. Often the whole simulation falls apart with less than a 5 percent change in some parameters.
Actually, this is true of almost all behavioral modelling. If you've ever done any reading in modelling of software systems, you know how hard that can be (try proving the safety of critical sections in a multithreaded system). Behavioral modelling has all of the great concurrency of software modelling with less determinism (or at least it should...). What makes anyone trust behavioral modelling is beyond me.
As a side note, many things in the real world are based on this kind of crappy science all the time. Check out the San Diego freeway system. The I5 805 merge was just recently redone to improve traffic flow. It failed miserably. Staffing levels on military vessels are done by models. The ships are always understaffed initially until trial and error fixes them.
Unfortunately, people think that computer simulations can solve all problems, even when there is no theoretical reason to believe that the model will even approximate reality.
UI design is a science. Apple used to be very well aware of that. Unfortunately, Apple seems to have ditched human factors engineers (cognitive psychologists) in favor of graphic designers. Pretty != Usable.
One of my pet peeves with OSX is that the mouse cursor hot point isn't the tip of the arrow. It is indented a few pixels. I find that kind of annoying given that the little resize window pad in the lower right corner of the window is smaller than the mouse cursor. I used to miss the window when trying to resize on a regular basis. Now I have become used to it.
My second peeve: The window resizer is on the lower right of the window, but the other window controls are all in the upper left. When using Flash or Office (applications with floating pallets), one of these two will invariably be covered up. Resizable Edges were an advancement in window management, not a problem. Actually, the new Office 2004 has a number of UI problems, but those aren't Apple's fault.
Third peeve: I click on a file on my desktop and hit delete. What happens? Nothing. No delete, no move to trash.
Fourth Peeve: How do you cycle through windows on the apple? Sorry, you can cycle through applications using Command-TAB, but even cycling through application windows within products made by apple isn't standardized. No key combo to cycle through all open windows. Fortunately they have the nice F9-F11 keys for "fast" window switching.
Fifth peeve: You are typing a report and want to listen to some music. Open up iTunes. Press play. Minimize iTunes, revealing your word processor. The cursor is blinking in your word processor. You press space to continue typing. iTunes pauses! WTF?! Oh yeah, you have to click on the word processor to give it focus (even though the blinking cursor looks inviting enough).
Sixth Peeve: There are two keys marked delete on the keyboard. Guess what? One deletes and the other backspaces.
Seventh Peeve: They still ship with a single button mouse even though nearly all applications make use of the context menu. Some applications which are fantastically easy to use mouse only with a 2 button mouse become keyboard bound on the default system.
OK, so that is probably all of my problems with the current mac UI. I find OSX increadibly usable compared to most alternatives, but I still feel that it is a step backwards in usability (but not attractiveness) from previous Macs (except for the part where old macs crashed all the fucking time and OSX doesn't).
Incidentally, pretty much all operating systems have UI guidelines. Sun has a UI guideline for Java. Microsoft has one for windows. I hear that the Gnome people have one as well.
"The said constitution shall never be construed to authorize congress to prevent the people of the United States who are peaceable citizens from keeping their own arms."
Samuel Adams
"No free man shall ever be de-barred the use of arms. The strongest reason for the people to retain their right to keep and bear arms is as a last resort to protect themselves against tyranny in government."
Thomas Jefferson
The whole purpose of the right to bear arms came as a backlash against the government taking arms away from private citizens. The US Military is not a well regulated militia. It is almost completely unchecked. If the government declared martial law, is there any potential for the citizens to defend themselves against the military?
The windows API is an ECMA standard. I happened to find it when looking for the CIL (MSIL actually differs slightly from the CIL standard by the way) standard docs. I think it is from the 3.1 days, though. Anyway, I thought it was interesting for two reasons:
1. Windows really *IS* standard. 2. ECMA really *IS* a standards for sale whore.
Also, to stick with the grandparent post topic, what does it matter if Windows are standard in business as far as a school is concerned? The students are likely checking email and browsing the internet, which really doesn't have much to do with what OS it is running on. It isn't as if the school is teaching kids to use MS Project so that they can all grow up to be good little middle managers.
Actually, my stupid comment just reminded me of something: does anyone out there use a Mac in a Windows dominated office? What do you use to open up MS Project files?
I am working on my PhD in CS and have had to do a decent amount of coding. I have a Dual G5 2 GHz with the 23" display which sits right next to my Dual Xeon 2.4 GHz WinXP Pro box with a 19" display (which sits right next to my Duran 1.3 GHz Linux box with no monitor). I have to say that getting the 23" display was worth every cent. I barely touch my WinXP box, and only to run my applications. One important thing... the new displays are DVI, not the Apple Cinema Display adapter (which carries power and USB), which means that you should probably be seeing a driver from NVidia for windows boxes at some point. Do it. You will not be sorry (just poor).
I wasn't trying to be an ass. Actually, the whole farming on Mars thing has always occurred to me as infeasible, at least on the timelines people discuss sending humans to mars on. So, the seed response is a bit of a zinger against my post. However, I am not aware of any plants grown from seed which do not also require a lot of oxygen (most plants have a fairly long respiration cycle as well). Most of the oxygen on earth comes from the ocean. Of course, since oxygen production is a surface phenomena, it isn't surprising given that the majority of the surface is ocean. I am by no means an expert at biology (ok, most things). Harvesting the O2 is still a formidable problem. Plants are not efficient in terms of the surface area they require for the quantity of O2 they produce (they may be the most efficient method that exists, but that doesn't invalidate the argument). The fact that there has been no successful sealed ecosystem experiment on Earth is more troubling. You have to build something to harvest resources on mars. First, you have to pay for designing that something. Then, you have to pay for getting that something to mars.
Mars is.38 surface gravity while the moon is.17 (relative to Earth). Still, that is not insubstantial in terms of fuel. I am aware that we are not sending a shuttle. In fact, my little sister is actually a rocket scientist at Lockheed working on (one of?) their next gen space platform project(s). My point was that off earth launch technology (including fuel refining technology) does not exist today and is a non-trivial problem. Even if you can get stuff from mars, you have to get the equipment to mars.
The majority of the arguments pro Mars seem to rely on the fact that since something is possible, it can be acheived through taxpayer funding. Have these people been asleep for the last few decades? This is the US Government we are talking about! Throwing piles of money into a barrel, covering it in gasoline, lighting a match, and salvaging what you can has a higher return on investment than most government funded projects. Heck, look at the post today about the magic laser stun gun.
I'd be a lot more for the whole mars thing if it wasn't pushing back science missions that actually have a chance of working. What happens if, after spending a trillion dollars on the project and absorbing the costs of all of the overruns and minor failures along the way, the public opinion of the project drops to a small minority? Then we will have lost all of the science missions that were cast aside AND the mars mission. What happens if the martian farming project takes five years and ends in failure? There are two government models: the first, fund an expensive project with good funding to solve the problem. The second, inadequately fund many cheap projects to solve the problem. Unfortunately, both models fail quite often (fortunately, many failures turn into successes after quadrupling their budget in overruns). The odds of everything coming together within the proposed budgets? Slim.
The taxpayers have balked at large projects before. Remember the supercollider project? That was a horrible waste of funding. Imagine if that funding had gone into magnetics (and high temperature superconductor) research. At the end of that initial funding, the supercollider could have probably been built for a fraction of the origional cost. Note, I've stolen this; this was actually the conclusion of a large number of people on the project, who attributed a lot of the problems to the size of the cooling equipment required by the magnets. Of course, they didn't want to lose their funding, and they didn't do magnetics research.
So, the question should not only be "Can we do it?" The questions should be "Can we do it within the funding constraints we envision?" and "Are there better projects to invest in that might reduce the cost of this project later on?" You might even want to add "Are there better projects that won't act as stepping stones for this project but are more worthy of funding anyway?"
Note: I had not heard of Zubrin's book. Thanks for the reference.
I worked as a subcontractor to a rather large defense contractor. They had no web site and their employees all had AOL addresses (and NMCI addresses). Oh, and they built web apps for the Navy. Go figure.
Actually, there is a vehicle mounted (or wall mounted) stun gun based off of a super soaker. A group of us got the idea when we were watching a bug zapper, holding a super soaker in one hand and a beer in the other thinking...
Anyway, my dad built the thing. It has been licensed for use in some non-US prisons for riot control.
http://www.jaycor.com/eme/watcan.htm
as far as small stun equipment, check out the same company's sticky shocker:
Slashdot Mirror
I actually have a friend who knows a great deal about the x86 ISA and does prefer the x86 architecture. I don't know how well he knows the PPC ISA, but would guess that he knows enough to form his opinion.
My preference for the PPC ISA has to do with the abundance of registers more than anything else (ok, decoding x86 instructions is kind of a pain). My preference for the G5 chip has more to do with power consumption than with the ISA. But, there are other systems that satisfy that need. Transmeta has a great position in the performance per watt and an acceptable position in the performance per dollar categories. I think those are the only thing transmeta has going for their chips (not that it isn't enough). It's the performance per watt and not the ISA or the performance per dollar that have made me consider getting the orion multisystem 12 node box.
I'm of the same opinion as the grandparent, so I will respond to your question. I would most likely buy a P5 OSX over a G6 OSX. Here's my reasoning:
1. Right now I favor my dual G5 over my dual Xeon because:
a. my dual G5 is plugged into a 23" display and my dual Xeon is plugged into a poor 19" display. If I upgraded to one of the new displays, ADC support is dropped and a P5 box will likely have hardware that supports the nice DVI display.
b. Most of my commercial software licenses are for my G5. This is actually a strike against switching to the P5 over the G6.
c. The dual G5 box is significantly quieter than the dual Xeon box. Again, this is a strike against switching.
2. Cost of hardware: Umm... my dual Xeon with SCSI hard drives cost 30% less than my dual G5 for approximately the same processor performance (better drive performance). I'll assume the same cost difference applies in the hypothetical scenario.
3. There exist very good compilers for x86. I ran some benchmarks against both of my boxes and the Xeon outperformed the G5 in areas where the Xeon should have lost hands down. Inspecting the assembly code I realized that gcc (for C++ especially, but even for C) generates much better x86 code than PPC code. In fact, the C++ PPC code generation was terrible (particularly for large array/martix math). Intel's compiler is even better for C++ than g++, and the lack of an alternative on PPC makes the gap even larger.
This really irritates me. The PPC architecture *does* have advantages over x86. However, they are almost completely unrealized due to the lack of a decent compiler (metrowerks on PPC is decent, and fares well againt g++ on x86, but doesn't compare to intel's x86 compiler).
If (3) was fixed, I may be willing to accept (2) and then change my answer, particularly because my commercial software licenses more than make up the difference.
I really like OS X. As far as driver support, it works with my wacom tablet, my microsoft mouse, my sandisk usb drive, my iRiver MP3 player, my Canon camera, and my Epson scanner. It failed to work with my Labtec usb headphones/microphone (which doesn't matter since I can't play CounterStrike on it anyway), my Canon scanner (which happened to be one of only two Canoscan models not supported--that irritated me), and my USB wireless adapter (which I tried just to see if it worked. I already have a wireless card. Besides, the USB fucker bluescreened my windows laptop on multiple occasions, so I wouldn't call its windows support great either). I would recommend OSX iMacs to people like my grandmother as and OSX powerbooks to fellow CS students.
Well, it most likely would have roughly double the transistors. Most of the difference is in the model it presents to the system. The system would need to be able to identify the single chip as two processors, kinda like hyperthreading. Unlike hyperthreading, this would really be two processors: double the number of registers, double the number of ALUs, double the cache, etc.
An OS is supposed to MANAGE programs, not be AFFECTED by them.
True unless the program is not segregated from the OS memory footprint (like a driver, or a debugger that allows poking all memory locations). These programs generally require administrator priveledges to install (and setuid patented equivalent to run). Unfortunately, 90% of programs on windows require admin install, and for no good reason. They spew dll's in system directories instead of application local directories, write to HKEY_LOCAL_MACHINE instead of HKEY_CURRENT_USER, or simply use restricted API functions instead of less capable alternatives.
I'm not absolving Microsoft of all of their sins, but most crashes are caused because most windows applications do not follow the microsoft guidelines, and users still install unsafe programs. As an analogy: if all Linux software required kernel.h and required setuid to run, it too could take down your linux box if it crashed. However, most linux programs do not require such priveledges, and if one does, the user generally tries to find an alternative program. So some of the blame should be shouldered by the developers of the applications and the users who install them.
Of course, on the flip side, any linux development guide will tell you that you do *not* put kernel.h in userland programs. Microsoft makes it very developer friendly to do the equivalent (by providing better documentation on all of the naughty stuff I stated above). Well, they've changed on that recently, but they can't erase all of the old articles and books that teach the bad form.
By the way, since you seem knowledgable: Is there any NP-C reduction that is conjectured not to be a log space (LOG) reduction?
From what I remember, most polytime reductions are trivially in LOG. I am also aware that LOG!=P is open, but it is conjectured that it is true.
heh. You still have the problem of implementing the "feasible" algorithm. Take triangulation of simple polygons. It was shown that a linear time algorithm exists -- in the decades that have passed nobody has ever implemented it. People typically use a randomized algorithm that is linear expected time with nlogn worst case.
Yes, I know that nlogn is still "feasible", but the point is that even the discovery of an algorithm doesn't imply that it has an easy implementation.
Ok, then maybe my "really boost the odds in your favor" was an overstatement. The parent is right, some people are just ignorant, and there is nothing you can do about that. I think even if it "doesn't count," it should still show something about the character of the person: they actually found work while in school.
I found that it helped *me* in my job search. Also, and when *I* was hiring people, it mattered. In fact, on more than one occasion I hired a current student instead of hiring a recent graduate with no experience. Three reasons (maybe not great, but it was my thinking at the time):
1. The student is making an effort to get work experience.
2. I could pay them less for approximately the same skills (which I estimated at "Zero, but trainable" for both).
3. I could contact professors I trust at the school and get their personal opinions of the current students. Only one of the recent grads was even recognized by the professors I knew.
The downside was having to work around a student's school schedule, but that only bit me in the ass twice in 2 years (and both times I was able to put in weekend time to compensate).
That said, the way I found the students I hired was through the school directly, instead of passively receiving resumes. So, more advice to the students: find which professors have industry ties and let them know you are interested in work positions. Professors are frequently approached by alumni looking to poach talent.
I worked for 3 years after graduating from college before heading back to grad school. My toys generally trump those of undergrads. As an undergrad, I was also happy enough to afford school without building debt.
I didn't like the old iMac myself. Let me rephrase: I hated it initially, but the design grew on me with the release of the 20", when the proportions actually worked. Then I liked it until I played with one in CompUSA. When I tilted the monitor down the screen went black! The connectors were shoddy, and tilting it back fixed it. Maybe it was a particularly bad display model, but I usually expect the premium price on Macs to buy higher quality.
I have had zero problems with my dual G5 or my G4 powerbook. I was looking at the G4 iMac as an office computer, and decided against it because of the display model. Instead, I bought an LCD projector (hey, cheaper than a 23" display), mouse, etc. to dock my laptop at the office.
I have to buy a computer for a coworker, who needs a mac set up by the end of September. I was hoping (more than a little) that G5 powerbooks would be coming in time for the new school year, as that would be ideal for my coworker. So now he and I get to decide between a G5 iMac, a G4 powerbook, or a bigger budget for a G5 desktop (which isn't likely). Of course, if I go out and buy the 30" display, I could loan the office my 23" display. Hmmm... I'll need to check with the boss (at home) on that one.
On a more personal note, I was sort of hoping that the new iMacs would be pretty aluminum instead of white plastic.
Thanks for posting that. She almost went with that option. Actually, after having a couple credit card applications rejected, she just went to Washington Mutual, where she has had a savings and a checking account for a few years (and just started a ROTH IRA). Washington Mutual was willing to give her a card with no credit history because she had an account history.
Bank of America required the secure card, because BofA outsourced their credit application process (no, not to India, just to another company). I find this part funny, my girlfriend used to work at BofA and had accounts there longer than at WaMu.
Another thing to check when leaving college (not highschool) is that there exist many loan programs with special rates for new graduates. I got my first car with a 3% loan because of one of those programs.
Oh, and more advice for people in college is to get summer internships or jobs. The job market may be tough when you get out of school, and having experience in the workforce will really boost the odds in your favor.
Heh. Earlier this year I was stuck in an undergraduate course (until I passed out of it). A buddy of mine was taking the course but ditching regularly. I was taking notes on my laptop (12" Powerbook -- the battery lasts forever). Unfortunately the class was the computer architecture class, and diagrams were sometimes necessary. Enter cell phone camera! I drew the diagrams on some scratch paper and took pictures of them. Bluetoothed them to the laptop and placed them in the notes. I had better lecture notes than the professor. Gratuitous use of technology.
As an added bonus, I was able to email my buddy the notes when he was out of town without having to scan in my terrible handwriting.
The follow up to this advice is to get a credit card anyway, and make sure you buy your books using it (and then pay them off). You will still have a credit history this way. With no credit card, you have no credit history. It makes it harder to buy a car years later (as my girlfriend is finding out at age 24). You won't have a great credit score--like having paid off a 10,000 dollar debt, making minimal payments, but never missing the payment--but it is better than nothing (and a lot better than bad credit).
Visio was better before Microsoft bought it and rolled it back one version (by cutting a lot of functionality that was in the last version). It's great again--even better than before the purchase--but the version rollback was sad and pissed off a lot of Visio users. The only thing they added was the MS logo.
Also, I spent 399.99 for Office:mac 2004 standard edition (no upgrade). No project, no access available. Unfortunately, I needed it for work (otherwise I just accept the few problems I have with OOo). The kicker was when OOo was able to open a file that Office:mac reported errors on.
Why do you think Project is great? I will grant that it is cheap, but have you ever used any other project management software? The *only* thing project does well is produce Gantt charts.
Powerpoint on the mac rocks. It's all about the rotating cube transition. Seriously. Also, have you ever used Powerpoint with a projector connected as a second monitor instead of as a mirror of your primary display? Doing so will allow you to see your speaker notes and outline while the audience just sees your slides. AFAIK, OOo doesn't have this capability (somebody please correct me).
When was the last time you used Access? I just trained a Psychologist to use it rather effectively with about an hour of training. I think Access is extremely usable, for what it does. Also, recent versions of Access allow you to use MSDE instead of Jet as the database engine. MSDE is SQL Server crippled, but it is still extremely capable.
I'm guessing no. I just visited in both safari and mozilla on the mac and got the Real 10 for OS X.
I actually like the real codec. A pity nobody likes the company.
Congrats! What are you going to do next?
Your post makes me happy and sad at the same time. I am happy that you learned to enjoy the theory aspects of your education. I am sad that the applied curriculum was so outdated. I used to think that Computer Science schools had all become code monkey factories. That was my (ill informed) opinion 10 years ago.
Now I have gone back to school at UCSB in the PhD program. I have been quite impressed with how current the applied curriculum is, both at the graduate and at the undergraduate levels (there are a couple excellent theory courses for grad students. I know nothing about the undergraduate theory).
At the graduate level, a course was offered on language implementation. Topics included bytecode verification of java and CIL, garbage collection techniques, virtual machine implementation, JIT compiling optimizations, etc. Quite interesting, and with a fantastic instructor.
I was the TA for the Operating Systems class. The first assignment was to write a shell using readline. The next assignment was a basic "let's learn pthreads" simulation. Those two assignments were cute and had theoretical importance, but mostly dated as far as practicality goes. The third assignment was to modify the proc system in linux (using User Mode Linux), and write a loadable kernel module. I'm not saying that linux is modern *wink*, just that a "modify what you need in these thousands of files that you didn't write to add feature X" assignment is quite a practical experience.
I am enjoying almost every aspect of being back in school. The only problem I have is balancing my time between being a full time student and working full time.
Oh, and VB has got to be one of the greatest languages ever designed. The exception handling is amazing: "If an error happens, just pick up where you left off (On Error Resume Next)" or "Hmm, the code hit an error. Let's reset all the values to default and retry the code from before the error happened". That's just awesome! (And to figure out if I am serious or not, you should know that I hate VB.Net)
I would say no. If there is no copyright, then the GPL could not be enforced (*ignores arguments about enforcement of GPL*). Therefore people could still take otherwise free software, modify it, and release only the binaries. You would no longer have the freedom to inspect and modify the modified code (except in any form you could derive from the binaries).
Sure, you could freely distribute the binaries to others, but that's only one of the freedoms you get from the GPL. Of course, you now have the additional freedom of not being bound by the GPL, but I digress...
Have you ever tried to program a traffic simulation? There are few simulations more trivial, so don't be impressed.
Second, traffic simulations based on human behavior are always post hoc analysis. Twiddle the parameters until it looks right, then make up the behavior that fits the parameters. If you've ever had a chance to play with one, they are a lot of fun. Often the whole simulation falls apart with less than a 5 percent change in some parameters.
Actually, this is true of almost all behavioral modelling. If you've ever done any reading in modelling of software systems, you know how hard that can be (try proving the safety of critical sections in a multithreaded system). Behavioral modelling has all of the great concurrency of software modelling with less determinism (or at least it should...). What makes anyone trust behavioral modelling is beyond me.
As a side note, many things in the real world are based on this kind of crappy science all the time. Check out the San Diego freeway system. The I5 805 merge was just recently redone to improve traffic flow. It failed miserably. Staffing levels on military vessels are done by models. The ships are always understaffed initially until trial and error fixes them.
Unfortunately, people think that computer simulations can solve all problems, even when there is no theoretical reason to believe that the model will even approximate reality.
UI design is a science. Apple used to be very well aware of that. Unfortunately, Apple seems to have ditched human factors engineers (cognitive psychologists) in favor of graphic designers. Pretty != Usable.
One of my pet peeves with OSX is that the mouse cursor hot point isn't the tip of the arrow. It is indented a few pixels. I find that kind of annoying given that the little resize window pad in the lower right corner of the window is smaller than the mouse cursor. I used to miss the window when trying to resize on a regular basis. Now I have become used to it.
My second peeve: The window resizer is on the lower right of the window, but the other window controls are all in the upper left. When using Flash or Office (applications with floating pallets), one of these two will invariably be covered up. Resizable Edges were an advancement in window management, not a problem. Actually, the new Office 2004 has a number of UI problems, but those aren't Apple's fault.
Third peeve: I click on a file on my desktop and hit delete. What happens? Nothing. No delete, no move to trash.
Fourth Peeve: How do you cycle through windows on the apple? Sorry, you can cycle through applications using Command-TAB, but even cycling through application windows within products made by apple isn't standardized. No key combo to cycle through all open windows. Fortunately they have the nice F9-F11 keys for "fast" window switching.
Fifth peeve: You are typing a report and want to listen to some music. Open up iTunes. Press play. Minimize iTunes, revealing your word processor. The cursor is blinking in your word processor. You press space to continue typing. iTunes pauses! WTF?! Oh yeah, you have to click on the word processor to give it focus (even though the blinking cursor looks inviting enough).
Sixth Peeve: There are two keys marked delete on the keyboard. Guess what? One deletes and the other backspaces.
Seventh Peeve: They still ship with a single button mouse even though nearly all applications make use of the context menu. Some applications which are fantastically easy to use mouse only with a 2 button mouse become keyboard bound on the default system.
OK, so that is probably all of my problems with the current mac UI. I find OSX increadibly usable compared to most alternatives, but I still feel that it is a step backwards in usability (but not attractiveness) from previous Macs (except for the part where old macs crashed all the fucking time and OSX doesn't).
Incidentally, pretty much all operating systems have UI guidelines. Sun has a UI guideline for Java. Microsoft has one for windows. I hear that the Gnome people have one as well.
Errr...
"The said constitution shall never be construed to authorize congress to prevent the people of the United States who are peaceable citizens from keeping their own arms."
Samuel Adams
"No free man shall ever be de-barred the use of arms. The strongest reason for the people to retain their right to keep and bear arms is as a last resort to protect themselves against tyranny in government."
Thomas Jefferson
The whole purpose of the right to bear arms came as a backlash against the government taking arms away from private citizens. The US Military is not a well regulated militia. It is almost completely unchecked. If the government declared martial law, is there any potential for the citizens to defend themselves against the military?
The windows API is an ECMA standard. I happened to find it when looking for the CIL (MSIL actually differs slightly from the CIL standard by the way) standard docs. I think it is from the 3.1 days, though. Anyway, I thought it was interesting for two reasons:
1. Windows really *IS* standard.
2. ECMA really *IS* a standards for sale whore.
Also, to stick with the grandparent post topic, what does it matter if Windows are standard in business as far as a school is concerned? The students are likely checking email and browsing the internet, which really doesn't have much to do with what OS it is running on. It isn't as if the school is teaching kids to use MS Project so that they can all grow up to be good little middle managers.
Actually, my stupid comment just reminded me of something: does anyone out there use a Mac in a Windows dominated office? What do you use to open up MS Project files?
I am working on my PhD in CS and have had to do a decent amount of coding. I have a Dual G5 2 GHz with the 23" display which sits right next to my Dual Xeon 2.4 GHz WinXP Pro box with a 19" display (which sits right next to my Duran 1.3 GHz Linux box with no monitor). I have to say that getting the 23" display was worth every cent. I barely touch my WinXP box, and only to run my applications. One important thing... the new displays are DVI, not the Apple Cinema Display adapter (which carries power and USB), which means that you should probably be seeing a driver from NVidia for windows boxes at some point. Do it. You will not be sorry (just poor).
I wasn't trying to be an ass. Actually, the whole farming on Mars thing has always occurred to me as infeasible, at least on the timelines people discuss sending humans to mars on. So, the seed response is a bit of a zinger against my post. However, I am not aware of any plants grown from seed which do not also require a lot of oxygen (most plants have a fairly long respiration cycle as well). Most of the oxygen on earth comes from the ocean. Of course, since oxygen production is a surface phenomena, it isn't surprising given that the majority of the surface is ocean. I am by no means an expert at biology (ok, most things). Harvesting the O2 is still a formidable problem. Plants are not efficient in terms of the surface area they require for the quantity of O2 they produce (they may be the most efficient method that exists, but that doesn't invalidate the argument). The fact that there has been no successful sealed ecosystem experiment on Earth is more troubling. You have to build something to harvest resources on mars. First, you have to pay for designing that something. Then, you have to pay for getting that something to mars.
.38 surface gravity while the moon is .17 (relative to Earth). Still, that is not insubstantial in terms of fuel. I am aware that we are not sending a shuttle. In fact, my little sister is actually a rocket scientist at Lockheed working on (one of?) their next gen space platform project(s). My point was that off earth launch technology (including fuel refining technology) does not exist today and is a non-trivial problem. Even if you can get stuff from mars, you have to get the equipment to mars.
Mars is
The majority of the arguments pro Mars seem to rely on the fact that since something is possible, it can be acheived through taxpayer funding. Have these people been asleep for the last few decades? This is the US Government we are talking about! Throwing piles of money into a barrel, covering it in gasoline, lighting a match, and salvaging what you can has a higher return on investment than most government funded projects. Heck, look at the post today about the magic laser stun gun.
I'd be a lot more for the whole mars thing if it wasn't pushing back science missions that actually have a chance of working. What happens if, after spending a trillion dollars on the project and absorbing the costs of all of the overruns and minor failures along the way, the public opinion of the project drops to a small minority? Then we will have lost all of the science missions that were cast aside AND the mars mission. What happens if the martian farming project takes five years and ends in failure? There are two government models: the first, fund an expensive project with good funding to solve the problem. The second, inadequately fund many cheap projects to solve the problem. Unfortunately, both models fail quite often (fortunately, many failures turn into successes after quadrupling their budget in overruns). The odds of everything coming together within the proposed budgets? Slim.
The taxpayers have balked at large projects before. Remember the supercollider project? That was a horrible waste of funding. Imagine if that funding had gone into magnetics (and high temperature superconductor) research. At the end of that initial funding, the supercollider could have probably been built for a fraction of the origional cost. Note, I've stolen this; this was actually the conclusion of a large number of people on the project, who attributed a lot of the problems to the size of the cooling equipment required by the magnets. Of course, they didn't want to lose their funding, and they didn't do magnetics research.
So, the question should not only be "Can we do it?" The questions should be "Can we do it within the funding constraints we envision?" and "Are there better projects to invest in that might reduce the cost of this project later on?" You might even want to add "Are there better projects that won't act as stepping stones for this project but are more worthy of funding anyway?"
Note: I had not heard of Zubrin's book. Thanks for the reference.
I worked as a subcontractor to a rather large defense contractor. They had no web site and their employees all had AOL addresses (and NMCI addresses). Oh, and they built web apps for the Navy. Go figure.
Actually, there is a vehicle mounted (or wall mounted) stun gun based off of a super soaker. A group of us got the idea when we were watching a bug zapper, holding a super soaker in one hand and a beer in the other thinking...
Anyway, my dad built the thing. It has been licensed for use in some non-US prisons for riot control.
http://www.jaycor.com/eme/watcan.htm
as far as small stun equipment, check out the same company's sticky shocker:
http://www.jaycor.com/eme/nlp.htm