Slashdot Mirror
Slashback: Pricedrops, Honor, Games
Making your computer worth even less. Acid-F1ux writes: "Advanced Micro Devices has slashed prices of its desktop and mobile Athlon processors just days after a similar move by rival Intel. The cuts range from 17 percent to 52 percent for mobile Athlon XP chips and between 11 percent and 32 percent for desktop Athlon XP chips. On Sunday, Intel dropped prices of its Pentium 4 processors by as much as 53 percent."
Progressive Education strikes a blow. darnellmc writes: "According to this Atlanta Journal-Constitution news article GA Tech had so many students violate the school's "honor code" that they have decided to change it.
"In the wake of the investigation, Tech officials have decided to allow students in introductory computer science courses to share information and collaborate on homework, previously prohibited under the school's academic honor code."
Of course code sharing also teaches the value of Open Source ;o) . Maybe now some young Computer Science student can spend more time on developing a good overall program, instead of spending a bunch of time writing simple things like their own sorting routine."
How many letters will the next big threat have? matthew writes: "LWN is carrying the notes from the Broadcast Protection Discussion Group (BPDG) conference (more info at the EFF). The BPDG is the body that will be suggesting future technological control measures; they make the DMCA and CBDTPA seem like trivial problems. The BPDG conference was last week and it was open to the public so anyone could call in. You can read about what the FSF's Bradley M. Kuhn digitalspeech.org's Jonathan Watterson thought of the conference. The basic summary is that we're screwed if people don't start fighting against this kind of injustice."
This is what's called taking license. infochuck writes "Back in January, this story on Slashdot focused on Borland's licensing PR fiasco, and how they promised to remedy the situation (in short, their license permitted them to search at any time any of your computers looking for stolen software). Well, here we are, five months later, and their license hasn't changed one bit - at least not the two most unreasonable clauses, 12 and 14.4, and not in the license included with the Windows version of the Personal Edition. Download for yourself to see, but be warned, you'll have to register, which involves many questions and no less than 5 checkboxes to uncheck, as well as at least a 25MB DL. I believe pr@borland.com is still the place to write..."
Playful is good. If the last month of pre-hype hype, pre-hype, actual hype and post-hype weren't enough, you'll be pleased to read that E3 coverage continues, at Gamespy (some cool reviews), Gamegal (good photos) and other sites beginning with "Game."
If enough people break the rules, they'll change em!
Never argue with an idiot, he'll just lower you to his level and beat you with experience.
"Back in the day", they used to have one of the friendliest licenses out there (the "treat this software like a book" license) and a great many people bought their stuff because of the low price and reasonable license. You could even install it on a another computer, and as long as two people weren't using the software at the same time, it was ok.
All the software I wr[oi]te at my school includes the GPL copyright notice. The nice thing about the GPL is that you can share with fellow students to your hearts content but if the students use any of your code, they have to clearly mark that it is your code if they use it.
As far as I can tell, this protects me in the event that a student is accused of cheating while still allowing me to show anyone my code. I personally think that software licensing should be a part of every CS program and the GPL should be encouraged to be used for all assignments.
int func(int a);
func((b += 3, b));
-WS
An operating system should be like a light switch... simple, effective, easy to use, and designed for everyone.
When you are writing software for a company or a non-classroom project, reinventing the wheel is usually a bad idea. However, in the context of learning how to program, I think it is important for students to do their own work and not just copy code from the Internet or from the smart student down the hall.
The new rules are just a license to cheat.
I can understand GT's reluctance to allow "student-student assistance" when I consider the reputation that they have as an engineering school (and not just computer / electical engineering, either). I seem to recall hearing that the Textiles lab at GT came up with about half of the specialty cloths used in NASA's spacesuits. I'm sure that with that sort of history they thought that clamping down tight on the CS students was a good PR move-- not only do we have brilliance, but we pull it out of one student at a time.
Maybe this was good PR. But it was a stupid educational move. And I don't care how good a business they do with NASA or anyone else, or what sort of press they put out; above all else, isn't college supposed to be about learning?
Do you like Japanese imports?
Maybe now some young Computer Science student can spend more time on developing a good overall program, instead of spending a bunch of time writing simple things like their own sorting routine."
Noooo!!!! I don't know how many "groups" you've worked in at college level, but almost every group has one or two people that do the work and understand the material. The others don't have a clue or are freeloading. Even if I do understand the material, jumping into a group and only doing a portion of the work sets me up to not know part of the material very well come test time or when I'm working at a real job.
For large projects, working in groups makes sense in a couple of cases. First, the project is cool, but too large for one person to complete in the alloted time is a prime candidate for a group project. Second, group projects teache the group members to deal with the frustration of working on software with another person. (yeah, cheap shot...)
But please, oh please, don't make it a policy to allow group projects at every level. You'd think we didn't already have a job market saturated with poorly trained CS people.
Personally, I am sorry to see that Ga Tech has decided to back off on its no-collaboration policy for the classes in question. The policy was only in place for "Intro to Programming"-type classes, where learning basic programming techniques, not group-environment programming skills, is the purpose of the course.
Can you imagine the hell of being given a group assignment in a higher level class where half of the members can't remember how to create a for() loop or use pointers, because they "collaborated" that part of their work in the intro classes?
It looks like GaTech will now be offering different Intro-level courses for computer-related majors and non-computer-related majors, but it looks like they won't be enforcing the "no-collaboration" rules even for the comp-sci major classes. With a shift in emphasis to quizzes and tests, rather than actual coding, I can only see this as working to lower the quality level of students' programming skills.
CC-licensed translations of Japanese fiction: http://tonygonz.blogspot.com/
Maybe now some young Computer Science student can spend more time on developing a good overall program, instead of spending a bunch of time writing simple things like their own sorting routine.
You're missing the point of a CS education. Students are supposed to learn how to write "simple" things like sorting routines. (Of course, if you think sorting is simple, you're either inexperienced or you're extremely well versed in computer science.) That's why professors always tell you to not use libraries for assignments you're supposed to do yourself. Otherwise, we're talking about short typing exercises. The more simple algorithms you write, the more experience you gain for making the jump from higher complexity algorithms to code.
Students should not be swapping code on assignments. That's called cheating. These kids need to write as much code as possible, even if it's been done before.
Why bother.
> E3 coverage continues, at Gamespy (some cool reviews), Gamegal (good photos) and other sites beginning with "Game."
;-)
So, am I the only one who misses all the obligatory pictures we used to get from these conventions of 3dfx's latest Lara Croft booth babe?
They may have fallen behind in the video card market before their demise, but they sure had *showmanship*. Teenage girls in skimpy fantasy-wear and video cards the size of...
Chasing Amy
(We all chase Amy...)
"The more corrupt the state, the more numerous the laws"-Tacitus
I dont care if its Gamegal or Gamespy!
If she managed to shoot a pic of the "Harry Potter" booth, she damn well could have managed to snag a pic of the DoomIII booth! Fscking chic!
Yeah! I could go to some other website, but still..
Rapid Nirvana
Treason never prospers,
What's the reason?
For if it doth prosper,
None dare call it treason!
I've forgotten who said it and I'm too tired to go search for it.
You cannot apply a technological solution to a sociological problem. (Edwards' Law)
My E3 photos. Note, there are more than just E3 photos there, but some 60+ of the show.
People think Microsoft is the answer. Microsoft is just the question, "No" is the answer.
Strange, that a similar statement for, say, an English course would be that now English students can more time plagiarizing Shakespeare, and less time learning the difference between "your" and "you're". The thing about a CS course, is that it is meant to give you a good theoretical underpinning to your coding skills. Knowing there is a black box called sort you can use is worthless. Knowing how the black box works is not.
Learning to use others work is a useful skill, sure, but unless you have something in your own brain that you can put into the process, you will never get anything but rehashes of previous work out. Maybe that's ok for an English student looking to write middle-of-the-road sitcoms, or a CS student who is going to churn out the same web application for the rest of their lives.
Summary, there are some things you need to learn for yourself. It's no good knowing that a calculator can add, if you don't know what addition is.
not_cub
q='echo "q=$s$q$s;s=$b$s;b=$b$b;$q"';s=\';b=\\;echo "q=$s$q$s;s=$b$s;b=$b$b;$q"
No, CS courses are not about learing to write sorting algorythms, compilers, operating systems, or any code at all. Computer Science courses should focus on the theory topics such as "this is how you determine the time it would take for the algorythm to complete" (since this is a very limited, deliberate subset of the Turing Halting question, it is possible). Most students won't understand the theory they know how to write code and put the theory to use writing (or analyzing) code-- but that shouldn't affect the course's focus.
You're confusing discrete mathematics with computer science. CS involves code to apply the algorithms. The algorithms are abstract. Most of them cannot be used in anything except theoretical constructs (such as those that appear on a computer system). You can learn everything about computer science without any computer technology, but you will be unable to apply it to anything meaningful. As a result, I the ability to code is critical to acheiving a useful computer science education. Just the same as physics students apply the theory of numbers to real world objects, computer science students apply the theory of algorithms to operating systems, compilers, and what have you.
So if when you talk of teaching "how to write... sorting routines" you mean teaching "this is the optimal sort algorythm" not "this is how you determine the efficiency", you are the one who've missed the point of CS education.
This doesn't oppose my argument. Code is extremely useful for determining algorithmic complexity. Obviously not required, except for a niche. If you don't know about code, you won't know where to use which algorithms. On top of that, if you can't code them, you probably do not understand them.
As for your statement that I do not know the point of CS... I can only respond by asking you what the point is. Computer science is a broad field. It includes everything from linguistic analysis to software development. It's not specific and myopic as you're making it seem.
Why bother.
The limit wasn't changed because of speeding.
In the 1970s the Federal Government mandated a 55 MPH speed limit, over the disagreement of the Western States (Arizona, New Mexico, Montana, South Dakota, North Dakota, Wyoming, Texas, Utah, Nevada.) The States sued in Federal Court and won. Then the Feds decided to withhold Federal Highway funds. The states moved to 55.
Then in the early 80s, the states also went to Federal Court over the 21 year old drinking age, and won that they had the right to have it at whatever age they decided, then the Feds threatened to withhold Federal Highway funds.
In the middle of the Reagan administration, thanks to a mostly Republican States-rights movement of Senators and Reps from the West, the Federal Governement allowed states to first raise the speed limits on two and four lane state highways, then on the Federal Interstate system.
I once passed a Montana State Trooper on I-90 doing 105 in my Beretta GTZ and he waved.
IMO, speed limits, drinking ages, and blood alcohol levels for DUIs should be a state issue, not a Federal issue.
No two ways about it. If you insert a solution into your homework that did not come from your own mind, then a) you haven't learned anything b) it will just bite you in the ass later on an exam where (I assume) you are still not allowed to look at your neighbor's paper.
This is ridiculous. At my school the honor code is shoved down our throat. And I mean shoved. Every class I take the professor takes a whole class meeting to tell everyone that even if you 'see' someones code, for a minute, you are cheating.
Look, I don't like it when one guy does the work and everyone else cheats either. Guess what though, that's how the world works. I'm sorry, but we pay these professors quite a good deal, and a lot of them give me little to no direction, and I am going to talk with my classmates as much as possible. No, I am not going to *copy* their code, but I see absolutely no problem with cooperation.
And then there is groups, this is perfect. Yeah, there is always some guy who does all the work. I say too bad for that guy, because if he let's everyone do the work now, he'll be doing all the work for the rest of his life. Go talk to the teacher and tell him/her what's going on, then leave it up to the teacher to mediate. THIS IS HOW THE REAL WORLD WORKS.
I'm one of those 'too old to be in college' guys, so I've seen both sides. In the real world, you almost always work in groups. You have to learn how to manage them, and how to be a part of them.
Let student's cooperate and enforce more tracking on who is contributing how much, moderated by the teacher.
-- I have fans? Wow.
Advanced Micro Devices has slashed prices of its desktop and mobile Athlon processors just days after a similar move by rival Intel.
I purchased my new AMD XP 2000 just a week ago, and now the price drops. It seems like every time you buy any computer part, the next week prices go down.
given the choice of Intel's price cut to 262$ on a 2.2 gigahertz Xeon processor, or AMD's linked price cut to 224 on its 2100+. I can honestly say that i feel that Intel has far better value for the cash right now.
:)
the competition however is great, we all stand to benefit from it
We had a similar sitution in regards to my physics course. The homework was turned in over the internet. This lead to several groups forming. Each group, however, ended up with several people doing the work and the rest 'collaborating'. Out of the 40 people the class started with, 12 remained at the end of the semester and only half of those students passed. It seems as though the students who were 'collaborating' didn't fare all that well on the IRL tests. Strangely, the one test conducted over the internet had an excellent pass rate ;-). Collaboration has great value in teaching students why it's not a good idea to screw yourself over by taking the easy road.
The best way to accelerate a windows box is at 9.8 meters per second square.
My old school deteriorated even further than that. Last year's "Student of the Year" who hardly ever attended classes, was caught using website cut and paste (without referencing) to supply almost all the material for his final report. Other students did the cut and paste thing all year. Some were caught.
Nobody was disciplined. None. It didn't make me feel all that much pride in my degree to know my school handed them out that indiscriminately.
My own pointless vanity vintage computing page
Of course they are. At some point, the CS student should learn the relationship between all this theory, and actual code.
Every CS class I took had a component of theory. Many also had a component of coding, with the objective being to see how the theory applies. You don't just talk about parsing theory, you write a compilier.
Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood
Maybe now some young Computer Science student can spend more time on developing a good overall program, instead of spending a bunch of time writing simple things like their own sorting routine."
Yeah! Last thing we need is fresh grads who actually know how stuff works. Give 'em to me thinking everything is a black box. Why, best thing is if they view the computer, the company, the government, the whole world and everything in it as a mysterious black box. That way Mr. Scorpio and I can fill their heads with our Mantra of Death(TM) while they lounge in their business hamocks. I sure hope this works better than our last endeavor.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
That's not what I meant to say, and I'm sorry if I was unclear.
"Computer Science", for me, is both the math to determine whether an algorythm runs in exponential, polynomial, or linear time, but also the theories that drive design (such as monolithic kernels versus modular). Coding is a vital skill in learning to accurately apply the CS theories to real applications, but it does not comprise all of CS; the merging of good coding practices and CS forms "Software Engineering". Teaching only coding, the lesson plan reads "sort a list like this", while a CS lesson is about many "optimal" sorting algorythms (binary trees, QuickSort, etc) and when one would be preferred over another. Software Engineering would have the students write a sorting algorythm and then apply the CS theories of efficiency to analyze them.
When I first read your initial post, it sounded to me like you were proposing teaching coding skills over CS or Software Engineering-- I read the statement "Students are supposed to learn how to write "simple" things..." as making coding skills the focus of the lesson, instead of being simply the application of basic theories to a concrete example.
Do you like Japanese imports?
yes xeon mobos do cost more, but that is because they are for the very high end systems in general, and support dual CPUs generally have 8 or more DIMM slots for massive RAM, etc.
:)
just saying that Intel is gettin pretty darn competitive in the high-end market. that $262 cpu was almost $500 a few days ago.
it's almost to the point where i could build a dual xeon box running at 2.2 GHz x 2 for under a grand, if i tossed my HDD and CD-RW in it from one of my old boxes... that is the kind of system i used to look at and drool wondering how long it would be till it would be within my reach financially.
gotta love price wars
I noticed others talking about the robots behind Sony, and haven't seen anyone post pictures. So here they are (along with some others):
http://www.eqteam.com/e3/
If you want to see something at higher rez, let me know...
I came across this one day and book marked it because it was worth it. It sort of pertains to the licensing story and to the college story:
l
http://www.gnu.org/philosophy/right-to-read.htm
T Money
World Domination with a plastic spoon since 1984
I'm proud to say that Wisconsin held out on the 21 drinking age for a while. Unfortunately, Minnesota and Illinois spoiled it for us -- their kids would drive across the border to get drunk and crash on the way home.
oooo Game theory.
Requirements:
1. If you take ballots, make sure the ballots are secret to avoid personal conflicts.
2. Make the system an accurate measurement of actual work.
3. The system should encourage more work, and less freeloading.
In any system, the person will be aware of any demerit or bonus due to participation grades. They will either see what the other group members got, or they will know that the project wasn't deserving of the grade they got.
Since they will be aware of the demerit or bonus, there is no way to implement this in groups of two people and still fulfill requrement 1. Groups of three are almost as bad.
The larger the group is, the better a system like this works.
Another alternative would be to use CVS and count lines of code checked in by a certain person. Of course this ignores possible planning/design work that one student did.
Yet another alternative would be to interview each member of the group, asking them what parts of the program do. This could be bluffed though.
I don't see an easy answer. Do you?
I've had enough abrasive sigs. Kittens are cute and fuzzy.
When the college says "Do not share code", they mean "Do not share code."
But how far do they want to take it? Can't use libc because you're "sharing code" written by the glibc authors? Can't use GCC because you're "sharing code" generated by the compiler?
Will I retire or break 10K?
The GPL requires that:
The original BSD license, in contrast, has an obnoxious clause requiring credit in any advertising, not just in the copyright notice and source code.
If a thing is not diminished by being shared, it is not rightly owned if it is only owned & not shared. S. Augustine
That's why professors always tell you to not use libraries for assignments you're supposed to do yourself.
Are the professors saying you're supposed to master talking directly to the kernel for I/O in every little program, without even using the abstraction of #include <iostream> (or its equivalents)? How far are the professors willing to take their "no libraries" stance?
Will I retire or break 10K?
Like GaTech, the UW-Madison CS department used to have strict "no collaboration" rules and policies for intro level CS courses. They had code analysis tools to find copied code among student's assignment submissions.
They found that cheating was so rampant, it could not be enforced. They got sick of dealing with it.
The new policy (which has been in effect for a couple years at least), is that collaboration is permitted, but assignments make up a smaller portion of your overall grade now.
Typically, there are ~3 big exams counting for ~90 % of your final grade. The last ~10% is made up of several assignments given throughout the semester (YMMV depending on the course).
The rationale is that if you collaborate in a bad way (to the extent that you're not learning the material), you will surely fail when exam time comes. Seemed fair to me, though I personally never collaborated (didn't know anyone, didn't trust others abilities, etc.)
Maybe now some young Computer Science student can spend more time on developing a good overall program, instead of spending a bunch of time writing simple things like their own sorting routine.
Yeah, that's great. Maybe now all the new CS grads will be so clueless that I'll have more job security.... </sarcasm>
Unfortunatly this isn't a perfect world, and if students stop learning the basics like writing a good sort routine it just means that in a few years I'm going to have some really obnoxious coworkers, and I'm going to end up doing a higher percentage of the work. Good job Georgia tech!
If you learn what you are supposed to from a computer science education, you learn that all the problems can be broken down to layers of simple soultions. Then you learn to understand the simple solutions. If you really understand the simple solutions, you already know how to put the parts to gether into a good whole. If you can't understand the simple stuff, what good are you? It's important to make CS students prove they understand the simple stuff. The rest isn't nearly as important.
The problem with this is, exams arent the best way to test someone's knowledge. By definition, they only test a limited subset of knowledge and skills, and they are usually timed, which imposes a different set of demands on the student. Assignments are much closer to a real-world evaluation.
If the school can't enforce no cheating, it is their own fault. All you have to do is start handing out Fs or kicking kids out, and things will change rather quickly.
Many people have lamented the fact that American students are educated in a school system that refers to students learning to cooperate on a task as "cheating".
Maybe we'll have one exception now.
Yeah, right.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
Don't worry the "new" price has reacehd the distribution market a few weeks ago. Look at this story from theInquirer for more details.
Help fight continental drift.
All you have to do is start handing out Fs or kicking kids out, and things will change rather quickly.
It's not quite that easy. There was a lot of red tape to go through each time. Also, even though students saw other students being caught, some assumed that they just cut-n-pasted code. They thought they could just change a few variable names and comments to get away with it, not realizing that the code analysis software would still detect the cheating. Over time, this could be alleviated perhaps. I'm just betting that it's much more difficult to enforce than we could imagine.
I agree with the general sentiment of your post, however. The exams certainly weren't a true test of knowledge. There were usually code-writing sections near the end of the exam that were worth a lot of points, though. My biggest complaint is that I believe some professors had difficulty coming up with good questions for the exams, resulting in an exam with questions that would be much more practically answered in the real world by looking it up.
Civil disobedience is when you break a rule, then proclaim, "I have broken this rule, and I have broken it to show you all how unfair the rule is! I dare you to punish me!" then wait for the world to notice. Then you make more noise.
Nope. That is not civil disobedience. See my response a little above in this thread for clarification on what constitutes civil disobedience. Publicility surely doesn't constitute civil disobedience (considering Thoreau, who coined the term more or less, was incredibly silent about his disobedience).
Your confusing civil disobedience with modern protest or Ghandi's satyagraha.
int func(int a);
func((b += 3, b));
What a great message!
If enough people break the rules, they'll change em!
Well - maybe they do this with some college rules. But laws generally don't get thrown away when nobody obeys them. They just tend to not get enforced.
The problem is that this puts you at a whim of policement, judges and anyone willing to sue on obscure laws. That's not how a judicial system should work.
im not talking big iron here my man, and a dual or quad xeon *is* high end to the average guy who is comparing prices with an athlon processor for a server or something. and thats what theese comparisons we were talkin about were in reference too.
:P
"high end" is always relative, nowadays you could be a UNICOS master working with old Cray's and say you work on a *high end* system and someone else could come along and laugh at you, but if you were comparing your system to many other systems you would be correct in your terminology.
it's all in the context of what you are comparing it too
Come on -- in order to be able to do the analysis, you need to understand how the algorithms work, and a great way to understand how they work is to implement them.
... any code at all", especially INTRO courses, is totally retarded.
I'm a real theory weenie myself, but to say that CS courses are "not about learning to write
Um, E3 => booth babes, so how can this possibly be off topic?
b abes/big/TwoInPinkBikinis.jpg
http://www.chickshardware.com/html/expo/ects2000/
(hey, you've got to spend karma somehow)
But on a typical undergraduate-level exam there will be no questions that involve truly original thinking. There *will* be questions that require the student to put together previously learned algorithm A with previously used data structure B to produce a little tool that does C. Being able to quickly identify the algorithms and data structures which solve a particular problem is the bread-and-butter of programming, and is pretty easy to test quickly. If a student is floundering at test time under such a simple load, they didn't do their homework :-)
But those tests are also not at all useful in the real world. Those sorts of tests are designed to certify that hundreds of kids know what Djikstra's algorithm is, but they are not designed to get at the heart of what CS is and how to apply what you know to do something original. I don't think we should just accept that an undergraduate program isn't challenging kids to think creatively. After all, that is what differentiates someone truly bright and original from the rest of the herd.