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.
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.
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.
They haven't used that "like a book" license in a long time.
I don't know anyone who's used their products in a long time.
They haven't made money in a long time.
Coincidence?
If all this should have a reason, we would be the last to know.
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.
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.
I think it all depends on how you set up the group work...
When I did my course, the core programming/technical subjects were individual - these subjects give you the basic skills you will need to build on later...
More higher level subjects mixed in team work as well as individual tasks and this develops your ability to interact as a team...
We then did a full-year industry project (our course actually does two, bust the std IT course does one) which was totally team based...
The risk with team tasks is that you can coast on through if you want to. I believe that technical subjects should be individual tasks because this lays the foundation upon which everything else is built. Teamwork is good, but it should be brought in after the basics are covered...
I definitely agree that this dilutes the competency of grads coming into industry - we had one noteable case recently who began work where I do and it turned out he went through a 3 year IT degree with virtually zero programming skills. We didn't expect the earth - but we did expect the Basics - he didn't last long.
I'm sorry, but I feel that unis are just trying to fill spaces (to get cash) and create warm-fuzzy degrees that don't prepare students at a foundation level.
I think Enough of the warm-fuzzy because it damages the prospects of those grads that actually *did* the work because an employer goes "don't hire grads from X - they're all under-equipped" (over exaggerated, but I will say in the context of my work, it did damage the reputation of grads from that uni - people hiring are now much more cautious about taking grads from there...)
-- Dan,
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?
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.)
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.