Slashdot Mirror

I can assure you our state is a little messed up sometimes. Catch this:

The state of South Carolina for years uses one-time funds for multiple years projects (not the brightest bulbs). This year the one time money didn't come in, and the state had a budget shortfall of $800M. The state decides to account for this shortfall they will cut funding to all state programs...except education...except colleges, because apparently colleges don't count as education. Tuition for instate residents at Clemson University just went up 40% this year to make up for the "we won't cut education, except for those rich colleges" decision. This is increasing ironic as last year Clemson University was named "Time Magazine's Public College of the Year" and this year we won a couple more awards. Apparently, in South Carolina, if you college wins a national award, you cut their funding. After all, we wouldn't want people to think South Carolina actually has GOOD schools! (I for the record do not mind the tuition increase. I personally support it as I feel the college had no choice. I fault the State, not the school.)

It does not suprise me in the slightest that South Carolina is having a little trouble figuring out what the law with regard to filtering should be. At least they made a decision here that tends more to the libertarian side.

you can argue by analogy I just don't think it is ever effective in the context of law.

I posted an article a few months ago about academic dishonesty at Clemson University. I can vouch from personal experience, all that it takes is a gratuitous "threat" and that pretty much takes care of any issues of dishonesty; being in the position of getting brought up on these charges is deterrent enough. Even as someone who was almost brought up on dishonesty charges, I do agree that there needs to be some way of catching those who always seem to slip though by "borrowing" work from others. I personally know many people who always wait until the last minute to do assignments and then try to bum stuff off of others; ironically these slackers have some of the highest GPAs in the major. Bottom line, cheating has always gone on and will continue, until it one day catches up with those who can't do anything themselves; these people eventually just end up cheating themselves.

I felt that the first book in the series, Red Mars, had more fruitful discussion of the role of space elevators in a changing economic system.

This page is a worthwhile discussion page on the issues raised in Red Mars, for those interested.

CDC built large mainframes in competition with IBM in the 60's. When they started, IBM didn't have anything near as powerful as the CDCs. So IBM announced the System 360, then tried to build it. The 360 hardware might have been on schedule, but the OS was several years late -- but still, many IBM customers waited for it instead of buying CDC. CDC filed an antitrust suit, claiming that selling vaporware was unfair competition. Long before the suit was settled, the full 360 line was out and delivering the promised power, and CDC was bankrupt.

This sounds like a mangled version of what I've heard was the story of a specific model in the System/360 line, the Model 91 (or maybe it was first called the Model 92).

The Model 91 (or 92) was announced as a number-crunching supercomputer model, perhaps before it was actually read; this note on another IBM supercomputer project says

In August, 1961, IBM started planning for two high-performance projects to exceed the capabilities of Stretch. The first project, called "Project X", was assigned to the IBM Poughkeepsie plant, and in 1963 the design became a member of the S/360 family. This machine was announced as the IBM System/360 Model 92 in 1964 and was delivered as the Model 91 (with core memory) in late 1967 and the Model 95 (with thin film memory) in early 1968. Project X had a goal of 10 to 30 times the performance of Stretch, with an initial cycle time goal of 50 nsec (the Model 91 shipped at a 60 nsec cycle time). The Model 91's floating-point unit is famous for executing instructions out-of-order, according to an algorithm devised by Robert Tomasulo.

This may have been a preemptive strike against the CDC 6600, which, according to this note, was announced in 1964 and shipped in 1965.

As far as I know:

1. IBM's announcement of the S/360 wasn't timed to get in the way of CDC, although the announcement of the S/360 model 92 (later to be the 91) may have been;
2. CDC wasn't bankrupted by that announcement - they eventually got out of the computer business, but that happened, as I remember, some time in the '90's.

if you are into Beowulf clusters, there is the Parallel Virtual File System. Basically it is something that allows you to configuration the drives from many machines into one large drive.

You can find added information here on other similar systems

Check out the Vinny the Vampire comic strip

The biggest problem in attracting well-qualified candidates is salary (or stipend). In my field, the going rate is about $18k for grad students and $25k for postdoctoral researchers. Not at all competitive with what industry can offer a talented programmer.

The only reason we can attract anyone at all with these stipends is that we're not hiring employees, we're training students and researchers. They're here for the education, not for the earnings. Or at least they should be.

It's unrealistic to expect someone who already has all the skills I need to be interested in working in my lab for peanuts.

By the way -- if anyone is interested, I'm currently looking for a postdoctoral researcher to work on a comp. chem. project involving object-oriented parallel processing in F90. Email me or see web page below for details.

-Steve Stuart
http://radar.ces.clemson.edu

OK, maybe these guys cheated, maybe they didn't, but there seems to be a pattern here--transfer out already!

I agree. Plus, if you look at the syllabus of the course ConcernedStudent is concerned about (which is posted here) the rules are clear:

All work on quizzes, tests, design assignments, and labs sit to be wholly your own. Possessing, using, providing, or exchanging improperly acquired written, verbal or electronic information will be considered a violation of the academic honor code. Violations will result in a grade of F for the semester.

Ask yourself, if the assignment was to write a web browser that does x, y and z and you brought in the Mozilla source code, would you expect a good grade? No.

The professor may be a stickler, but he did lay things out in the syllabus.

By the way, I'm an occasional teaching assistant. What you really want to do if you aren't absolutely certain about the right thing to do is ask the TA or professor. They'll be glad to help you out.

I actually did my master's research on a comparison of multi-user operating systems. http://www.cs.clemson.edu/~cfreeze/research.html . I had fun doing it, and was suprised.

This post sounds a lot like the reaction I get when I offer people Coke Cake. Yes, it's got Coke in it.

While it was a nice design, it was hardly the "most revolutionary CPU ever built". If you want something revolutionary, look at the IBM Stretch (7030), delivered in 1961. Most of the techniques used in today's microprocessors were pioneered in the 1960s. They are just a lot cheaper today.

This article is sort of silly. "In effect, trying to patent the instruction set itself" is a fairly vague notion; in fact, what Intel is doing is patenting some of their software techniques (expressed usually in small groups of instructions) for prefetching and control/data speculation. Right or wrong, this happens all the time. If some company has a nifty new caching algorithm, they will patent the idea; not the gatelist and implementation.

For example, if you could implement a IA-64 clone by (say) ignoring all prefetch instructions, and just fetching the data when it was needed (effectively turning the chk instructions into the actual loads, for those who are aware of this stuff - you could do it with binary translation). While this may not be a very good idea, it wouldn't infringe their prefetching patent, even if you used the same instruction mnemonics and produced a chip that could run the same binaries.

Personally, I think these patent are potentially disturbing, but put it in perspective with common practice. Read the back of Microprocessor Report sometimes; there are lists upon lists of patents being granted for techniques in exactly the same fashion as above.

As for the patents, I haven't read them, but I suspect that they'll have a tough time with them. IA-64 didn't spring out of nowhere, and a lot of the ideas that went into it follow a fairly predictable (no pun intended) path of development in academia and industry. A fairly stacatto summary of these paths can be found at Historical background for HP/Intel EPIC and IA-64 - if you don't already know something about computer architecture, don't expect to be illumined. The point is, Intel (or more accurately Idea or whatever the Intel/HP collab. is called) hasn't necessarily added that much to prior art here, so the patent may be too broad and subject to either legal attack, or too narrow and easily worked around.

Oh, and to the people cheering on the failure of IA-64, I beg to differ. Some of us write compilers and binary optimizers and code generators, and the death of the x86 architecture would make us very, very happy. The fact that the first IA-64 is going to be a dog isn't really that suprising - it's a huge engineering task and the first chip was always going to be a reference chip more than a serious performance model.

turn each node of your cluster into an I/O node and make use of both extra disk space and parallel I/O. It's quick, fast, and good.

And of course we have the inevitable flames. Every single Katz article seems to end up looking like this sooner or later.

I attend Clemson University in South Carolina and my freshman year, a friend of mine created a page which does the exact same thing. The site, http://hubcap.clemson.edu/PSP/ryp.htm is located on the Phi Sigma Pi web site. A national Honors fraternity.

I believe it was a Dr. Li in the Math department (MTHSC) who threatened for several months to sue the web-site creator and the fraternity for libel. The school paper picked up the story and there was a big stink for a couple of months.

Now the lawsuit's basis was that the former web master (it is now automated via Perl) was attempting to remove profanity and total flamage from some of the posts. Dr. Li threatened to sue because of the fact that the former webmaster edited some of the submissions... just not enough b/c there was still profanity and the like. In the web masters defense, there were way too many submission for only one person and he had a hard time keeping up w/ both the site and his classes.

The KICKER here is that I was with this former web master when the idea presented itself. The presenter was another college professor who was meeting with students from the Honors College. He/she was one of the best professors I have had to date so its obvious why he/she pushed this concept. As a matter of fact, he/she remains one of the top ranked teachers on the site.

He/she was also a wonderful connection to have because any conversations between the faculty were forwarded to the former web master and friends and eventually to the web master's lawyer.

So, after several months of B.S., the buffon finally got over the whole liberal thing and the web site was transformed into an automated message board w/ a little math functionality to compute rankings. Go BigBlue!!!

I attend Clemson University in South Carolina and my freshman year, a friend of mine created a page which does the exact same thing. The site, http://hubcap.clemson.edu/PSP/ryp.htm is located on the Phi Sigma Pi web site. A national Honors fraternity.

I believe it was a Dr. Li in the Math department (MTHSC) who threatened for several months to sue the web-site creator and the fraternity for libel. The school paper picked up the story and there was a big stink for a couple of months.

Now the lawsuit's basis was that the former web master (it is now automated via Perl) was attempting to remove profanity and total flamage from some of the posts. Dr. Li threatened to sue because of the fact that the former webmaster edited some of the submissions... just not enough b/c there was still profanity and the like. In the web masters defense, there were way too many submission for only one person and he had a hard time keeping up w/ both the site and his classes.

The KICKER here is that I was with this former web master when the idea presented itself. The presenter was another college professor who was meeting with students from the Honors College. He/she was one of the best professors I have had to date so its obvious why he/she pushed this concept. As a matter of fact, he/she remains one of the top ranked teachers on the site.

He/she was also a wonderful connection to have because any conversations between the faculty were forwarded to the former web master and friends and eventually to the web master's lawyer.

So, after several months of B.S., the buffon finally got over the whole liberal thing and the web site was transformed into an automated message board w/ a little math functionality to compute rankings. Go BigBlue!!!

To the heart of the story, the wait is to have a committee of students, faculty, and DCIT staff determine how to integrate this technology into the University's existing long distance service. Access to the sites will most probably be restored, but the impact to the various services at the University must be examined. Hardware may need to be upgraded and prices changed. Right now the switchboard is self sufficient, remove long distance revenues and it must be made up somewhere or disconnected.

To the heart of the story, the wait is to have a committee of students, faculty, and DCIT staff determine how to integrate this technology into the University's existing long distance service. Access to the sites will most probably be restored, but the impact to the various services at the University must be examined. Hardware may need to be upgraded and prices changed. Right now the switchboard is self sufficient, remove long distance revenues and it must be made up somewhere or disconnected.

Warning: This post is LONG.
As a former student of Clemson, I'd like to share a few unhappy facts about the university's allocation of resources.

Fact #1: DCIT is the Micro$oft of Clemson.
They have traditionally been so clueless that the bigger colleges formed their own computing support infrastructures. A good example is the College of Engineering and Science. They have their own support infrastructure known as Engineering Computer Operations. They have all their own routers, computers, etc... All because DCIT couldn't deliver on competence and facilities in the past. To be fair, DCIT has improved service over the years-Cisco everywhere, for example. Campus-to-campus connections are now nice and fast.

Fact #2: Clemson's administration == clueless and deaf
It is the administration and leadership (Duckenfield included) that live life without a clue.
Just some background, and information, so that everyone has a little history. I arrived at clemson in 92 as an engineering major. I discovered these cool Sun workstations on a network NOT managed by DCIT that I could use to do gopher, and even some web browsing.
I think that sometimes, you could go to a DCIT lab, and the dot matrix printers there might work.
In 97 or so, they started introducing "resnet", which was ethernet in the dorms. It was only available in the two most expensive dorms (20-40% > other dorms), and cost $40/semester. The service was lousy but hey, it was a start. In 98, DCIT completed wiring the other dorms on campus, and elimintated the $40 setup fee. Suddenly there were around 5000-6000 new nodes on the campus network, full-time, surfing, ftp'ing mp3's, etc.

Clemson only had three T1's at that point. You do the math.
You were lucky to get 800 *bytes per second* between 7am-3am. Imagine downloading Solaris patch clusters with that handicap.

Also in early 98, DCIT proposed (and the administration approved) a mandatory $50 "technology fee," to be paid yearly by all students in the interest of improving campus computing resources. Multiply that by 17000 students... $850000/year income from this alone!

Sounds like a great qualifier for funding more bandwidth, right? Wrong.

Everyone from the students to the deans fought tooth and nail to have the pipe upgraded. And in TWO YEARS (think internet time...) it never happened. "Not enough funds." (in the background, CHA-CHING CHA-CHING $850000/year).
Many resnet users switched to faster 33.6k and 56k modems, using local ISP's, just to get better performance than on-campus ethernet.

When I left in may.99, they still only had three T1's. I understand they upgraded recently, not sure to what since I'm not there any more. In the meantime, DCIT has been blissfully upgrading all lab PC's every 6 months. I guess we know where the $850000 goes. Terrible management of resources!

Well I guess I'll end my rant. I'd go ahead and post about Clemson putting the firewall in between the dorms and the internet backwards (protecting the internet from the dorms!) and required all off-campus traffic to be authenticated agains their NDS tree--but I'm sure someone else will do it. If there's enough demand and I don't see it appear I'll post it later.

Thanks,
DragonWyatt

'Fraid I don't have time to go into detail today, but fortunately there's plenty of resources available: Several books, of which my favorite for the beginner is "Four Days With Dr. Deming : A Strategy for Modern Methods of Management (Engineering Process Improvement Series)", and websites, you can start with the W. Edwards Deming Institute Web Site. Also try Deming's Fourteen Points and his Seven Deadly Diseases. Don't be discouraged that you've seen some of these before in failed methods -- there've been many attempts to create a more management-palatable "Deming Lite", with disapointing results.

One last scrap: RE: STUDENT REQUEST: Deming's 14 points applied to a software engineering organisation.

If you're not interested in using NFS behind a firewall (just because it's slow, insecure, ugly, unreliable, fault-intolerant, and buggy doesn't make it bad, right?) you might be interested in PVFS. It sounds like what you are looking for, and under the "Files" link are the sources.

Personally, I like the "adventure" of Coda, but haven't tried setting it up in a few months. Now that my roommates have agreed to be guinea pigs for the Windows client, I figure I'll set it up behind my NAT box and play with it again. It's overkill for everything but a big installation, but I still think it's kind of fun. The thought that terrifies me is working with a multi-GB datafile or such over Coda -- but since my roommates will probably be more interested in playing Dopewars and moving around small files on a FE network, I'm going ahead with the grand master plan anyways. Besides, I have a laser printer and a burning desire to experience the frustrations of Samba...