i would just like to share with you guys that I....

[sakky]

<p>

I just don’t see it as a big issue. I don’t see gross inequities or massive quantities of students left out of their desired majors. The balance I’ve seen is relatively good.

The stats for those people who tried to get into the L&S CS program are published. You will see that a rather significant percentage of people who tried to get into CS were denied, including some people who had technical and overall GPA’s of up to 3.5, were denied from the major.<br>
<a href=“http://www.eecs.berkeley.edu/Peer/lowerdivision/admissionsstatistics_data.html#sp02_tech[/url]”>http://www.eecs.berkeley.edu/Peer/lowerdivision/admissionsstatistics_data.html#sp02_tech&lt;/a&gt;
Seriously, how would you feel if you got both a technical and overall 3.5 GPA, and still couldn’t get into the major you want? What’s up with that? </p>

<p>

I don’t know what the variances are, personally. I would imagine 10% would be within range. For a major with 500 people, that gives you a total spread of 100 students. Do you budget for 500 or 550? Would you rather have extra or not enough? It’s a fair question, certainly, and someone has to decide. In a purely pragmatic sense, 500 should be the choice, but then you’ve got a potentially impacted major. You’ve optimized and you’ve still got impacted majors. What now? If you were selling bananas, you’re either going to have to take some home to let them rot each day, or you’re going to have to say sorry to some of your customers (okay, in this case you could just eat the bananas yourself). Depending on the cost of producing and selling a banana, one situation may be more profitable than another.</p>

<p>Further, historically, majors can and do fluctuate a lot. Take a look at EECS over the past decade. The dotcom bubble affected EECS enrollment hugely. As an example, CS61A was up to 400+ students at the peak, and nowadays is around 100. While actual EECS major numbers didn’t vary by 400%, you can imagine that it was at least significant enough that if changes weren’t made, accomodating that many students would be impossible. That means either hiring more faculty to teach these courses, or kick some students out of the major

</p>

<p>Obviously there is some buffer zone. Nobody knows EXACTLY what the demand is going to be. That is why you have to build in some slack capacity. </p>

<p>The question is, how much, and this, again, gets back to a well-understood topic of operations management. Depending on the service levels you want, you budget a certain buffer for each queue. But that doesn’t mean that you just create huge buffers willy-nilly in one station while other stations are starved. Good operations management means properly apportioning capcaity buffers.<br>
That’s what those lecturers are for. If you suddenly get a huge influx of demand, you hire a bunch of lecturers. EECS (just like any department) has plenty of post-docs, some of whom wouldn’t mind working as a lecturer for a few semesters to make extra cash and build a teaching record. If demand disappears, you just don’t offer any lecturer contracts that semester, and those postdocs can just go back to being regular postdocs. Or, again, if CS demand ramps up, you get some of the Math and Physics profs to teach some of your classes. If demand dies down, then you can just have your regularly scheduled EECS profs teaching those classes again. The point is, there are ways for you to dynamically ramp capacity up and down.</p>

<p>

Otherwise, it would be iffy to take from Math to give to EECS if Math needed what it had. If you’re taking from Math research funding to pay for lackeys to teach lower level EECS, you’re basically taking from the top minds so that the students who were deemed unqualified for EECS can now take courses in EECS, which doesn’t seem optimal at all.

</p>

<p>This has nothing to do with “research”. I don’t know why people keep bringing up research. I am not talking about research. I am simply talking about having the teaching capacity necessary to handle the interested students. I never said anything about taking any research money from one department to another. Teaching more students doesn’t require an expenditure of research money. Those math profs who would be teaching CS classes are still free to do all the math research they want.<br>
Besides, how much money do you really need to do math research? Let’s face it. Most math research consists of proofs, and while proofs are obviously very difficult, they don’t require a lot of money to do. Some math journals, a whole bunch of paper and pencils, and a whole bunch of coffee, and that’s all you really have to pay for. Paul Erdos once remarked that a mathematician is a machine that converts coffee into theorems. The only thing I can really see is if a math prof is doing computational/algorithmic proofs that require large supercomputers. But then, if that’s what you’re doing, that basically makes you, in effect, a CS prof, which means you have even less excuse not to be teaching CS classes. </p>

<p>

Not necessarily. There is something bad about using unexperienced lecturers, and most untenured lecturers aren’t as experienced as tenured lecturers. That’s not to say there aren’t many bad tenured lecturers and many good untenured lecturers, but if you’re going to pick between someone with 2 years of teaching experience and someone with 10 years (without any more information), I think you’d probably pick the 10-year veteran

</p>

<p>Yeah, but the 10-year veteran in what? Keep in mind that plenty of profs don’t exactly have a lot of teaching experience either. Who is going to be a better teacher - a newly hired EECS assistant prof, or Mike Clancy, a lecturer who has been teaching for at least 20 years? </p>

<p>Let’s face it. Right now, Berkeley has plenty of courses taught by inexperienced assistant profs, and taught poorly. So if that is happening already, is it really any worse to use inexperienced lecturers (i.e. those postdocs who get called up when demand increases unexpectedly)?</p>