|Re: RPN for Kids: Why it's WRONG !|
Message #24 Posted by Bruce Bergman on 23 Aug 2007, 1:33 p.m.,
in response to message #17 by Valentin Albillo
Harsh or not, my assessment is an accurate one [...]
Your opinion it might be, but I disagree that your assessment is accurate.
For efficiency reasons, the teachers need to standardize and due to a number of important factors, not the least of which is the fact that all mathematical texts use standard algebraic notation, they chose algebraic calcs. This is a fact, and going against it is comparable to driving by the wrong lane, you're going to crash and burn.
Instructors (and schools) clearly need, and benefit from, standardization. There is a huge value in that, both in instruction and in merely saving money. However, textbooks are not as "standard" as you think.
In fact, the recent trend in higher-ed math/stat/finance textbooks is to offer "how do I do this with my calc". They list a couple of different ways of accomplishing examples using 2-3 calculators as the model of choice. There's no reason why HP couldn't aggressively and actively court the textbook publishers to include HP calcs using RPN as an option. Mostly, the publishers and authors just want someone to write that for them. They resist putting HP calc explanations in there because (1) up until recently, HP shunned the educational market, publishing included, (2) they don't have the time or incentive to write each example themselves, and (3) they don't know the technique of RPN. BUT! If you offer to write it for them, they're all for it. I contacted an author of a statistics text for higher-ed last year and discussed it with him, and he was quite open to me submitting an HP-specific approach to the examples. The publisher appeared to be open as well, although I never followed up on it (um, I got busy elsewhere, unfortunately).
Going against the flow of traffic isn't a BAD thing if you are talking of concepts. Use of algebraic isn't a fact of nature, or of law. We're talking about concepts and ideas here. If everyone went with the flow, we would have ONLY Microsoft software in the world, now wouldn't we? Instead, we have Linux, or replacements for Microsoft Office. We have Firefox instead of MSIE. We have the Mac instead of Windows XP.
o There's also the fact that, due to market pressure and other factors, the number of RPN models in the market is very small, a negligible minority in the calculator market to say the least. Essentially, only one company still makes them, and there are very few models to choose from, namely the HP-12C, HP33S, and HP35s, if we stick to classic RPN, all of them subject to some serious limitations such as an absolute lack of I/O, for instance. So if you're used to classic RPN or want to introduce it to your children, your options are few and, mostly, relatively expensive compared with the algebraic offer.
I agree that the numbers are small, and choices aren't the best. But, it CAN change, as we are seeing with the HP-35. In addition, I notice that you conveniently excluded the HP-50g/49/39 families, which have extensive I/O functionality. And compare favorably to the upper-end TI calcs with applets.
Did you even read the Texas Math Challenge posts earlier this week? Those kids go out and buy used calcs from eBay and craigslist, and use those when the "current" models aren't what they want. A lot of them use the 33s and 50g models. Yes, there aren't as many HP calcs as Casio or TI, but if you want to use RPN, you can certainly find a wealth of options.
There's also the distinct possibility that the only company making them decides that they're not worth the effort, their sales aren't enough to justify the costs, and so the powers-that-be might as well decide to write them off the market, thus completely terminating their actual, precarious availability for good.
It's a risk, I admit. At one time in the past, I thought HP did give up on the market. I hope they never do in my lifetime. I choose, however, not to scare myself into using one platform simply because of the possibility of something happening.
o It's also a proven fact that RPN tends to generate what could be termed as 'addiction', in the sense that once some people are exposed to it, they'll get hooked to the point where they won't feel comfortable at all having to use any normal, non-RPN calculator, and would be highly inefficient using them, if at all, and would do whatever it takes to avoid using them, as they simply can't cope with algebraic machines anymore, at least not without feeling extremely uncomfortable while doing it.
And this is a reason NOT to use RPN? C'mon, you're reaching... A "proven fact"? I'd love to see the clinical research study you are referring to. Could you please send it to me?
o Combining all of these factors, you can see that by introducing your children to RPN, you risk they getting hooked on a paradigm which doesn't resemble what they see printed in their books, doesn't coincide with the procedures and keystrokes taught in class, and forces them to either adapt and convert on the fly what's being taught to their RPN ways, or else burden the teacher with the task of dealing with them and their RPN way of doing things, thus generating inconvenience for everyone there: the teacher, the rest of the class, themselves.
I don't buy it. Even as much as I respect you and love your comments, challenges and articles, this one has me puzzled. It's such a short-sighted, narrow-minded attitude. I am really surprised that your opinion is as you outline it. Sure, you've got some good points here. Some of which I agree with. But I just don't feel the same way as you and I'm kind of surprised that someone as smart as you is so obviously dissing RPN. Choice is good! I'm not a over-the-top RPN fanatic, but I feel strongly that this "choice" needs to get right back into the thick of things in the education market.
Look at it this way: Darwin is right, in that the stronger will survive. The ultimate survivor of this question may be algebraic. But I am not willing to concede the fight this early on. I say we give it to the kids, and let them decide the outcome. I'm willing to trust what comes out over the long run. Give them the choice and let them decide with an educated decision.
You may say "we've already done that!" and yes, that's partially true. But it *has* survived, even after being beaten down and almost exterminated. And it seems to be getting a second chance. I don't give up on my beloved football team simply because they didn't do well the past five years. Give them the opportunity to try again.
I know of *many* people, friends of mine and colleagues alike, that suffer from the RPN bug: they simply *can't* use an algebraic calculator with any decent proficiency, to the point that they'd rather take their NIB collectible HP-15C out from the box and into the dusty construction site if necessary if contemplating the possibility of being forced to use any algebraic calc.
I know the same kinds of people. I think, however, if you dug deep enough into their head, you'll find it's not a function of "can't" use it, it's that they don't WANT to use it. Big difference. I don't WANT to use TI calcs, and I don't have any in my house. I would prefer to use HP calcs with RPN, and it does annoy me when I have to borrow an algebraic calc. But I can certainly use them, and proficiently. I just choose not to. There's a big difference.
If, tomorrow, someone took away RPN from every written word, and magically erased every RPN calc from the world, I somehow thing all your friends and mine would manage just fine. They might not like it, but they'd get along.
I see this sheer reliance on an obsolete paradigm which surely faces extinction as a dangerous and unnecessary niche adaptation, which can be no good in the long run for the person aflicted with it.
And sliding right down that slippery slope with you, I should give up on Linux, because it can't possibly replace Windows. And I should only learn one or two current programming languages because the others will be extinct in a few years. And I should stop sending emails now because the younger generation says that email is dead and text messaging is going to be it in the future.
Whoa. Pretty free-thinking opinion there...
o Also, there's the myth of RPN being perceived as "superior" to algebraic systems, offering this or that "advantage", which "absolutely compensates" and overcomes any alleged disadvantages.
Truly a myth. I don't tell my folks to use RPN because it is superior, but because it's a different way of doing something, and a way that they might find not only easier, but more helpful. I never say to them that it is superior, because that only breeds arrogance and resentment towards RPN.
As I've posted in several threads in years past, this is essentially a myth, mostly based in the fact [...]
But 35 years have passed by, and while classic RPN hasn't barely changed one iota, algebraic models have experienced authentic quantum leaps.
Classic algebraic is the same as it was 35 years ago too. The advances you speak of are -- in some cases -- much like that 4-deep stack. There are accelerators to help with expressions. You can see more on the screen, but that's not a change to algebraic.
What about the changes we've experienced in RPL, or in solver equations? Or that both classes of calcs have seen with CAS? Those "model" changes are as quantum as any of the TI calcs.
o Lastly, one of the usual arguments in favor of teaching RPN to kids (or of using RPN in general) is that it allows you a more intimate understanding of the expression you're evaluating: you can see intermediate results, you can check them on the fly, the fact that you must frequently pre-decide the order of evaluation makes you more aware of it all ...
I don't necessarily push these aspects either, but it's just as easy to come up with "usual arguments" as to why algebraic is better, and that doesn't make it any more right either. In fact, you were using many of them yourself. They are not necessarily wrong, but they are not necessarily right. Same for algebraic. I don't see these as big factors in the discussion anyhow.
In reality, this is akin to making virtue of necessity. It's not that you can do that with RPN, it's that RPN forces you to do things that way: you must see intermediate results, whether you want or need them or not, you must spend time and effort to pre-decide the order of evaluation whether you want it or not, lest you risk losing items out of the top of the stack, ....,
Algebraic "forces" you to use parenthesis, doesn't it? Algebraic "forces" you to ignore intermediate results, doesn't it? For almost every necessity you can say RPN requires, algebraic has one too.
This is similar to the arguments rised in favor of doing calculations with slide rule when the first electronic calculators arrived: many insisted that doing them with the slide rule required a more thorough understanding of the problem, what with re-scaling being constantly needed, and that forced you to think about your numbers and prevent mistakes or erroneously scaled results.
Agreed. I remember those very same discussions when the first calcs came out, and I agree that those discussions are similar to the whole "RPN helps you understand the math" arguments. I don't buy them any more than I bought the slide rule vs calc arguments back then. I had a slide rule and understood it as well as any other basic slide rule user, but I dumped it in a heartbeat for the calc. Not because it was archaic and handicapping, but because the calc was a more productive tool. I'd rather be more productive, getting more done, with more powerful tools, than to say that I "understand" the math more. The TI calcs aren't necessarily any more productive than the HP calcs. Sure, they do some things easier, but I can find a half-dozen things that the HP calcs do more productively.
The bottom line is: who cares about the calculations ? About whether intermediate results are seen or not ? About how we do our operations, whether algebraic and parentheses or RPN and stack ? The important thing is that the children understand the problem, understand the algorithms used to solve it, and can correctly write down the necessary expressions that get the results from the initial data.
I don't care about the calculations either. I DO care about choice. About giving kids the ability to push their envelopes, even if it isn't the norm. I'm not saying every kid should learn RPN, or that it should be the standard taught in school. I AM saying that it should be an option.
How these expressions are evaluated is absolutely irrelevant. It could be by hand, with an abacus, with a slide rule, with an RPN calculator, with an algebraic calculator, or by saying aloud "Computer: evaluate this, please" in a near future. Who cares !?
In all of those cases, I agree with you. In many others, I hear cynicism and...something else. Not sure what. Not pleasant though. I just hope and pray that our next generation of students don't have the same narrow-minded attitude that you appear to have.
Now THAT is harsh.