Gakken Newsletter #158

I wasn’t expecting it this time, but the Gakken newsletter showed up in my email a few hours after I posted the announcement of the Electromagnet speaker kit. And you can guess what the lead news is this time.

It starts out by explaining why “Plus” was added to the speaker kit’s name. Basically, the editors at Gakken want to include school students to “Adult Science” kits’ audiences. The subtitle for the kit is “let’s enjoy experiments”.

Also, the editors have extended the deadline for the USB camera “15 second video contest”.

01 – Special edition “Otona no Kagaku Magazine Plus” on sale today, Nov. 28, 2013
02 – Now Taking Entries for the USB Camera 15-second Video
03 – Have Fun In Your Own Home Kit, On Sale Soon
04 – Adult Science for Women – “The Science of Getting Beautiful Skin”


01 – Special edition “Otona no Kagaku Magazine Plus” on sale today, Nov. 28, 2013
This kit is specifically designed for adults and kids to have fun doing experiments. The base kit has wireless LEDs that change color and flash while the music plays on the speaker. It uses cutting edge technology to provide transmitted power to the LED units. You can use it in both speaker mode and experiment mode, since the LEDs can be removed from the base to play with.

On sale Nov. 28
64 page magazine
2,800 Yen

02 – Now Taking Entries for the USB Camera 15-second Video
Due to overwhelming request, the Gakken USB camera 15-second video contest deadline has been extended to Dec. 23rd.

03 – Have Fun In Your Own Home Kit, On Sale Soon
The next Petit Handmade Kit will be coming out soon, aimed at anyone that wants to make stuff at home but doesn’t already have the tools for it. This one seems to be based on the porcelain dishes designed and created by Naomi Ashida.

34 pages, A5 paper size
On Sale Dec. 17
1,900 Yen

04 – Adult Science for Women – “The Science of Getting Beautiful Skin”
Yeah, you could argue that this one is sexist, but this will probably be a best seller here in Japan.

On sale Jan. 9, 2014
108 pages
850 pages.
No URL for it yet.

Electromagnet Experiment Speaker

(Image taken from the Gakken site for review purposes only.)

Gakken pulled something of a fast one on me. Back on Nov. 3rd, they made a short announcement of a new kit on their Facebook page, but I didn’t feel like translating the text and the picture looked like something related to the USB special effects camera. So, I just assumed that they were still talking about videos made using the camera. However, the main Otona no Kagaku page has just announced a new kit – the Denji Jikken Supi-ka- (Electromagnetic Experiment Speaker). It’s marked as “Otona no Kagaku PLUS”, for people that like doing experiments, and it’s not part of the numbered kit line.

The kit consists of a plastic case, circuit board, battery holders, volume knobs, speakers, 3 regular tri-color “LED units”, and one LED unit with a coil in the base. Suggested assembly time is 60 minutes. The idea is that you can put all four LED units in the main case, and use that as a flashing external speaker for your MP3 player. Or, you can mount the LED units in other holders and use them to play with. One suggestion is to suspend the units by threads and take strobed photos of the light paths.

The kit does come with a thinner, 64-page mook, and has an article on Nikola Tesla. Official release date is Nov. 28, so I won’t be seeing it in Kyushu for another 2-3 days. Cover price is 2,940 yen ($30 USD).

No new information on the next numbered kit, the Auto-Writer doll.


Electric Bike

Just a short blog entry this week, as I’m in the process of putting together some video tutorials on synthesizers for beginners. If you want an idea of how to spiffify your bike, you could try to outdo this guy.

A second look at Kasoku Kids

(All rights belong to their owners. Images used here for review purposes only.)

Back around September, I saw the announcement on the Gakken Facebook page for the release of the first volume of the Kasoku Kids (Accelerator Kids) manga. I wrote a little about it in a short review. It’s an official comic sponsored by the Japanese KEK accelerator project, and is intended to inform Japanese school children of work currently ongoing at KEK, as well as to teach them the fundamentals of particle physics. I bought my copy right after it hit the shelves, with the intent of getting back up to speed on college-level physics as simple self-study. I figured that getting an illustrated Japanese comic would also help me with my Japanese studies. My plan was to just type up the dialog and narration into NJStar, and go from there.

(Screen cap of NJStar.)

Windows supports Japanese character sets, and uses IME for entry of both Japanese and Chinese kanji. However, this is only useful if you know the pronunciations of each kanji. If you don’t, you need to look them up in a separate dictionary based on the stroke counts and then just wade through all of the kanji with the same stroke counts and sub-components (usually referred to as “radicals” in English) for the one you want. It’s very time-consuming and occasionally frustrating. Then, when you know the pronunciation, you can type that into IME and wade through another bunch of kanji that have the same sounds to try to find the one you want again. There’s a reason why I don’t do that.

(NJStar, with the pop-up dictionary turned on.)

NJStar is the work of a professor living in Australia, designed to make Chinese, Korean and Japanese wordprocessing easier. It has several features I really like, including kanji look-up based on radicals or pronunciation, and a pop-up dictionary that gives short definitions of each word match it can make, starting with the longest-matchable word down to the individual starting kanji. This makes translation a lot easier if the document is in electronic text-readable form, such as when copied from a Word or PDF file.

(Trying to find a kanji given its radicals.)

The problem is when the document isn’t electronic text-readable, such as with a paper book, or a flat jpeg image. Then, I have to type it up first, and I hate having to do that. And, guess what – I bought Kasoku Kids as a paperback book, and the online version at the KEK site is flat jpeg images inside an HTML file running some kind of javascript. So, there’s LOTS of typing involved. Some of the more text-heavy chapters can take 3 days to type up 16 pages. But, and this is important, it is forcing me to memorize more kanji pronunciations to avoid having to look them up by radical all the time.

(Entering kanji by typing the romaji spelling.)

The science in Kasoku Kids, volume 1, is mostly overview – no real hard math, with the only exception being a proof of Einstein’s E=mc^2. But there is a lot of terminology, with discussions of the different types of quarks (up, down, top, bottom, strange, etc.), the existence of gluons, an introduction to quantum physics, particles that act like waves, and Schrodinger’s Cat. Some of this stuff I didn’t hear about until I was in university, and now we have a Japanese comic aimed at younger school kids. I don’t think this bodes well for America’s reputation as a leader in math and the sciences.

(The only real math-heavy page in the book.)

I’m not planning on doing a full translation or scanilation, and you can already see the original manga online. But, it’s still taking me weeks to get this far. I finally completed Part 2 (up to page 129) last week, and now I’ll take a break. I’ll tackle part 3 some time later, which will take me to the end at page 208. I was thinking that I’d like to sporadically do the chapters already online before they get the book treatment, but that may not be feasible. Volume 1 contains the first 17 chapters, plus a special on the Higgs-Boson. “Season One” has 5 parts (part 4 is chapters 18-23; 5 is chapters 24-30). Then there’s “Season 2”, which adds another 2 parts of 13 chapters total. My book has 17 chapters, and there’s already 26 chapters waiting to go into volumes 2 and 3. Sigh.

(An example of one of the discussion-heavy pages. This took me a couple hours to just type up the Japanese text part. Note that I’m taking a number of liberties with the translation to make it sound more natural. But, most of the explanation is true to the original text, which may make it “less scientific” than it should be in English.)

I think I’m going to use this as my excuse for why I haven’t done anything further with the Rockit synthesizer.

(This is as much math as we usually get in this book. Prof. Kobayashi is professor emeritus for KEK, and shared the 2008 Nobel prize in Physics with Toshihide Masakawa for their work on CP Violation in the 60’s.)

Actually, I’ve been in the middle of several large projects (all personal projects which don’t pay for any of my other hobbies) and I can only do one at a time. When I finish some paying-work projects, I’ll go back to the synths. I still need to troubleshoot the PAiA Fatman kit, learn how to hack the Rockit, and figure out how to make the Arduino MIDI shield work as a sequencer.

A cautionary tale

I bought the Gakken Wind-Powered Generator around Aug., 2009. At the time, my apartment in Tokyo didn’t have any good locations for taping it to a balcony, so the thing mainly just stayed on my bookshelf. Then, after I moved to Kyushu to get away from the Fukushima reactor meltdown, the wind generator stayed in a box in storage. Finally, a few weeks ago, I was able to get to that box, and I pulled the wind generator out, attached it to a soda bottle, and used electrical tape to mount it on the front balcony. Occasionally, the wind would get strong enough at night to move the propeller to the point where the LED would flicker on for a few seconds. It was nice having it set up, but the LED is only visible in low light, so it’s not that interesting to look at during the day.

Unfortunately, after about 3 weeks, a typhoon swept in through Kyushu, and we got near-gale-force winds. That definitely got the propeller spinning at a good rate. The next morning, the propeller was gone, blown off the generator spindle. Given the strength of the winds, it could have traveled blocks, maybe even up to a mile or until it slapped into an apartment high-rise. Regardless, I’ll probably never see the propeller again. What’s funny is that when I got the generator out of storage, the double-sided tape holding the fins to the spindle core had degraded and the fins were falling off. I retaped all the fins and made sure they were firmly stuck in place. If I hadn’t done that, I’d have lost a couple fins, but I’d still have the spindle core to tape new fins to (make the fins from soda bottle cut-outs).

Moral: If you live in an area with gale-force winds, make sure you put up your wind generator in such a way that you can take it down again quickly when you need to.

The poor thing looks so naked now…

Primes – Not quite set yet

I know, I know. I just can’t let it go. This is just the way I am.
Unfortunately, this time I don’t have any nice graphics to make all the text more agreeable.

Let’s take the numbers from 2 to 21, and then group them into the set of even numbers.

s1 = {2, 4, 6, 8, 10, 12, 14, 16, 18, 20}

And then do the same thing for every other number up to 10 (after 10, the sets only include 1 value).

s2 = { 3, 6, 9, 12, 15, 18, 21}
s3 = { 4, 8, 12, 16, 20}
s4 = { 5, 10, 15, 20}
s5 = { 6, 12, 18}
s6 = { 7, 14, 21}
s7 = { 8, 16}
s8 = { 9, 18}
s9 = {10, 20}

Now, if we look more closely at s3, we can see that it is made up of even numbers and is actually a subset of s1, as are s5, s7 and s9. We could simply say that s3, s5, s7 and s9 are all subsets of s1 and could therefore be eliminated from the discussion right now. Instead, let’s factor 2 out of the 3 sets.

s3 = 2 * { 2, 4, 6, 8, 10}
s5 = 2 * { 3, 6, 9}
s7 = 2 * { 4, 8}
s9 = 2 * { 5, 10}

If we were to extend the sets past 21 up to 1,000, it’d become more obvious that s5 is just a subset of s2, while s7 is a subset of s3, and s9 is a subset of s4.

s3 = 2 * s1
s5 = 2 * s2
s7 = 2 * s3 = 4 * s1
s9 = 2 * s4

Multiplying a set by a scalar doesn’t actually add new elements to the set. It’s kind of like multiplying a number by 1, or adding 0 to a number. In terms of set difference, (2 * s1) \ s1 = {} (the empty set). For my purposes, s1 and 2 * s1 are equivalent (this means that I really don’t know all that much about set theory) and now I’ll combine them into 1 set.

s3 = 2*s1 -> s1
s5 = 2*s2 -> s2
s7 = 4*s1 -> s1
s8 = 3*s2 -> s2
s9 = 2*s4 -> s4

Leaving s1(set of multiples of 2), s2(multiples of 3), s4(multiples of 5) and s6(multiples of 7). If we then write the sets as:

s1 = 2 * {1, 2, 3, 4…}
s2 = 3 * {1, 2, 3, 4…}
s4 = 5 * {1, 2, 3, 4…}
s6 = 7 * {1, 2, 3, 4…}

We’re just regrouping the numbers as they’re applied to the sieve algorithm. The prime numbers are the minimums of each set, which is defined as that value * 1.

Again, SO WHAT!?!

Why beat this poor dead horse any further?

Answer, I grew up as a city boy. I don’t like horses.

Actually, this takes me back to rectangles. It’s like the positive counting numbers are inherently unstable. They don’t like being in a unit rectangle state (1 by N, or N by 1 shapes). They want to collapse into smaller rectangles that minimize perimeter length to area. If perimeter = 2*W + 2*L, a rectangle of area 18 can be drawn as rectangles that are 1 x 18 (p = 40), 2 x 9 (p = 22) or 3 x 6 (p = 18), plus their rotations. Make a unit rectangle that is 1 x 18, it will want to curl up to be 3 x 6, like a snake coiling to stay warm.

A unit rectangle that is 1 x 7 can’t collapse to form a smaller rectangle. It’s unstable, and it’s only choice is to flip rotation, going from 1 x 7 to 7 x 1 and back, like a compass needle that can’t align N-S.

Unfortunately, the only way to tell if N, a positive integer greater than 1, is prime is to specify the value of N. In a way, it’s the same as Schrodinger’s cat. You can give probabilities that N is a member of any given prime set (s1, s2, s4, s6, s10, etc.), but you won’t know if it’s element 1 of a single individual set until you’re told what the value of N is and you can test for membership and if it’s the minimum of that set.

I wish I had a copy of Flash to work with. I’d love to make an animation where you have a 1 x 1 rectangle and add another 1 x 1 to it. The 1 x 2 would flip rotations unstably. Add another 1 x 1, and the resulting 1 x 3 would keep flipping. Add another 1 x 1 and the subsequent 1 x 4 would immediately curl up into a warm little 2 x 2 shape and sit that way forever, or at least until the next 1 x 1 is added and the 1 x 5 would go back to flipping again. Adding another 1 x 1 would let the poor thing coil up into a 2 x 3 and finally go to sleep.