What I am trying to learn.

By: Janus | Date: December 8, 2013 | Categories: Uncategorized

It sounds like a simple statement, but it is not really. Learning is hard for me unless I can break a ting down into its simplest pieces.

I am trying to learn many things, some hard, others relatively easy. I have a hard time using online guides for the same reason I had trouble in school, my brain works differently. To me, and I am fully aware the limitation is mostly mine, almost all tutorials start out fine, then begin using circular definitions. I get frustrated because they are all based on memorizing stuff until it sticks, and they assume it will stick quickly. For me that doesn’t work. I lack the memorize and regurgitate gene, I actually have to learn a thing in order to retain it. Once I know what it is, once I have a clear mental picture, once I Grok it, it is just there, always and forever. Failing that however, the words fade in a self referential haze it is gone in a few days at best, and I have to start completely over.

For me, I need to have a thing defined. Telling me that star positions are based on a particular moment in time is great, it gives me a starting point. I know that RA & Dec give me a point on a sphere that was directly underneath the star with those coordinates relative to the center of the earth at that time. I even know that the equinoxes, (Spring and autumn), are when the earths tilted equator lines up with the suns equator, which is defined as earths orbital plane, so that is when measurements are taken relative to for consistency.

What starts giving me fits is eastward at the vernal equinox. (Vernal? come on, just say northern hemisphere spring equinox and leave the latin out of it). North is easy, select and mark or highlight Polaris and you have north. What I don’t get is east, where exactly is it pointing. How exactly does this relate to the vernal equinox point.  Is there an eastern star, like the north star but pointing to the east? This is part of what I am trying to figure out.

I am trying to make a 3d ‘T’ diagram with North, South and East marked from earth at J2000.0 as self help guide for visualizing where things are. I also intend to include a ring demarcating the equator. Once I get that made I will be putting it up as a starting point for other people like me. I will be able to see where RA starts, and what the equator is that DEC is above or below.

First among the things I am trying to learn is astronomy obviously, which is something celestia excels at, but has almost all of its knowledge locked up in circular logic cryptography despite the source. It all seems to depend on keeping a bunch of relative things with no hard definitions and no starting point in your mind all at the same time. Once again I know the limitation is mostly mine.

Telling me to do a corrective Eigen space normalization and rotation of the positional matrix for celestial to equtorial conversion is just word based cryptography to me. On the other hand telling to rotate the coordinates around the , RA, Dec, xy, xz or yz axis leaving it alone but adjusting the others is exact. Shorthand obscures when learning, so I am striving to be exact, even if long winded.

I am also expanding what I already know about programming. Mostly I have always worked on small projects, normally cleaning up after a collaborative effort has failed, folded, or to fix a specific problem. I have not worked on a large scale project because I see the world differently than those around me. Normally I break things in a controlled manner to see how something really works compared to how it is documented. I then take difference between what I expect and what I get to refine my next round of tweaks.

I am doing the same thing here in order to learn large projects and astronomy. As stated already though, I am not an astronomer. I find the night sky fascinating. I find stars fascinating. I enjoy playing with the structure of things, taking them apart mentally to see what makes them tick.

With the last couple of SVN versions here I have made a few small internal alterations to the stars database in order to better understand how celestia works. The purpose of this is to help me envision where the stars in the sky really are.

By where they are I mean in relation to earth. I already know the earths orbit, its orbital plain, is tilted compared to the MilkyWay’s own plain, and the earth’s equator is tilted in relationship to the its orbit around the sun. I began investigating this by locating Polaris, the north star, to give me an idea of where it and we are in relation to the MilkyWay itself.

When I tweaked Celestia to show me where it was in relation to use by displaying its coordinates it felt great. I was able to tweak the binary database to include the original RA, Dec & Dist, and the routine for reading it as well. Though it required learning enough perl to tweak the script.

Once I had that done I further altered Celestia to show the xyz, and that gave me a surprise. My carefully calculated positions of things were off, way off. So I began investigating. Using the default celestia as a starting point, assuming it was accurate, good enough for NASA, good enough for me. Me being off meant I had something to learn, and I am, slowly.

So regular users and real astronomers may see some positional glitches while I work through my learning phase. All changes are commented and dated, so all can be reversed easily. I found a different set of formulas in the createstar routine than was used in buildstardb.pl, so I am investigating that first.

It is highly likely that I will find something simple, probably a single step I missed, that gave me erroneous results. However that is how I learn. I break things in a controlled manner and watch it happen. Once I understand the logic, I will be able to do what I need to later without issue. What I learning however is the why, not the what. Once I know the why, making it part of the what is automatic, just like breathing.

So using Polaris: RA->37.946 Dec 89.264 Dist 432.58 I calculated

x:   4.381  y: 432.546   z:     -3.416  in light years but I got

x:   4.381  y: 395.494   z: -175.191 instead.

Once I understand it, I will put it all back and a guide up for those like me who need things put plainer without resorting to area specific terminology that feels like linguistic cryptography.

Janus.