PRIORITY - Brainstorming on Novien reports

[Jim Eshelman]

I forget where the Novien is going to appear - I think it's 0.6? - but thought I'd brainstorm a few notes. There are unique problems (not of great consequence, but definitely things to treat differently).

First, what are usable reports we might want? I want two things: Aspects within the Novien, and Novien-to-natal aspects. I don't know if the latter should be optional (turn it on or off), I'm comfy either way, some people might not want it, but I don't anticipate a loud outcry if it's there. It should have its own aspect options, of course. I'd default both tables to Class 1 major hard only (conjunctions, oppositions, squares within 3°), user configurable. (Natal to natal could be turned on/off the way mundane aspects are, by having a separate aspect table that can be left blank.)

NOTE 4/28/24: In a preliminary deployment (without a separate Novien options set), this can default to the natal aspect orbs, pulling Class 1 and 2 for Novien-to-Novien and Class 1 for Novien-to-natal.)

The Chart Options page won't need some features. People should still be able to turn on Eris, Sedna, and Node, but I don't see a need to include Vertex. I'm sure angles don't exist in harmonics charts. (If they did, that would mean angles take all sorts of aspects, which they don't.)

The background Chart Options settings are irrelevant. The concept doesn't exist.

Mundane aspects also are meaningless: that section can go (or replaced by a separate Novien-to-natal table)... but that's part of a larger discussion in the next post.

[Jim Eshelman]

Where "are" Novien positions. That is, in a 3-D world, we've always calculated their longitudes but never any other coordinate. Do they have latitude?

I think the answer is that they do not. They are all on the ecliptic. At the very least, that is the case within the range of current knowledge. There is no way that Navamsas or Noviens have been used that has to do with anything but longitude, and probably the 'harmonic' (to use the word for lack of a better one) operates along a specific plane. (All aspects work in a single plane.) I think all Novien positions should be defined as having 0°00' latitude.

This simplifies the first table enormously! There's no need for RA, Dec, Azi, Alt, PVL, or an angular measurement and Grounds marker. And, of course, you don't need to mention latitude either. (No mundane aspects. No lots of stuff.)

It would be really neat to have Speed, which is just 9x the instant speed given in the natal. We've never had this feature before (other than estimating it ourselves), and it could prove extremely useful.

If Asc, MC, and/or Vx are allowed and selected (which I discourage, but others may want), they should be listed in this table just like other planets.

I think the columns should be the planet, the Novien longitude, the Novien speed, and the natal longitude. This still leaves a lot of room in case other stuff can be moved up; I can't think of anything that Siderealists have genuinely asserted is relevant to the Novien that would flesh it out.

[Jim Eshelman]

At first I was thinking, oh, you only have three things to show - aspects in the Novien, aspects to the natal, and planet positions. Maybe one table with three columns will do it! -- But then I realized (duh!) this doesn't allow for more than one class of aspect, which is not a stretch at all. - Personally, I probably wouldn't bump it up to Class 2, but I might (just to have slightly larger orbs than usual for a flow-over. or to see Fagan's traditional pentade or 5° orb for them without having it intrude into my normal Novien orbs). The default probably should be Class 1 only (but we might decide to give it something else). Or, maybe, given the class distinctions, the default should be more generous. Brainstorming...

So, arranging columns could be complicated. In particular, there will be a LOT of extra room in the top table, even with the natal planet longitudes listed.

Oh, I hadn't even thought about the basics of the wheel. [NOTE 4/24: This is probably irrelevant now. We did away with the idea of a wheel (unless later discussion brings it back).] The PVL needs to go - it's meaningless. The bigger question is whether the Novien by default is a one-wheel or two-wheel chart. Personally, I like to look at it alone, then add natal planets for a deeper look, but Fagan (in the short time he was working on this) usually showed the natals as a routine feature. Maybe it's a routine two-wheel? (With the aspect lists, this isn't a burden, since ultimately we're only going to look at the aspects and, sometimes, the placements themselves.

The design, of course, should be equal ecliptical houses from Novien Moon at 1st cusp.

I don't see a need for the Cosmic State report. It's too limited with no angularity, meaningless sign positions, and few aspects - although one could argue that sign placements are at least important for determining dignity (or not). And LOL, don't get me going on midpoints! (The gist of my opinion is: Too much data makes things LESS clear, not more. At the very least, they should be turned off by default.)

I'll reread my notes from Fagan on the Novien, but I think that's all the input I have.

[Jim Eshelman]

My summary of Fagan's positions on the Novien are here:
viewtopic.php?f=61&t=142

In rereading these, I only found one feature that maybe should be in the report if there is a spare column. I'll candidly say that I think it's worthless, but, on the other hand, it is something that mattered much to Fagan during the year or two he was investigating the Novien before his death, and that deserves respect by default.

Here is the matter: Historically, Moon's Novien (traditionally Navamsa) sign-position was given great emphasis. Fagan considered it significant (it's much of what excited him about the Novien). After years of spinning in circles between Novien and Navamsa (where my Navamsa Moon's sign position fit me, but there's no way the Novien Moon sign fit me), what released me from the whirlpool was realizing that I really don't think sign positions matter in the Novien at all. This is also consistent with the whole body of sign studies that imply (by their construction) that the nature of a sign is homogenous from 0°00' to 30°00', and not subdivided in meaning by decanates, dwads, noviens, and the like.

Nonetheless - ignoring my opinion in the matter - Fagan thought the Novien sign position was important and, furthermore, that successive Noviens of Moon were important. (The Novien of the Novien. And then the Novien of that. And so forth.)

So... if you're looking for another column on the top table, it could be successive Noviens of Moon. This is hard (well: tedious) for most people to calculate and might be of interest. Since there has been so little real work done with the Noviens since Fagan's death in 1970 (over 50 years ago!), it makes sense to make it easy to have one of his often-repeated theories be testable from the report.

He also felt this was useful for rectification (as successive Novien Moons change signs). I thought it was a cool idea when I read about it in the early '70s and, in principle, he's right IF these Novien sign positions are, in fact, meaningful. If they aren't meaningful, then it's just something else to trip over. But, again, I lean toward letting Fagan's theories have their say and be open to testing.

Using Fagan's own natal Moon, here is where successive Noviens fall. (Believe me, the following is incredibly tedious to calculate with the calculator on my computer.) Let's presume we know Moon's accuracy to the nearest second (using TMSA).

8°20'27" Virgo - natal max error 0°00'00"
15°04'03" Cancer - 1st Novien max error 0°00'09"
15°36'27" Pisces - 2nd Novien max error 0°01'21"
20°28'03" Pisces - 3rd Novien max error 0°12'09"
4°12'27" Taurus - 4th Novien max error 1°49'21"
7°52'03" Gemini - 5th Novien max error 16°24'09"

NOTE 4/28/24: In practice we will not have a birth time more accurate than the nearest minute. On average, Moon moves 13°10'36"/day or, in one minute of time, 0°00'33". Keeping all the decimal places, this means that the 1st Novien could be 0°05' wrong, 2nd Novien 0°44' wrong, 3rd Novien 6.67° wrong, and 4th Novien 60°02'10" wrong. So the 4th Novien isn't practical EVEN if this theory is legit (which I doubt). It should cap at 3rd Novien.

So there you have it: I don't know how many Moon-signs somebody needs, but, by his own theory, there are Fagan's first six Moon-signs It seems the 4th Novien is the practical limit based on reliability of astronomical calculations.

By knowing the mannerisms, disposition, and idiosyncrasies of the native fairly intimately, an astute astrologer can speculate successfully as to the possible zodiacal constellation of the 1st, 2nd, or even the 3rd noviens of the Moon and without much ado obtain the correct longitude of the natal Moon and hence, the true time of birth.

FWIW, here are mine, which do not impress me. My Moon is a round minute (27°24'00' Aquarius) so I can go the easier route of Fagan's published table:

27°24' Aquarius - natal
6°36' Libra - 1st Novien
29°24' Pisces - 2nd Novien
24°36' Cancer - 3rd Novien
11°24' Gemini - 4th Novien

(Since TMSA is also powerful as a teaching tool, and interested newcomers will certainly use it that way, successive Noviens [if present] should be turned off in the Students options IMHO and, perhaps, in the Advanced form.)

[mikestar13]

I think I may actually do noviens somewhere in the process of 0.5 development. I'm unconvinced of the value of noviens of noviens, but this will likely be available as part of a generic novien mechanism. Single Wheel, Bi Wheel or both will be user selectable on a per chart basis. I'm wondering about having the novien be the cornerstone of a generic 0 Taurus Sidereal based harmonic mechanism for research. Haven't made up my mind yet about that last. The creation of the two modes of display will be simple: copy existing code and remove a lot of unnecessary parts.

[Jim Eshelman]

I shudder at the concept <g> but do we support people having choices where it doesn't take away from standard functionality we know we want? Yes, for other harmonics you'll have all the basic code in place with only one variable to change.

I look forward to being able to see what 10,000 asteroids on the 496th harmonic looks like. Perhaps it will finally be the answer to my unsolvable problems. <vbg>

[Jim Eshelman]

Geeking out a bit... digressing a bit from the purpose of this thread (but perhaps indirectly serving it)...

I'm persuaded of the Novien over the Navamsa because of indications that its positions are somehow objective in the same sense as natal planets. Transits to Novien Moon work (at least mildly); transits to Navamsa Moon do not. Interchanges of Novien Moon, Sun, Venus, and Mars to lovers' natal planets are stunning; those of Navamsa positions have a higher noise-to-signal ratio and ultimately are unpersuasive.

But there is one stunning mathematical property I miss about the Navamsa: It gave me a structural definition of the zodiac that I haven't found anywhere else, one that (for example) honors that "Taurus was the first sign everywhere" and yet identifies the crucial measurement points as 0° Rim!

Here is the interesting mathematical property: Even though a 9th harmonic wraps the entire zodiac around the zodiac nine times, it actually creates a structure of eight equidistant points where the 9Harm zodiac and the root zodiac align.

The fiducial of the ancient zodiac wasn't 0° Taurus, even if that's where the zodiac began. The fiducial was Aldebaran marking 15° Taurus. If we structure the 9th harmonic so that 15°00'00" Taurus the stable marker, we get two remarkable features:

(1) There are precisely eight points around the zodiac where the 9Harm longitude is identical with the root (1Harm) longitude.
(2) These occur at 15° Hub and 0° Rim!!!

I thought that was pretty {bonking} awesome.

But, trusting the data that eventually caused me to leap back to the Novien model, I think I have to file this {bonking} awesome result in the same folder as the solar apex.

[Jim Eshelman]

BTW, Mike, have you noticed that your Novien Venus is 7°58' Pisces, suggesting where in Sagittarius Terry's Moon might be? (For my spec time, the orb is 1°16'.) [2023 Jan 2 note: For the final actual time of her birth, 2:10 PM, Moon is 8°46' Sag. It's partile!]

It goes both ways, though. Venus moves 1°15' on her birth date, meaning Novien Venus moves 10°55'. For my spec time, her Novien Venus is 0°21' Aries, 1°57' from your Moon. [2023 Edit: For the actual time once obtained, her Novien Venus is 29°58' Pisces, in Mike's Sun-sign but 2°20' from conjunct his Moon.]

Both of these suggest that at least the general time of day (mid-afternoon) is right. - It's, at least, one theory on the matter.

A few other Novien-themed aspects:
Her N Sun op. your r Sun-Venus
Your N Sun to her r Mercury-Venus-Jupiter (minutes)
Her N Mars precisely conjunct her r Moon [2023 note: It's still partile with the real time]
Your N Mars co. her N Sun 13'
Her N Mars (and r Moon) square your N Venus-Pluto (~2°)

[mikestar13]

I shudder at the concept <g> but do support people having choices. Yes, you'll have all the basic code in place with only one variable to change.

I look forward to being able to see what 10,000 asteroids on the 496th harmonic looks like. Perhaps it will finally be the answer to my unsolvable problems. <vbg>

Can you share what you were smoking when you wrote that one? Even us programmers need to get high now and then. <vbg>

[mikestar13]

I'm very confident Terry's 3PM speculative birth time is a reasonably good one, and the noviens are another piece of the puzzle.

[Jim Eshelman]

I'm very confident Terry's 3PM speculative birth time is a reasonably good one, and the noviens are another piece of the puzzle.

I've added update notes to the post on this above - with the real time (within an hour of 3 PM), they show very well and serve as an example of one way Noviens can be useful in rectification.

[mikestar13]

I'm thinking for first pass noviens of natal/event charts. Other harmonics can be added later if there is demand. Single/bi-wheel with be user selectable per chart and the default will be saved in the chart option file. There will be an easier way to select single or bi-wheel for solunars as well (and single/bi-/tri-wheel for anlunars).

Question looking ahead to the transit module. Are any/all of the following valid?

1. Ecliptical (1H) transits to novien positions.
2. Novienic (9H) transits to natal positions.
3. Novienic transits to novien positions.

All would be easy to implement.

[mikestar13]

I'm thinking of adding a "default choice" or similar section to the chart files as well as a "parans" section". I'm thinking I may designing a different format for novien option files, with it's own default choice section.

[Jim Eshelman]

1. Yes, from what I've seen. I conclude yes.
2.-3. Nobody knows AFAIK. Never checked. Seems to fast for all but the slowest planets in the best situation. If TMSA were a research tool, it would be useful I suppose to one day test that.

[Jim Eshelman]

FWIW, I just ran the Novien of the chart for arguably the most important minute of my life (with transits precise to the minute). The Novien to my natal chart had only one contact, generically appropriate to any turning point, and nothing else. Novien-to-Novien had no partile conjunctions, oppositions, or squares other than a possibly/unclearly Novien Neptune transit to Novien Sun 45'. At best, I didn't need this transit to tell me anything about the event (and, if interested, could have seen it from a 0°04' 10° multiple).

[mikestar13]

Ok, first pass will be ecliptic transits to novien position only, other possibilities may be added later.

[Jim Eshelman]

Here's the proposed mock-up I promised. I think I got everything from the discussion above, and an added item to fill out the space.

Code: Select all

Pl  Nov. Long   Speed   Natal Long  
Mo   6Li35'54"  132°36' 27Aq24' 0"  Moon's Noviens
Su  22Sc 9'14"  + 8°54' 22Vi27'42"  1  6Li35'54" 
Me   6Cn 9'21"  + 6°44' 17Li21' 3"  2 29Pi23' 6"
Ve- 16Sc55' 7"  + 4°28'  1Sc52'48"  3 24Cn27'54"
Ma  20Ar18'02"  + 5°30' 28Sg55'21"   
Ju+  2Sg30'46"  + 1°01'  3Cn36'46"  Sun's Noviens
Sa  14Ge29'25"  + 1°02' 14Li56'37"  1 22Sc 9'14"  
Ur  29Sc59'33"  +11'33"  3Cn19'58"  2 19Ta23' 6"
Ne  12Aq 3'34"  +19'57"  1Li20'24"  3 24Li27'54"
Pl  18Le55'13"  +12'00"  2Le 6' 8"  
------------------------------------------------------------------------
  Class 1 Aspects        Class 2 Aspects       Novien-to-Natal
Mo sq Me  0°27'100%    Su co Ve  5°14' 46%     Mo sq rJu  2°59' 81%  
Ve sq Pl  2°00' 91%    Su sq Pl  3°14' 78%     Me co rJu  2°33' 86%  
Ju co Ur  2°31' 86%    Ve sq Ne  4°52' 52%     Me co rUr  2°50' 83%  
                                               Ma op rMe  2°57' 82%  
                                               Ur co rMo  2°36' 86%  
------------------------------------------------------------------------
Created by TMSA 1.0-pseudo  (10 Feb 2023)

[mikestar13]

That is an excellent format and I will use it with appropriate options, perhaps with slight modification. I'm assuming there will be a Novien Moon on first cusp equal house wheel, though personally I doubt I'd look at it, since it has no use for me. All I really want is the planetary positions and aspects, in contrast to a natal charts, solunars, or ingresses, where the wheel gives me a useful gestalt of angularity and mundane aspects before I delve into the planetary data and the aspectarian.

[Jim Eshelman]

Yes, Novien Moon as 1st cusp, with equal division. I agree, there aren't really houses, but that's the traditional display, trivial to calculate, and plays up the presumed lunar nature of the chart. - At least the "angles" aren't misleading, since they are major hard aspects to Novien Moon, which reads quite similarly.

I found myself yesterday tickled about the Speed column in this. It's not that I think I'll end up using it much, but it does clearly show things most people probably just would never think about. Some of these can be important. I'm cautiously curious about what this could do for rectification.

[Jim Eshelman]

That is an excellent format and I will use it with appropriate options, perhaps with slight modification. I'm assuming there will be a Novien Moon on first cusp equal house wheel, though personally I doubt I'd look at it, since it has no use for me.

Mike, perhaps this is a radical idea... and you may already be far enough in your development that this would be a big backtrack.... but I'll throw it out there anyway.

Does the Novien need a separate page - a separate chart? Or, since it is ALWAYS paired with a root chart, perhaps it makes more sense to make it an add-on option at the bottom of the root chart.

Nobody will ever calculate a Novien without having an underlying chart. But, if Novien is selected in the underlying chart's options, the tables could be added under the Cosmic State report. (You could always go back later and spin it off as a separate chart with a developed wheel if you were so moved, or if demand were high.)

Like you, on a Novien I probably will always cut right to the tables. Maybe they're all we need? (At least for 1.0?) - This also communicates that it's a subsidiary matter of the root chart, and may make things easier: If people want to always look at the Novien, they can just turn on the feature in the natal options without having to run two charts every time (or forgetting that they meant to always look at the Novien).

[mikestar13]

That's the answer, an optional add on to other charts. Will still need a separate options page as the novien will almost certainly use different aspects/orbs than the underlying chart, but this probably should be a sub form under chart options rather than a button on the start page. The listing of natal planets then becomes redundant, but of course novien to novien and novien to natal aspects would be retained. The positive/negative sign indication fits seamlessly.

This also reminds the beginner that the novien is not really a chart in its own right (though at times convenient to read as if it were) but rather an additional way of understanding the underlying chart. Most of your suggestions for TMSA have been quite good, but this one is positively brilliant. I will start working on it as soon as I finish redoing the aspects (adding parans) and the cosmic state report. And it's no problem calculating the novien of the outer wheel of a bi-wheel or tri -wheel chart, which should be a useful tool for delving into what significance noviens might have in solunars. I'm not optimistic, but willing to be surprised. Not really any extra work as the framework for charts is fully generic for one, two, or three wheel charts, the biggest internal improvement from 0.4.

[Jim Eshelman]

That's the answer, an optional add on to other charts. Will still need a separate options page as the novien will almost certainly use different aspects/orbs than the underlying chart, but this probably should be a sub form under chart options rather than a button on the start page.

Agreed.

The listing of natal planets then becomes redundant, but of course novien to novien and novien to natal aspects would be retained. The positive/negative sign indication fits seamlessly.

I wondered about that. My thought, though, is that retaining the (duplicated) natal list is the one place that Novien and natal will be viewable side by side (and, of course, there is a lot of space), so it's not only, "Oh, why not," but also usable.

This also reminds the beginner that the novien is not really a chart in its own right (though at times convenient to read as if it were) but rather an additional way of understanding the underlying chart.

Exactly.

Most of your suggestions for TMSA have been quite good, but this one is positively brilliant.

Simplicity seems that way sometimes . Most of the time, the best rewrite involves deleting words, not adding them.

And it's no problem calculating the novien of the outer wheel of a bi-wheel or tri -wheel chart, which should be a useful tool for delving into what significance noviens might have in solunars. I'm not optimistic, but willing to be surprised.

Excellent point and approach. - It seems to me that the toggle for "Novien on/off" should be separate in each chart type (Default_Natal, Default_Ingress, Default_Return).

[mikestar13]

I'm inclined to have the default start as off but allow it to be turned on with a button, and of course the user could have two option sets, say Default_Natal and Natal_With_Novien (I might pre-create that one, user editable of course). I already have a design in mind for the chart options page. Option files are simply short .json files, very economical in terms of both space and processing time. (Precisely why the .json format, developed originally for JavaScript, has spread so far beyond it.)

[Jim Eshelman]

A mildly radical suggestion: As I add more Options files, I find I forget what I've created because they aren't grouped together in a way that I see easily. I've started naming them with the chart type first instead of last; for example, I have Return_Outstanding and Return_Closer along with the basic Default_Return. I know decisions were made at version 0.1 when neither of us probably had any idea where this was going.

Might it make more sense to reverse these and make Ingress_Default, Natal_Default, Return_Default, and Natal_Student the format at this juncture? I think this will make life easier for people creating personalized special-purpose option sets.

[mikestar13]

Yes, and folks with already existing option files could automatically have the filename changed to put the word Default if it occurs lasts, and the words Natal, Event, Return, or Ingress placed first. A very simple operation in Python, and a radical change to the internals such as 1.0 will be is a good time to make the change. I'm also going to hard code the values that will be distributed in Natal_Default, Return_Default, and Ingress_Default, which will be used if no option files are found. I've developed a custom dictionary type for Python that will make this quite easy, also make it easy to supply default values for new settings such as parans and noviens as well as such new settings as will be added in subsequent versions, or to cope with most defective option files. I'm very open to breaking backward comparability in such matters if I can remove/greatly ease pain points for users. Well worth the programming time involved.

[mikestar13]

This is the new Natal_Default.opt option file with new options added for parans and noviens. This is hard coded for the values TMSA will use for natals if Natal_Default.opt is not found in TMSA\options or it lacks the particular entry:

{
"use_Eris": 1,
"use_Sedna": 0,
"use_Vertex": 0,
"Node": 0,
"show_aspects": 0,
"partile_nf": 0,
"angularity": {
"model": 0,
"no_bg": 0,
"major_angles": [
3.0,
7.0,
10.0
],
"minor_angles": [
1.0,
2.0,
3.0
]
},
"ecliptic_aspects": {
"0": [
3.0,
7.0,
10.0
],
"180": [
3.0,
7.0,
10.0
],
"90": [
3.0,
6.0,
7.5
],
"45": [
1.0,
2.0,
0
],
"120": [
3.0,
6.0,
7.5
],
"60": [
3.0,
6.0,
7.5
],
"30": [
0,
0,
0
]
},
"mundane_aspects": {
"0": [
3.0,
0,
0
],
"180": [
3.0,
0,
0
],
"90": [
3.0,
0,
0
],
"45": [
0,
0,
0
]
},
"midpoints": {
"0": 0,
"90": 0,
"45": 0,
"M": 0,
"is90": "d"
},
"parans": {
"use": 0,
"major": [0, 0, 0],
"minor": [0, 0, 0]
},
"noviens": {
"use": 0,
"lum_second", 0,
"0": [
0,
0,
0
],
"180": [
0,
0,
0
],
"90": [
0,
0,
0
],
"45": [
0,
0,
0
],
"120": [
0,
0,
0
],
"60": [
0,
0,
0
],
"30": [
0,
0,
0
]
},
}

This will easily be extended in future versions. Is anything 1.0 will need included and have I put anything superfluous? Note degree number keys are in quotes, Python doesn't handle non-string dictionary keys well. I will make similar modifications to Return_Default.opt and Ingress_Default.opt.

[Jim Eshelman]

Cool. I think I can make all of these out

A few thoughts:

"90": [
3.0,
6.0,
7.5

Mostly based on how the scores come out on the curve (and partly from just seeing a few too many weak aspects in the second column), I've switched this (and trine and sextile) to 3.0, 5.0, 7.5. It fits the curve better. Your choice, of course, whether to change, I just wanted you to know where I've been for a while.

"mundane_aspects"

FWIW, I always put in Class 2 - same orbs as ecliptical - figuring that if there is a mundane one closer than its ecliptical counterpart, I'd like to know.

"parans": {
"use": 0,
"major": [0, 0, 0],
"minor": [0, 0, 0]

I agree these should be turned off by default (Use = 0). However, do you want to put default orbs in? This would be different from other aspects since historically On/Off has been based on whether there is an orb given; but here you have a Use option separate from orb.

"noviens": {
"use": 0,
"lum_second", 0,

I'm curious what that last variable is.

Similar question on orbs: Since you have an On/Off variable, do you want to prepopulate default orbs (to be used in the event it gets turned on, to set a base standard), or still leave blank?

Is anything 1.0 will need included and have I put anything superfluous?

Not that's occurring to me. My head is deep into being a few paragraphs from finishing a LONG and demanding chapter of The Book, so it's not in variables at the moment. (Lol, just wrote a page on balancing attention to precise mathematical matters with other brain functions.)

[mikestar13]

I think I will use your recommendations for the class boundaries. BTW what class boundaries are you using for conjuction/opposition? The use variables allow us to include/exclude parans/noviens by a single test (the orbs will be ignored if use is 0, but it's easier to save them in the file in either case to avoid handling different formats. I intentionally make it difficult to not use ecliptical or mundane aspects (have to set every orb to 0 or blank) because I really want to discourage this. lum_second controls whether 2nd, 3rd etc. noviens of the luminaries are printed. (I might default this to 1 = "show them" unless user indicates (s)he doesn't want them.

[Jim Eshelman]

BTW what class boundaries are you using for conjuction/opposition?

I keep going back and forth between 3° and 4°. (I think the threshold is in between somewhere, maybe around 3.5 - I calculated it and forgot.) At the moment I have 4°. I don't like that, though, because if I use 3/7/10 then they're exactly the same as the major angle foreground boundaries and, when it comes down to it, those angle-contacts are probably the same concept as a conjunction. (Not that angles make aspects yada yada but it does look like the contact curves are the same.)

lum_second controls whether 2nd, 3rd etc. noviens of the luminaries are printed. (I might default this to 1 = "show them" unless user indicates (s)he doesn't want them.

No strong opinion, but moderate opinion FWIW is to leave it 0. I think it's a waste, but it was extremely important to Fagan and we haven't given it a chance. I'll turn it on for mine, but would rather people have to ask or find how to turn it on, rather than have people assume we're saying it's a real thing.

[mikestar13]

Good idea to have it off, the phrasing that appears on the option form will be more expressive than "lum_second". Re conjunctions/oppositions, I've been increasingly convinced that orbs of 3 degrees or less are the same strength as for squares, but the allowable orbs for classes 2 and 3 stretch further. Not hard to model mathematically, but I'm doubtful if I should.

[mikestar13]

BTW some theoretical considerations of class boundaries. For squares the formula is cos(12 * orb) which gives the class 2 lower bound of 3 degrees of a strength 83% or 5/6 while the class 3 lower bound of 5 degrees gives a strength of 50% or 1/2.

For conjunctions, the formula is cos(9 * orb), so 83% is at about 3.75 degrees and 50% at about 6 2/3 degrees, approximating the 4/7/10 you've been using but not exactly. (Both formulas from ISR.) Assuming aspects follow a pure sine curve. Angularity doesn't, it is a modified sine curve, so the boundaries need not be identical.

Using a formula of (10 * orb) for conjunctions would put the class boundaries at 3 1/3, 6, and 9, which if we put the cutoff for class 1 to 87% would give us 3 degrees for class 1. However this would make the class boundaries for a square 2.8, 5, and 7.5. So if we use pure sign curves, there will always be a mismatch of some kind.

If we hold that the class boundaries for conjunctions are 3, 7, 10 degrees, we must use the formula cos(12* orb) for orb <= 3 degrees and a modified sine curve (not worked out yet but not hard to derive) for orb > 3 degrees.

In my own work (not charts published on solunars), I use the same orbs for conjunction and square, which simplifies the matter considerably.

[Jim Eshelman]

I found my spreadsheet so I can speak a bit more on it. (I'm using percentages differently from you, as I know you'll see at once but others might not.)

Start with the trine-square-sextile curve which zeroes out at 15°00' orb. The 50% mark is 7°30', which I take as the operative cut off based on the idea that something is more than 50% likely to happen above that point (and less than 50% likely if wider than that). The next critical-seeming threshold is "50% above the line" (50% of the space we consider "active") or 75% on the whole curve. This occurs at precisely 5°00'. [50% of the scaled curve]

The 3° (or sometimes a little larger) threshold was discovered by Bradley's statistical studies and IMO confirmed in practice. This is about 90% of the whole (which also intuitively feels important). The exact 90% score is 3°09'. [80% of the scaled curve]

Functionally, we can call "partile" the (theoretical) "place of 100% certainty." If I take that as "rounds to 100%," or at least 99.5%, that's 0°41'. The 9.0% mark is 0°58'.

By this reasoning, using 99.5%, 90%, 75%, and 50%and putting everything in the nearest round degrees, Class 3 ends at 7.5° [exact], Class at 5°00' [exact], Class 1 at 3°09' [call it 3°], and partile at 0°41' [but let's call it 1°].


Applying the same logic to a curve with a base of 20° for conjunctions and oppositions, we get: 50% is 10°00' [exact], 75% is 6°40' [call it 7°], 90% is 4°05' [call it 4°], and 99.5% is 0°54' [call it 1°]. BTW, 99.0% is 1°17.

[mikestar13]

Yes, I looked up the formula I'm currently using and it's p = (cos(m * orb) - .5) * 2 (but not less than 0) where m = 60 / maxorb to expresses the numbers in terms of % above baseline (50% -- where it's a coin flip whether the aspect will manifest or not). This deviates by a maximum of 4% from the simpler formula and is within roundoff error for most of the range. I will continue to use the slightly more accurate % from this formula.

This strongly suggests that if maxorb is 10, the other class boundaries should be 4 and 7, and if empirical evidence shows that class 1 needs to be narrower, the class 2 boundary and maxorb need to shrink correspondingly. Again relying on the assumption that aspect strength is an unmodified sine curve.

[Jim Eshelman]

Yes, I looked up the formula I'm currently using and it's p = (cos(m * orb) - .5) * 2 (but not less than 0) where m = 60 / maxorb to expresses the numbers in terms of % above baseline (50% -- where it's a coin flip whether the aspect will manifest or not). This deviates by a maximum of 4% from the simpler formula and is within roundoff error for most of the range. I will continue to use the slightly more accurate % from this formula.

My brain's not geared to understand this on a quick read this morning. I'm not clear why this is slightly more accurate. (Nonetheless, I trust you know what you're doing )

This strongly suggests that if maxorb is 10, the other class boundaries should be 4 and 7, and if empirical evidence shows that class 1 needs to be narrower, the class 2 boundary and maxorb need to shrink correspondingly. Again relying on the assumption that aspect strength is an unmodified sine curve.

I agree with the argument. Plus, for conjunctions and oppositions, those are the orbs I'm using, though I'm aggravated it's a little different than angularity drop-offs are felt. (But on checking I see that the 90% threshold is indeed just past 4°, not close to 3.5° like I misremembered.) Conjunctions feel like a toss-up wherein the 3-4° range it goes, though the 7° and 10° marks seem exactly right.

Theoretically, I do think this curve is sinusoidal. I was more movable in this view before discovering that the five major (i.e., Ptolemaic) aspects are all defined precisely by thresholds in a sinusoidal curve, i.e., cosines equal to +1.0, +0.5, 0, -0.5, and -1.0 (the curve is halved and quartered exactly at these points). This leaves me feeling there is some innate connection of aspects to this curve that probably cascades in several directions.

[mikestar13]

Jim, I'm a bit confused.

The following is a listing of the values of cos(6 *x) for 12" increments (expressed as % rounded to nearest 1%):

0.0 100
0.2 100
0.4 100
0.6 100
0.8 100
1.0 99
1.2 99
1.4 99
1.6 99
1.8 98
2.0 98
2.2 97
2.4 97
2.6 96
2.8 96
3.0 95
3.2 94
3.4 94
3.6 93
3.8 92
4.0 91
4.2 90
4.4 90
4.6 89
4.8 88
5.0 87
5.2 86
5.4 84
5.6 83
5.8 82
6.0 81
6.2 80
6.4 78
6.6 77
6.8 76
7.0 74
7.2 73
7.4 71
7.6 70
7.8 68
8.0 67
8.2 65
8.4 64
8.6 62
8.8 60
9.0 59
9.2 57
9.4 55
9.6 54
9.8 52
10.0 50
10.2 48
10.4 46
10.6 44
10.8 43
11.0 41
11.2 39
11.4 37
11.6 35
11.8 33
12.0 31
12.2 29
12.4 27
12.6 25
12.8 23
13.0 21
13.2 19
13.4 17
13.6 15
13.8 13
14.0 10
14.2 8
14.4 6
14.6 4
14.8 2
15.0 0

This gives 99% for 1 degree, and 95% for 3 degrees, as discussed above, however 75% is at about 6.9 degrees (call it seven) and 50% is at 10%. Clearly classes 2 and 3 are too wide for squares and trines (indeed more appropriate for conjunctions and oppositions) so the question arises what exactly is the formula you are using to get 5 degrees as 75% and 7.5 degrees as 50% while maintaining the class 1 boundary at 3 degrees and 99% for partility?

[Jim Eshelman]

I'm not sure why we have constructed this differently. My 15° base curve uses (cos(12x)+1)/2. Here is the output for half-degree intervals.

0 0 100.0%
0 30 99.7%
1 0 98.9%
1 30 97.6%
2 0 95.7%
2 30 93.3%
3 0 90.5%
3 30 87.2%
4 0 83.5%
4 30 79.4%
5 0 75.0%
5 30 70.3%
6 0 65.5%
6 30 60.4%
7 0 55.2%
7 30 50.0%
8 0 44.8%
8 30 39.6%
9 0 34.5%
9 30 29.7%
10 0 25.0%
10 30 20.6%
11 0 16.5%
11 30 12.8%
12 0 9.5%
12 30 6.7%
13 0 4.3%
13 30 2.4%
14 0 1.1%
14 30 0.3%
15 0 0.0%

For the 20° base, it's: (cos(9x)+1)/2

[mikestar13]

Ah yes, the +1 /2 wrinkle explains it and seems more accurate than mine.

[Jim Eshelman]

Ah yes, the +1 /2 wrinkle explains it and seems more accurate than mine.

It seems that the correct way to map to the curve is to spread it along the whole range (in this case, multiplied so that it reaches the 30° from -15° to +15°). That produces a result between -1 and +1. To scale it, one adds +1 (making it run from 0 to +2), then divides by 2. Yes?

[mikestar13]

BTW your formula generalizes to (cos(mx)+1)/2, where m = 90 / maxorb (7.5 for squares, etc.), but scaling is done as 2(raw strength - .5), returning "no aspect" for negative scaled values, thus 50% for maxorb becomes 0%, etc.

Edited after rechecking.

[mikestar13]

I have decided to adopt your formula, as now is the perfect time to make the change.