Celestia Forums

Finished! Minerals/Industries has been released. Note: For the tipology of add-on, "Minerals" name must be considered generic. See "Limitations" below.

The database on which the add-on is built is here:
http://tin.er.usgs.gov/mrds/find-mrds-intl.php

That database has been robbed country by country and commodity by commodity. Since the world's gold/silver databases are so large that the server doesn't allow the download, the U.S. have been checked even State by State.
The datasets have been rendered "weighted up", that is, commodities haven't been "taken and put" as they are, because LOTs of sites would have shared the first, second and third commodity causing unreadableness.
For instance, often the location for unique lead producer is the same for unique zinc, and their overlapping would causes the impossibility to read which they are; of course in this case the locations have been gatered as Pb-Zn. Another one: as first commodity is easy to find lead and zinc altogether as well as zinc and lead; in this case one of the two has been omitted (i.e, is hold Pb-Zn instead of Zn-Pb). A far difficult case is that in which the same site share more commodities. Here the adopted criterium is both the connective "-" that the relevance of commodity as well; in particulary whether the commodity belong at the same of thats closest to itself: i.e, Ni, Cu, Co shares the same site and are sourrounded by lots of Pb's in which one of these Pb share the Ni, Cu, Co's locations, the Pb is neglected while is taken Ni-Cu-Co (yes I know, it's a bit arbitrary but is readable).
Thus, the DBFs datasets have been meticulously processed (with a Perl script) and the results dealt with in the ways showed below - for each country its commodities (except occurrences, that are gathered apart). In this manner it might also be more easy accounting for up-to-dating the data.
-------------------------------------------------------------------------------------------------------------------------------------------------
The symbol conventions are:

Au/Ag (yellow) = precious elements;
PGE (cyan) = Platinum Group Elements, also referrings as PGM, Platinum Group Metals (Pt/Pd/Os/Ir/Rh/Ru);
U (magenta) = fissionables;
REE (maroon) = Rare Earths Elements (La/Ce/Pr/Nd/Pm/Sm/Eu/Gd/Tb/Dy/Ho/Er/Tm/Yb/Lu);
C (gray) = graphite;

Ax (white) = amethyst;
Dx (white) = diamond;
Ex (white) = emerald;
Jx (white) = jade;
Sx (white) = sapphire;
Rx (white) = ruby;
Tq (white) = turquoise;
Tx (white) = topaz;

gems (white) = other precious/semiprecious gemstone;
Qz (white) = quarz;

* (yellow) = Au/Ag occurrences;
* (cyan) = Pt/PGE occurrences;
* (magenta) = fissionables occurrences;
* (maroon) = REE occurrences;
* (white) = gems occurrences;
* (gray) = graphite occurrences;

Al = producer, mine (all types), first produced commodity;
Al-Mn = producer, mine (all types), either first or first-second produced commodity;
Al-Mn-Fe = producer, mine (all types), either first or first-second-third produced commodity;
Al<Se> = producer, mine (all types), first-<n-commodity>, very secondary produced commodity;
<Se> = producer, mine (all types), very secondary produced commodity;

(Al) = producer, mine/plant, plant (all types), first produced commodity;
(Al-Mn) = producer, mine/plant, plant (all types), either first or first-second produced commodity;
(Al-Mn-Fe) = producer, mine/plant, plant (all types), either first or first-second-third produced commodity;
(Al<Se>) = producer, mine/plant, plant (all types), first-<n-commodity>, very secondary produced commodity;
[<Se>] = producer, mine/plant, plant (all types), very secondary produced commodity;

[Al] = plant, processing plant only (all types), first processed commodity;
[Al-Mn] = plant, processing plant only (all types), either first or first-second processed commodity;
[Al-Mn-Fe] = plant, processing plant only (all types), either first or first-second-third processed commodity;
[Al<Se>] = plant, processing plant only (all types), first-<n-commodity>, very secondary processed commodity;
[<Se>] = plant, processing plant only (all types), very secondary processed commodity;

--------------------------------------------------------------------------------------------------------------------------------------------------
Advantages:
- Fast and smooth, no performances reduction.

Limitations:
- Periodic table fashioned, no compounds;
- Many commodities are absents (no enough symbols: stone(?), crushed stone(?), limestone(?) cement(?), etc.)
- No prospect sites;
- No past-productions;
- Several countries lacks of actual data, just either past-productions or occurrences;
- Cote d'Ivoire seem "censored"; a threatening warning appear when one try to download the dataset;

Standard usage: just locations: "RU" (rupes);
LUATOOLS users: selectable add-on; use your own text/images. For use with LUATOOLS ver. before the 1.2 beta 7, please remove the ending "s" from the locationtypes directive within minerals.lua file. It must be locationtype


http://marauder.webng.com/files/Minerals.zip (300 Kb)

Massimo,

I think this is a very useful add-on, in principle. Yet I see an incredible "overpopulation" of green location names, which should not happen for small sizes of our globe. A long time ago I have derived an algorithm that avoids overcrowded location labels at ANY zoom level perfectly. The bigger the globe gets, the more locations will be appearing...

Here is a detailed example of my algorithm for the Moon locations with illustrations!
viewtopic.php?f=2&t=8561

There is also a PDF output of my extensively commented Maple derivation. There you should easily be able to find out what to do.
http://www.celestiaproject.net/~t00fri/images/moon_weights.pdf

To avoid overcrowded labels at any zoom level, one needs to exploit the so-called "importance weights" that need to be derived according to some simple criteria...

Fridger

t00fri wrote:Massimo,

I think this is a very useful add-on, in principle. Yet I see an incredible "overpopulation" of green location names, which should not happen for small sizes of our globe. A long time ago I have derived an algorithm that avoids overcrowded location labels at ANY zoom level perfectly. The bigger the globe gets, the more locations will be appearing...
.
.
.
Fridger

Well, indeed...
First, thanks for the links because now I know more about the Importance settings and how it works (in the 2005 I wasn't joined).
Second, I'm a bit perplex whether is best to adopt either an Importance values based upon the square km of the countries, the index rating of its human development (economy) or upon the quantity of the single commodities. For instance, Sudan and Mongolia are big contries but with relatively few produced commodities; from distance they would shows its commodities without to be "truly" important (in the sense of large productions). Troughtout the databases the quantity of productions are specified, but the sites with large or medium productions' outputs are very (relatively) few. Several contries ought its big productions as sumns of small productions due to closest sites. A way to do the thing could be that of account for the number of a specific commodity: i.e. in the U.S. there are LOTs of Au, Ag and U. From distance some of these Au, Ag, U can be showed. Its "filtering" would be due to the complessive numers of Au, U rispectively; to say, if they are 100 each one, then 10 will be display for small size of the globe, and so on. Thus your algorithm would be fitted to the commodity number for each country. Can this reasoning be mathematically sustained within your "Maple enquires"? I think yes, but of course I'm a bit far from be a "Gausslike man" Any suggestions here are very appreciated.
Third, in the PDF is made a reference to a "Perl script": is this script too available?

Massimo,

you should best have a look into my respective derivations for Earth and then judge for yourself what is the best variable upon which to base the calculation of importance weights. Starting already in 2003, I did a LOT of work with this stuff. Unfortunately, sorting all material out again is somewhat time consuming...

Here is another thread of mine from 2003 refering to some relevant issue in the context of city population:
viewtopic.php?f=2&t=3932&p=27731

Altogether, I did derivations of importance weights for Earth, Moon, Mercury, Venus, Mars... These are also implemented in the official distribution. In most cases you can find the respective Maple files or PDF/html printouts.

Here is the stuff for earth, for example:

http://www.celestiaproject.net/~t00fri/earth.html

If you scroll that html file half-way down, there is a specially enhanced paragraph containing the detailed reasoning and formulae for the importance weights!

In case of un-inhabited planets/moons, I needed and used of course another variable on which the importance weights were based. E.g. crater size! Which is the best ,depends on the particular problem and is best decided by you.

Yes, I still have most Perl scripts. But these are so numerous that you best get specifically "involved" first in that problem and then tell me which one you want.

Good luck,
Fridger

You want rescue my mathematical concerns... The last time that I've dealt with math systematically Led Zeppelin were together...

P.S.
I'm computing, I'm computing, I'm computing...

Fenerit wrote:You want rescue my mathematical concerns... The last time that I've dealt with math systematically Led Zeppelin were together...

P.S.
I'm computing, I'm computing, I'm computing...

YEEEES! Actually here is a respective quote of a shatters user from 4 years ago:

buggs_moran wrote:Very cool Fridger!! However, you are destroying my math student's beliefs that math is not used in the real world and especially not for leisurely endeavours... Indeed man, have you lost your mind.

Fridger