Friday, July 29, 2011

One blog to rule them all

have rather a lot of projects.  To try to keep them all moving (although this one is still breezing along, although invisibly as I program stuff) I am starting (and blogging, I'm hooked) a central plan here.

Sunday, July 24, 2011

Back from vacation

Northern Virginia is a lovely place, but back to the stars.  I am resuming work on the interpolation object.  It will let me interpolate basic star data (remember the datum we need most is mass) from the Spectral and luminosity classes, and also interpolate spectral type from data like color index and luminosity.  And give any "made up" data a shot of verisimilitude.

The initial HYG dataset contains 111168 apparently single stars; of these I have complete spectral and luminosity data for 54611.  This is going to be a slog!

Thursday, July 7, 2011

A bit of a break

It's time for a bit of a vacation, which will be spent in sites more related to the ACW than far future wars.  In the meantime, software progress has been good.  I am working on writing as generic as possible an interpolator for star data.

Monday, July 4, 2011

More database progress

I've found a repository of multiple star catalogs


This includes catalogs of visual binaries which are going to quite close; and other fun stuff besides.  I am making good progress on techniques for working with the Hyg database initial load to track change and to get close to complete for as many stars as possible.  My next step with that data will be to nail as much as I can for the solitary objects, then see how I can integrate this binary data.

I've also been looking at how easy to parse the Bayer/Flamsteed name is.  If you look at some samples, it is just:
25    Psc
7Rho Cas
26    Psc
Gam2Oct
Eta Tuc
84Psi Peg

<0,1,or 2 digits of Flamsteed's Number><3 character abbreviated Greek letter name, left justified, may be spaces><1 digit distinguishing number, may be space><3 character abbreviated constellation name>.


I was thinking of parsing these out, but why bother?  You can search using like and get the right results, and the real fun is having that nifty Greek glyph output on a map.

Friday, July 1, 2011

Dog stars

Part of the fun working with data is discovering the "odd bits"

One of the odd bits in this database is Sirius.  We all know that Sirius is a binary, right?  However, the Hyg dataset lists it as A0m... -- and no, there is no distinct entry for its white dwarf companion.  I'm going to assume that "..." stands for an "n-ary" system and process accordingly.  Might as well assume the same for ".."

Also, I will have to list all the "famous stars" when I am done and double check (and patch) the results.  I am not too worried about the great mass of stars I might as well nail down the easy ones.

Monday, June 27, 2011

BlueMax blog - realistic spacecraft design

Have a look here.(Update: this link no longer has the original destination.  The designs are still available in places like RPGnow, the the Zan Qing link below)

This gentleman has thought through spacecraft design very deeply.  I have picked up one of his very reasonably priced designs on RPG now (I love instant gratification).  The Zan Qing is a very plausible orbital patrol craft; I am certain I will have direct use for it, and it is great inspiration as well.  I believe it is designed for D6 Space; but reality is not bound by a rules system.

Manually organizing the database

I am learning a lot from using SQL to interpret the strings in which spectral information is provided.  I have, in fact, interpreted spectral class (thats the OBAFGKM part) on more than three-quarters the stars, and luminosity class (the sd/Ia-VII part) on more than a third.

I am now asking myself if I want to take what I have learned and go back a couple of steps to clean up some details and to code the thing as a single program.  I think I do.  There is no deadline or business case standing in the way of a high-quality result and I will have to live with this for a while.  The thing to keep track of is that I need to get to system mass data so that I can plot transit routes between the interesting stars.

Thursday, June 23, 2011

A bit more progress on the database

I have managed my first trial load of the database.  It is certainly easier to assess the distribution of things with SQL available.  The load was simplistic so took quite a while, but query response is quite solid.

Unfortunately I have miss-coded the association of desingations with stars so a bit of re-work will be called for tomorrow.  The key next step for this are:
  • Spectral parsing
  • Star detail generation
  • The visualization tool.

Wednesday, June 22, 2011

FTL Limits and otherwise

One thing I have been trying to get my head around is a sensible distance limit to FTL travel; something that can be vaguely justified in the same way as the FTL rules around flat space and potential energy can be.

One thing I will try is not limiting the distance, buy limiting the precision.  In other words, a rule something like:  "You can jump to any star you want provided that there is no star closer to your start point and within 20 degrees of arc of your destination as seen from your start"  Angle would go up or down with risk, technology, prep quality, etc..  Once I have the database ready (getting close) I will see what this means for communications maps.

I have also been thinking of having the actual jump take zero time.  I want to have as few rules as possible for "jump space" and not spending any time there makes that simpler.  One of the implications is that one could have FTL ships (or drones) jumping back and forth between stars transmitting and receiving message traffic at both ends; a process that could make communications time to a distant colony as little as 24 hours assuming FTL transit points are 2-3 light-hours apart.

I'm not sure I see this as a problem.  It would not be cheap, it would be easy to cut off, and it does provide some justification for a "strategic player" getting some information from the frontier in a timely manner.

Sunday, June 19, 2011

Another one for your reading list.

Packing for Mars, by Mary Roach is a delightfully entertaining but still serious and accurate book about human factors and space flight.  I picked it up and finished it yesterday; hilarious but I learned a lot about social interactions in confined spaces, the need for gravity to preserve bone mass, the possible application of the hormones produced by hibernating black bears, and a dozen more points that I will have applications for; and I laughed out loud more than one while I was reading it.

Rethinking the timeline

I gave a set of future-history criteria in an early post.  What I have started thinking through is a contradiction between the requirement to have cultural continuity (so that current nations are recognizable) with a major ecological crisis including large-scale population die-off and large scale migrations, all with a 500 year horizon.

So, a slightly different progress:
  • Climate change, freshwater exhaustion and population rise are severe stresses 50 years from now.  The situation is serious but (just barely) manageable.
  • About 2070, an alien species that has migrated permanently away from planetary surfaces establishes a fuel processing center around Saturn.  Part of their strategy is to increase custom by contracting to locals for food supplies (still cheaper to grow on a planet) and to trade for technology that will get those customers into space and thus buying fuel and equipment.
  • The drive for space is to establish humanity on other worlds so all our eggs are not in one basket.  Colonists would be people with a wide range of objectives from political and religious idealists to those planning to become the privileged aristocracies of new kingdoms.  The bulk of the people eager to migrate, however, would be landless and displaced peasants of the developing world and displaced industrial workers of the rich world.
  • First colonies by 2100 (if not earlier).
  • Now add a few hundred years; but not too many.

Looking more at rockets

First, I dug out my spreadsheet of acceleration and potential energy that drove the rules for jump from earlier posts, and found that the radius I had set for Sol earlier was 1.5e15 meters (10 AUs) so just outside Saturn's orbit.  Lets use that, and the average chord of a sphere of that radius is (4/3)r or 2e12m. as the transit distance.

That gives our revised time vs acceleration table, with delta-v added:


g deta-v time (days)
0.1 2.73E+06                  32.74
0.2 4.00E+06                  23.15
0.3 4.90E+06                  18.90
0.4 5.66E+06                  16.37
0.5 6.32E+06                  14.64
0.6 6.93E+06                  13.36
0.7 7.48E+06                  12.37
0.8 8.00E+06                  11.57
0.9 8.49E+06                  10.91
1.0 8.94E+06                  10.35
2.0 1.26E+07                    7.32
3.1 1.55E+07                    5.98
4.1 1.79E+07                    5.18
5.1 2.00E+07                    4.63

Rolling out the Tsiolkovsky equation  we can work out the engine exhaust velocities we need for a range of mass ratios (that is, mass at the start of the transit over mass at the end)


m0/m1   0.1 g    1.0 g 
                 2 4.0E+06 1.3E+07
                 3 2.5E+06 8.1E+06
                 4 2.0E+06 6.5E+06
                 5 1.7E+06 5.6E+06

These are high. but not inconceivable, velocities.  Basically, we have to master some pretty fancy fusion technologies to make it work.  I'll have to do a detailed wok-through to see if I am on the right track, but so far the only outright impossibility I am talking here is still FTL.

Friday, June 17, 2011

Progress report

Last night, I completed the DDL for the tables to hold the basic HYG data.  Tonight I will proceed with the code to populate them.  That should give me enough information to start playing with visualization tools.

Thursday, June 16, 2011

Time and travel

Lets say for the sake of argument that the average distance to the FTL transit limit is equivalent to the orbit of Jupiter; we'll also assume that in each system we have to cross the system (that is, travel twice the Jupiter semi-major axis) to get to the right place to transit to the next star, and that we start and end that crossing at zero velocity.  Lots of assumptions there that the final software will do in detail; and we are ignoring trade stops and refueling.  Given all of that, here's a data table:


Time .01G .1G .2G .3G .8G 1G 2G 3G
sec 8E+06 2E+06 2E+06 1E+06 9E+05 8E+05 6E+05 5E+05
min 1E+05 4E+04 3E+04 2E+04 1E+04 1E+04 9E+03 8E+03
hour       2,192     693     490     347     245     219     155     127
"days"             91       29       20       14       10          9          6          5









max V (kms / sec) 4E+02 1E+03 2E+03 2E+03 4E+03 4E+03 6E+03 7E+03

This gives transit time for various accelerations, as well as the maximum velocity reached at turnover. Our trip to Iota Persei, from this earlier post and taking the actual FTL time as zero would take eight times the numbers above.  If we have a military-grade engine and remass tankage that can sustain 3Gs we can make the colony from Earth in about 40 days.  Slowboats had better freeze the cargo: at 1% of a G it will take 2 years.  This suggests that to have the scale of empire I want, even traders will need sustained boost capability over .2G, and the more the better.

It also means that if interstellar war (or even pirates) are a real threat we will need adequate local forces to deal with a threat and provincial governors with enough discretion to use them.

Wednesday, June 15, 2011

Knowledge is power

Or at least a trade-able commodity.  The problem is what knowledge is available to whom, where and when.

Lets say Explorations Inc discovers a nice habitable world (lets call it Thrixx) with a thriving non-human population on the rimward edge of Human space.  The ExpInc ship returns to a human world, and sells this information "Knowledge is Us" a known-space-wide information consortium.

A bit later Uplift Unlimited wraps up a technology uplift contract on the coreward edge of known space.  The local UU rep contacts the local KiS office and submits a data request for possible customers.  Will the ExpInc data on Thrixx and all those excellent potential customers be there for the UU rep to buy?

This is an interesting problem.  We have a database with all sorts of information in it, plus all kinds of other events going on.  Part of the interest of a realistic star-spanning environment is dealing with the propagation of information (and dis-information while we are at it).  A solution to this will (IMHO) be a critical success factor for the game aspect of this little enterprise.  For a novel, you can keep notes on the few things that matter; for a game how to you anticipate what will be important?  I expect you have to track anything that *might* be important.

Tuesday, June 14, 2011

Seriously cool little program

At work, I use visio to toss together small design diagrams.  But that costs money - money I could use for important things like miniatures.

But it looks like I have found my home use tool -- yEd.  It is the first free program I have found that lets me toss together a good-looking crow's-foot notation ER diagram.

Sunday, June 12, 2011

Coordinates

Note to self: The systems in the Hyg database are in earth equatorial coordinates.  Ill convert to Galactic and center at Sirius (about the brightest local star) for a less human-centric world view.

Risk and space travel

Looking at some air safety stats, it seems to me that travel has to be pretty risk-free for people to consider it safe.  To have a functioning interstellar economy, especially given the length of voyages and the fact that all the time is spent in in-system travel apart from the FTL risk, and jumps are going to have to be very safe indeed - of the less than 10^-7 deadly failures per jump.  Indeed, I am inclined to think of the FTL drive is a fail-safe device.

However, there is something that has been bugging me about the FTL-drive concert  I have now.  Once you reach a range limit for the technology (you note we have not nailed that down yet) the borders become very certain and solvable.  Two technology levels might offer very different graphs of the transit routes, but in the end it is a very certain graph.  However, if we let risk enter the transition equation then the borders become uncertain.

Insured private ships will only operate at low-risk ranges (say one failure in 100,000 with one in 1000 of those "fatal").  So will military ships in normal operations.  But if I can push an extra parsec out of the fleet by taking a one-in-10,000 risk and so effect surprise would it be worth it?  Is it worth taking a series of one-in-1000 risks will get word of the alien invasion to the nearest base 3 months earlier, is it worth the risk?  If bypassing three border system and transiting right to a friendly location is all that will save your battle damaged butt, is it worth a 1-in-10 risk?

The price has to be sharply exponential, or we might end up with no frontier regions at all; but a fuzzy frontier is worth a bit of work - IMHO, YMMV.

We interrupt this future history blog

for a brief gloat about an alternate past.

yes, I played Turkey.

Saturday, June 11, 2011

When in doubt....

I expect that the 26th century will still see Marines.  They will be found wherever nations whose traditions include distinct land-operating military (British Royal Marines, Russian Naval Infantry) have ships and project force ashore from them.  That would be Earth, and any planet that has enough water to be worth building a warship there.

For operations in and from space, you will not be "sending in the marines."  For Cislunar forces the exact terminology we will use will be "Space Assault Forces." I am basing it on the prototype of the Russian Vozdushno-Desantnye Voiska mostly because this is a distinct arm of service within the Russian military.  Since the forces would originate from the contributing polities, each with its own traditions, individual formations may hold that their transitions originate with Marine, Airborne, or Army units; but new traditions  would doubtless form rapidly.

In this model, the mission of the SAF would center around:
  • Operations involving planetary assault or raiding from orbit against primarily human-habitable targets.
  • Conduct of planetary operations on habitable but unearth-like environments.  This makes adaptability and flexibility a key requirement.  SAF aircraft, for example, would have to be configurable for multiple gravities and atmospheric pressure profiles; and adaptability that would probably come at a cost compared to craft optimized for the environment.
  • SAF would probably also have formations (and elements within larger forces) that are true space combat specialists, concentrating on shipboard, zero-gravity, and domed-colony assault work.  These would probably also form teams for ship security and boarding actions.
By the way, this does not mean that I think that ships in space battles are going to pull alongside one another and swing across with cutlasses in their teeth.   Ship boarding operations happen right now.  That mission is not going to go away when it comes to spacecraft.

Neither, for that matter will search and rescue, or support for civil authorities.  The exact combination of forces that will perform such roles will depend on the polity.  Many states have certain military (like the US National Guard) or paramilitary (the French Gendarmerie nationale, the Internal Troops of the Ministry of Internal Affairs of the Russian Federation) forces that perform more civil-directed action.  Which a given polity or colony world might use would depend on population and on the traditions of the main founding polities.



The space-based polities - Jupiter and Saturn and their subject objects - would probably not have extensive ground troops, while individual combat specialists would be purely of the space-based types.  Large domes, like Luna Farside, would need internal security troops, but these would be more of the Paramilitary Police type.

Thursday, June 9, 2011

Visual presentation of astrographic data

Lets assume for a moment that we will be dealing with something like 100-200 thousand star systems.  Each important system will have data about what is there, but lets ignore that for now.

Looking at the star maps in Project Rho, it's my feeling that  much more than 30-40 stars in a single map of 3D space projected onto 2 dimensions is pushing the boundaries of human comprehension.  It is also a requirement here that we use flat 2-D images - I am not making assumptions about technology available to players or readers.  Really, some level of the presentation would have to work as a map in a printed book.

With say 150K objects, splitting them up into "subsectors" of average 30 systems each, we would be talking 5000 individual subsector maps - it it were all cubical, then 17x17x17 subcubes.  Of course, we don't want to issue players with a 5000 page atlas to try to work their way around.  We can generate any where we really need detail on the fly, but we need top-level maps.



So here is a "top level" map of Europe, courtesy of the CIA. What does it tell us?  Well, quite a lot.  It shows centres of population, oceans and land, mountains and rivers (strategically significant since they constrict and direct movement) political boundaries and relative national sizes.

All the key stuff, strategically.  We can get a rough approximation of travel times, of constricting terrain, of key objectives and obstacles.  If it were 1944, you can't plan D-day from it, but you can get a pretty good idea of the progress of the liberation of Europe.  It excludes, but could add, symbols for transportation routes, industrial and resources regions, and lots of other types of data.

So, for our star map, what sort of data could be displayed?  Well, the simplest thing is Polity capitals.  Transportation routes (with width indicating flow volume and color indicating polity) are another.



For multiple polities it might be too dense, but the boundaries of a single species could be shown by all the systems with worlds inhabited by that species.  Or if we consider a graph of a Delaunay tessellation in three dimension of all stars controlled by an empire, we could display small  dot in the middle of a line connecting a controlled start with an uncontrolled.  This also bring up a question of what might constitute "controlled" or "inside a sphere of influence.  Good thing Graph Theory is fun.

For ship routes, although most would look more like a railway timetable, strip maps are worth a glance.  Here's one from the AAA. In case you think the AAA invented them, comparable maps were used to describe routes in Europe from at least the 16th century.  The advantage of the strip map is that not only does it show the core route, but it also gives diversions from the route.  This could be useful for a planner who had developed a core approach, but wanted to look at local alternatives to avoid particular problems.  This sort of data could be presented as a 3-d map or as a classic shape and line graph - which in fact is what a road map is.

Anyway, quite a few ideas to try -- I must get back to that star data -- but I am open for more suggestions.

Software, support, and what is this all for (yet again)

OK, so I want an star/system/logistic management system to manage notes for a novel, notes for an RPG, and data for a strategic game.

There are a number of functions that are obvious, such as route and time determination software and system creation.  For a game there is certainly logistics tracking.  Also subsector map generation and the like.  Those are functions I need on my own desktop; data products like travel itineraries or images of maps can be produced and distributed as needed.

For a low-intensity function like an RPG the software is probably fine just on my desk.  For a full, formal strategic game I would have to pick one of two choices:
  • Abstract the system to the point where results can be worked out "in your head" -- which, when you get down to it, would not be a horrible bear.  It is probably safe to assume that I could also distribute player data in the form of spreadsheets - there are simple APIs for spreadsheet output.
  • Provide software for player use to allow full calculation, but only with data he can see.  The program could also generate order files that could be processed by my GM software.  The real downside to this is development time and support effort.  Do I want to take it on myself and do I want to impose it on others?
There is a third choice -- a "strategic RPG".  The players would have strategic roles within various polities and interact with one and other and the GM as they would in an RPG, but one of the outputs would be the strategic inputs to what would basically be a solo game at a low level.  I think this would be a valid option.  The players would essentially  (in addition to other RPG play) receive briefings and proposals from their subordinates and review and approve or modify operational plans at a high level.

A strategic RPG is also less sensitive to player coordination -- as a "solo game with outside inputs" instead of a conventional multi-player it can clog on without direct player supervision.  This is especially true if each side has multiple players.

I like this last one.  What do other folks think?

Wednesday, June 8, 2011

Data modeling

Two things.

One is that I did a bit of poking into the state of Object Relational Mapping.  The conclusion that I have come to (which is slightly odd for someone who has been working with OO concepts for more than 15 years) is that I want to focus on a relational design for the core star data.  SQL is a mature and widely understood technology that is superior for exploring data than the object based approaches that I am familiar with.  While maintenance and mapping apps will need object models, these might be best customized for purpose rather than aimed at generality.

The other is that, while a blog is a great place to discuss ideas in general, it is not a great tool for organizing and re-organizing design documents.  I have two choices there: I could use standard word-based design documents, or I could set up a Wiki.   Wiki interfaces are very good but are not the same bandwidth (at least for me) as a word processor.  On the other hand, they do allow easy sharing of results, much as a blog does.

Wednesday, June 1, 2011

Force shields

In an early post I speculated about "shields" as an acceptable technology; it prolongs space combat, gives the heroes more survivability, and also provides a way to justify otherwise pretty unreasonable engine exhaust velocities.

I have been re-reading the "defenses" page of Atomic Rockets and re-thinking that position.  Whatever hand-wave we might want for the engine, keeping the weapons systems effective and dramatic is probably a pretty good idea.  I have been reading sections of The Fighting at Jutland and I was impressed at how the most dramatic stories were those of the most damaged ships.

If there is a question, it is what are the most viable platforms?  Large ships with extensive defenses or small craft, but many, with big-ship-killing weapons?  I must think through my technological assumptions and come up with something that works for me.

Tuesday, May 31, 2011

Thoughts on an alien species

I want to get down a few thoughts about an alien species.  The basic idea is a species who's individual members are too small too possess the raw brainpower to be highly intelligent,  but are linked by a natural sense (electrically or magnetically based) to form a multi-body intelligence.

The idea is hardly new.  Olaf Stapledon has at least one collective species who evolved from bird-like flocking creatures in Star Maker, while a there is a Poul Anderson story (sadly, I cannot recall which one) in which intelligences are formed of three members distinct symbiotic species linking physically.  And there are always the Borg.

I'd like to pose something that is individually large enough for a single body to have intelligence (perhaps at the level of a chimpanzee) and a member of a small enough collective (perhaps half a dozen members) that it cannot be looked at as disposable the way an intelligent ant colony might think of a single ant.  Currently I find myself picturing it as small (dog sized) and with a body plan like a compact centaur.  In fact, one my sources for this idea is a discussion the possible uses of Baccus 6mm Centaur infantry as 25mm figures mounted 3-4 to a base and claiming that they used the group mind to concentrate the fire of individual small lasers.

I expect something like this might evolve from pack hunters who used the ability to coordinate hunting.  In the ecosystem where this evolved I would expect it to be a widespread capability, like magnetic orientation in many species of bird.  Special pleading would be required to explain how a "bioradio" of sufficient bandwidth to distribute cognitive processes is present in this ecosystem; however it is not hard to see how it could confer useful advantages once evolved.

The interesting questions are social.  There are a number of dimensions on which "micro-centaurs" could organize themselves socially.  Lets express them as diametric opposites.  Let us also assume that the individual units are either male or female, and that there is some level of sexual dimorphism and a distinction of capabilities (although ranging widely between individuals) roughly aligning with 20th century human gender stereotypes -- because, after all, humans are all we have available to play head games with:
  • Genderist vs Multi-Genderists:  Genderists believe that a single collective should all be one one gender, while multigenderists believe that an collective can be made of any combination of unit genders.  I can see genderism being considered natural where sexual division of labor is commonplace such as hunter-gatherers while more settled societies may want to mix capabilities within a single individual.
  • Immortalists vs Discretest.  Immortalists believe that the collective should take advantage of its potential immortality by socializing offspring as part of the collective with older members; new collectives would be formed by older members separating along with some younger members.  While units would die, the collective would like on.  Discretests believe that each collective should experience the full cycle of life and death.  Offspring are socialized within age groups and raised as distinct individuals, dying when all units have passed on.
  • Extentionists vs Independentists.  Extensionists believe in forming super-collectives to work together on serious problems or simply for the self-transcendent experience.  Independentists believe that the right to self-determination of the individual collective is foremost.
Of course, not being micro-centaurs we can view these three dimensions abstractly.  For the m/Cs themselves each culture consider the three choices to be "normal" and "disgusting perversions of nature and God's will" whatever the choices their culture makes are.  The closest analog to human would be Genderist, Discretist, and Independentist -- mostly because our biology offers us no options.

And this does not get at questions like reproductive and child rearing practices.  Do multi-genderists reproduce with themselves, or are there the equivalent of incest taboos?  How do inheritance laws work?  How do class structures work within each society? Are domestic animals somehow integrated in the collective in an evolutionary path like the one that led to dog/human symbiosis or would that evolution stop at a lower level of integration?

And what is a collective like to interact with?  Do they use voice as well as bioradio to communicate with each other?  How about with other collectives?  If they can speak, if only to talk with humans,  would there be one spokesbeing or might conversation rotate around the members like a well-timed comedy routine?  Would this all vary by the individual (a word that rapidly loses its utility)?

What would an isolated unit be like?  Sure, it he or she has the intelligence of a chimp but chimps are *not* stupid and our unit would have the habits of "civilized" behavior.  Would the last unit of a dying Discretist collective be like a human with Alzheimer's, remembering that it once though larger thoughts and knowing that it is more prey to its emotions than ever it was when it was in its prime.

Heck, if they speak, what sort of pronouns might they have for multi-genderists?  Would they have a family of pronouns that expressed two-thirds-female-one-third-male? Different pronouns for units and collectives?  Genderist cultures would probably have different pronouns from muli-genderist, and would have to use borrow-words from one-another's languages to hold a conversation.

Add to all this the fact that it "takes all sorts to make a world" and the mini-centaurs would have their own managers and workers, criminals and heroes, artists and engineers and I think you could have a very interesting society to explore.

Saturday, May 28, 2011

What data do we need about stars and star systems

For each star the bottom line numbers we need are Luminosity (preferably visual and bolometric), and mass.  It's also nice to have radius.  It is also useful to have some potential characteristics, like variability or flares.  It's good to preserve the conventional spectral classification in a form that people are used to seeing it in.

Do we need the temperature?  Might be worth keeping.

For groups of stars - binaries and the like - we need the base data required to construct orbital elements in case they become important.  Ideally this will be consistent with at least the well known binary system parameters.  For that I will need a binary or variable star catalog with the appropriate cross-identifications.  And all of this needs to be compactly encoded.
  • Using bit fields, 16 bits will be plenty for the "classic" spectral type including interesting features
  • Real (4 bytes in Derby) will be plenty of precision for floating point numbers.
  • Integer (4 bytes) will be plenty for "object IDs"
I will work up individual star data 1st, then look for other characteristics.

    Friday, May 27, 2011

    Spectral types in Hyg

    These are interesting data.  We all know Oh, Be A Fine Girl Kiss Me.  It used to include "Right Now Sweets" until the R, N and S types were demoted.  However, spectral classification is a lot more complicated than that.  Thankfully, with a few exceptions, the data look to be reasonably amenable to parsing.  Which is good, because aside from nulls there are at 4013 distinct values in the 119000 catalog entries.  We need to develop (and encode in the database) enough understanding to provide input to a  game-grade star system generation model.

    There are a couple of interesting peculiarities in the data, and I am not sure yet how to interpret them.  For example A1III/IV is clearly an A1 which is somewhere between a super-giant and giant.  We can just drop it into one of the two categories or encode it as is and assign an intermediate mass.

    How about A8/A9V? I think, judging from other entries, is a main sequence A type intermediate between A8 ad A9 I am fairly sure F6IV + F6IV is a binary with a pair of F6 giants. G0/G1:+... on the other hand would be a start somewhere between G0 and G1 of uncertain luminosity with a fuzzy spectrum in a pair with -- well "..." just means Damned If I Know.

    So, the two next tricks are:
    1. Develop a coding that incorporates what we need to know.
    2. Implement a parser that takes one of these descriptions and plops out our code.

    Handling bad catalog data

    I am making progress with processing the "Hyg 2.0" dataset.  One of the challenges in dealing with any large body of data is dealing with imperfect data reduction.  The Hipparcos mission is a dismaying introduction to this  issue for people accustomed to dealing with hand-editable volumes of data.  The author of the "Hyg 2.0" file made this decision:

    Distance: The star's distance in parsecs, the most common unit in astrometry. To convert parsecs to light years, multiply by 3.262. A value of 10000000 indicates missing or dubious (e.g., negative) parallax data in Hipparcos.
     There are 705 such stars in the data.  Only two are brighter than 6th Magnitude; only 1 has a Flamsteed number.  Amazingly, one is in Gliese.  The brightest are hot giants and so will be well out of our range of interest.

    For the solitaries with simple spectral types it's not hard to compute a magnitude-based distance and recompute the distance.  For multiple star systems it gets a bit more complicated.  My thinking right now is to ignore the lot.  Nothing there is going to be familiar enough to draw attention to itself for anyone but an expert; I may re-digest them later but I think that the "Backfill" approach will be close enough for gaming and fiction purposes.

    Thursday, May 26, 2011

    Do we really need FTL?

    With one very simple nudge, we can turn our magic FTL drive into something that lets causality sleep at night: make it into a light-speed drive.  The travelers experience no time in transit but planet-bound observers, who are pretty much all in the same frame, will agree that 1 second passes for every light second traveled.  Isaac is not our friend, but at least Albert is.

    Now, I'm working on this as an environment for fiction, as a potential RPG setting and as a strategic gaming environment.

    There's lots of fiction that manages without FTL travel, and provided the gaming party stuck together an RPG could still work fine.  However, can a strategic game -- or for that matter comprehensible military SF -- function without FTL travel?  Can you, indeed, have meaningful wars?  Conflict, certainly, but anything we could recognize as move/countermove?  Strategy?  I don't really know.

    I am not sure I want to start down the no-FTL path but it is interesting to think about.

    More FTL limits

    Lets impose another limit.  The summary of what we have:
    1. Limitations that make sense
    2. Jump Drive 1.1 - What I don't like
    3. An implication of the FTL system
    4. Travel to another star
     So far, I have not placed much restriction on where one enters hyperspace in relation to where one emerges.  Lets do that now.   The restriction is that one travels on a straight line between the centre of mass or the origin system and the centre of mass of the destination.  Further, that line cannot pass through a point where the "flatness of space" criterion is invalidated (a safe enough bet, given the sparsity of stars, but still).  We can assume that the parameters are reasonably loose, say a volume roughly equal to that of Jupiter; still, it means that you do have some travel in system between jumps to get to the next departure point.  This may be enough to allow a relaxation of the distance related time penalty to allow simply instantaneous jumps.

    It may imply that we have instantaneous knowledge of the movement of mass -- at least at a stellar mass level -- within a binary system.  Which starts to give me problems I must think through.

    Also, Ken Burnside has made a good point in a post in this discussion:

    The real question in any kind of strategic campaign game is this:
    How do you prevent Pearl Harbor style attacks from ending the game?
    If you have something like Traveller jump numbers, you get something like hard frontiers. At the extreme end, you get something like Starfire warp points.

    If you have the ability to go from any star to any star, with only travel time being the deciding factor, you get nebulous frontiers.

    If you can go from any star to any other star with all travel times being the same, you get no frontiers.

    A problem with 'campaign games' is that hitting with overwhelming force as the first move can be a winning strategy, not because you've delivered the decisive blow to your opponent, but because you have made the game not-fun and your opponent quits.

      Tuesday, May 24, 2011

      An interesting body layout

      Brachiating terrestrial cephalopods ( posted on youtube Update: And now pulled.  Link rot is a horrible thing).  From a BBC program about hypothetical life in the future, but also an interesting   body layout for an alien intelligence.

      Sunday, May 22, 2011

      Read this book

      Court of the Crimson Kings by S.M. Stirling.  A good friend game me a copy when I was in hospital a couple of years ago, and it is one of the best pieces of SF I have every read.  For our purposes, you must have a look at the Martian biotechnology.  Absolutely 1st rate; an alien species with this sort of technology is a must-have.

      Click for some sample chapters!

      Saturday, May 21, 2011

      What were we trying to do again?

      First, I want to put in a link to a nice chart of the closest dark nebulae.  How much complexity this might add I am really not sure at the moment; but at least we are good out for 300 light-years or so.

      Lets go back to 1st principles on the star map front.  The overall objective is to produce an interesting future history that can provide a good base situation for a strategic wargame, an SF novel or two, and an RPG campaign, possibly all at once. The secondary objective is to enjoy the journey, which may mean giving the any particular attention that it deserves.  This is my hobby, not my job, and playing with this is fun by itself.

      The role of the star map in this exercise is to provide the stage upon which the play is acted out.  It has to be large enough that, even though the human action is crowded at the centre, there is a sense of awesome extent toward the wings.  It should be large enough that other species may have their acts on the same stage without a sense of confinement.  In other words it has to be big enough to be empty and impersonal.  Not just stated to be, but experienced to be in the sense of "show, don't tell".

      So, how do we manage the data if we want to push the boundaries out, say, 150 parsecs to include a round million things?  And where do we get the data if we want to do that?

      Part of the solution (as I picture it at this moment) is "make it up"; the other is "broad strokes where detail is not needed".

      I don't want to contradict reality, but at the same time I am willing to sweep some annoying realities under the carpet.  So, lets say we start with a catalog down to 9th magnitude. That particular catalog has just shy of 120,000 objects - it is all of Hipparchus plus all of the Yale Bright Star. If an object is dimmer than about 6th or 7th magnitude the human eye can't see it from Earth.

      So what I propose is this (using a competent database such as mySQL or Derby and lots of Java programming):
      1. Take all of the "good" systems -- those with clean positions -- and build a starting dataset with absolute minimum information compactly encoded with each cataloged system.  By compactly encoded I mean about 20 bytes/system -- mostly be relying on putting bulky information in secondary tables that are only used when detailing is needed.  This minimal table will start with 120,000 rows.  It will end up with lots more.
      2. Cleanup.  The initial data will have real-world grit.  Astronomical catalogs are never as simple as the undergrad textbooks let on, especially for entries such as spectral type -- generally because the observation has never been made.  This has to be purged so that everything is expressed with an entirely synthetic cleanness that will fit well with a computer implementation of some reasonable random star system generation rule.
      3. Backfill.  We have base object density for nearby objects.  As we move out from Sol, more objects will fall below the magnitude limit of the catalog, whatever that is.  We make up for it by completely randomly generating objects up to more-or-less the expected density that are dimmer than the required magnitude at this distance.  That means that out ratio of made-up to measured entries would go up as the distance from Sol increases.  While someone with a better catalog will soon discover the fake bits, it will never contradict what someone (even in a dark place) can see when they look up.
      4. Big Picture.  In an approach familiar to Traveller GMs, identify the star systems that are good candidates for habitable worlds, and randomly generate them along with any key infrastructure.  Then identify the systems on the minimum cost jump routes between them and generate big picture items -- in Traveller, for example, the presence of a Gas Giant was always important.
      5. Make choices and detail where important.  Pick major species homeworld and detail those systems, for example.
      6. Drill down when needed.  Until the players or plot are about to enter a system, or the fleet is about to have a battle in it, the details don't matter.  Until the players are planning to land on a planet, the weather does not matter very much.  ("Lets see, your flight plan calls for a landfall on Earth.  Near Winnipeg.  In February.  Got a parka in the locker?").  Once something is generated, it goes in the database; but we are never going to have to develop complete data for a million stars, 10 million planets, 100 million moons and odd rocks.
      You know, I think this might work.

      Cork 1: Space is crowded

      In Habitable Planets for Man (Elsevier, 1970) Stephen H. Dole identifies 42 stars within 22 light years of Earth that are candidates -- although only 14 of them have (in his estimation) odds of having a planet in excess of 1%.  One way to confine exploration is to plop Earth-like worlds down at all of them, give all of them sapient inhabitants, let half of them advanced technological civilization, a quarter of those superior to ours; and put No Trespassing signs everywhere.

      There are a couple of things I don't like about this.  The most significant is that space shoudl feel empty.  It's one thing to say that it takes 6 months to get to Alpha Centauri; it is quite another, and more powerful, to play through plodding through systems with nothing in them but rock-balls until you get to something interesting.

      Still, some parts of this idea is good.  If Space is just plain empty of complication and competition it can get tired pretty quick.  There has to be variety somewhere -- in one direction, open frontier; in another, unpleasantly advanced aliens nearly on our doorstep.

      Keeping the Djinn of exploration corked in the bottle of data

      One of the interesting problems in trying to use realistic data and a reasonable projection of the future is that, unlike the 2D closed planet we are used to, the amount of data - the number of interesting or strategically significant places we should consider - goes up with the cube of the distance traveled.

      The related problem is that space is (effectively, for our purposes) unbounded.  Magellan's expedition sailed for 3 years in as straight a line as they could manage, but they still ended up back in Seville.  Such are globes.  While space may curve on itself over a sufficient distance, for practical purposes if we are going to concern ourselves with individual suns and planets, you have to turn around if you want to get back.

      So, lets say we want to take about six months to get to Iota Persei.  Lets also say that for plot reasons we want to have a 30 year old colony there.  That means that we could have data back from expeditions that have traveled at least 15 times as far from Sol.  Since i Persei  is about 10 parsecs away, that means our dataset should carry out 150 parsecs, and we might have some brave souls out as far as 300 parsecs.

      The dataset I am currently using extends about 15 parsecs.  This is a volume of 14 thousand cubic parsecs.  Now the catalog I have been playing with has about  a thousand objects, for a density (using less rounded numbers) of 0.701 objects per pc^3.  Projecting that out to 200 parsecs give about 2 million objects.  At 300 parsecs we are talking 8 million.

      And its not like we are talking about invisible stuff here.  There are catalogs right now down to an apparent magnitude of +21.  If we consider an M6 at the dimmest star we want to talk about having a planet that's absolute magnitude +12 -- which a catalog down to +21 we could see such a star out to 600 parsecs.  Even M8, which is getting down there, is visible to 100 pc.

      So, I want my earth-origin protagonists to have a good sized sandbox to play in; but I need reasonable ways to keep them in the sandbox.  But right now, I need some coffee.

      Wednesday, May 18, 2011

      Travel to another star

      I've been digging through my old graph theory books, and the HabHyg near star list, to look at the interstellar movement rules we discussed before.  I have also been looking at Iota Persei, a nice solitary G0V star that I have decided will have a nice, earth-like world in orbit around it.  I have also decided to add a cost in time to jump travel proportional to e^distance (in pc).

      Given that, here is the shortest (in time) path from Sol to Iota Persei.


      N HabHyg Name Sector Type Mass (Sol) Abs Mag distance time
      0 0 Sol 0:00:00 G2V 1.00 4.85 3.11 22.34
      1 12 Gl 905 0:00:00 dM6 e 0.13 14.79 1.42 4.14
      2 31 DO Cephei 0:01:00 M6 V 0.13 13.29 2.51 12.30
      3 98 Gl 34 B -1:1:0 K7 V 0.74 8.64 1.63 5.12
      4 161 Gl 53 B -1:1:0 M8 0.10 11.61 1.77 5.88
      5 257 Gl 51 -1:1:0 M5 0.21 13.88 1.47 4.35
      6 318 NN 3126 -1:1:0 M4 0.38 11.03 1.93 6.86
      7 354 NN 3182 -1:1:0 DC9 0.50 15.25 1.10 3.00
      8 378 Iota Persei -1:1:0 G0V 1.10 3.94



      Now, I will confess that this is output is not that pretty straight off.  Each row shows the star we are jumping from, the basic specs for the star, and the distance (in pc) and time to the destination star; that destination is on the next row.


      The mass is estimated (by hand at the moment) based on the spectral class.  Gl53B has no spectral type in the dataset, so I plugged it based on the magnitude.

      NN3128 is a continuous degenerate star (a cool white dwarf with no discernable spectra lines).  The data to determine the mass probably does not exist yet.  White dwarfs are very old star between .5 and 8 solar masses.  The "C" types are probably very old.  I have no particular faith the the assigned mass; but I doubt that anyone will pay any great attention to it in the near future.

      One of the things you have to be willing to do in SF is make decisions that fit very imprecise data -- decisions that no scientist would every be willing to stake a reputation on.  If we are not willing to do that, we have to either waffle our game and plot lines around the fuzzy bits -- you can do that in a novel but not in a game -- or stuff everything to the far side of some convenient dark nebula.

      The next thing I need to do with this table is work out what happens at each jump according the the rather complex "jump radius" rules I defined early in the blog.