How the Penny maps

5th May 2013 – 3.51 pm

Every now and again a comment is made about how I make my maps. As they are not screen-grabs and aren't part of the EVE Online's mechanics it's not immediately obvious how I do it. The simple answer is that I use a basic Paint application to create each map. But the simple answer glosses over much of the nuance that goes in to the process, which I hope to now correct.

Before I get to Paint, I first must decide I want to draw a map of the w-space constellation. This decision occurs either because the constellation is extensive, interesting, or illustrative. That is, the map is big, there is something uncommon about it, or both combined illuminate the narrative. A simple home-to-C3a-to-low-sec constellation is neither out of the ordinary nor difficult to imagine. A winding route through multiple systems with their own forking wormholes, however, helps illuminate how bookmarks must be created descriptively, why we use system identifiers, and how long getting from point A to point B can take.

Once decided that I want to draw a map, I sketch it in to my notepad. This is for two inter-related reasons. First, I normally realise I want a map when I am still part-way through scanning, and second, I hand-draw each map. I will recognise that the constellation is interesting when I have a handful of systems uncovered and still have more wormholes left unexplored. And because each map is hand-drawn I can't reasonably pause each time a new connection is discovered to take time to add to an image. At least, not the way I draw my maps.

I've seen some tool-generated maps. They serve their purpose well, but not mine. I try to tell stories about adventures, and although I believe appropriate images can help the story I also don't want to interrupt the flow of the text, which is why I restrict the vertical size of each image. This includes the map. As much as possible, I optimise the constellation image with vertical height in mind. This is not the only restraint I have on my maps, though.

I draw my maps such that:

  • vertical height is optimised
  • wormhole direction is indicated
  • static connections lead directly out to the right of the system, where possible
  • spacing, arrow lengths, and system sizes are regular, where possible

Each restraint is not always possible to follow, so I break my own rules where necessary. But I try to maintain the restraints to make the maps easy to read, which I like to think I achieve with my self-imposed guidelines, though they have only been implied up until now.

The preliminary sketch is as important to the finished map as my scribbled notes are to a finished post. In both cases, any self-imposed guidelines can be ignored for the purposes of detailing information quickly and accurately. The sketched map still needs to show wormhole directions and, ideally, static connections, but there is no great need to limit vertical space in my notebook. Accuracy is important, though, because once the bookmarks are deleted then all I have to create the map is the sketch and my notes. They'd both better be right.

My sketch begins from the home system. As I've already said, the sketch will generally being part-way through a constellation, so the initial few systems I can normally draw from memory. Even so, bookmarks are useful. Each wormhole is bookmarked in such a way to show system information and wormhole direction, and experience and my w-space notes will inform me of which connections are static. But even once finished, I will often check my sketch as I travel backwards on my way home, pausing in each system to ensure I have the right number of connections, pointing in the right direction, to the right class of system. I still catch the occasional error this way.

I transfer the sketch to Paint only after I have drafted the story the constellation is attached to. This is partly because the sketch will be half-way through the notes for the story, but mostly because without a story I don't need a map. But it is normally drawn immediately after the story is drafted, because I need to link the text file with the image file in some way. I don't write directly in to Wordpress, and neither do I have the means to upload image files to my server, not that I would want to fiddle around with that at the time. When I write, I write. Images are added later, even after the editing process, when I no longer need to feel creative.

The actual map creation process is pretty basic. The first map took more time and effort, as it the first, and created from scratch. It also became a template, of sorts. I drew the boxes for the systems, defined the arrow lengths and arrowhead sizes, and decided on a typeface for the text. Once that first map was completed, all I had to do for the next was to copy the file, and make liberal use of copy-and-paste. With this method, I just need to move the boxes around and link them with the same arrows, adding specific system text either once the constellation is complete or at the point all the links become confusing without it.

Normally, creating the constellation map from the sketch is fairly straightforward. This may be because I have got the hang of doing so, and craft my sketch already following my guidelines. But more likely is because the constellation, whilst not always strictly linear, does not meander too much. Random connections terminate quickly, or don't lead to more random connections or K162s. Other times, the map will overlap horrendously if I don't either increase its vertical height by adding extra levels of systems, or extend arrows linking systems to give more horizontal space. I consider a longer arrow preferable to a 'fatter' map.

Sometimes I need to be clever. I'll look at a map, drawn faithfully from my sketch, and realise I can compress it without breaking my guidelines or losing information. I love those moments. I flip a diagonal arrow, drop a chain there, extend this arrow, and I can lose twenty vertical pixels from the image. Now I feel like I am creating something, not just recreating it. It's not always possible, but being able to visualise such an improvement is pleasing to my sense of the aesthetic.

This post would be incomplete without an example. And so, for the first time, I offer a peak at my actual notes.

Scanned sketch of a w-space constellation

Judging from the sketch's position in my notes, I probably reached C3b when I realised I may be creating a map. Little did I know it would expand by so much, nor what a mess I'd get in to. I even ran out of room to get C1a's static exit to high-sec properly on the page. Also witness my terrible handwriting. I really do scrawl when writing my notes. They're just for me.

I have the finished image created from that sketched constellation map. Three versions, in fact, which is unusual. The first was more-or-less what was sketched. The second came after one of those revelatory moments when I saw how to compress it beautifully. The third came, curiously enough, when comparing sketch to image for this post, and saw how to better reflect a couple of static connections, breaking one guideline in favour of another.

But I'm not going to include the final version, or any of the images, in this post. Not yet, anyway. It's no coincidence that I chose a map that already overlaps. If you're interested in how I make my maps, maybe take a few minutes to decipher the image and try to decide how you would draw it according to my guidelines. And if you have some time to spare, I'd be keen to see what you come up with.

Update: the post with the completed map has been published, so it's time to share the full creation process. The first version of the schematic was pretty much a straight copy of my hand-drawn map.

First version of the w-space schematic

There's nothing particularly wrong with the image in itself. The question mark at the bottom is because, at least for the first of the two posts that use the schematic, the connecting wormhole was unvisited and so the destination class of system undetermined. Whilst I'm okay with missing that detail, because the map is big I preferred filling it out completely, once the missing piece was known. That gave me a problem, though, once discovering it led to more w-space. Where would I fit the system's static wormhole's destination? I would need another vertical level, in a map already four-tall. Luckily, staring at the graphic for a while gave me one of those clever moments.

Second version of the w-space schematic

I saw that if I translated C3c so that the vertical direction of the link between C3b and C3c pointed downwards, and C3c ended up where C4d was, the outward chain would continue as before, but the chain heading backwards would tidy itself up and give room for C1a and its exit. Even better, one of the diagonal lines, for C4d's static connection, would become horizontal, as I prefer them to be represented.

I have to admit that this is one of my best modifications to a map, for letting me add information and clarifying the rest in one minor change. But it didn't end there. Whilst reviewing the map for this post, I noticed I could make a second change.

Final version of the w-space schematic

It's a small change, not really necessary, and breaks one rule to fix another. Still, I flip-flopped between this and the previous version for longer than I should have, before deciding that I liked the long horizontal chain of static connections, C2a fitting in to place by extending the arrow down to C1a.

But the iterations are generally small, or noticed early, because all the work is done in Paint. Every system bubble, wormhole arrow, and system text is bitmapped, so every change takes time. Maybe I should find some better software, but it's not like I do this a lot. Anyway, in the fiddliest case, that's how I go from scanning w-space to creating a schematic for a post.

  1. 13 Responses to “How the Penny maps”

  2. You have the handwriting of a physician. Not a meta-physician.

    By Firstly on May 5, 2013

  3. Your handwriting ceased to be critiqued over 20 years ago. Let go of the stone and swim.

    By Kename Fin on May 5, 2013

  4. Doesn't seem to be illegible or anything to me, altho it is hard to say with such small sample and only couple actual words.

    By Mick Straih on May 5, 2013

  5. There are still occasions when I spend a minute trying to work out what word I meant to write, but scribbling notes after a fight tends to have me scrawling quickly to get everything down.

    Anyway, this isn't about my handwriting. Let's focus on Rampart the maps, people.

    By pjharvey on May 5, 2013

  6. Seriously the best you can do is complain about the hand writting ?

    Penny how much time would some thing like that take to map ?

    By Rom on May 5, 2013

  7. The joke, it seems, has sunk. My apologies good people; I meant no offense.

    Your diligence really does continue to amaze, Ms. Harvey. Thank you for sharing. :)

    By Firstly on May 5, 2013

  8. i think every joke has been missed, Firstly. Don't worry about it.

    Rom, I assume you're asking about how long it would take me to scan so many systems, and not how long it would take to draw the map from my notes. Well, it's easier to answer the latter, so it takes maybe fifteen minutes to convert the sketch to an image, maybe longer with the more complex maps.

    As for the time it takes to scan so many systems, naturally it varies, depending on how untidy each system is. To start with, I reconnoitre the system, looking for and locating towers and ships, before beginning to scan for wormholes. After that, I will scan the system thoroughly, looking for as many options as possible.

    But once I realise I'm diving down a multiple-forked constellation I will end up recalling my probes the moment I've resolved the static wormhole, even if that means resolving one signature and ignoring thirty. If systems are inactive and empty, my priority becomes moving to the next to look for targets rather than wasting time resolving sites I don't care about.

    The above constellation must have taken two hours to map, maybe more. And did I find anything? Well, the story hasn't been published yet.

    By pjharvey on May 6, 2013

  9. I've started incorporating your wormhole techniques into my own. I've also started keeping a record of systems in spreadsheet form. I may develop this into a proper database that records capsuleer activity eventually.

    By Mekhios on May 7, 2013

  10. I started that... then passed out as pilot #3,329 in a row was idling in their force field. Log off or Fly out where we can scan you down!

    Seriously, most pilots in wormholes seem content to babysit their tower.

    By Kename Fin on May 7, 2013

  11. Sounds good, Mekhios. Fin has a point, though, and the irregularity of visits makes individual efforts somewhat frustrating. There are many systems I return to only after a year or more.

    But if enough people did the same, and there was a centralised repository of information, maybe the data could become useful. Mind you, I would be surprised if the same people who would be exposed by the data would willingly contribute to it.

    Even so, I hope that keeping details of visited systems pays off occasionally. It certainly helps us from time to time.

    By pjharvey on May 8, 2013

  12. I am not sure why, but I felt compelled.

    By Kename Fin on May 28, 2013

  13. Compelled to forget to actually include the comment it seems:

    By Kename Fin on May 28, 2013

  14. Phew, good job, Fin.

    By pjharvey on May 28, 2013

Sorry, comments for this entry are closed.