Great movie. Great scifi wrapped up in robots (who behave a lot like humans, but eh).

Let’s just get one thing straight. EVE does look like she was engineered by Apple, so you could say she’s like Mac or whatever. But WALL-E sure as hell ain’t Windows. He’s Linux straight up.

Ogame? This — is — SPATRAS!

Spatras has been approved as our senior project so we don’t have to do bioinformatics like the rest of the sheep.

What is Spatras you ask? I’ll tell you! Here is the email I sent to Marc Miller, creator of Classic Traveller in asking for his permission to use the CT ruleset:

Dear Mr. Miller:

Hi! Thanks for writing me back.

Our initial focus will be on space trading, ship-to-ship combat, and exploration. Also character generation will keep things like pistol skills and (maybe not initially) we would like to use that for boarding. At some point in the future we would like to implement some sort of planetary adventure as well.

Our concept is a web-based game similar to (but far better than) ogame.org. This game has huge flaws. There is no exploration to speak of, combat is static (you can determine the exact results of any battle before you engage using a player-made tool), there is no trading, no visual representation of other players, etc. As a long-time fan of Classic Traveller I felt that the flexibility and adaptability of these rules would enable us to create a more satisfying game. Our vision is a game that allows a playing model similar to the following:
* Create characters (using Book 1)
* Enter Space Trade/Combat/Exploration (using Book 2 and Book 3)
- via Scout mustering out with Scout Ship.
- via teaming up and pooling resources to buy a ship.
- via working for other players,
existing player-run companies,
or for the Imperium (which would be us, the developers).
* Allow for Ship Purchase, Building, and Combat (using Book 2)

This overall gameplay would (hopefully) have something like the following mechanics:
* Exploration. Worlds and other errata are randomly generated when explorers venture into unknown map regions. Then they are maintained as part of the persistent universe, meaning other players can visit there. We would use the rules in Book 3 to handle this. As of now we don’t intend on using the existing canon maps, everything will be generated.
* Trade. Book 2 rules.
* Imperium. Either ourselves or possibly ai to be a governing force in the spirit of Classic Traveller and the era of wooden ships. This also means of course that players can become pirates and prey on other players or tangle with the Imperium.
* Visual recognition of other players and communication keeping in the spirit of CT. This means that long range messages are handled via courier (or possibly drone) and short-range is real-time (via radio). Initially this may mean that you see their ship on some sort of map. We hope to have this feature be so basic as to allow players to form companies of mercenaries, traders, or pirate bands in a way that is integral to the game.

In the future, if and when we continue to work on this project, we would like to implement more books from Classic Traveller; especially High Guard, Merchant Prince, and Robots.

To summarize, we would like to implement Books 1-3 as much as possible. The game itself will have as many players as want to play, similar to an internet forum. Therefore, while my colleague and I may be something like a referee, there won’t be any specific story-telling unless it is performed by the players themselves using the game. I have not played this game, but the old Trade Wars BBS game might be a similar concept; EVE Online is also a loose sort of example, but ours probably would not have quests and certainly won’t be as graphics oriented (we’re limited to webpage-type interface and graphics). We want a system where players can run around on their own trading and fighting and interacting and what-have-you and we feel that Classic Traveller rules are a great match for implementing this. The only things that we would want to implement that are canon are the Imperium and some of the basic ships and items. If you only allow us to use the rule set we can, of course, use our own items and governing body. Additionally, and this is a much later if ever goal, we would like to use ads or some similar model to support the server. This might mean royalties of course, or some sort of licensing agreement. I’m only speculating, I have never done this before so I don’t know what it would involve. I just do not want to do anything illegal, much less infringe on something that I respect and enjoy so much.

Allow me to restate that we do not now or ever intend to infringe on your works.

I feel it’s also worth mentioning that some parts of CT probably will not work in our implementation. The fair-use faq I read at travellerrpg.com mentions that rule rewriting is forbidden, and I understand why. However, especially since we do not wish to do game time compression if possible (i.e., one minute in real life is one hour in game time) and since all players have to be doing things in the same time-frame; there are many elements that would have to be changed. Jumping for example shouldn’t take a week. If we gain your permission I hope you can give us some advice on this.

Thank you very very much for your reply, interest, and time.

Note some formatting and context issues. Woot!

For OS X go here.

If you have any trouble, keep reading. I may have placed a note covering your problem. If not then just post a comment here so I can help and/or update the article. Please note that the scope of this article is to get you setup with a developer environment in Windows that can also compile, for Eclipse help (C/C++) go here. That being said, I’ve included a couple of notes on using Eclipse at the bottom.

1. Download the Eclipse IDE for C/C++ Developers.
   1. Extract it to your hard drive, I just put it on the top level (c:\) so I ended up with c:\eclipse.
   2. If you want, make a shortcut to eclipse.exe.
2. Setup a compiler.
   1. Download and run this. Use ‘Current’. Install all or at least the g++ compiler. Do not install to a custom folder, use the defaults.
   2. Download and run this. Use the defaults and follow the script instructions. When it asks for your installation directory the slash is reverse of what it should be, this is ok. If you used the default just give it “C:\MinGW” . Note: if the script doesn’t run properly or doesn’t run at all, you can rerun it by navigating to c:\msys\1.0\postinstall and double-clicking pi.bat .
   3. Append the following to your ‘Path’ environment variable: ;C:\MinGW\libexec\gcc\mingw32\3.4.2;C:\MinGW\bin;C:\msys\1.0\bin
      1. To do this, right-click ‘Computer’ (Vista) or ‘My Computer’ (XP), click ‘Properties’, click ‘Advanced System Settings’ (Vista), then click the ‘Advanced’ tab and then the ‘Environment Variables’ button. Find the ‘Path’ variable and click edit. Be sure to add the line at the end, the whole entry is automatically selected so you could erase your entire path. That is bad.
3. Start Eclipse and try a ‘Hello World’. It should get your compiler paths automatically.
4. Get and install glut.
   1. Copy glut32.dll to your Windows\System32\ folder.
   2. Copy glut.h to C:\MinGW\include\GL.
   3. Copy libglut32.a to C:\MinGW\lib (overwrite existing file).
   4. Delete C:\MinGW\lib\libglut.a.
   5. When starting a new Eclipse project you need to add the glut32, glu32, and opengl32 libraries. Be sure the MinGW toolset(s) are checked. If you have Cygwin it will also be visible, but do not use it unless you know what you’re doing.
      1. In Eclipse, go to Project -> Properties -> C/C++ Build -> Settings -> Tool Settings. Under ‘MinGW C++ Links’ or ‘MinGW C Links’ click Libraries. In the Libraries field, click the icon with the green plus sign and type in a name mentioned above. Do this for all the libraries.
         Note: When using a ‘Managed Make’ project, click C/C++ Build and check the box for ‘Automatically generate make files.’ Then click Settings and the Tool Settings tab will now be visible. Libraries are placed under ‘MinGW C++ Linker’ as above.

Before compiling save your files or you will get errors. When distributing a program be sure to compile in Release mode and include glut32.dll.

Sources:

1. GLUT Eclipse CDT Tutorial
2. Using GLUT with MinGW
3. Setting up a compiler

Some notes on working with Eclipse:

• When working with a Makefile project, checking ‘Generate make files automatically’ under Project -> Properties -> C/C++ Build makes things worlds easier (you can also use this option by, when creating and after naming your project, hitting ‘Next’ and ‘Advanced Settings’).
• If you’re having compile issues, double check your paths. These can be found in Project -> Properties -> C/C++ Build -> Environment. PATH is inherited from windows but you can fix minor typos here. This will avoid any relogging issues sometimes associated with such changes, although you should also fix them in your system path per the 2-2-1 instructions above.

SYSU

Idea for this script borrowed from a friend. Found the original here. This is pipe-able if you add a ‘-’ after ‘vim’ in the alias, but this will break regular file viewing (or you could just use ‘less’ since this won’t give you syntax highlighting on your piped stuff anyway). I added an ‘!’ to the keybinding for ‘q’ so vim won’t complain, it’s not like you’re going to be writing to anything anyway.

” alias vmore=”vim -u ~/.vimrc.more”
” No compatibility — necessary for mappings to work.
set nocompatible

” Status line
set laststatus=0
set cmdheight=1
set nomodifiable ” Only in version 6.0
set readonly

” Syntax colouring — lines taken from syntax.txt discussion on colour xterms.
” See ‘:help color-xterm’. Use appropriate lines for your own set-up.
if has(”terminfo”)
set t_Co=16
set t_Sf=[3%p1%dm
set t_Sb=[4%p1%dm
else
set t_Co=16
set t_Sf=[3%dm
set t_Sb=[4%dm
endif
” My xterms have a navy-blue background, so I need this line too.
set background=dark
” Turn syntax on
syntax on

” Key bindings.
nmap b <C-B><C-G>
nmap q :q!<CR>
” To type the following line, type *two* C-V’s followed by two spaces. This
” is how you map the spacebar.
nmap ^V <C-F><C-G>

While searching for a link to back up my assertion in Samat’s blog that it is impossible to prove code bug-free, I ran across some amusing anecdotes here. Highlights follow:

1. Programmer produces code he believes is bug-free.
2. Product is tested. 20 bugs are found.
3. Programmer fixes 10 of the bugs and explains to the testing department that the other 10 aren’t really bugs.
4. Testing department finds that five of the fixes didn’t work and discovers 15 new bugs.
5. Repeat three times steps 3 and 4.
6. Due to marketing pressure and an extremely premature product announcement based on overly-optimistic programming schedule, the product is released.
7. Users find 137 new bugs.
8. Original programmer, having cashed his royalty check, is nowhere to be found.
9. Newly-assembled programming team fixes almost all of the 137 bugs, but introduce 456 new ones.
10. Original programmer sends underpaid testing department a postcard from Fiji. Entire testing department quits.
11. Company is bought in a hostile takeover by competitor using profits from their latest release, which had 783 bugs.
12. New CEO is brought in by board of directors. He hires a programmer to redo program from scratch.
13. Programmer produces code he believes is bug-free.

100 level 3 bugs in the source
100 level 3 bugs
You fix the code,
compile, and reload
104 level3 bugs in the source

104 level3 bugs in the source
104 level3 bugs
You fix the code,
compile, and reload
115 level3 bugs in the source…

This isn’t news, but in an attempt to share the fixes I’ve spent time finding, here is one from this site:

If you’re getting the long filename error when extracting Eclipse in Vista, you simply need to use a third party program instead, like WinRAR. The default Vista zip handler is just as crappy as it was in XP.

Reiser. A sensationalist article with interesting use of code. It is true that in very good, long, complicated code there is a huge measure of artistry that makes the the program something that is very close to the soul, or at least the subconscious. In any case, several things stood out.

The evidence against Reiser is very strong and is further served by his refusal to explain key details. The whole atmosphere around his family seems strange and on several points I do disagree with his philosophy, at least as they are represented in the article. For one, 6-year old boys should not be playing such realistically violent games as Battlefield Vietnam. I agree that some aspects of modern society seem to hinder more than help young male development, the removal of recess and using medication to treat ADD come to mind; however I do not believe it is appropriate to ‘train’ a 6-year old boy into manhood with violent video games. Just as we are horrified to hear stories of children being pressed into combat in war-torn countries, I was saddened at Reiser’s seeming attempt to ‘battle-harden’ his son. Regardless of your stance on violent video games it seems that Reiser was deliberately trying to mold his son out of his childhood innocence into a man. At 6! Let him keep his childhood! I admit that some ratings are an educated guess at best but come on! At least use them as a very strong guideline.  I think it was clear in this case that it was unhealthy for the child.

Is Reiser guilty? I don’t know. The evidence is very strong and I have a feeling he will be convicted (although as a friend pointed out, OJ walked despite even stronger evidence). The primary thing that can save him is if his ex-wife turns up alive, or at least dead with no injuries that can be connected to what police have found. If he is convicted its also most likely the end of ReiserFS as we know it. I’m not familiar with it but the concepts seem brilliant; at the least the loss of another option, another developing and continuing concept will leave open source/linux a little poorer. Hopefully its best concepts will live on as inspirations or implementations in other systems.

I also agree with Reiser that there is in inherent fear mechanism fueled by media frenzy whenever violent video games are mentioned (in reference to his own video game playing), although his perceived prejudice against pale nerds (as represented in the article at least) seems overblown. There was another case recently where this has been an issue, I will try to find the links and comment on them later. Video game violence is something I’m very interested in, especially from a societal standpoint. Like many things in the media these days the resulting frenzies cloud or even deliberately obscure real facts and details. Truth and honesty are sadly optional qualities next to personal agendas and self-interest.

To be clear, I believe that adults are capable of disconnecting themselves from the fantasy; or at least in making educated decisions about what is appropriate for them to play. After all there is much in this world that is considered unhealthy to consume and it is left to the adult to decide for themselves how to consume it. I agree that it is true that some things result in dangerous or extreme actions that may result in crimes, however, we tried banning alcohol and it didn’t work. Why should games be any different? Every time the government steps in to control some aspect of life we lose another freedom. Life is dangerous but is also more delicious the freer we are. Improve the educational system, make support groups, encourage parents to be more invested in their children, do something to help; but don’t remove our ability to choose. Actually parenting your children is especially important — after all, the lessons they learn now will ultimately shape how they run their adult lives. Take into consideration the mounting evidence that allowing people to choose for themselves often results in a safer environ. I admit the link is somewhat unrelated, but the concept is sound.

So at work we’re using these little Dymo label printers to do the smaller labeling jobs departments need. They are cheap and durable but don’t use standard formats or language — you can’t send plain text to them and hope to get anything useful. You have to use their software. Fortunately they provide you with code examples for accessing the coms and dlls and other junk the software installs when you first set up a printer. Oh, and you have to have a visible label printer in order to install the software.

So why am I having to code a custom app to use the little buggers you may ask? Why, it’s because we’re managing all of them out in their separate departments via a single computer! And we’re sending jobs to them from Unix! My job has been to take the label jobs, which are parsed and hacked up plain text files from student records, from Unix to my poor little Optiplex GX270 (not the one with Ubuntu) running WinXP and then send them out to the appropriate departments in a language and format the Dymo’s understand. Yes, I have to use WinXP because the people who will be operating and maintaining it are students who may or may not have computer experience and most very exceedingly likely will only know WinXP. So I found a program that will handle the queues, found the printers, modified the code the printers’ sdk came with to work for our purposes, made the crazy batch file that handles calling the program from the server, documented everything… . Now, I discover that the way the Dymo software knows a printer is available is from the local list of printers. Which may or may not reflect which computers at the various departments are actually powered on and receiving for their little counterparts. Yes, I had to connect the printers to an actual computer and then share it over the network. If you can find a label printer that is stand-alone network capable for under $100 I will cry in frustration cuz I already looked.

So I have to call net print from my batch file and sort through the resulting vomit to find error code 1231 or whatever. Yeah, the batch file. Since there’s no way to easily configure my program that prints to the printers everything is being handled in the batch file. The alternative is to hard code crap into my program. I suppose I could make an editable text file and call it from the program for settings, but gmeh. I’ve come this far. Besides, batch files have known syntax and breaking and such so that would be one less thing I’d have to document or watch out for.

If you do know a better way, or a relatively easy way, let me know. Right now the only thing the program does is take the printer name, a path for the filename, and an optional switch for landscape or portrait labels. If you get the printer name wrong it spits out a list of acceptable printers. It returns 0 in success, 1 for bad printer name (and a couple of other things irrelevant to this discussion), and 2 for failure to print. Which I haven’t been able to generate yet. The batch file attempts to handle everything else. It will keep trying until it times out and then it will send an email to the operator outlining the failure and which file it was so the job can be repeated. The server leaves a copy of the job in the queue directory (for reprinting) so the batch file also deletes anything older than ten days to save disc space. Am I an idiot? I’m not skerred, just give it to me straight.

Update: I also tried finding the printers in my program but I couldn’t for the life of me figure out how. I think the main problem is that I have to edit and compile based on an existing Visual Studio project and therefore have no control over the headers (the one I want is already included but I cannot use the functions I need). So much of the coding that could go into the program is out of my inexperienced hands. Fortunately it turns out that net print has return codes so adding it to the batch file was much simpler. Now the batch file checks for the remote printer for about a day, and if it fails it sends an email to the operator with enough info for the job to be repeated.

Unicon is my first high-level programming language. Having used C for most everything up till now, the experience is interesting. Things I used to have to always do in C I cannot do at all in Unicon since it’s all taken care of automagically. For example, malloc is a thing of the past. Unicon even supports dynamic lists and strings, I can concatenate to my heart’s content (or to memory’s end).

However, the drawback is that I have to think in a more abstract way, something I’m not quite used to. I was especially partial to Assembly since it was very straightforward and it was possible to memorize all the commands. From there, solving a problem was an enjoyable creative process since one knew all the rules. In Unicon there are hundreds more rules to know and solving a problem requires longer and more careful consideration, and elegance and style matter. Multiple solutions mean that sometimes one way effects other aspects of the program so thinking ahead matters too. On the other hand, the problems one tackles in Unicon are far more complex that in Assembly, and it could take one line to do something in Unicon that would take 15 in Assembly.

Of course if you program you already knew all this, so if you’re unfamiliar with Unicon here are some of the highlights :

1. Easy 2D and even 3D integration. Unicon includes a large number of manipulable primitives, one can draw scenes with relative ease. It’s designed for building games, so this is a must.
2. Die rolls are built in.
3. Windowing. I’m still learning how events are handled, but Unicon has seemingly robust window handling. The binaries include a simple gui designer that makes it easy to generate interfaces.
4. Dynamic memory management. So far, I’ve not had to allocate, deallocate, garbage collect, or any of that.
5. Easy file management (just remember to close).
6. Easy string and list manipulation.
7. Based on Icon, which translates to 99% compatibility. That being said, I’ve already discovered a couple functions from Icon that don’t work in Unicon, however it’s mainly a matter of naming convention or simplification.
8. Automatic typing of variables. Variable is type of whatever you assign it.
9. Lots of unary operators.
10. Automatic null arguments for functions.  Everything is automatically null so there’s no garbage memory.

Setting up the binaries is relatively easy, especially in Windows, but the builds are old. Newer builds are available but haven’t been posted on the site, so I’ve attached my Linux source. Instructions are pretty straightforward, if you can’t get it to work try rm src/*/*.o and/or changing -O to -g in the top-level makedefs. The zip file is annoying, sorry about that, I never bothered to re-archive it.