Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Bot Land, a game where you fight others via code (bot.land)
444 points by Adam13531 on Sept 29, 2019 | hide | past | favorite | 87 comments

Hey all,

I've been working full-time on a game called Bot Land for the last four years, and it just launched: https://bot.land/ In the game, you can make bots with a Scratch-like interface called Blockly or use a subset of JavaScript. Either way, bots are fully automated.

Here are some screenshots of the game: https://imgur.com/a/Rs2OWFg

Almost all of the development process (~5000 hours) was streamed live on Twitch: https://www.youtube.com/watch?v=RS5Va996Exg&list=PLSM9PbIe-9...

It's free to play, so please try it and tell me what you think!

I’ve been wanting to play/build this exact game for years!

Congrats on launching. After four years that’sa huge accomplishment.

One problem I'm having, when manually editing the code, is using Math functions. Ideally, I'd like to calculate the square root of an expression, but whenever I try to use Math functions I get the error "The Callee's type is MemberExpression which is not allowed".

Any advice?

Hey, so one thing to keep in mind is that BotLandScript is a subset of JavaScript. Member expressions (e.g. "entity.life") are only allowed in extremely narrow scenarios.

Second, all available functions are accessible by clicking the question mark that shows in the script editor. Some Math functions like "abs", "floor", and "ceil" are exposed to users, but "sqrt" is not. This means that you would have to either write your own implementation or convince me to add it to the game. When I hear "sqrt" in Bot Land, I assume you want to find the distance between two objects, in which case you may be able to use the built-in "getDistanceTo" function. It can either take in an entity or two arguments representing (X, Y) coordinates and returns the Manhattan distance between the bot and the target.

If there's another scenario where you'd like the square root, I'd be glad to consider either accommodating the scenario or just adding access to Math.sqrt directly. If we go that route though, you wouldn't have access to whatever results from it before ~Tuesday of this week at the earliest.

What's the rationale for excluding sqrt? And, do you allow Math.pow (since that's the same thing)?

> What's the rationale for excluding sqrt?

That's a great question, and one with a non-obvious answer! For any potential function in Bot Land, including utility functions like Math.sqrt and Math.pow, I purposely wanted to provide as few as possible until someone expressed a desire for them, that way I could ask about the scenario they were trying to accomplish (as you saw in this thread). From there, I try to figure out if there's an even easier way of supporting their scenario so that everyone (i.e. even Blockly users) could enjoy it.

For example, if we find out that Math.sqrt was needed solely to find distance and that getDistanceTo was sufficient, then I could treat it as an education problem on behalf of Bot Land. In other words, I may want to include more snippets that use "getDistanceTo" or maybe make more tutorial content around that function.

Another philosophy would be to just add whatever functions I can think of and see what users do with it. That would probably have worked too; I'm not saying mine is better, it's just what I was thinking.

Thanks for the reply and the explanation. You're right that I was trying to calculate distance.

One thing that confused me though, when I was trying to figure out how to square root is that one of your code blocks seems to translate into Math.pow, which made me think that I could use it too.

For the same reason you gave, equality between block-based and text-based programmers, I'd suggest any function that the blocks translate into should be allowed.

If it would help you I can find the specific block I'm referring to latter today.

> You're right that I was trying to calculate distance.

Oh good! Thanks for following up on that.

> one of your code blocks seems to translate into Math.pow, which made me think that I could use it too.

The way that the blocks work is that they convert into JavaScript under the hood. This means that there's no way to do something through the blocks that you can't do in code (although the converse isn't necessarily true). This means that if there is a block that converts to Math.pow, it's broken, so I'd definitely like to know about that if you find it.

Got to say, I love your approach here.

> Another philosophy would be to just add whatever functions I can think of and see what users do with it.l

Well, the middle ground could be allowing the basic PEMDAS mathematic functions and their inverses, just to allow common manipulations

Awesome. This one goes in my collection of programming games. Thank you. I hope it gains traction.

The visuals look great, will play it in a bit. And wow 5000 hours-- incredible!

Congrats, Adam! I've been following your stream for years, and I'm always impressed and inspired by your programmers' log and note taking.

How the hell did you have the ability to drop 5000 hours on this?

I think the other comments were interpreting this as "where did you find the time?" rather than "where did you find the other resources to support that time?" (e.g. money and motivation). I've been maintaining a huge FAQ since I've gotten these questions a lot over the last four years, and this one in particular should answer the question: https://github.com/Adam13531/BotLand/wiki/FAQ#how-are-you-af...

I linked that rather than copy/pasting it because you may find other answers interesting on that page. :)

EDIT: I should clarify that 5000 hours is roughly the number of hours that I streamed, not the number of hours that I worked on Bot Land. There were way more hours spent in the background doing all sorts of things (including coding), but the majority of the actual development was on-stream.

I've been in the industry 8 years and worked on maybe a half dozen games but none of the projects was >400 hours.

I don't see how to make the commitment to spend multiple years without like an income coming from it and a team to work with. By myself it just becomes unbearable after a while.

I want to make a living off games. I'm almost 30 and things aren't getting easier.

Ahh, I see what you're saying. I think the answer to that is that I had set up my entire life around Bot Land. It was my full-time job, I had the stream for accountability, and I'd dumped my savings into the production of the game. I couldn't just walk away from that. Plus, it's been a lifetime goal of mine to make a successful game, and I'd always said that the one true way to definitively fail was to not finish Bot Land (I hadn't had such a definition of failure for other projects).

Without getting too rambly/preachy here, I'll say this: there are lots of things in my life where I think, "I'd like to do that, so I'll get to it eventually". After a long enough amount of time, you realize that "eventually" doesn't just hit you in the face—you have to carve out time for the things that you want to do. At the end of my last traditional job, I realized that my career goals weren't being met. Primarily, I wasn't learning. I talked to my wife about what I should do, and she suggested that I productize Bot Land. That's when I had to realize that the concept of "eventually" was peaking through the cracks probably trying to hit me in the face, and I had enough momentum to just go with it.

You mentioned that you're almost 30 and that things aren't getting easier. I can't say that they will get easier. What may help you take one of your many games to completion is simply scoping it down way further than you think you should. Cut out multiplayer, cut out innovative AI, cut out randomly generated dungeons, and just focus on a couple of solid aspects.

I don't know if any of that advice resonates with you, but if not, perhaps defining what barriers you feel like you've hit would help identify what patterns may be stopping you.

The barriers i repeatedly hit seem to be due to my lack of competence or inexperience at the tools im using.

I feel like most people have that much time assuming they cut out time wasting things like Facebook/YouTube/etc.

Self-care and wasting time or not equivalent things.

True. But for a lot of people, Facebook is the opposite of "self care"...

5000 hours is a long time, but you can do it in 5 years with under 3 hours per day. That's within most people's means though you would have to make sacrifices.

Actually about 4 hrs/day...when figuring out approx billable hours in a year a good estimate is 50 weeks @ 40 hrs/week, so 2000 hrs/year full time (8 hrs/day) or 1000 hrs/week half time (4 hrs/day)...

Where did your EULA come from? Seems a bit brutal :)

A combination of a lawyer and some minor touch-ups I've made over the years as things have changed.

Do you mind clarifying what you find brutal? I am very far from being a lawyer, so I just wanted to make sure I'm protected, which is why I contracted the work out in the first place.

Why do you have the videos unlisted?

This is amazing, congrats.

> Why do you have the videos unlisted?

Two reasons:

1. By making them public, they show up for all of my subscribers on YouTube, but I assume most subscribers on YouTube want to watch "major" update videos like announcements, recaps, etc. 2. I didn't want a new person coming to the channel to get lost in the ~800 development videos rather than seeing the important ones.

> This is amazing, congrats.

Thank you! It's surreal to have spent this long developing something and finally launch it.

What about the two channel system that is so common on youtube? One for major productions, the other for faster-release-but-narrow-appeal videos like behind the scenes or development progress? That way super fans can follow the bonus channel without having to go through your website to find videos.

That probably could have worked, although I don't think I was the best at advertising things. :( As a streamer, I felt like one of the most difficult things to do was to alert people about news/changes for basically anything (my vacation schedule, a bug, the current task I was working on, etc.).

Now that the game has launched and my streaming schedule is up in the air, I don't know if I'll be changing anything with the YouTube channel. It's been around for so long now, and I don't think many people want to watch the past VODs anyway beyond skimming to see what's changed in that amount of time.

really like the scratch style snapping chunks of logic! nice trailer too

Nice. I had a idea a few years ago, but never got around to working on it. Let me know if you ever want to buy the aifighter.com domain off of me.

Cool, a modern day Core War!


Came here to make the same comment.

I also played - I guess back in the late 90's, with an autonomous Asteroids-like game where you'd write code to control your "space ship" that had "radar" to show you nearby opponents and had pretty much just Asteroid-like controls, which ran as a networked screensaver. It came out of a bunch of people working at Canon - a friend of mine worked there writing printer drivers...)

(Now I'm wondering what you'd get if you let loose inside a modern device, like a cellphone for example, with two opposing bits of fighting code with full root access, but strictly equally time and resource shared? Would Core War even make sense any more with Gigs of ram, hundreds of gigs of flash, asnf stuff like wifi and bluetooth?)

You probably played crobots: https://tpoindex.github.io/crobots/ It was quite popular at that time!

Thanks I just looked into this, I've never heard of it before.

Here is more on its language http://vyznev.net/corewar/guide.html

This is nuts!

Congratulations, looks awesome.

I had an idea for similar kind of game where you are a wizard and program your own spells. The mechanics would be inspired by how magic works in The Dresden Files[0], where you draw in magical energy, transform it and then use it in some way. Maybe one could connect it with linear logic and programming.

[0] https://en.wikipedia.org/wiki/The_Dresden_Files

That reminds me of 'Midnight at the Well of Souls' by Jack L. Chalker:


the minecraft mod Psi works just like that https://psi.vazkii.net/

CodeSpells combines magic and programming. It gathered quite a bit of hype early on, but I didn't really follow it's progress since.

Shame about the loot boxes.


Are you talking about the monetization model, or is something broken with opening salvage packs?

I assume the former. The goal was always for Bot Land to be free-to-play, and there were many different approaches to monetization discussed throughout development. In the end, I settled on the one you see in-game now, but the tenets were always the same: https://share.bot.land/monetization.html

For reference, salvage packs are the single monetization point of Bot Land. I'm not planning to include advertisements, and I don't want to do subscriptions or anything. Salvage packs contain only cosmetic items, and the rarities are included up-front (which I would have done even if certain platforms like Apple didn't require that).

You can play for free and obtain every item in the game. The game took not only a substantial amount of time to develop (4 years), but also a huge chunk of my savings that I'd built up from my past jobs. I do need some way to make money, and I don't think people would give a game like Bot Land a shot if it weren't free.

I appreciate your intent to make the game is not P2W and use IAP only for cosmetic items.

However, why not allow users to purchase those items directly? The use of "loot boxes", even for purely cosmetic items, can create dangerous and unhealthy addictive behaviors as people buy again and again trying to get the item they want. A number of countries [0] have made "loot boxes" illegal or placed restrictions on them, so this isn't a purely humanitarian concern.

Additionally, since you can sell those cosmetic items for "Botcoin, which you can then use to purchase anything else in the game." doesn't that end up giving your game P2W functionality? It seems like moving to direct purchases of cosmetic items would eliminate the need for re-sale and prevent concerns about your game being P2W.

[0] https://en.wikipedia.org/wiki/Loot_box#Regulation_and_legisl...

I think that with the current climate of gaming, this is a minefield to answer, and I don't think it does me any good to answer it. It feels like a "gotcha, I knew it!" kind of question rather than an honest inquiry into how Bot Land's monetization came to be. However, I've always wanted to be transparent where I can, and I'm sure some people are wondering about this, so here's my thinking that led to cosmetic-only salvage packs:

For any game to be monetarily viable, it needs to pull in an average of $X for each player. Any free-to-play game (including freemium but not including something like shareware) skews how that average is formed because typical free-to-play games are lucky to have 5% of their playerbase spending any amount of money. This means that a large portion of the playerbase is actively costing money. Thus, they tend to rely on whales to spend huge amounts of money to compensate for all of the people spending nothing (or next to nothing). For this to work, the spending ceiling needs to allow for purchases of this size to even take place.

I wrote a salvage-pack modeler that would compute how many items of each rarity you would get by opening X packs, and how much Botcoin you could convert those items into should you want to sell them. I determined that at the highest number of packs that you can buy (70 packs, i.e. $100), you unlock nearly every cosmetic item in the game except for the rarest ones (which you actually unlock very few of). However, the average Botcoin value is enough for you to purchase several specific items that you may have wanted. I never wanted players to spend $100 for a single item and not be able to get it somehow.

By allowing direct real-money purchases of individual cosmetic items, I'd have to skew prices so heavily to make the same average revenue that I don't think players would want to purchase them. Also, there wouldn't be the fun or mystery of opening a salvage pack. I know what that sounds like. I even did a parody of EA's famous words for April Fool's last year (https://youtu.be/cCmj2hKbWeQ). But I can unravel a bit of this at least:

1. I'm not a businessperson. I don't know much about the psychology behind sales and how to make various business models work.

2. I have somewhat conflicting goals with Bot Land. I want the game to be free-to-play, I don't want to exploit users, and I want to make a living off of the game.

3. The model which I know can fit those goals is the salvage-pack system that you see in the game now.

4. The payment model is not set in stone. So far, Bot Land has grossed significantly less than $1000, and a large portion of that is from users who already had accounts before the game even launched, likely indicating that those users wanted to support me moreso than get something in the game. It's easy to be ethical until challenged, but I'd like to think that if I found users spending ludicrous amounts of money in the game, I'd do something to curb that behavior.

5. There's been an absurd amount of work just getting to launch, and I'm not positive that I can even continue Bot Land's development beyond 2019 without greatly changing the plans my wife and I made (more about that here: https://docs.google.com/document/d/1aCE4s5UvVLH7dHuO1OA2m5Sw...). This bullet point's relevance is that I have so few resources to even be able to focus on the game alone right now, let alone any other issues that may pop up.

You may take this as incoherent rambling or mental gymnastics to be able to justify what is typically a predatory business model in gaming. I will be reevaluating as time passes to figure out whether something needs to be changed. I highly doubt Bot Land will be successful enough for me to even be able to make those decisions though.

> doesn't that end up giving your game P2W functionality?

Functional items are supposed to be balanced such that they can be situationally better than other items of the same category, but not outright better. Arguably, being situationally better allows you to be outright better the more situations you come across, but I think that fully exploring the nuances of this would be a much larger conversation when I think that the heart of your questions is around loot boxes. In short: I don't think that even if you could purchase Botcoin directly would the game be pay-to-win.

Finally, one last note that's not the most salient: I never really intended for salvage packs to be convertible into the other items in Bot Land, just cosmetic items. The one monetization point that I've written down for the future is to introduce hardware coupons, that way you can directly purchase functional items. I don't know when this would happen though, if ever.


There are parts in this post, specifically with the five points I enumerated, whose connection to the original questions and even to my own supporting points may not be clear to a reader. I took time and care to type this, but I'm sure that I could have made things clearer. I don't mind participating in an honest discussion about this or other aspects of the game. I'll likely stream tomorrow (Monday) if you want to ask me in-person (although I can't guarantee that I have the required amount of time to devote to the conversation).

Thanks for taking the time to reply in such a thoughtful manner!

> I determined that at the highest number of packs that you can buy (70 packs, i.e. $100), you unlock nearly every cosmetic item in the game except for the rarest ones (which you actually unlock very few of).

It is these rare items that tend to drive the addictive behavior behind loot boxes. Is there a particular reason you feel like you want to have these rare items be so hard to obtain?

If you don't mind me asking, what is your target revenue per player? $100 seems pretty high...

> By allowing direct real-money purchases of individual cosmetic items, I'd have to skew prices so heavily to make the same average revenue that I don't think players would want to purchase them. Also, there wouldn't be the fun or mystery of opening a salvage pack.

By using a randomized loot box (especially one without published odds) you are effectively tricking players into paying more than they would otherwise wanted to. This seems abusive to me.

> For any game to be monetarily viable, it needs to pull in an average of $X for each player. Any free-to-play game (including freemium but not including something like shareware) skews how that average is formed because typical free-to-play games are lucky to have 5% of their playerbase spending any amount of money.

There are monetization choices that I think do a good job of balancing that distribution. Loot boxes is not one of them, nor is any mechanism by which players are given the option to repeatedly purchase consumable items that aid progress/grinding. This can actually discourage low-value players from paying since they can't pay enough to make a real difference.

I personally think that one-time purchases for ($5-20) that give a permanent bonus to grinding (10-20% boost to rewards earned by playing) are actually more player friendly. They encourage more players to spend a little bit of money on the game while reassuring them that you will not keep raising the bar and pushing for more money to keep progressing. I will almost always buy a package like this (as long as it is reasonably priced.)

> a large portion of that is from users who already had accounts before the game even launched, likely indicating that those users wanted to support me moreso than get something in the game.

Don't under-estimate this. The desire by fans and serious players to support your game and keep you afloat can be a significant factor in the decision to spend money on your game. Predatory monetization can destroy that goodwill, which will in turn make you more dependent on your whales. When I feel like a game is pushing me to spend money, it decreases my respect for the game and the developers and reduces my desire to support them.

> It's easy to be ethical until challenged, but I'd like to think that if I found users spending ludicrous amounts of money in the game, I'd do something to curb that behavior.

An ethical approach to whales can significantly mitigate the predatory effects of monetization. However, it is also a fairly invisible monetization technique that doesn't do as much to engender goodwill across your player base.

> In short: I don't think that even if you could purchase Botcoin directly would the game be pay-to-win.

P2W comes in a couple of different flavors and doesn't really mean just one thing. There are some heavily P2W games where F2P players struggle to compete and spending money becomes necessary past a certain point of the game. (I think of these games as P2W PvP) (Good matchmaking can push a game that would otherwise fall into this category down the the one below.)

There are also a lot of P2W games that don't disadvantage F2P players directly, but just allow players to pay to progress faster. Any games that allows real money to directly convert to a fully unlocked/upgraded account falls solidly into this category. It sounds like your game is one of these. (I think of these games as P2W PvE) (Games with synchronized server starts that would otherwise fall into this category can easily fall into the above category.)

Then there is the whole spectrum of games that provide some in-game advantage, but limit some forms of progression to gample/grinding.

Finally, there are the games that are truly not at all P2W, where real money purchases can only buy cosmetic items and there is no mechanism to provide any in-game advantage to paying players.

I don't like P2W PvP games. I don't mind P2W PvE games, especially when the montization is well done and the money/grinding ratio is high ($1 to save 100hrs of grinding is unacceptable, $100 to save 1hr of grinding is great).

One of my favorite indy P2W PvE games is Hades' Star which is almost entirely unlockable by spending money, but has an extremely high money/grinding exchange. The game also offers a very reasonable permanent package that provides a 10% boost to a couple of specific things.

I hope your stint on HN gives you a boost and helps you reach a sustainable level of income from Bot Land. Good luck!

Should gambling be illegal? If you believe in human freedom then no, it shouldn't, so maybe you're barking up the wrong tree here. Let's focus on the fact that A) this practice is generally turning the AAA industry into grindy garbage and B) this stuff is targeted towards kids

Dude spent 5,000 hours just on the live-streaming code on this project and you are worried about loot boxes?

Don't buy one.

They have to make money somehow. With a free to play game there aren't that many working monetization strategies to choose from.

Sounds interesting, but didn't work for me: it got stuck on the "bots are busy assembling the game". There is no indicator of progress, no errors, no logs. It's a good idea to add a feedback button that would upload logs on demand. I'm using Firefox Android with uBO. Only GA was blocked, but there's no way I'm enabling it.

Do you mind trying either the app or a different browser? I don't know why it wouldn't load at all. I agree that there should be some form of errors, logs, etc. I believe that will eventually happen if the game makes enough money to continue development. Launching on four platforms was only possible due to being able to thinly wrap the site, and unfortunately, now there are a lot of other important tasks to tackle that will probably happen before this. :(

Nice. RobotWar for a new generation.

[0] https://en.wikipedia.org/wiki/RobotWar

I grew up playing RoboWar for the Mac... looks like it was based on the one you linked. Guess every generation has theirs!

Woah congrats. I have watched a stream or two of yours and it looks really interesting.

I really liked your marketing document and think others might find it really useful, hope you don't mind me linking it here: https://docs.google.com/document/d/1TiSmB7IaWpIf_7cUWzk7GAtR...

(let me know if you are not ok with that and I'll edit the comment)


The marketing document is fine to link. It hasn't really been updated much, but then again, I'm going to try more ideas from it now that the game has launched and the fires have died down a bit.

Congrats man!

The Google code blocks integration is quite nice. as a developer, I wouldnt want to play this though.

You'd probably like Screeps then. It's aimed for developers, but it is JavaScript (Con to me, pro to others). Essentially a bot coding MMO.

You can code in a subset of JavaScript too. Even with that in the game, I find Blockly a little bit more usable given the current feature-set in Bot Land right now (mostly the existence of snippets and premade scripts in Blockly). It's nice to get large chunks of code that you can just drag around.

Alternatively, maybe you were referring to some other part of the game when you said you wouldn't want to play this, in which case I'd love to hear your feedback!

I think they were just saying that since they code for a living, they don't want to play games that are also code.

Oh yeah, you may be right. I took it to mean "given that I know code, I wouldn't want to use a drag-and-drop interface". I've had a few pieces of feedback like that before, and sometimes even pointing out the existing coding interface doesn't please those people. To each their own, for sure!

Screeps is also a cool automation game like that (and 100% FOSS!). (Judging from the looks of this it's a very different game though, so try both!)

Enjoying it so far, small bug report:

Loadout weapon drag window can get permanently stuck on the screen, requiring a refresh. See here https://i.imgur.com/cL4x2Vk.png

Steps to reproduce: Go to Edit Blueprint Window. Start dragging a window. Press backspace to go back to the game field. The window is now permanently stuck.

Thank you for reporting this! I say this with no sarcasm: I think this is a completely unique bug.

Also, it's better that I get reports than not, but if you've got any future bug reports or feature suggestions, there is an in-game bug reporter that you can use via the button at the upper right of most screens in the game.

Bit of a shameless plug I guess but it's especially fun for me to take a look at this because I worked on a pretty similar game about 3-4 years ago making bots battle each other for the students I was tutoring. Right after that I made another game that was a more single player experience and decided it would be fun to learn OpenGL so I obviously got a lot less done. This definitely looks a lot more polished than what I produced and I hope it turns out well! For those interested in the projects I worked on, you may find them here: https://gamejolt.com/games/masjien/80426 https://gamejolt.com/games/proxy/221194

On iOS. Signup Email input should be of type email. Password length feedback should be inline - not after typing pw twice and submitting.

Gave up at the ‘place two bots’ stage. Couldn’t work out how to place a bot. Love the idea. Needs UX love!

Thank you for the feedback! I've added all of this to the list of things to look into.

I've heard sporadic reports about not being able to place bots on iOS, and I'm still not sure if it's a bug (I only have my wife's iPhone to test on and it works there, but I know it's not representative of everyone's experience) or unclear instructions. This is the flow that should be happening: https://i.imgur.com/EalfL7J.png

The steps, as text:

1. Open your bot selector

2. Select a bot

3. Tap in the Arena to place one

Is that flow showing up for you? Are you on an iPad or something that shows the "desktop" view of things (i.e. the bot selector is expanded by default)?

Alternatively, in that mobile tutorial, should you try dragging the bots from the selector, that should work too, although it tends to take a little bit longer to deploy multiple bots that way.

Same here, android, unable to figure out how to place a bot. Everything that made sense (select robot, tap in arena, tap bottom bot name and tap arena. Even double tapping didnt work. Gave up too.

Update. Got two bots (at least based on count), I have no clue how, started fight, but there is nothing to be seen, just black background. Battle finished and again everything black except the menues.

I am fan of bot wars and would love to play it, but this is just too frustrating to use.

Thank you too for the feedback! Would you mind either sharing the following or submitting a bug directly through the game? If you'd like to share this outside of the game but still privately, my Discord ID is Adam13531#3531.

• What device are you on?

• Are you using the app or the site?

• If you're on the site, what browser are you using, and have you pinned the site to your home screen, or is the URL bar showing?

• What OS version are you on?

The mobile versions of the game are thin wrappers around the web version (https://play.bot.land/), and there weren't very many mobile users before launch, so bugs certainly made it through the testing process.

Yeah I suspected it is just a web game. I am so sorry you have decided to make it web based. I have webgl disabled as it is common for websites to use it for fingerprinting and I will not enable it.

A for loop seems to be supported but there is no documentation on how to use it. I'm getting the following error when trying to use a loop index i:

Your for loop's body section is not legal. You can only use literals as a refinement on objects (when trying to index into an array[i] returned from findEntities).

I looks like only the two built-in variables "array1" and "array2" can be used like actual arrays. someBot = someArray[0]; will not work but array1 = someArray; someBot = array1[0]; will.

If you like these kind of games I recommend this one:


How was streaming development of the game? Are you concerned about your source being visible? Or feeling like you are being watched or judged while coding? When did you start streaming development? I'm worried my side project is too early in development to stream... Did you end up having to spend a bunch of time making extra stuff for your audience, like the faq you linked?

Oh boy, do I have a huge answer for you!

> How was streaming development of the game?

It was great! People ask me a lot about what I'd change if I could go back in time or start again, and while I would certainly change some technological decisions, I would not change the streaming aspect. There are so many great things that have happened over the years, and I feel like the culmination of all of it can be summed up in this launch-day highlight I took: https://www.twitch.tv/videos/486320333

That one video felt like it made the whole journey worth it.

> Are you concerned about your source being visible?

I wasn't super concerned for a few reasons:

• Code is copyrighted as soon as you write it, so if someone were to take it, I would be able to take legal action against them.

• Perhaps more importantly, I didn't think anyone would want to copy Bot Land's code. Automation games are growing as a genre, but rarely will they include scripting as the main component. To my knowledge, there wasn't a scripting-based game ~4-5 years ago that proved that they could be successful.

• See the next answer...

> Or feeling like you are being watched or judged while coding?

Impostor Syndrome runs rampant with coders. Did I feel like I'd be judged? Yes. Was I? 100% yes (by some people). Were some people mean about it? You'd better believe it. However, I started streaming for accountability and marketing, and I wouldn't have gotten those benefits if I didn't stick with it.

Also, remember that as a streamer, you control who gets to stay around in your chat. People are entitled to their own opinions, but I think the #1 rule as a streamer is that of entertainment: the show must go on. As a programming streamer, this doesn't mean that you have to be entertaining all of the time since you have to be productive while streaming. Either way, in order to keep the show moving, you have to be in a good enough mood to do things. When a troll gets in the way of that, you purge/timeout/ban and move on.

If someone is judging you for your coding style or standards in an unconstructive way, you have tools at your disposal to handle it.

> When did you start streaming development?

I started on September 8th, 2015. I believe I had one follower on Twitch at the time, a real-life friend of mine.

> I'm worried my side project is too early in development to stream...

I don't think anything is really off-limits for a development stream. I've spent entire days streaming myself milling around entirely in a text document (i.e. not even code). I've spent days writing automated tests, triaging bugs, debugging, etc.

> Did you end up having to spend a bunch of time making extra stuff for your audience, like the faq you linked?

Yes. If I had to estimate how many full work days I've put into just streaming-related tasks, I would say it's about 10-15. Granted, you can just start with practically no preparation and see where it takes you, but I don't advise doing that. My advice on streaming hasn't changed too much in the past years; it's mostly encapsulated here: https://blog.bot.land/2016/10/being-a-development-streamer-o...

Streaming is just as saturated as games are nowadays. There are a million streamers trying to make it big who "don't care about the numbers, they're just doing it for themselves". If you want to get started, I'd say that you should come up with a goal pretty early on, and that goal should not be to earn money from the stream. I won't share exact numbers, but I make less than minimum wage as a streamer. As a developer at a traditional job (in the U.S.), I'd make plenty of money.

I'd be happy to answer more questions you may have, although I think you'll have to see for yourself how streaming goes in order to consider whether it's worth it. And be ready to come to the conclusion that it's not worth it since it can affect your productivity and possibly your mood/energy.

A little more context -- I stream playing games (because it's entertaining to have people drop by). I have no interest in being a streamer for a career, so I don't feel I need to compete with anyone, but if folks found my side project interesting and dropped by to chat I think that could be a nice break.

Do you find it difficult to converse with folks while programming? How do you maintain flow state while streaming? (Maybe this is just a skill you pick up?)

> Do you find it difficult to converse with folks while programming? How do you maintain flow state while streaming? (Maybe this is just a skill you pick up?)

This comic (which you've probably seen!) sums up context-switching as a programmer: https://heeris.id.au/2013/this-is-why-you-shouldnt-interrupt...

Imagine that but multiplied by 100. Every 3 seconds, someone is saying something that you can see out of the periphery of your vision. How difficult it is to interact is up to you; you can hold off on chat and handle it every X minutes, you can let them interact with themselves, or you can interrupt your flow every few seconds.

I try to take extensive notes about what I'm doing at any given moment so that I can always pop the context back off the stack and resume my work. It's a skill you'll hone as you go.

Full on programming as a mechanic seems overwhelming, but I like the idea of automation in strategy games

I’ll give your game a try, maybe I’ll be proved wrong

There are three levels of scripting you can do:

* None - just ignore the scripting and let the default AI take care of everything. There are some items that can't be used this way, e.g. Teleport and EMP, because the default AI wouldn't know how you'd want to use them. * Blockly - use the block-based scripting to control your bots. This is what I suggest as the step for newcomers who still want to customize their bots' behavior. * BotLandScript - this is a subset of JavaScript that I only recommend for people who already know how to code.

Blockly and BotLandScript should have feature parity (not that they do right now, but the differences are somewhat minor given the target audiences for each), so you don't necessarily have to code (or even know how to code) to play Bot Land. Then again, scripting will certainly give you an advantage.

For the record, the "figure it out" block is what kept me in the game instead of trying it once and moving on. A lot of these strategy coding games like to throw you in the fray and start you off with a blank slate.

I know how to code, but I don't know how to win a game of BotLand. Having the ability to just start a match right away is killer.

I played a dozen or so matches so far. Really excited to see how development progresses. Also surprised at how well the mobile app runs considering it's a web wrapper (but it doesn't feel like most web wrappers do!)

Best of luck, and thanks for the game!

Blockly (or similar methods like Blueprints in Unreal & Simulink) is how most people will program in the future.

You’re going to get negativity from “developers” about using Blockly, but unless they’re your target audience don’t let it get to you.

Lots of EE’s use graphical tools to layout the circuit boards anyway... so text is really just an unnecessary abstraction over schematics & you’re bringing programming back to its roots :-)

Want to make a timeframe for that prediction? I'd like to record it... While I agree to ignore the criticisms from people who won't even play the game (which includes me, so I haven't commented on it, even though I've played a number of variants on this during high school) I think your prediction here is a stretch. First a quote to consider:

"Linux supports the notion of a command line or a shell for the same reason that only children read books with only pictures in them. Language, be it English or something else, is the only tool flexible enough to accomplish a sufficiently broad range of tasks."

-- Bill Garrett

Now a point about EEs/CEs... it's way nicer to write:

    module dff (

    output q;
    reg q;
    input d;
    input clk;

    always @(posedge clk) begin:
        q <= d;

than to actually lay out and route a flip flop. Of course you're going to simulate such modules, and use graphical tools (the least of which is just to picture signal waveforms), this isn't to say graphical tools are going away or are pointless, but I would readily take the other side on a bet about your prediction since I don't think programming will even in 100 years be something people think of as manipulating graphical elements rather than typing/chording expressions satisfying some grammar/syntax we'd call a language.

Has anyone ever played Robot Odyssey? It was one of most beloved games as a child. It's somewhat of a precursor to this project.

Looks awesome, I signed up! I'll try it out over the next few days. Looks like it could have a lot of potential

I accidentally removed my bot's head/torso/feet during the tutorial and could not add them back.

I always loved the classic PRobots and later CRobots, its a great way to learn programming.

Very interesting. It looks just fun at first but become more and more addicting. Good job!

If you like the genre you should also check Gladiabots on Steam: https://store.steampowered.com/app/871930/Gladiabots/

looks like a great game! Thanks for working on it!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact