I'd already made plans for today, so I wasn't able to make this one, but I hope it happens again sometime in the future. I'd definitely attend.

I like the premise (if done somewhat akin to Grindhouse), but I think that the omnipresence of  indie bundles have really desensitized people to this kind of idea. If a pair of games were notable enough as a unit to pierce through that boredom that many critics have towards "bundles", then I imagine they'd be sufficiently good that they'd do well as stand alone games. I think that thematically linked game releases would do well within the indie community itself, but wouldn't necessarily impact their reception by wider audiences.

I generally use Jekyll (https://github.com/mojombo/jekyll) to generate a static site. I find that, even for a blog, dynamic content is so rare, that it's wasteful to have a processing step on every hit of the page, and any database accesses.

There are exceptions to this, but I suspect that most game sites will be static enough that using this imposes no greater burden, and will be much more capable of handling traffic, or being hosted anywhere.

I live in SF, and I'd definitely come to these. Most of the gamedev meetups I've been to in the city don't really focus on either getting stuff done, or showing stuff off. Something like what you're proposing would be a lot more substantial and useful.

Hi, my name's Stephen, and I've recently started doing game development in my free time. I've spent the last year and a half working at an iOS game dev shop doing server-side development. I decided that I wanted to get more involved in the client end of game development, and started making my own stuff.

I've mostly been working with Lua and LOVE 2D, and am thinking about learning some Actionscript to be able to do flash development, so that people can play my games without downloading them.

I put all of my stuff here: http://evilrobotstuff.com

I don't know what platform you're developing on or for, but I know that the last time I messed with it, there wasn't any support for 64-bit versions of Python, which was quite a hassle. That pretty much killed it for me.

If sound isn't critical (unlikely, but maybe), what I do is just make the game write the buffer to disk every other frame. Then I used ffmpeg to string the screenshots together into a video.

Put it on a different subdomain. As otherwise cookies from your domain will be hijackable - which could compromise any login systems you have on the same domain.

Even if it were on a different subdomain, it'd still have access to cookies written for ".domain.tld". I'd suggest a different domain entirely.

I don't think you can even do that on iOS - isn't it against Apple's TOS or somesuch to have apps interact with each other like that?

I haven't read it in a while, but the TOS did forbid offering an incentive for app installations. While Apple's language in the TOS was broad, enforcement was specifically targeted at things like Tapjoy, where companies were paying for installs within other apps. They weren't pleased that their top rankings were effectively being bought. In my experience, these more direct promotions have been alright.

There have been a number of attempts to do something similar, and they've all fizzled.

One example that I could recall: http://indieappsalliance.org/

The main problem is that cross promotion really only works in volume, and most of the members in these networks do not have the audience to provide much of a benefit to the network. Conversion rates on these things, even between games with very similar audiences, is pretty minimal. What works much better is providing an incentive in game A for downloading game B. Those work, but require a deeper level of integration than is possible with a distributed network.

note: played in Android emulator on OS X.

I think the game looks fantastic. The only elements on the screen are those that need to be there, and they all are pretty effective at communicating their purpose. The only omission I can think of is the score. I'm particularly a fan of the red dots representing valid win states on the upper bar.

I had a couple issues with the main game mechanic, though.

There were a couple instances where all of the win states were crammed into the rightmost end of the bar. Players are scored based on finding the shortest route, but I don't think that finding the shortest path in a web like this is a task that humans are very good at. Even less so when all of the valid paths differ by so little.

Based on my experience, the game seems to have two phases. The first: I sit and stare at the screen trying to plot a valid route. Then: I enter the route to check that it works. I don't interact with the game in an exploratory way. My complete set of moves pretty much has to be decided before I touch the first node. Undo alleviates this a bit, but they're too few in number to mean much at higher levels (where there are 8 - 10 nodes).

There seems to be a good framework here for a game based around path finding. I just don't know that the mechanics as they exist now capture all of its potential.

I just realized that the encryption I did for the Password is pointless. I simpley used a hardcoded key. The server doesn't really need to decrypt the password, he saves the encrypted password. But if I use a hard coded key then someone who gets a hold of the password file can decrypt it, because he knows the same key is used for all the passwords.
Instead, I will let a user to create his own private key(different from the private key for the communication) that is specifically used to encrypt his password.
I finally figured why it is useful to encrypt the password, I didn' realize it before and just used a hard coded key, heh.
The communication is secure, except I am also using the same private key for all the clients. I didn't intend it to be like that and the key is stored on a file so I did plan to make a special key for each client. So I need to think how to let the client create his own key.
I didn't see games ask for players to create their own key, so how do players get their own private key? One way is to create a key according to your registration, or purchase. Each purchase send you a different private key.
Another way, is simpley let the player create his own key locally. That would be easier for me, but most companies probably generate a client private key upon purchase. I think they also generate a certificate for the client, so the server can tell if that is an authorized client(which is basically a DRM and I think what WoW and etc are doing to prevent from anyone with a client to play online).

What do companies do, and what you suggest me to do?

I suggest using something like bcrypt with randomly generated per user salts. For password checking, the salt will be retrieved from the bcrypt generated hash. bcrypt accepts a configurable cost parameter that determines the computational effort needed to generate the hash, which should be small enough to enable quick checks, but large enough to make brute forcing infeasible.