3 ms·

Ask HN: Review my startup - Social gaming framework for Android, iPhone

18 points by metachris 16y ago

We are building a social gaming framework for smartphone developers (Android and iOS, but also Flash and more), which solves many common problems of developing multiplayer games: networking, matchmaking, user authentication and attribute storage, security and scalability. The social services include cross-platform friendships and feeds, presence status and invitations as well as support for external services such as Facebook (authentication and the Graph API), Twitter, etc. The server infrastructure is hosted at Amazon EC2 which greatly simplifies the dynamic scaling of the network. Gameserver instances are created via a web-interface and provide detailled statistics about the games, the players and the connections (including a real-time display of active users). Our plan is to charge based on the used resources, trying to make a very good offer for indie devs in particular. Recently we've started private beta phase with the Android implementation, and will soon support more platforms. The first developers have deployed games on the Android market using the framework; everything works great and we are as excited as the devs are! It feels like a great moment to tell the HN community about the project - and to ask for feedback in particular. Thanks! http://www.flockengine.com

metachris 16y ago
Clickable: http://www.flockengine.com http://www.flockengine.com
l4u 16y ago
got a 404 when downloading flock-sdk-android.zip
- metachris 16y ago
  Thanks -- one link to the sdk was broken which is fixed now! We recently switched from PHP to web.py and this was hopefully the last orphaned link.
dirtyaura 16y ago
Interesting. At one point, we were planning to open source our gaming framework and build business around server-side functionality. However, after various discussions, I finally abandoned that plan and we are focusing to use it internally. First, in games particular, considerable amount of work is spent on the client side. Unless you are super-successful to encounter serious scalability issues, all the hard problems are on the client-side. Although gaming companies do not currently have that much back-end guys, they still have quite capable coders in general, who can learn back-end stuff quite easily. Thus, you really need to provide significant value to them (us). Looking through your APIs, you currently have solved some basic authentication stuff. That's okay as a start but definitely not enough to attract skilled teams. Second, game business is very hit-driven, thus you are likely to serve the less-profitable long tail developers. The best offering for them is not to provide server-side technology per se, but bring more buyers/players for their games. Thus, approach that e.g. OpenFeint has taken seem to serve those guys better, they basically are offering the stuff that you offer + the existing community. But best luck! I suggest that you talk to experienced game studios. Find out what are the biggest pain points and try to provide solutions to those. EDIT: Note, I didn't want to sound discouraging, although reading it now, it definitely came out like that. I think there's definitely opportunities in this space, I just wanted quickly to highlight some key issues. Note also that it will be a very competitive space, because successful mobile multiplayer game companies can also try to enter the space with their proven propriety solutions. That's part of our long term plan, i.e. we first try to create succesful mobile multiplayer games, and then consider platform-provider option again.
- metachris 16y ago
  Thanks for your feedback! I agree that using such a framework internally and focusing on developing games is definitely an interesting option. OpenFeint looks pretty similar but has a primary focus on highscore and leaderboards, which is a free feature of theirs. I couldn't find out how much they charge for their turn-based multiplayer feature though. We are trying to make a good offer for small developers; for the first developers it looks like < $10 per month with around 150,000 connections. For Android there are also a couple of highscore/leaderboard services such as ScoreLoop and ScoreNinja. We plan to provide interfaces to both and more. Apple's GameCenter (see WWDC Session 404) will be a big "competitor" on the iOS arena, but they probably won't expand to Android or other platforms. We are starting off on Android and see where it goes from here. Flash is an interesting option too. > Looking through your APIs, you currently have solved some basic authentication stuff. That's okay as a start but definitely not enough to attract skilled teams. Yeah, I agree. Right now we have a MVP, and many ideas. > But best luck! I suggest that you talk to experienced game studios. Find out what are the biggest pain points and try to provide solutions to those. Thanks, and I agree this is the best approach right now.
cmelbye 16y ago
You're aware of the established social web browser named Flock, right?
- metachris 16y ago
  yes. currently "Flock" is a working title; we looked for a name that developers can relate to while programming with the API (eg. to use "Flock.connect()"). Do you think that name would be too confusing for users because of the similarity? we probably will change to a better name sometime soon. suggestions are appreciated :)
  - cmelbye 16y ago
    Well, it might be confusing, yes, but the Flock browser might not be happy about you using the name (especially considering the similarity between the browser and your project).
zaph0d 16y ago
Awesome idea! Wish you all the best.
d0m 16y ago
I don't know about the idea, but I can give you some feedback on the main page. First, "Flock is an easy-to-use social gaming network for Android and Iphone". I have no idea what that means. Is it a tool I download to chat with my buddies in game on iPhone? Is it something that will help me as a developper to create and publish game? Is it a kind of steam for iPhone and android? I think the main page as too much "features" and not enough of "wtf is flock?". Like, "Integrated encryption with SSL/TLS and AES-256".. I really don't care about the encryption you use if I'm not to use your service because I don't know what it is. If you absolutely want to tak about security, say it in a: Flock is a secure and reliable way to blah blah. Keep it simple!
- metachris 16y ago
  Thanks, highly appreciated! I've updated the main image hoping it better conveys what the framework is about. Before: "Flock is an easy-to-use social gaming network for Android and iPhone" Now: "Flock is the easiest way to connect users on Android and iPhone to multiplayer games." That's really the core functionality right now. The social part is rather peripheral.
fictorial 16y ago
Awesome, I have been working on something similar for a long time. I used to greet news of similar services with dread, but as HN has taught me, having competition is a good sign that you are thinking of something less than completely hairbrained. I do have a different angle than Flock, and I am far down the path to launching. But, my daughter just arrived at 31 weeks and my time is now happily gone. So, good luck to you (and to myself! Heh). Edit: the teaser site is http://PlayRelay.io http://PlayRelay.io
drewcrawford 16y ago
I'm an iOS developer who does a few games a year. 1) I'm betting on Apple's Game Center. It will be popular with developers. You need penetration--how will you compete with Apple on numbers? 2) The other reason I'm doing GC integration is that people who don't have my game can get invites from friends who do for multiplayer matches. This is huge for discoverability. You can't do that. Again, how are you going to compete?
- fictorial 16y ago
  One does not compete with Game Center. One integrates with Game Center.
- metachris 16y ago
  It's probably impossible to compete with Apple GC - essentially they are building the same thing for iOS. There are other options though. The client implementation is not that complicated and could be ported to various platforms: J2ME, Flash, HTML5, etc. We think developers might appreciate a "GameCenter Light" which is easy to integrate, cheap and does does the basic things right. Discoverability is another interesting topic, and of course GC will be huge for that. I think that again it would be possible to start a light version, which offers additional exposure for developers and their games. There are many ideas that could be implemented in that direction.
  - drewcrawford 16y ago
    You're making the same mistake I made with my startup. From my comment you know: 1) I'm on HN, so I'm automatically sympathetic 2) I'm smack dab in the middle of your target market That should have caused you to swing into sales mode. You're telling me what "we think [hypothetical] developers might appreciate." Hypothetical developers will never pay you any money. I might, if you can convince me not to write the GC integration code next week (Not a joke, I am actually gearing up to test a new game with it this month). What would I do with a GC lite? In my experience (and from talking to the apple engineers at wwdc), GC looks really easy to integrate. It's also free. If your play is to do less than GC, how will you compete with free? There are some things that GC is doing wrong. Maybe you can improve on these. But you haven't touched on anything that specifically irks me about GC yet. So I have no reason to buy in.
arpit 16y ago
Interesting, though I wonder if apple could potentially block something like this to favor their own platform like they do for analytics and the Appsfire fiasco
- metachris 16y ago
  Good point and one we are also thinking about. I can't imagine a reason for Apple to reject such a framework, since it is really only basic socket connections with a bit of additional logic. They can't take sockets away from developers.
  - ericflo 16y ago
    Sure they could--provided you aren't using them in a way they approve of.

18 points by metachris 16y ago

metachris 16y ago

Clickable: http://www.flockengine.com http://www.flockengine.com

l4u 16y ago

got a 404 when downloading flock-sdk-android.zip

metachris 16y ago
Thanks -- one link to the sdk was broken which is fixed now! We recently switched from PHP to web.py and this was hopefully the last orphaned link.

dirtyaura 16y ago

Interesting. At one point, we were planning to open source our gaming framework and build business around server-side functionality. However, after various discussions, I finally abandoned that plan and we are focusing to use it internally. First, in games particular, considerable amount of work is spent on the client side. Unless you are super-successful to encounter serious scalability issues, all the hard problems are on the client-side. Although gaming companies do not currently have that much back-end guys, they still have quite capable coders in general, who can learn back-end stuff quite easily. Thus, you really need to provide significant value to them (us). Looking through your APIs, you currently have solved some basic authentication stuff. That's okay as a start but definitely not enough to attract skilled teams. Second, game business is very hit-driven, thus you are likely to serve the less-profitable long tail developers. The best offering for them is not to provide server-side technology per se, but bring more buyers/players for their games. Thus, approach that e.g. OpenFeint has taken seem to serve those guys better, they basically are offering the stuff that you offer + the existing community. But best luck! I suggest that you talk to experienced game studios. Find out what are the biggest pain points and try to provide solutions to those. EDIT: Note, I didn't want to sound discouraging, although reading it now, it definitely came out like that. I think there's definitely opportunities in this space, I just wanted quickly to highlight some key issues. Note also that it will be a very competitive space, because successful mobile multiplayer game companies can also try to enter the space with their proven propriety solutions. That's part of our long term plan, i.e. we first try to create succesful mobile multiplayer games, and then consider platform-provider option again.

metachris 16y ago
Thanks for your feedback! I agree that using such a framework internally and focusing on developing games is definitely an interesting option. OpenFeint looks pretty similar but has a primary focus on highscore and leaderboards, which is a free feature of theirs. I couldn't find out how much they charge for their turn-based multiplayer feature though. We are trying to make a good offer for small developers; for the first developers it looks like < $10 per month with around 150,000 connections. For Android there are also a couple of highscore/leaderboard services such as ScoreLoop and ScoreNinja. We plan to provide interfaces to both and more. Apple's GameCenter (see WWDC Session 404) will be a big "competitor" on the iOS arena, but they probably won't expand to Android or other platforms. We are starting off on Android and see where it goes from here. Flash is an interesting option too. > Looking through your APIs, you currently have solved some basic authentication stuff. That's okay as a start but definitely not enough to attract skilled teams. Yeah, I agree. Right now we have a MVP, and many ideas. > But best luck! I suggest that you talk to experienced game studios. Find out what are the biggest pain points and try to provide solutions to those. Thanks, and I agree this is the best approach right now.

cmelbye 16y ago

You're aware of the established social web browser named Flock, right?

metachris 16y ago
yes. currently "Flock" is a working title; we looked for a name that developers can relate to while programming with the API (eg. to use "Flock.connect()"). Do you think that name would be too confusing for users because of the similarity? we probably will change to a better name sometime soon. suggestions are appreciated :)
- cmelbye 16y ago
  Well, it might be confusing, yes, but the Flock browser might not be happy about you using the name (especially considering the similarity between the browser and your project).

zaph0d 16y ago

Awesome idea! Wish you all the best.

d0m 16y ago

I don't know about the idea, but I can give you some feedback on the main page. First, "Flock is an easy-to-use social gaming network for Android and Iphone". I have no idea what that means. Is it a tool I download to chat with my buddies in game on iPhone? Is it something that will help me as a developper to create and publish game? Is it a kind of steam for iPhone and android? I think the main page as too much "features" and not enough of "wtf is flock?". Like, "Integrated encryption with SSL/TLS and AES-256".. I really don't care about the encryption you use if I'm not to use your service because I don't know what it is. If you absolutely want to tak about security, say it in a: Flock is a secure and reliable way to blah blah. Keep it simple!

metachris 16y ago
Thanks, highly appreciated! I've updated the main image hoping it better conveys what the framework is about. Before: "Flock is an easy-to-use social gaming network for Android and iPhone" Now: "Flock is the easiest way to connect users on Android and iPhone to multiplayer games." That's really the core functionality right now. The social part is rather peripheral.

fictorial 16y ago

Awesome, I have been working on something similar for a long time. I used to greet news of similar services with dread, but as HN has taught me, having competition is a good sign that you are thinking of something less than completely hairbrained. I do have a different angle than Flock, and I am far down the path to launching. But, my daughter just arrived at 31 weeks and my time is now happily gone. So, good luck to you (and to myself! Heh). Edit: the teaser site is http://PlayRelay.io http://PlayRelay.io

drewcrawford 16y ago

I'm an iOS developer who does a few games a year. 1) I'm betting on Apple's Game Center. It will be popular with developers. You need penetration--how will you compete with Apple on numbers? 2) The other reason I'm doing GC integration is that people who don't have my game can get invites from friends who do for multiplayer matches. This is huge for discoverability. You can't do that. Again, how are you going to compete?

fictorial 16y ago
One does not compete with Game Center. One integrates with Game Center.
metachris 16y ago
It's probably impossible to compete with Apple GC - essentially they are building the same thing for iOS. There are other options though. The client implementation is not that complicated and could be ported to various platforms: J2ME, Flash, HTML5, etc. We think developers might appreciate a "GameCenter Light" which is easy to integrate, cheap and does does the basic things right. Discoverability is another interesting topic, and of course GC will be huge for that. I think that again it would be possible to start a light version, which offers additional exposure for developers and their games. There are many ideas that could be implemented in that direction.
- drewcrawford 16y ago
  You're making the same mistake I made with my startup. From my comment you know: 1) I'm on HN, so I'm automatically sympathetic 2) I'm smack dab in the middle of your target market That should have caused you to swing into sales mode. You're telling me what "we think [hypothetical] developers might appreciate." Hypothetical developers will never pay you any money. I might, if you can convince me not to write the GC integration code next week (Not a joke, I am actually gearing up to test a new game with it this month). What would I do with a GC lite? In my experience (and from talking to the apple engineers at wwdc), GC looks really easy to integrate. It's also free. If your play is to do less than GC, how will you compete with free? There are some things that GC is doing wrong. Maybe you can improve on these. But you haven't touched on anything that specifically irks me about GC yet. So I have no reason to buy in.

arpit 16y ago

Interesting, though I wonder if apple could potentially block something like this to favor their own platform like they do for analytics and the Appsfire fiasco

metachris 16y ago
Good point and one we are also thinking about. I can't imagine a reason for Apple to reject such a framework, since it is really only basic socket connections with a bit of additional logic. They can't take sockets away from developers.
- ericflo 16y ago
  Sure they could--provided you aren't using them in a way they approve of.