Sync Lock - A week in review
The past week has been one of the most “unique” weeks of QONQR’s existence. I doubt anyone reading this will find that surprising.
In the past 8 days I’ve chatted with hundreds of players via GroupMe, Forums, and email, including a friendly wakeup call from the UK Legion at 4am on Sunday morning, which turned into a nearly two hour chat. No worries, I’ve only slept past 4am twice in the past week, and the early morning chat was a very good conversation. The chats have been productive and influential. I’m confident Sync Lock will gain better improvements over the next few weeks.
Over the past week, we have simplified and refined Sync Lock instructions to avoid Sync Lock.
Make sure at least half your deployments are 2Km away from your friends, or stagger your deployments by at least 15 minutes.
We are also encouraging cross faction communication to encourage players to get to know each other, and play honorably by not Sync Locking people they feel confident are not multi-scoping. This good sportsmanship approach is having mixed results, but it is definitely creating cross faction discussions, which has also lead to some good feedback for Sync Lock improvements.
We feel like the next biggest change that can be made is to reduce the 2Km distance. This should have a big impact on many of the legitimate players who are currently suffering. Starting today I am “going dark” as I did last week for 2 days. I’ll be pushing hard to get the mobile clients updated. I’ve been very responsive to all requests over most of the past week to jump into group me conversations and respond to support emails. Beyond a few scheduled chats, I will be hard to reach in the next few days as we rush to get the next client release out.
I would also like to also announce that we are planning to add a few more items to the next release. Because we hope to release the next round of client updates at nearly the same time, we are going to take this opportunity to try and add some features that will require each platform to get an update in the same week (or two). We feel like these changes will bring the players some long awaited features, and should not delay a client release for Sync Lock improvements by more than an extra day or two. If the delay is any longer, we will probably hold some of these features for a later release.
Sync Lock updates:
- Show infected / inoculated on the HUD
- Add GPS accuracy information with launch request to reduce the 2Km range in Sync Lock formula
Additional Features we hope to add:
- Zone Search by name and Zone Id
- A list of zones currently visible on the map, and sortable by name, distance, or faction control
- The ability to “Attach” a zone link to an in-game message
Much of the plumbing is in place for these additional features and we hope we can push them out quickly so they can be included in the next release.
Once again, thank you to everyone for their contributions to the Sync Lock discussions. While the frustrations have been clear, the conversations have nearly all been moving toward a common goal. I am extremely grateful to the community as a whole. We build this game for you, and you make it possible for us to keep working on it every day.
-Scott (aka Silver)
Sync Lock Changes
We have posted a video on our YouTube channel to explain changes we are making to Sync Lock. The video shows a visual representation of how sync lock works now, and Silver talks about what we are working on for the future of Sync Lock. Thank you to everyone who has contributed ideas and feedback in the GroupMe groups that Silver has been visiting. We appreciate you and ask for your continued patience while we work to make QONQR better.
Sync Lock Day 3
An update on the community feedback. Yesterday I chatted with 100-150 players over an 18 hour period. I feel very good about the outcomes of the day’s conversations. I wanted to hear people’s concerns, but more importantly to collaborate on solutions. The conversations were tough, but constructive.
As a result of the conversations, Gaderson (support), GungHo (founder & originator of QONQR concept) and I worked through improvements to make based on feedback.
- We will simplify the formula for determining a sync lock between two players by dividing all launches (locations where you are sitting at time of launch) into buckets that are 2Km by 2Km grids on the map. We will count all the buckets that have Player A alone, Player B alone, and both players together. If over 50% of either player’s launches are in the shared bucket, a sync lock can be established. There is no longer the ability for you to closely coordinate launches while separated for a short amount of time to prove individuality. While the old math worked very well for us to find multi-scopers in our support tools, the guidance for how to avoid being locked was too complicated. Just make sure half your launches are away from each other and you will be fine.
- We will look back at your last 1000 launches for this new 50% rule, but will go back no more than 4 days. Going back up to 4 days will limit the impact of the weekends or working from home one day a week.
- Players will now get a message (and we will fix the current message bug) that says they are locked and which faction locked them.
- All players will be able to see a list of players they have locked who are still locked. In addition to this list, you would have the ability to “recall” the virus, to unlock the two players you previously locked.
- Sync Lock will only be effective as long as both scopes continue deploying in the same location. (More about this below.)
We want Sync Lock to be used to combat multi-scoping. Let me first address the question, “Why are you letting people buy their way out of Multi-scoping if you don’t want it at all?” People are going to multi-scope. We cannot keep up with the growing user base. We have seen that we cannot avoid this unfair advantage using the old model of reporting multi-scopers to QONQR support. The options are to let it happen for free or pay. If someone is going to use multiple devices to beat you, don’t you at least want them to pay for it? Since only 5% of players spend money, this takes 95% of your enemy out of the running for multi-scoping. While it isn’t stopping it completely the impact should be noticeable.
So why the 5 changes I have listed? We want Sync Lock to be used to stop unrestricted multi-scoping. Our experience over the past year has shown us you can’t build an algorithm to accurately tell the difference between one person at their desk with three phones, and a family sitting on a couch. This new method still makes it difficult for multi-scopers to avoid detection, but gives legitimate players some options.
In an ideal game, this is what I would like to see: QONQR returns to the mostly friendly and generally polite rivalry that was common a year ago, sportsmanship and good-natured fun. It is a game after all. Unfortunately much of the world has devolved to a mean and ugly place. Players have no reason to be anything more than foul to each other, because there is no incentive to be polite. I asked several people today, “Give me the name of one of your opponents you respect that you could reach out to for help with a problem. Someone who plays with honor.” I was saddened to find that nearly all said, there isn’t one.
We are going to add the ability for those who do want to play in the nature of good sportsmanship, the ability to do so. Now players will know another local player has locked them and which faction they play for. They can reach out to a friendly contact in that faction and say, “One of your players locked my son and me today. You know us, and you know we don’t multi-scope. Could you ask around, see if you can find who locked us, and see if you can help us get unlocked?” Knowing the GroupMe groups like I do, I suspect it will be common to find the players involved. Over time, perhaps players will slowly become more civil towards each other, now that there is a reason to have some friendly cross-faction relationships
What if that isn’t possible? What if players still refuse to be civil? Let’s say you are playing with a bunch of jerks who lock you and your son every chance they get, or perhaps your family are the jerks and that is why you get locked. Now what do you do? We are planning to make it so Sync Lock only has an effect while you are in the same zone as the person you are locked to. Leave the zone and your scopes are too far apart to be able to share overheat. So if you get synced with family, you will be able to deploy from work normally, but will by synced again when you return home, if your infection hasn’t expired.
This does not fix the issue for everyone. There are still a few moms and dads who stay home with their kids all day, husband and wife teams that work together, people who work within a mile of home, or students co-located on a college campus. It’s not perfect, but it’s a good start. We are working very hard to address the needs of our players, and we will continue to do our best. Until we can truly separate the individuals in the same location, you will have to use diplomacy in an attempt to avoid Sync Lock. We’ve given the community the ability to do the right thing, we can only hope some take the opportunity to rebuild relationships.
The work on these changes starts immediately. Thank you for the constructive conversations. Thanks for playing QONQR.
Sync Lock Day 2
I have spent many hours in the forums chat and in many GroupMe groups in the past day. In this post I will take the time to summarize the top conversations.
Q: I can’t believe you said it was OK to multi-scope. Why are you telling people to do this?
A: We are still against multi-scoping. We think multi-scoping creates an unfair advantage and we don’t like it, which is why we didn’t just let it happen for free. We are not promoting multi-scoping, we are saying we are no longer going to ban people for doing it. We want to continue to discourage multi-scoping, which is why we implemented Sync Lock.
Q: How did it get to this point? What caused you to think Sync Lock would be a good idea?
A: We hired a full-time employee, basically to handle the multi-scoping concern. For a company with 2 full-time employees, that is a massive investment. However, it was an investment important for us to keep the game fair. In the past few months, the public perception of multi-scoping has grown very dark. The player forums are toxic with accusations of cheating. Facebook and Twitter have seen an increase in the volume of negative comments, stating we allow cheating and the game has become a corrupted system.
What was seen publicly was only the tip of the hate and negativity that we saw daily. We spend 20 minutes to 2 hours for every multi-scoping report we receive. We review location reports, launch history, and player to player messages in an attempt to a do a fair and accurate assessment. Unfortunately in recent months, most players began to disagree with our assessment, calling us incompetent or accusing us of supporting cheating.
This negativity had become a cancer in the community. Despite our continued dedication to maintaining a fair environment, the community began to reject the outcome. The perception of fairness was gone. Our support costs were growing rapidly, but the positive impact was no longer there, making the continued investment of our limited resource a bad business decision.
QONQR has always thrived because of the positive gamer community. We saw this community degrading rapidly and felt this change was necessary. We knew it was going to be a black eye for us for a while, but felt like it was necessary for the long term survival of the game.
Q: Why not just allow multi-scoping and tell people do deal with it instead of hurting families?
A: Whether it is “legal” or not, one person using multiple devices will always be considered an unfair advantage. Players are willing to accept being beat by someone who cubes, but won’t accept defeat by those that are “cheating” or not playing fairly. We needed to remove the “cheating” aspect of multi-scoping. Now we can take our resources from the cheat-to-win problem and put them on building game expansions that offer new things to do in QONQR where buying cubes has little or no impact.
Q: This is just a money grab. You guys are greedy.
A: We did not implement this change to increase revenue. This change was about reducing costs and the managing long term culture of the community. Multi-scoping was a huge drain on our time and cash flow and was hurting the enjoyment of the game in an unmanageable way. If we could have found a motivation to reduce multi-scoping without using cubes we would have taken that route.
Q: This isn’t going to work because all the multi-scopers will just pay for their second device.
A: Only 5% of the players that play QONQR spend money. This eliminates 95% of the potential multi-scopers from being able to have an unfair advantage. While some in that 5% of spenders may play on multiple accounts, the total number of multi-scopers will decrease.
Q: Other games allow one person to have multiple accounts. Why can’t you?
A: The best examples of player-versus-player or player-versus-everyone games that allow multiple accounts are also games that require a monthly subscription. We do not want to be a subscription based game.
Q: This is a huge impact on the game. Everyone I know is Sync Locked.
A: In the first day of Sync Lock, 7% of the players who launched during the day, were Sync Locked. There were more failed sync attempts than successful locks. Approximately 100 “couples” were infected with Sync Lock. Only 39 Anti-virus packs were purchased. The Chicago area was the most impacted with just over 20 players synced. However most states/regions had an average of 4 players that were Sync Locked. Although, being synced is a huge impact for those that suffered the lock, overall it was not nearly as big of an impact as players predicted.
Q: Fine, I get why you felt you had to do it, but why are you anti-family?
A: We hate that legitimate friends and families are hurt by this change. Unfortunately, we could not devise a way for our software to know the difference between 4 people at the kitchen table each with a device, and a multi-scoper sitting alone at the same table with everyone’s devices. It is the later that is exactly the reason we are in this position. Multi-scopers claiming to be a family, but launching on four devices in the middle of the night. As with all things in QONQR, this game evolves based on community feedback. We are open to new ideas for how to solve the family issue. In four months of contemplation, we could not come up with a solution that couldn’t be tricked by a cheater, and did not involve an email to QONQR support (a flawed system that got us here). We welcome all suggestions for improvement, but we request that you include with your description of the change, a list of all the ways a cheater could get around your system and continue to multi-scope unrestricted.
Q: You guys are going to lose so many players. You are killing your game. Everyone is quitting.
A: We see the messages that people are quitting. It is disheartening and we hope that people give Sync Lock a chance. However, what people haven’t seen is that many more people have emailed us over the past two months saying they are quitting because they are tired of spending money on a game where cheating is allowed. The threat of losing players is not new. The difference is that nearly all of the players that threatened to quit previously were paying players. With only 5% of our player base spending money, this was a much bigger threat to the survivability of the game. We don’t want anyone to quit QONQR. It may seem like this change is causing people to quit, but multi-scoping had been forcing people to quit at a greater rate. It just wasn’t as visible to the community as it is now.
The new change will force some people to adapt their gameplay. Some people won’t adapt and will decide to stop playing. We wish we had made multi-scoping work as it does today with Sync Lock from the very start. But we didn’t know then what we know now. We felt this change was necessary for the long term survival of the game. We expected the issues Sync Lock would cause, but are looking to the future. We hope you stay with us, but we understand how hard change can be.
We have always tried to be open with our players. I’ve spent most of my past day talking directly to players in the forum chat and GroupMe. I am happy to directly answer any questions. If you would like to include me in your GroupMe discussions, please contact support@QONQR.com and we will connect. As always, thanks for supporting QONQR.
-Scott (aka Silver)
Sync Lock - The End of Multi-Scoping
Can’t stand reading these long blog posts? Read the summary in the forums.
Today QONQR is announcing we have lost the battle against multi-scoping. We have spent several hundred hours of developer time over the past year building tools to detect and analyze player behavior, tools to gauge the likelihood of someone cheating and tools to penalize those that multi-scope to gain an advantage. Even with all this technology, in reviewing Christie’s time card for the first 6 months of her employment, we were disturbed to find that over 50% of her support time is spent researching multi-scoping claims.
This alone isn’t a problem. It is very important to keep everyone on an even playing field. However in recent months public sentiment has become very dark against QONQR support. When we investigate a claim, most of the time we find enough individual activity to respond to the person reporting that the activity doesn’t look suspicious, or looks as if a family or work colleagues are playing together. While many people respond to our notices with appreciation for checking, many more have begun to hurl claims of fraud, incompetence and bribery. Some have threatened lawsuits. In some cases the people who are reporting cheaters and unwilling to accept our answers, are hurting the QONQR community more that the cheaters themselves.
In response to this business challenge, we are stopping all player bans for multi-scoping. Any accounts previously banned for multi-scoping can now be reactivated by emailing support@QONQR.com. Accounts banned for other reasons will stay banned.
Today, we are also announcing a new feature called “Sync Lock.” Each day at midnight UTC/QST (5 PM Pacific Daylight Time) all players will receive a free Sync Lock Packet. This network packet pair is a QONQR Virus (not a real virus that will hurt your phone) that you can use to infect two other players in QONQR.
Scopes that spend too much time in proximity become susceptible to Synchronization Lock. Sync Lock is a virus that can infect scopes, causing them to combine overheat metrics, resulting in rapid overheat on both scopes. Scopes become susceptible to “Synchronization Lock” when deployments happen together, within the same GPS proximity. Scopes that have been jail broken or rooted have greater risk of being infected based on common launch patterns. All players are issued a Sync Lock packet daily. Players can attempt to sync lock two opponents by sending the sync packet pair to the targeted players. If the scopes can be infected, the packet pair will returned indicating the outcome of the Sync Lock attempt. Returned packets can be reused to infect additional players. If the two scopes cannot be paired for sync, the packets will be destroyed in the process. There is no limit to the number of players that can be infected with the same packet, if it is not destroyed. Players can purchase a new packet for qredits if they are unwilling to wait until a new packet is issued the next day to infect more players. Players can protect themselves from Sync Lock infection by purchasing inoculations in the depot.
Players that rarely play in the same location as other players should have no concern for being infected by Sync Lock. However multi-scopers will find this feature a new challenge to their previously easy road of holding their zones by getting “their wife and 3 children” to play with them 12 hours a day. For families who legitimately play as individuals there are two options. Purchase anti-virus in the depot, or make sure at least 80% of all your deployments for both players are 2 Km (a little over a mile) apart or that 20% of your “same time” launches are over 2Km apart. “Same time” launches are launches that occur within 30 seconds of each other. It may take some good coordination, but family and work friends can avoid sync lock without buying anti-virus through well-coordinated attacks when apart. There is also the option that you stop being a complete jerk to your enemies, meet them for beers after work, show them that you are a fun loving group of work friends and perhaps you won’t get Sync Locked every day.
Scopes qualify for Sync Lock based on the previous 24 hours of activity, and Sync Lock infection lasts 24 hours. Sync Lock Anti-virus comes in 24 hour and 30 day protection. Attempting to Sync Lock two players can currently only be done through the web portal at https://portal.QONQR.com/SyncLock.
Antivirus can be purchased through the portal or the phones. If you attempt to purchase anti-virus but are already inoculated, the purchase is blocked (you won’t waste your funds). In future version of the clients on iOS, Windows Phone and Android, an indicator will make it easy to see if you are infected or inoculated. We are working on the client updates now. You will receive an in-game message if you are infected. The web portal will show who is infected or inoculated. Player profiles will look the same as always if they are neither infected nor inoculated. http://portal.qonqr.com/Player/Search
Still don’t get it? Here is an example. Scott and Christie play QONQR together all day. Someone discovers this and Sync Locks their scopes. In the past hour Scott has deployed 5 times and Christie has deployed 21 times. However since both scopes are now Sync Locked, it is as if both scopes have had 26 deployments in the past hour. Christie is in maximum overheat anyway, since she has already deployed over 20 times in the past hour, and the extra 5 deployments from Scott’s scope has no impact. However, Scott, who had previously no heat on his scope with only 5 deployments, is now also in maximum overheat and suffering from the painfully slow bot and energy refreshes.
Q: What if my kids only play 1 hour a day, when they sit with me and watch TV after dinner?
A: Well, if your kids are a big threat, an enemy will probably Sync Lock you every day, but an hour after they stop playing, their scopes will not have any impact on your overheat. Since your overheat is calculated based on the deployments of both scopes in the past hour. Your deployments + no deployments = your deployments.
Q: So you are saying if I keep guessing right, I can infect dozens of players for free every day?
A: Yes, until you guess wrong and pick two players that don’t qualify for Sync Lock, you will be able to use the same Sync Lock Packet over and over to infect more players. Once you lose your packet by selecting two players who don’t qualify for Sync Lock, you have to wait until midnight UTC/QST to get a new packet for free, or buy a new packet.
Q: Can I be synced to more than one device at the same time?
A: Yes, there is no limit to the number of devices that can be Sync Locked together. However if you are attempting to Sync Lock players A,B,C and D all to each other you must sync each pair. A to B, A to C, A to D, B to C, B to D, and C to D. If all players are synced, emptying the scope once for each player would put everyone in (or near) maximum overheat.
Q: My friend and I are infected, I bought an inoculation, is my friend safe too?
A: No, your overheat will no longer be impacted by anyone else’s scope, however your friend will still be impacted by your scope. Additionally, other people can still be locked to your scope despite your inoculation, but only they will be impacted.
Q: If I try to infect two players and they are both protected from Sync Lock, will my Sync Packet be destroyed?
A: No, the Sync Packet will establish a connection based on the proximity vulnerability, but fail to embed in the scopes. The packet will be returned to you if the connection is established, regardless if the infection works.
Q: I’m going to a party with a bunch of QONQR players, how do we avoid getting Sync Locked
A: Deploy all day long before you get together, while you are still apart.
Q: Do you have any suggestions on how to best use Sync Lock?
A: Yes, create a “hit list” and share it on One Drive, Google Drive, or Dropbox. Every day work as a team to sync lock your opponents by taking turns working down the list until you lose your packet. Then let the next person take over on the remaining players. Re-organize your list to maximize lock potential. Avoid buying more packets by working as a team to Sync Lock your enemies. As a multi-scoper, randomly take days off for one of your scopes, to cause your enemies to waste a packet on you.
Q: My husband and I buy a lot of cubes in QONQR. This sucks that we have to spend even more to play together. Did you realize that you are hurting us?
A: Yes we did. If you are buying the 550 cube pack the cost for you to keep enjoying each other’s company in QONQR every day is $0.18 a day or $3.67 a month for those that buy the 30 day inoculation. We know this new feature hurts legitimate players who play together. As a cuber, we hope you find this as a new weapon to use to keep the game fair. Many players have emailed us saying, “I’m a paying player. I pay your bills and I don’t cheat. You are letting people who don’t help QONQR survive as a company, push me out of the game through cheating. If you don’t stop the cheaters, you will lose all your paying customers and your company will go out of business.” We have been living the pain of stopping multi-scopers for a long time. We don’t think there is ever going to be a perfect solution to the problem, but we hope you agree this is better.
Deciding to make this change was a difficult choice. Many people predicted that as QONQR grows we would never be able to stop multi-scoping. While you may not like this new feature, we hope you agree removing the multi-scoping restriction was an inevitability. We hope you agree this solution is an acceptable one. To the families of free players, we spent many hours working through the scenarios to make sure there was a way for you to continue to play for free and avoid Sync Lock. It will take you more effort now, to have the same attack ability as before, but it was important for us to maintain the fun of QONQR, without the necessity to spend money.
As always, we appreciate your support and will endeavor to make QONQR better every day.
-Scott (aka Silver)
Battle Engine Rewrite Completed
We’ve been quiet about it, but Mondays update was not just a data migration, but was also the introduction of the new and improved battle engine. As I said in the January blog post where I outlined that the battle engine was the top priority for 2014, no changes to the formation-vs-formation battle outcomes were made.
We have thousands of unit tests that test one formation against another, tests groups of deployments against each other, every faction against every faction, and all those permutations against each other at different upgrade levels and formation levels. All our tests showed the same battle outcomes within a couple bots of variation due to rounding of fractional bots. In other words, Zone Assault 5 against Deflection 5 with 25,000 Hardened Lattice 2 bots boosting shields has the same outcome today as it did last week, within 1 or 2 bots.
It is unusual for us to keep secrets like this, but we kept the introduction of the new battle engine quite for a few days to avoid “mass hysteria”. We’ve learned from experience that following an update, some users jump to the conclusion that we’ve introduced a bug the moment they don’t see something they expect. With an update this important it was critical for us to reduce the false positives, so real bug reports could get our focus following the release. We are happy to report that only one bug was found, and it was not critical. Switching factions didn’t update the zone counts immediately if the switch resulted in the zone being completely cleared. However, the first person to hit the zone following the switch faction, causes the counts to be updated, and the issue self-corrects. This bug was fixed the day after the release.
The new battle engine is lean and mean. It runs on a streamlined database table that reduced our storage for tracking bots from 28 million rows down to 1 million. This isn’t the last update we could do to the way we store data, but it was a huge leap.
This update consumed nearly all my development time since mid-December. It is by far the largest single feature we have undertaken (other than Android) since the game was launched. I’m now looking forward to finally working on new features.
The battle engine did introduce some changes that aren’t related to bot against bot fighting.
- Who was selected to fight in a battle (choosing your enemy) was previously based on how many bots an enemy had in the zone, and a “Threat” attribute for each bot based on the formation. This was originally intended to identify who had the most defensive stamina to hold the zone. It made it so the little guys could get in the zone and build up their strength while the big players in the zone held off the attack. In reality this was more complicated than it needed to be and quite frankly not very accurate. The new battle engine looks at the total amount of shields a player has in the zone and selects an enemy this way. In this way, the person who is most threatening is the person who has the most defensive shield strength. If two players both have 100,000 bots in the zone, but one has full Hardened Lattice and Shield upgrades, it is that person that is the bigger threat, and will be targeted first. Were there any other reasons to change threat to be based on shields? I’m glad you asked. Besides simplifying our code, it also makes some future features such as “Tactical” much easier to implement.
- Divert works a bit differently too, out of necessity. Divert no longer lowers total threat as a separate attribute, but instead “masks” the shields of your most valuable bots. Divert hides the shield of your bots making your force look weaker than it is, so you are less likely to be selected. The result is the same. It still reduces your risk of being selected for battle, but now it is much more explicit. It is much easier to figure out exactly how many Divert bots you need to “hide” your support bots. You can still have “negative threat.” More information here on how Divert works.
- In my best Oprah voice, “You get an upgrade, and you get an upgrade, and everyone gets an upgrade!” In the new battle engine, when new players graduate from Zone Assault 1 to Zone Assault 2, all bots already in the field become Zone Assault 2 bots. The same goes for upgrades. Increase your scope’s Shield Enhancer, and all bots you already have deployed get the extra bump to shields. Many people are surprised to find out that it didn’t work that way before. It does now.
- In the old battle engine, it was possible that two launches, which were milliseconds apart, could target and kill the same enemy bots. While this could happen in real war, it is annoying in a game like QONQR, especially if both players used a DPF core. It happened infrequently and usually only multi-scopers ever noticed it because the timing had to be precise. In the new battle engine you will now see a response saying that your launch failed because the bots you targeted were already destroyed by another operative. Don’t worry, launch again and move on to the next enemy.
There you have it. These are the changes that you may have noticed in the game since Monday. Overall we are very excited about putting this behind us. It was a large barrier to our scalability, and no new formations or battle features could be added until this massive rewrite was made. Although the migration on Monday took much longer than expected, our queries show that no bot was left behind. A quick check of the leader board confirmed that even those with tens of millions of bots in the field have the exact same bot counts before and after the migration. We also ran queries checking not just totals, but checking and double checking individual bot counts to make sure Seeker bots stayed Seeker bots, and so forth.
As always, thanks for your patience and support. We already have two new features queued up to debut in a week or so. I’m excited to see what you think.
-Scott (aka Silver)
Looking Ahead to 2014
Many players have requested that I share our plans for QONQR in 2014. Those of you who are also software developers are probably familiar with the anxiety of doing this. Promising features is often a bad idea in software, as priorities change more frequently than the Minnesota weather, and you never know when some unexpected issue is going to side track you for weeks or months.
So rather than saying what is going to be done in 2014, let me share what we have started, and might actually finish.
Android. Yes, we might actually finish it. We have implemented a few cheater detection algorithms on the server to watch for people who cheat their location, and have more tests planned to deal with the litany of issues with the hacker friendly phones. We know many of you are concerned with Android and we are doing our best to keep the game fair. A recent survey of some of our Android beta testers indicated they believe we are 90% of the way to being ready for a general release. Let’s hope the last 10% comes soon.
We’ve had a long-time dream of having custom maps for QONQR, something that would extend our brand. We started down this path with an awesome GIS consulting firm in September and after some exploration, decided we wanted maps that shaded the world the colors of the factions that controlled the zones. We later discovered that we have so many millions of map requests every month, outsourcing the custom maps would raise our hosting costs by nearly 400%. We did more work and set up our own mapping server using an open source product called CartoDB. Hosting maps ourselves would be a huge monthly cost savings, but much more effort to implement. That is where we are today. As with most tech projects, the costs for this feature are now double our early estimates, and we are about half way to our desired product. We are working to show what we have so far through the web portal, and this will probably be available in the next month. Work on maps will continue, but probably at a slower and more affordable pace. Eventually, we hope to allow for players to be able to see personalized maps of what areas impact them, or even be able to do time-lapse animations of major battles. These last two are much longer term goals, but if we can get the mapping server setup correctly, it should be possible.
Free cubes? Many months ago, iPhone had the ability to earn free cubes by installing other software, or watching video advertising. QONQR made advertising revenue from this offerwall. We were incentivized to work with a partner firm on the iPhone early, and put it into the iPhone while we searched for a Windows Phone equivalent. We only found one partner for WP, and reviews of that product were not positive. The reality of the revenue and the quality of the iPhone offerwall wasn’t good, and as a result we pulled it from iPhone. The Windows Phone users hated that iPhone users could get free cubes, and iPhone users hated that we took it away. We believe we may have found a solution to provide those who cannot spend money on cubes a way to earn them by fulfilling offers through the web portal. We have some preliminary code written for this. If the quality and implementation seems good, this could be a feature we release in 2014.
If you follow the Swarm comics, you may have seen a hint of changes to the game that allow short range weapons to be deployed over a great distance. That is all I will say about this. Perhaps the comic is foretelling a new feature to be released, or maybe the story will evolve into nothing more than propaganda by Nagumo Industries.
A new and improved help website is on the way. Getting into QONQR can be a big challenge for new players. We are working on better help to make it easier for new recruits to find their way through the many features in the game.
We had made a partnership with a good friend and developer in November. The plan was for him to join QONQR fulltime and his first order of business was to build us an in-game chat engine that would scale to thousands of users in any chat room at once. Ten hours before his first day, he called us to say his old company made him an offer to stay that he couldn’t refuse (we agreed it was very good), and so the task was shelved. Chat is still a top priority for us and we hope to see this happen in 2014. Along with this, we hope to implement new communication systems that will make it easier for new players to get connected with the veterans. We want the veterans to be able to bring the new people under their wing sooner (even getting them to switch factions while in training).
Our intern, Cole, began work on Facebook integration, to make it easier for you to share your achievements and leaderboard rankings on your wall, including the images to go with the awards, and leaderboard charts. The technical requirements of Facebook development required us to postpone this until Cole was off the college campus and had more control of his internet connection and network configuration. We hope to complete that work this year.
Push notification for Apple and Android, with more reliable push notifications for Microsoft. We now have a direct line of communications to Microsoft on a new cloud services offering that allows us to have a single place to push notifications to all platforms. Implementing Apple push notifications is a mystery it seems few have unraveled. We are hoping that this new service offering will allow us to finally get this working.
QONQR lite? With every Apple iOS update, some people’s QONQR app broke. It happened with iOS6, iPhone5, and now iOS7. We’ve also begun to see some issues with Windows Phones in our last update, with 1% of WP players. With Android, broken applications are to be expected, given the fragmentation of the operating systems. We don’t want anyone who has invested time and money to be out in the cold if a new update breaks them. We are exploring the idea of creating a backup scope for all platforms. A second application in the app store you could install alongside the normal QONQR app. This would be a very slimmed down version of the game — only the bare essentials needed to launch, harvest, and create bases. Basically it would do the things that cannot be done through the web portal, and almost nothing else. The idea would be that if normal QONQR broke, hopefully the lightweight version would still run. It would take extremely low memory and not rely on many phone features. We hope a QONQR lite version might save some people who find themselves in a position where the regular app won’t start. We may release a prototype version of this to the general public on one platform to test it. If it is well received, the other platforms will follow.
Finally the biggest coding challenge for us in the first half of 2014 is going to be a rewrite of the battle engine. We plan to keep all the existing formations the way they are. The purpose of the rewrite is to avoid the equivalent of the “Twitter Fail Whale” in QONQR. The outages we had in August were a result of our systems being unable to handle our traffic, just like Twitter experienced in their early days. Specifically, our database hit its limits, which cannot be solved by adding more servers. We fixed those August issues, but the solutions were only temporary. We are starting work now on a new system that moves us past “startup company mode” and into a more robust and scalable system. We now have the experience necessary to see how millions of battles a week impact our systems, and where the weak parts of our solution are.
The battle engine changes will simplify the code for launching nanobots and remove the bottlenecks that cause some launches to fail during peak times. It will also open the door to creating new formations and options for attack. Previously, we were hesitant to add new features to the already complex battle engine, knowing a rewrite was needed. This will remove those handcuffs. However, this will be a major rewrite, and will take some time before fully implemented. The good news is, it shouldn’t change the way you play. The bad news, if done correctly, you won’t notice anything, meaning you might think we spend all our time playing QONQR, instead of working on new features.
We have been saying throughout all of 2013, that QONQR 2.0 will be a massive game expansion that centers around bases as a core component in the game. This is still the plan. We hope to start the 2.0 features in the second half of 2014, but it is unlikely many features will see a release during 2014, unless our development staff is able to increase by a few more developers.
We have several other things we hope to do in 2013, but I will leave those as a surprise. As I said, I will only talk about what we’ve started and might finish. We have a very long list of new features on the white board. Some of those should happen in 2014.
I can’t promise what you might see in 2014 from QONQR, but I pledge that we will continue to work hard to make QONQR better. Thank you for supporting QONQR in 2013 and we hope you stay with us on our startup journey.
-Scott (aka Silver)
A Reflection on 2013
2013 was an amazing year. Let’s start with what didn’t go so well. We had 3 major outages. I can’t remember the details of the few hours of downtime in the Spring, and there was an overnight outage in December when our database couldn’t keep up with zone updates. However it will be hard to forget August 8, 2013. This was my birthday. On the 7th my kids asked me what I wanted. I said, “For stuff not to break.” Sadly, I didn’t knock on wood.
August 8-10 were 18-20 hour days of manually killing database deadlocks and furiously rewriting critical code features while completely removing less critical ones. It was the first realization that our systems were still running in “startup mode.” It was our Twitter Fail Whale. Over the next few months, we rewrote features, and added mechanisms to take the load off our database. For the most part, it forced us to completely put new features on hold while we worked to improve our scalability. The outages put a fairly big dent in our revenue for the next few months, and impacted our plans for growth. It was also the same week we lost our 5 star rating in the Windows Phone store, after maintaining 5 stars for more than 6 months.
For those of you who might not have been following QONQR for very long, you might be surprised to find out that QONQR is a tiny company. For 9 months of 2013, I was the only person working on QONQR full time. Well, not exactly full time. My timecard says I worked 3,215 hours in 2013, which is about 9 hours a day 7 days a week. I’m guessing that is probably below my actual time in chair by at least 20%. To say that 2013 was exhausting is an understatement.
We encountered delay after delay in Android due to the difficulty of the platform, especially in the areas of reducing the risk of cheating. The iOS 7 SDK caused us to rework much of the iPhone application to fix memory issues and comply with Apple’s new best practices. Windows Phone anomalies with advertising and purchasing created a distraction from more productive tasks.
Now that the biggest of the bad stuff is out of the way, let’s talk about what went right. We made two really smart decisions in 2013. We hired Christie (aka Gadgerson) in October. Christie was a friend of a friend and as I found out, a die-hard gamer with online media experience. I hired Christie for a few hours a week in early 2013 to manage our Facebook and Twitter accounts. By October, Justin and I had decided it was a bad idea for me to be spending 10-20 hours a week managing our support email. We hired Christie as our Community Engagement Manager to manage our online presence, support, and QA efforts. Christie has made a hugely positive impact on QONQR and helps make sure the developers stay busy writing code.
The second smart decision we made was to hire an intern over the summer. We took a big chance here. Interns can be a blessing or an albatross. The biggest risk is that they would take more time in guidance than they would provide in output. Our intern, Cole, quickly proved to be a huge asset. He developed nearly all the new web portal features. In fact, the web portal would not have been created without him. We now have an alternate place to do many of the game features -and some portal exclusives- should we encounter a problem that causes features to break in the mobile applications. Now that he has graduated, Cole has accepted an offer to work for QONQR full time, and will start his career with QONQR in a few days.
QONQR players have provided great success for QONQR in 2013. By the end of 2013, our players had launched nanobots (and plasma) over 60 million times. Nanobots have been deployed in nearly ¾ of a million zones in over 200 countries and territories around the world. QONQR now covers approximately 20% of the populated earth. There are now over 30,000 posts in the QONQR player forums.
In 2013, we released the initial version of our public API. This super small API allows developers to poll for zone data on a limited schedule. We wanted to allow our developer players to get just enough data to coordinate major attacks and periodically pull statistics for tracking. We have seen some amazing applications come out of this little API and we hope to add more features in 2014.
2013 may be remembered as the year of GroupMe. There is a common adage in game development: if you don’t build something, the community will find a replacement. Our players did that when it came to private group messaging. GroupMe changed the dynamic of the game significantly this year. Massive attacks of many players from multiple states have occurred throughout the year. Unfortunately, this resulted in hundreds of emails to support, “These 5 accounts have to be multi-scopers!!!” Some had not discovered GroupMe yet, and could not fathom how so many players could attack the same zone at once, then move to a new zone in unison. By now, it is hard to find players that don’t know about GroupMe and we feel confident it is likely to be a QONQR player tool that is here to stay.
The QONQR comic books came to life this year. We were very lucky to find a talented artist that could bring our vision for a QONQR comic book to life. Thanks to your contributions in the “Save the Nanobots Campaign”, Joey Vazquez has been very busy trying to keep up with the pages that have been funded. The profile pics of Scott, Christie and Justin are just a sample of his talent. Be sure checkout the comics.
QONQR extended our part time contractor family by a few members in 2013. It is quite a juggling act to track a half dozen people who are contributing to QONQR 5-10 hours a week on various iOS, Android, HTML, marketing, comic books and other miscellaneous tasks, but it has worked out very well to have the help we need, when we need it. We don’t have the revenue yet to hire dedicated people to fill each role, but working with such flexible and talented people has made the chaos of remote part time development much easier. Many thanks to those who contribute their extra time and talents to QONQR on a part time basis. We appreciate your flexibility and willingness to help move QONQR forward.
Finally, I would like to send a sincere thank you to the many players that have spent money in the game. I received my first paycheck in March of 2013, after 15 months of living on savings and debt. Your contributions have allowed me to consistently draw a moderate salary every month since. You made it possible to hire Christie and Cole, and we believe the revenue will allow Justin to return full time in the first half of 2014. I know many of you feel that the cubers hurt the balance and enjoyment of the game, but those that spend money make it possible for us to continue working on QONQR.
I close with a thank you to my business partner, Justin. QONQR was his idea, without which I would not have something to consume my every waking moment and many of my dreams while asleep. But seriously, although he is not at QHQ every day, he has always done an amazing job of balancing the needs of QONQR with his day job. I’m optimistic our revenue will continue to grow and we can finally bring Justin back to QONQR full time.
-Scott (aka Silver>
The End of the Recruiting Bonus
Over the past few months we have been seeing an increase in recruiting cheaters. This caused us to completely overhaul our medals system last month, only awarding the recruiting achievements for recruits that had completed training (level 100), and only awarding recruiter cube bonuses once per device.
Today, I’m saddened to report that we have suspended all future referral bonus cubes for players that recruited their friends. Ultimately it is the cheaters who caused this problem to bring us to where we are today, but it was Apple that made this necessary.
Without going into technical details, we have found that Apple’s imposed limitations on applications in iOS7 makes it impossible for us to reliably stop cube farming programmatically, and we lack the human manpower necessary to continue to combat this problem on an ongoing basis. So far, the problem has only manifested itself in a handful of players, but recent developments with an abusive player and his friends have forced us to take action today.
We continue to work hard to keep QONQR fair for all players. We receive numerous reports of bad players every day. It is a frustrating and demoralizing part of our day to say the very least and unfortunately takes us away from adding new features to the game. We thank those who play QONQR with honor and integrity. We hope you understand our decision today. Had there been a reliable technical solution, we of course would have endeavored to keep the bonus in place. Your recruits will still receive 2 cubes for using you as a recruiter, at the time of their enlistment.
Thank you for your continued support of QONQR. You make this game great, we are working hard to keep it that way.
-Scott (aka Silver)
The Passing of MrBizzy
Last week, I received a shocking and surreal support email from one of our players in Michigan. A fellow QONQR player, MrBizzy, had died unexpectedly. The person contacting me had asked if we could offer any clues for his friends and family, because MrBizzy was an avid QONQR player and was probably playing much of the day.
It was a bit shocking to see deployments many times an hour, for 6 hours straight, and then nothing. I didn’t expect the feeling of emptiness I felt looking at that last log entry. I’ve read his obituary and wish I had known Vuthy Pen. He was my age, was a co-founder of an IT firm, and leaves behind 4 children. I didn’t know Vuthy, but just the same, I’m saddened by his passing. The entire QONQR family extends our deepest sympathies to Vuthy’s family and friends.
A fellow QONQR player sent this, and asked us to post it.
On Monday, November 11th, Michigan Swarm lost an exceptional player named MrBizzy.
He was all one could ask for in a teammate: kind, humorous, helpful, and never hesitant to rain down missiles or stack a zone to a million bots. He welcomed new players, frustrated the competition, and kept us all laughing. We lost a true friend and a great leader.
However, Swarm does not mourn him alone. An outpouring of support was shown by Legion and Faceless in a temporary cease-fire that began almost as soon as the news of his death was publically announced.
Many players, from all factions and multiple states, changed their in-game avatars to one honoring Bizzy.
The Qonqr forum had a thread this morning discussing a nation- or world-wide way to pay him tribute.
Our team members have received many messages of condolence and support… and all of this means the world to us.
Michigan Swarm feels like a family, yet we recognize that Bizzy’s actual family has lost a father, son, husband, sibling, cousin, and uncle. His game-play was a true reflection of a life filled with caring for others, leading by example, and being a joy to those around him.
Bizzy, you will truly be missed and we are all better players and people for having known you.
The QONQR players on the forums have suggested we make Thanksgiving Day in the United States (November 28th) a day of remembrance and thanks in Honor of MrBizzy. The players have suggested that we “Stack to the Heavens” for MrBizzy, deploying only defensive bots and avoid battling for the day. We agree this is a great way for the community to give tribute to a fellow QONQR player, and show how thankful we are for the new friends we have all gained through QONQR. I will be displaying the MrBizzy tribute image as my profile picture through Thanksgiving, and I hope you will too.
-Scott (aka Silver)