Today I have some sad news to share; we’re shutting down WallaBee.
Before I get into the details, I want to assure you that this decision, whilst only planned in the last day or so, has not been taken lightly. One of the things that I pride myself on is the open way in which we conduct this business and take time to explain when we need to. This will be a lengthy post but I’ll be and frank and honest as always. You will be able to contact me directly, as always, should anything be left unanswered.
The key problem, as I’m sure you’ve noticed recently, is that we are hitting upon some fairly serious technology limits. You’ll have felt over the past year that things have been getting slower and releases have had more timeouts and errors. Whilst we’ve spent a huge amount of time and money trying to resolve this, the truth is that we are running on software that we wrote 4 years ago when things were very different. We now have hundreds of millions of records in our database and this means that things as simple as rendering your profile page are hitting the limits of our infrastructure. I myself am not an expert in server technology and what worked for a small app doesn’t work when you have a huge burst of traffic around an item release. To fix these issues requires a complete rewrite of our application and this is something that we just can’t do in parallel with running what exists already - I also fear that the time required to do that would extend beyond the amount of time we have until the server outages become more severe.
A secondary, but no less important problem, is that we are severely resource constrained. I am the only developer on the team and I manage both the app and the servers in addition to dealing with support. As we’ve grown, the number of support issues (both technical and personal) have increased to a point where I’m spending a lot of the time I have available dealing with either fixing up basic technical problems or arbitrating the latest number collector problem. This has left me with less time to work on the crucial things that actually need doing (such as re-building our server code and the app) and also adds an emotional burden which saps my desire to keep working on this project.
Finally, whilst WallaBee has always made revenue, we are not always a profitable business. Our growing infrastructure has led to substantial costs (upwards of $4000 a month) and we have always paid the market rate for our excellent designers as we refuse to compromise on quality of design. We have no investors; everything has been entirely funded by myself. When we started WallaBee, I knew it would take at least 18 months until we hit break-even and I was prepared to take the financial risk. Whilst we are pretty much break-even now, myself, Simon, and Alan take no salary which makes it hard to justify the huge amount of work we put in each day. The truth is that we simply don’t make enough for me to recoup the substantial investments of money and time that have been spent and it is not something that is sustainable. In addition, we each have fulltime jobs outside of WallaBee and so without the ability to take a salary it is becoming increasingly difficult to be able to balance both roles.
Hopefully that explains our current position and the difficulty we find ourselves in; we have a great product but it is slowly dying due to hitting tech limits; the amount of time we spend on trying to patch it up whilst also dealing with player support means we aren’t making any progress on a rewrite; this is alongside the fact that we are essentially doing it for free and have been for a long time.
Our plan at this point is as follows:
- Today we will remove the app from the App Store so that no new players can download the app. We will also disable our online store and the in-app purchases so nobody can spend any real money. The public websites will be redirected to this blog post.
- On Friday 28th August, we will turn off the public API which will effectively disable 3rd party apps such as Honeyblend and ItemBrowser.com (although I have more details on that shortly)
- On Sunday 20th September, the WallaBee app will be put into maintenance mode for the last time and we will do an export of all of the data which will then be stored securely. Once complete, the servers will be turned off and the website will be replaced with this blog post.
Alan and Simon have long said that I have been taking on too much by working the hours that I do on this project whilst also funding it. My plan is to take a few months break completely from WallaBee and then evaluate what is next. Our current thinking is that we will start a rewrite of the infrastructure and app free from any expectations or support issues and then relaunch in the future with all of our previous data intact (i.e. all of your items would be as you left them). I can’t promise that this will even happen, let alone when it will be, but my expectation and hope is that this is not truly the end.
I’m sure you have lots of questions and so I’ve done my best to answer what I expect to be the most popular ones.
Why can’t you just buy a bigger server?
This is what we have been doing over the past 4 years, slowly improving the servers rather than the code that sat on them. We’ve now got to a point where even with a ridiculously powerful server, our code is just not good enough to work with the huge amount of data we have. Whilst there are ways around that (switching to separate servers for reads and writes, partitioning the database, etc) they are not trivial tasks and require substantial time, effort, and cost and may not be successful. My fear is that we are fast approaching a point where our servers won’t cope and we could start losing data which would be disastrous. Unfortunately there is no simple fix for this; the only solution is a complete rewrite and it isn’t possible to do that alongside maintaining what is a quickly collapsing application.
Can’t you keep everything online but just not release items?
Whilst that would remove some costs (mainly artwork), the scale of our database is such that our server costs would still be prohibitive especially as it would be highly unlikely we would make much trade in selling honeycombs and locks with no item releases. Whilst we would love to keep the community alive whilst we are shutdown, that is something that will need to be done elsewhere. It may be possible that we bring back the messaging and forum features as a completely separate app on a smaller server as that would not cost much to run and would keep the community going; however, I would not start on that until I’ve had some time off and so it likely would be 6 months away. If there is demand for it though, I’ll think about it and I’d be able to restore all existing forum posts and chat sessions in such an app.
Why are you doing this today? Why hasn't it happened before?
One of the major factors in this decision is that we removed one of our last contractual constraints. When we had some serious server problems around 18 months ago, we entered into a very expensive contract with Rackspace to host our infrastructure and provide round-the-clock support; this contract had an 18 month term on it. This ended a little while ago and so we were able to move to a new host that had no contract and was on a more flexible "pay as you go" arrangement. Whilst the support and service they provided was outstanding, it became apparent that we were reaching the constraints of our code base, not the infrastructure we were using (which should be able to cope with millions of users at once). This means that no matter what host we use, we will always have these problems when we release an item due the sudden influx of connections and the way in which our application is structured (meaning the only solution is a rewrite from scratch). After spending 2 solid weeks of moving everything to a new host and still running into performance issues, it has become evident that now is the time to stop as we can't continue to spend as much time as we are on patching something that is doomed to fail. Other time related factors have been an increase in support issues (especially the more difficult cases of bullying, number collector angst, and so on) and the impending release of iOS 9 which could cause issues.
What about the v2 app?
We started on a new version of the iOS app back in November 2012. Since then, it has been restarted numerous times as new technologies have become available and things have happened such as the complete shift in design as a result of iOS 7. Unfortunately, we've never made a huge amount of progress on the app as other issues have taken over such as spending time on server fixes or dealing with support problems. That said, a new app is sorely needed; whilst most of the problems are around our server application, we also have issues with the app itself as we're reaching a point at which we can't update it anymore. The WallaBee app was originally build when iOS 5 existed and many of the newer technologies that Apple are promoting (such as AutoLayout) weren't around at the time; this means something like making the app work natively on the iPhone 6 and 6 Plus is a massive undertaking. The upcoming release of iOS 9 may have caused issues and whilst we've been doing band-aid style fixes over the last year or so, the app just isn't scalable at present.
Surely it doesn’t take that long to deal with support?
You’d be surprised! As we’ve grown, we’ve added lots of new features such as commenting, trading, and forums. I’ve long joked that things were better before we did that but from a support point of view that is true; every day I have to deal with everything from technical problems and bug fixes to requests for information or sorting out the latest number collecting spat. As well as taking up a lot of time, it is emotionally draining; it’s very hard to spend years working solidly on a project and investing a huge amount of money and energy to then have people descend into petty arguments over digital pixels. This is one of the reasons that I’ll be taking time completely away from the project for a few months.
Why not get investors?
This is a personal choice on my part. This is my baby and I’m not keen on being beholden to someone else, either creatively or financially. I already find it a large weight on my shoulders that there are thousands of people using this app; adding some financial responsibility to people other than myself on top of that is not going to do anything for my stress levels.
How about crowd sourcing?
This is similar to the above in that we don’t want the added responsibility of investors be it a single venture capital firm or thousands of individuals. In addition, this is something that needs constant investment to keep it running and whilst we may be able to successfully get enough to run for a month or so, we wouldn’t be able to generate enough to keep us afloat and also do all the work that needs to be done; we’d just be extending the inevitable.
Can’t you make more money by selling unique items again?
Whilst we love unique items, they aren’t actually a profitable venture for us. We lost money on most of the unique items that were sold and whilst we’ve spent a lot of time over the past few months creating an app for the creation of future uniques (along with guidelines and some automation to speed up the process and avoid problems) it would only make this a break-even aspect; it would not come close to the amount we would need to keep going as we are. On a related note, there are plenty of ways we could generate money within the app. We could fill it with adverts, hearts, and timers and use psychology and data analytics to increase purchases. We could create premium markets that require real money for items. We could limit items to 100 issues and allow people to buy and sell with real money on the market and take a cut. We could do all of these things and more… but we never will; this is abhorrent to us and we’d rather shutdown than become like other freemium games which are nothing more than casinos where the house always wins.
Will there be any more items?
Yes, we still have the rest of the Beanstalk set to finish along with an August and September item from our “Women of WallaBee” calendar set. However, we’re going old school and we’ll be releasing these after the public API shutdown and with no push notifications. Why? It’ll avoid having hundreds of people coming online at once making everything slow and it seems nice to us that we go back to the way the game was played originally as we come to the end.
What about the money I’ve already spent on honeycombs and locks?
All of our purchases are for honeycombs, locks, or pouch upgrades and we don’t guarantee that WallaBee will be around forever when selling those. We are continuing to release items for the next month and if we relaunch at a later date then all of your honeycombs and locks will copy across (most likely with added interest). If you have made a purchase in the last 14 days (August 12th onwards) from the App Store, you can request a refund directly from Apple by going to your previous purchases section of iTunes. If you have made a purchase on our online store in the last 14 days (August 12th onwards), then you will automatically receive a refund; this will take 5-10 days but you should get an email confirmation in the next 24 hours. We are turning off the ability to purchase in app and online to avoid having players purchase something whilst unaware that we are shutting down.
What about my unique item?
We don’t guarantee that WallaBee will be around forever when creating your unique item and the cost of the item is all in the design. If you don’t have a high resolution copy of your unique item already, let us know and we’ll supply you with the artwork which you can use for your own personal use. If we relaunch at a later date, then your unique items will still be available to you. The unique gifting portal will remain active until we shut down on the 20th September so if you wish to gift any items before then, you can.
What are you going to do with my data?
Why are you closing the public API so soon?
A simple reason; cost. We are turning off all ability to buy anything in the app and so the last month of app usage will be coming directly out of our pocket. Whilst we’d like to keep the public API running longer, it simply isn’t feasible for us to do that in addition to keeping the game going a little bit longer. If you are a 3rd party developer, we will have all of the data backed up when we shut down on the 20th September so if you want to run a particular query against our dataset, just get in touch and we’ll be glad to help you out.
Hopefully that has answered the majority of questions you may have, but you can comment here on email me directly via firstname.lastname@example.org should there be anything we haven’t touched upon. As I mentioned, this is not a decision we have taken lightly and this is not the outcome we wanted for this project. However, I would prefer to have a clean break with the possibility of coming back in the future rather than watch it slide into a slow and painful decline. Sometimes you need to know when to end a good thing; we hope we have done the right thing.
We had no idea what to expect when we started this crazy journey in the back of a pedicab in Texas. WallaBee has grown to be so much bigger than we imagined and has become a huge part of our lives. I’d like to thank Simon and Alan for all the time and effort they have put into this project and in trusting me enough keep it going for so long. What you may not know is that I have thought about giving up on this project in the past but they have always been there to put me back on course. I hope you’ll also join me in thanking Andy and Robbie; their hard work over the years has been the magic that keeps this game great and I’m sure that their artwork will be appreciated for many years to come.
It only remains for me to thank you for your support over the past 4 years; we have always strived to do what is best for our players and to provide you with the game we wanted. This has been an amazing journey and whilst this is a painfully sad moment, I’m hopeful that you’ll remember the good times you’ve had with WallaBee. I have met some incredible people and good friends through this app and I’m sure many of you have as well.
Gaberbocchus veniet cito,
CEO and Founder of WallaBee