Answers to Assorted Cloud Service and Sync Questions

[rob]

This thread attempts to answers assorted questions pertaining to the cloud and syncing that have appeared in various threads and support questions. I’ve assembled them all in one place so that interested parties can more easily find them. Please read through and you’ll hopefully have a better sense of what the plan is regarding the cloud and syncing. If you have not yet seen our recent announcement about cloud service and corresponding FAQ update, please start with those, as this builds on the information provided there.

Before I launch into actual answers, I must first make a request. Some of the concerns expressed in other threads and support questions were truly surprising to me, as they interpret things in the worst possible way and then extrapolate to alarmist predictions about how we’ll handle things. We’ve managed to build a successful business by creating products that tens of thousands of gamers have flocked to for nearly 20 years. We never would have accomplished that if our practices came anywhere close to some of the assumptions being drawn. So please give us the benefit of the doubt and ask questions before leaping to conclusions.

1. From the very beginning, users have been able to backup their data independently of the cloud. It’s a simple process that’s managed via the Launch Dialog. The cloud provides an alternative means of backup so you don’t have to worry about it yourself, but that’s not its purpose – that’s just a side benefit. If your only purpose for having cloud access is for backups, you probably don’t need the cloud service and can simply choose not to continue it. Just make backups of your data like you would for any other program on your local computer.

2. Syncing is a carefully choreographed interaction between the desktop client and server. The only content that gets sent is the content that has actually changed. If you have 1GB of data across all your realms (some users do) and you only change the name of a topic, the only thing that gets synced is the change to the topic. If you were to use Dropbox or another file storage service, you’d copy the entire 1GB every time. For users with bandwidth concerns (either speed or total transmit size), this is a huge consideration.

3. The future of software like Realm Works is ultimately the web. Users will want to access their content on desktops, tablets, wherever. They will also want to share content with players across the full spectrum of devices. That means the web. The desktop solution is necessary as an initial step for a few reasons, such as processing power, offline access, bandwidth limitations, and others. However, those concerns will be diminishing over time. That’s why we started on the desktop and mapped out Realm Works’ evolution to the web from the beginning.

4. Due to the web evolution being planned from the beginning, a number of the especially complex aspects of the product have been exclusively targeted to the cloud. Yes, we could do them on the desktop as well, but it would be significantly more complicated to do so and we have very limited resources. The additional delays would set us back even further, and the platform priority for users would continue to shift even further to the web while we did it. That doesn’t make any strategic sense as a business. This is not an “artificial” constraint but one of practicality that was part of the design from the beginning.

5. Access to anything via the web will fundamentally require cloud service. This applies to GMs wanting to access their content via the web and players accessing revealed content via the web. Without the cloud service, GMs will have access via the desktop application and can move their data between computers through the use of backups. In addition, the only means of sharing content with players without cloud service is through the integrated Player View mechanism.

6. Access to the Content Market will not require cloud service. However, what you can do with the content you’ve purchased will be impacted by whether you also have the cloud service (see the FAQ for more details). I’m sure that some users will have differing opinions on what they should or shouldn’t be able to do with or without the cloud service. That’s a reality for any business on the planet. Our focus is on the majority of consumers who play RPGs and bringing them a solution that meets their needs, keeping in mind our resource constraints as a small business. While our desire is to create something that everyone will want and find ideal, the reality is that’s impossible, and we have to be OK with that.

7. There will be two levels of web-access for players. There will be a free level that all players can access and the Player Edition. The free version will provide access to topics, maps, plots, and everything else. The difference is that the free version will only provide access to a subset of revealed information, while the Player Edition will provide access to the entire realm – just like it does on the desktop. The GM will be able to control what subset of information is made available to players through the free version, and it can morph as the game progresses. The focus is on something that casual players will find useful, and we expect this solution to work well for more than 50% of all players (based on our own experiences and the anecdotal data from numerous users).

8. If you discontinue the cloud service, you haven’t lost anything unless you delete everything off your local computers and discard your backups. All that happens is you lose access to the cloud. You can still utilize the full product on your local computer, and you can still make backups of your content exactly like you would with any other program on your local computer.

9. The website for Realm Works has been in place since before we sold the product. The “standard tier” reference on the pricing page and elsewhere is there for one specific reason. There are some users who utilize the product heavily and others who utilize it much less. As with anything in life, there is a wide spectrum of usage levels across users. So there may be a need to introduce one or more additional “tiers” of service at some point to better accommodate users with divergent usage patterns. It’s also possible we won’t need to do anything, but we’ve made sure to properly clarify the “standard tier” – just in case – so we can easily introduce new tiers.

10. The projected storage limits of the “standard tier” are way more than necessary to accommodate multiple Pathfinder Adventure Paths, assuming you don’t do something unusual like include all the images at abnormally high resolutions. As technology advances, we’ll hopefully be able to utilize services that give us more storage and bandwidth for the same price over time, which would allow us to correspondingly increase the limits for users. That’s our expectation, at least. At the moment, we have imposed no limits. Before we ever do so, we will give users appropriate information about their usage and provide mechanisms that users can leverage to reduce their usage.

11. Comparisons between the Realm Works cloud service and file-centric services like Dropbox are akin to comparing an elephant to an elephant seal. Yes, they both have the word “elephant” in them, but that’s about where the similarities stop. The same with Realm Works versus file-based cloud services, where they both involve the “cloud” but do so in dramatically different ways. With a file-based service, you copy the file, as a whole, every time, and that’s about all they do. If you want a smart backup solution, they work well for that, provided you don’t mind copying the entire file every time (about 1GB for some users). In contrast, the Realm Works cloud only entails copying the individual elements that you’ve modified when syncing. It detects collisions if you make changes in different places and warns of the conflict. It allows GMs to share their realms with players – via Player Edition today and web access in the months to come. It allows players to make notes that directly link to content created and shared by the GM, and this will extend to journals and other similar capabilities down the line. And it will enable GMs to access their realms via the web later this year. If none of these capabilities is of interest to you, then you have no need for the Realm Works cloud service. However, we believe that one or more of these capabilities will be attractive to the majority of GMs out there. [For even more examples of how the two differ, @Viking2054 provides some insights in this post.]

12. Creation of realms has no requirement for cloud service. We have hundreds of Kickstarter backers who opted not to get the cloud service. They have been creating realms and fully utilizing the product since it was first released. Creating realms DOES require an internet connection for a few seconds, but there is no cost and no need to utilize the cloud service in any way. [Answer 3b in this thread from @Viking2054 accurately summarizes more technical details that we've also outlined in our FAQ.]

13. Product updates have zero relationship to the cloud service. All of our Kickstarter backers who opted not to get the cloud service have received every product update that every other user has received – in the exact same manner. As @Viking2054 theorized in that same post, Realm Works product updates will be free unless we opt to introduce some major capabilities that would present added value options to users. The cloud service is separate in this manner, as might be map creation tools.

14. If you have the cloud service and choose to integrate purchased content into your realms, then you let the cloud service expire, your realms have already been created and the content has been integrated. Consequently, all of that integration has already been performed and the results are already in place. Absolutely nothing happens to the content you’ve created, provided you sync it down to your local computer. Unlike many other products and services, this is a not a subscription service where you’re effectively “leasing” the content and it gets taken away from you if you stop paying the rent. Obviously, if the cloud service expires, you will no longer have the ability to leverage the features provided for new content or realms, but existing content and realms remain intact. For example, let's assume you have three separate realms and purchase ContentX, at which point you integrate ContentX into two of those realms. You sync everything to your local computer and let your cloud service expire, and both realms continue to have ContentX fully integrated into them, exactly as you had before the cloud service expired. If you now decide you want to integrate ContentX into the third realm, that won’t be possible due to the lack of cloud service to orchestrate the process, but the first two realms have it all.

15. Duplication of realms is distinct from “inheritance” of realms. Simply duplicating – or cloning – a realm makes a snapshot of the original realm that is wholly decoupled from the original realm. The “inheritance” of a realm allows you to create a realm that continues to inherit changes from the original realm. This “inheritance” mechanism is the identical mechanism used to integrate purchased content into your realm. In effect, you “publish” one realm to yourself and then “import” that material into another one of your realms. So it’s all orchestrated through the exact same mechanisms used for consuming purchased content. There is nothing artificial about the requirement for the cloud here. It’s just a design approach that allows us to keep the highly complex logic in one place – the server – where we have complete control over all the variables. It’s a matter of practicality.

16. When you “import” content into your realm, you can decide whether it should be a snapshot or “inheritance”. That decision will apply to the entirety of the content you’re importing – it’s all or nothing – and there is no changing your mind. In addition, “inherited” content will not just magically appear in your realm. You’ll need to trigger an “update” operation to import whatever changes have occurred. So you could import a realm using “inheritance” and at some point choose to stop applying any updates, effectively leaving the imported content frozen in time. The “update” logic will similarly operate on an all-or-nothing basis for a particular body of content. If you purchase ContentX and ContentY, importing both into your realm, you will then have the option to “update” the material from ContentX, ContentY, neither, or both. However, you will not be able to pick and choose whether to update individual bits of content from ContentX.

17. The Content Market should not be conflated with the cloud service – they are entirely distinct. We expect many users to purchase material without having the cloud service. We also expect many other users to leverage the cloud service without ever buying anything. Users that have both will simply be able to leverage additional capabilities.

Alright, it’s time for me to get back to work on the product. I think I’ve covered all the major questions and concerns expressed thus far. If I’ve missed something or something isn’t clear, let us know. Either Liz or I will provide answers where we can. Please understand that there are some details that we can’t delve into yet, so we might not be able to answer everything. Hopefully, the above provides a lot of useful information.

Thanks, Rob

[Parody]

It's good to see more of how you plan on it working. Most of this is about future features, so I don't have much to say about it at this time. I do have one side comment, though:

Dropbox does partial updates. It can't do them on the application level (which allows for much more optimization) and it only tries to do partial updates on larger files, but for many changes to large files Dropbox uploads much less than the entire file. In your example, unless changing the name of a topic results in many widely-scattered bytes of the file changing, Dropbox would only upload 4 MB of the 1 GB file.

[rob]

Very cool! I didn't know that about Dropbox!

[ibecker]

Rob,

Thank you for taking the time to articulate all of this in such a clear and concise manner. Hopefully, this will rein in some of the rampant speculation that happens across these boards (though that might be asking for too much... ).

I, for one, am very excited about the direction you've laid out here!

-Ian

[wilcoxon]

3-5) Web access to content in no way necessitates cloud. It is often the most convenient for the user but it is certainly not required. Given the choice of paying for an ongoing cloud subscription and spending a few hours configuring things so that I could serve content without the cloud, I would choose the latter every time. However, we are not being given this option.

15) I am HIGHLY disappointed that I can not have realm copies still linked to inherit changes without cloud service. I have not used this functionality yet but anticipate needing to do so in the future.

17) Without the integration feature (without cloud service), I will likely not be buying anything from the Content Market. I can live with that.

Thank you for the explanations of the reasoning behind some of the decisions. I suspect many are still upset about the limitations being imposed without an active cloud subscription.

[rob]

Originally Posted by wilcoxon

3-5) Web access to content in no way necessitates cloud. It is often the most convenient for the user but it is certainly not required. Given the choice of paying for an ongoing cloud subscription and spending a few hours configuring things so that I could serve content without the cloud, I would choose the latter every time. However, we are not being given this option.

15) I am HIGHLY disappointed that I can not have realm copies still linked to inherit changes without cloud service. I have not used this functionality yet but anticipate needing to do so in the future.

17) Without the integration feature (without cloud service), I will likely not be buying anything from the Content Market. I can live with that.

Thank you for the explanations of the reasoning behind some of the decisions. I suspect many are still upset about the limitations being imposed without an active cloud subscription.

You're correct that web access could be provided without requiring the cloud. We made that choice based on practicality, in exactly the same way as other choices I outlined above.

The vast majority of GMs and players lack the technical expertise to configure their environments in the ways you're referring to - you're obviously an exception. Of the small percentage that do have the necessary expertise, many of them would prefer to not bother with hours of configuration time and simply have something work without any hassles for a few dollars a month. That means only a tiny percentage of our overall user base would find it desirable to spend those hours configuring their environment appropriately.

From our end, it would require significantly more work for us to make that possible. Assuming we did add that capability, we would then have to spend significant amounts of time providing technical support to users who want to get it to work but have a myriad different equipment configurations that complicate everything. So our cost increases dramatically both on the development side and on the support side.

What it sounds like you're saying is that our business strategy is unsound because we've chosen not to pursue this path. It sounds like you believe we should instead choose to invest significant resources into developing a capability that will only be used by a tiny fraction of our users, that incurs significant additional support costs, and that provides us with zero additional revenue to offset those costs. Unfortunately, that strategy simply doesn't make any business sense to me.

Our strategy is to invest our efforts into features that the bulk of our users want. And I'm confident that the vast majority of our users feel the same way. So that's exactly what we're doing.

Edit: Thanks to the extensive user survey we completed a few months ago, we have a concrete understanding of what the biggest priorities are for our users, and the results of that survey provide us with an excellent roadmap for the months ahead.

[MNBlockHead]

Rob, thank you for your detailed response. Very happy to hear how the integration will work (#14). I can understand RW not continuing to support updates if I'm no longer a subscriber, so long as I don't lose existing integration.

[Viking2054]

Wow, three references to my posts, I feel honored.

I have a small question about topic number 15 and 16. I hope my theory is correct and I hope I understand it.

suppose I create a unique realm of my own and want to keep it pristine for use in multiple campaigns. But I also want to be able to expand the original creation and have it propagate to the copied realms. This is when I would use the Inheritance feature. Would I have to create a blank realm first and then tell it to inherit everything from the 'pristine realm' or would I need to make a snapshot first and then tell it to also inherit? I'm also assuming changes, additions or deletions made in the inherited realm do not propagate back up to the 'pristine realm'. Is it to early to tell us exactly how this might work?

If I later purchase some content from the marketplace and want to integrate it into only one of my campaigns from the above example, I have the option of a snapshot or inheritance. Would the inheritance method allow corrections of the purchased content made on LWD's end to propagate to me through a simple sync or would I have the option of not updating that portion if I don't want the corrections? I think I get the gist that a snapshot would be a one time integration with no further migration of content. Again, if this is to early to clarify then I can understand that.

[Vargr]

Interesting reading, Rob. And thank you for taking the time.

It looks very promising - especially the inheritance system.
I am looking forward to have all these things in place.

[MaxSupernova]

7. There will be two levels of web-access for players. There will be a free level that all players can access and the Player Edition. The free version will provide access to topics, maps, plots, and everything else. The difference is that the free version will only provide access to a subset of revealed information, while the Player Edition will provide access to the entire realm – just like it does on the desktop. The GM will be able to control what subset of information is made available to players through the free version, and it can morph as the game progresses. The focus is on something that casual players will find useful, and we expect this solution to work well for more than 50% of all players (based on our own experiences and the anecdotal data from numerous users).

I'm not sure I understand the difference.

The quote above says that the free version will only display a subset of revealed information. How is that subset defined? Why not show all of the revealed information? It says that the GM will be able to control the subset, which is exactly what "reveal" does. Is there some new mechanism that will interact with reveal?

It also says that Player Edition will provide access to the entire realm, as opposed to the revealed info that the web client will show. Currently the player edition only provides access to revealed snippets. I don't want my players to have access to the entire realm. Why would this happen?

This paragraph is very confusing and unclear.