iCloud support

I’m sure you already are, but please consider adding icloud support. It looks like it will work even without iOS devices. And, it seems that it will solve some of the complexities of dropbox syncing (ie needed to ensure that younclose the file on one Mac before opening it on another). I would even pay to upgrade to 3.0 that has lion support and iCloud baked in.


Absolutely! And full Lion support. I couldn’t wait to purchase this software but watching the Dev. Conference keynote has me jonesing for all my favorite apps (Scrivener now tops among them) to be Lion ready.

Although keynotes are always exciting, try to remember that we developers have no more idea about what is going to be introduced than anything else. iCloud is brand new and I know very little about it yet. I’ll absolutely be looking into it but I cannot promise anything - I have yet to see whether or not it will be appropriate for Scrivener.

As for everything else, of course Scrivener will be taking advantage of certain Lion features, and a lot of work has already been into making Scrivener 2.1 Lion-ready.

All the best,

I doubt that iCloud will change anything regarding Dropbox “problems” like the need to close the file on one computer before opening on another. Those are inherent in pretty much any synchronization protocol. If iCloud doesn’t have a complete copy, it can’t give a complete copy to the second computer. And if the master copy lives in the cloud, local computers are going to be pretty much stuck if they don’t have connectivity.


Here’s a quick article that talks about data consistency in the cloud:


And if after reading/skimming it you don’t think you would have these problems, just think of the simple collaboration scenario: two writers trying to work on the same project at the same time. Won’t work. No magic bullet.

Now, I haven’t looked at the iCloud APIs and perhaps Apple has added the ability to lock a resource and to retry the mutex. That would help. But it still wouldn’t fix the problem. Data consistency is oftentimes a problem that can only be fixed by a human being staring at the differences between two files and editing a third to make it consistent. And Scrivener is especially difficult since (a) it requires knowledge of the semantics of the project’s XML description and (b) it requires the locking of more than one resource at the same time in order to maintain consistency. For example to add a new item to the binder, the project’s XML file has to be updated and the RTF file itself needs to be added in the Files/Docs directory.

Perhaps everyone knows about these problems. My apologies if I’m stating the obvious.

Our sour note about this brouhaha: everyone will have to upgrade to the latest, most expensive iOS devices. Already out: 3g iPhones. Soon to follow: the iPad 1. We bought this stuff only two years ago, shelled out obscene amounts on phone bills, and now we’re expected to make another upgrade? Humbug.

jravan, I think you’re spot on with the problems involved. The trouble with keynotes is that they - of course - are all about showing off the “wow” factor of new features and don’t have time to cover the practicalities. In this instance, the “It just works” refrain could easily lead many users, who understandably don’t realise the technical practicalities involved, to think that something like iCloud is capable of magically and instantly keeping all of their devices in sync all of the time. What it washes over is how all of those devices need to be turned on and connected to the internet, and how you have to wait for them to sync (so only small files will sync instantly - many of us have very poor internet connections, too; making Lion download-only, for instance, will mean many, many users in the UK will have to wait eight hours or more for it to download).

For a Pages document and the new versions feature, that’s going to work great; for a package-based format such as Scrivener’s there is always going to be the problem of dealing with a project being open at the same time on two machines trying to sync a structure and multiple files within it. Everything in the keynote showed files being pushed to other devices - nothing there suggested that it solves the basic problem of a program with complex and structured data reloading based on remote changes and dealing with conflicts. I could be wrong, because I haven’t had chance to look through the APIs much yet (and couldn’t comment even if I had, seeing as I’m under NDA), but so far I’m not sure that iCloud would provide a better solution than Dropbox for Scrivener documents. I will, of course, look into it though.

All the best,

druid - Yes, and we’re probably going to have to stop supporting Tiger and PPC machines in the near future, too, because Apple have dropped support for such things from their developer tools. I think they want everyone to upgrade…

To clarify that, by stopping support I think you mean continued maintenance and feature additions to the application itself, not tech support. We still support people using 1.x, and would support those on 2.0.x that can’t go any further with future updates due to dropped support for their platforms at the development tool level. It will become increasingly difficult to support older platforms and OS’s as it is harder to replicate bugs that only trigger on a particular chip, but on that same coin, a hot fix for that bug would not be feasible in an Xcode 4 environment, even if the cause were nailed down.

Of course, correct me if I’m wrong, but I think that is what you meant by support.

Yes, sorry, I just meant that updates in the future might not work on Tiger or PPC, but we’d still support those platforms using an older build of 2.0. Nothing’s definite yet, though, as I’m trying to find the best solution.

I have been using a Scrivener - Export Sync to Folder - Dropbox - PlainText for iPhone/iPod Touch combination.

Even if Scrivener never comes out for the iPad, if I can justify using an ipad to write I might consider continuing to use this set up and making an iPad my on the road writing device…

Not sure if typing on the screen is something I want to do though. Also, having a bluetooth keyboard kills the portability of the iPad for me… decisions decisions.

I think the fundamental issue is that the Cloud is not a remote hard disk.

It looks like one, the provider wants you to think of it as one, but it isn’t. There is always going to be a gap between creating data Here and saving it There. The more complex the data (Scrivener, DevonThink, et. al.), and the longer the gap (slow connections, intermittent connectivity, etc.), the more opportunities for disruption. Neither Steve Jobs nor anyone else can make that gap go away.