So far the release of the paid version of Jungle Disk has been going great. We’ve actually seen an acceleration of usage and downloads since the release. With that hurdle behind us it’s time to take a fresh look at our roadmap for the near future and beyond. While most Internet companies keep their future plans close to the chest, I think it’s important to share a roadmap with customers so they can help us make sure we’re on the right track – there’s nothing worse than building features that no one wants! Keep in mind that everything below is subject to change however.
In the near future (3-4 weeks) there will be one or two more non-beta releases to fix a few bugs that have come up and add some minor features. The most important bug we’ve fixed recently involves occasional spikes to 100% CPU usage by the Jungle Disk process following a loss of network connectivity or system sleep. We’ve also fixed some issues with proxy authentication and activity logs. In terms of features, we’re looking at adding an optional startup password as well as a dialog-based restore procedure to aid in robust restores of large amounts of data.
Mid term (2-3 months) our focus is on two key areas. First, on the client we are planning on adding native file system integration on all the platforms. Currently we rely on the WebDAV support in the operating system to integrate with the file system. This had the advantage of providing a single interface for us to support across all the platforms. However, we’ve also come to learn that the operating system WebDAV support can be flawed or buggy in some areas, and each platform has its unique issues. The majority of the support requests we see these days are due to problems with WebDAV rather than issues with Jungle Disk itself. While we can work-around many of these issues it can be difficult to ensure that everything works out of the box for all users. By integrating with the native file system we will bypass the WebDAV layer which will improve reliability and performance. This process involves some low-level work that is unique to each platform and so it will take some time to complete, but in the end we think the product will be more robust than ever.
Secondly, on the backend we are planning on offering support for backing up only the changed portions of large files (delta-updates) rather than the entire file. We’ve been working on this for a while but want to make sure it’s working just right before release – the integrity of your files is the most important thing.
Longer term we want to continue to build a robust backup and personal storage product, while expanding into some of the broader uses for online storage such as web-access to your files, public and private sharing of your files, and “groupware” support where multiple users can have their own accounts with a shared storage space.
What are your thoughts? Are we on the right track? Feel free to respond to this post, or if you’ve got suggestions of your own post them in our Feature Requests forum.
28 Comments
RSS feed for comments on this post · TrackBack URI
Subscribe to our feed
Luhmann said,
August 21, 2007 @ 7:39 am
I think the near and mid-term goals are spot on, but I think your longer-term goals are moving in the wrong direction. While I would love to have web-access to my files and public and private sharing, etc. and probably even requested these features myself at some point, I now think that this would move you two far away from your core strengths.
More useful would be to look at additional features that can be built upon your backup tools and tight integration with the various file systems. The one that leaps to mind is syncing. It would be great to be able to sync changes across computers quickly and easily. Right now I use a complicated setup using Unison (two-way Rsync) and I would love something simpler and more robust.
Another would be a good versioning system that would allow saving multiple versions of different files, not unlike Apples forthcoming “Time Machine.”
Irelandshope said,
August 21, 2007 @ 8:27 am
Hi Dave
I think you have really nailed the near and mid term goals in fact they are the things I would have requested.
I agree with Luhnann that further maturing of JungleDisk as a backup client would be great. Dont get me wrong you could stop development today and it would fill my needs but a few extra items and it would be great. Limited versioning would be a great benefit as would folder syncs across machines. I’m very much of the school of thought that really great apps are those that do one thing but do it very well.
At some point the backup portion will be pretty much fully mature with nowhere left to go and at that point sharing etc is the natural direction.
Just my two cents worth, you have always been ahead of the curve with features and I know whatever way you go it will work great.
Thanks for a great app. I sleep peacefully knowing a copy of all my important stuff sits waiting for me many thousands of miles away.
Tom N said,
August 21, 2007 @ 9:05 am
In terms of backup, it would be useful to have multiple backup ’sets.’ Right now, the automatic backup is scheduled all at once. It would be nice to be able to back up different sets of files at different intervals.
And for the long term, having a good way of web-accessing files could be nice. I’m envisioning something where I can upload photos and videos to S3 via Jungle Disk, and link to them from my blog and/or Gallery.
But even with where it is today, I’m pretty happy with Jungle Disk. Now if I could only get my cable modem to quadruple my upload speed, things would be even better…
mcglinch said,
August 21, 2007 @ 11:29 am
have been enjoying jungledisk and right now my main ‘want’ is to be able to configure upload speeds for different periods of the day (no limit overnight, and
Tobias Linder said,
August 21, 2007 @ 12:08 pm
The feature I’m missing most in Jungle Disk at the moment is versioning. I love it the way the mac appliation Synk does the backup-versioning. It creates in a separate archive-folder a folder for every backup and moves the non-existing file in the source to this folder with the same data-hierarchy. After a defined number of days it deletes this folders too.
Jeffrey said,
August 21, 2007 @ 12:25 pm
I add my name to the list for the backup sets / time machine / versioning concepts – though it’s be difficult to get the right balance between features and UI to expose the options.
I also add to the flexibility of backup list. Different upload speeds for different periods is a good start. Alternatively, to say “backup every hour during x:00 and y:00″ might be good.
A small feature that would have value would be the ability to get a single preview of all my backup options. That is, I have 2-4 backup rules and it’d be nice to see a preview of the result of all of it: a preview of everything to be backed up. I also think the preview could be summarized in some fashion to increase signal::noise and help with the feeling of safety for users.
Another possibility might be to look at where Microsoft is going with their stuff. The recent beta of Windows Live OneCare allows for “circles” where each computer reports back to a main server. I think the idea here is that power users act as IT support for a “circle” of people (usually a house – sometimes several houses): their family.
It’d be nice to be able to look at the current “backup status” of all the computer that I “admin” at home.
That’s at home… I’m not going to even start on features for the SOHO space where I also use this product.
James Woodruff said,
August 21, 2007 @ 12:31 pm
Seems like everyone is on the same page and you are moving in the right direction. The Help forum has been great and responsive, but for those not as blessed with PC data storage terms and knowledge a really good help file is in order. The new mapping workaround hasn’t missed a beat yet.
Make it a great day!
Steve Bush said,
August 21, 2007 @ 2:01 pm
I like your directions, I especially like that you *have* directions and are willing to share and discuss them! Not all developers let you know their roadmap.
File system access is not why I use JD. If I saw my JD as one big binary blob (like a tar or zip file) it would be ok with me as long as my disks are backed up and I can restore stuff if needed!
Why I *do* use JD is that it is fully offsite and gives me a measure of backup protection unparalleled by *any* local solution. Hopefully I will never need it, and I don’t need to access it in any other way (like as a virtual disk drive).
Although, backing down on that, sync of folders between machines is really great, a 2nd priority.
What Apple does with TimeMachine remains to be seen, but when all the details come out, it may change everything for me. I suspect *it* is what I want, and then all JD needs to do is backup the backup! I suspect it will set a new bar for the nature of backup programs, and I don’t want to see JD become obsolete or competing with that feature set.
What JD does (and does very well) is give me OFFSITE backup, which no amount of local software can ever do. So that is what I want from JD! (Although it also remains to be seen if Apple integrates .mac or other remote storage solution directly into TM. If that occurs, I hate to say, JD is lost).
Jungle Dave said,
August 21, 2007 @ 2:08 pm
A quick note on Time Machine – one of the main reasons we are looking at file system integration is to allow for possible TM support. From what we can see so far, TM doesn’t have any remote backup support (.Mac or otherwise) built-in and doesn’t even support WebDAV. So a FS driver will be the only way to support it.
Chris Judd said,
August 21, 2007 @ 7:42 pm
I love the product and could not ask for anything more…but…I want incremental backups more than any other feature. If I could get incremental backups, I would not have to keep running my own backup scripts.
-Chris
Vacilado said,
August 21, 2007 @ 9:36 pm
I can’t wait for the delta technology. It is IMO much more important than working on removing the WebDAV dependency!
The delta (differential) would mean vast improvement in speed and useability of the files stored on S3, while the WebDAV removal means hardly more than removing some instabilities in various operating systems. You will go into the different FS details and this will bog you down, where now you debug WebDAV, you will be debugging a number of various systems.
I really believe that if somebody else gets some kind of delta upload working with S3, JungleDisk will lose its so far promising future immediately… and that would be so sad for all of us who believe it it.
I think backups are not crucial right now in JD! There are many S3 backup programs emerging and they are better thank what JD can do in terms of backup (see e.g. S3 Backup). I am happy to keep using those, why not – they specialize on what they are good at!
I love JungleDisk because it can map a virtual S3 drive, this is your gem, this is what nobody can do better right now. It works fine for small files, but working with a few megabyte files, such as some images, mail storage files, etc gets so slow that it makes JungleDisk royal pain, unusable, with other concurrent up/downloads timing out etc.
Eric2 said,
August 22, 2007 @ 5:24 am
I would really like JD to be secure and stable for backing-up big files like mail storage files.
I am really happy that you plan on offering support for backing up only the changed portions of large files (delta-updates), that would also reduce the cost and time of re-transferring the same big mail storage file, only because of few more mails in the file.
SASniper said,
August 22, 2007 @ 2:44 pm
These three quotes sum it up for me.
Steve Bush: “What JD does (and does very well) is give me OFFSITE backup, which no amount of local software can ever do.”
Chris Judd: “I love the product and could not ask for anything more…but…I want incremental backups more than any other feature. If I could get incremental backups, I would not have to keep running my own backup scripts.”
Vacilado: “I can’t wait for the delta technology. It is IMO much more important than working on removing the WebDAV dependency!
The delta (differential) would mean vast improvement in speed and useability of the files stored on S3, while the WebDAV removal means hardly more than removing some instabilities in various operating systems. You will go into the different FS details and this will bog you down, where now you debug WebDAV, you will be debugging a number of various systems.”
Otherwise, keep up the excellent work
Luhmann said,
August 23, 2007 @ 7:33 pm
JungleDave said: “A quick note on Time Machine – one of the main reasons we are looking at file system integration is to allow for possible TM support.”
Cool!!! Now if I could only use Jungle Disk to boot up my mac!
Robert Accettura said,
August 25, 2007 @ 6:03 pm
I’d love to see internal compression. Would improve upload times, as well as reduce storage costs. What about adopting something like 7z? Vary the level of compression to allow for not sucking to much cpu. Would be an awesome way to reduce costs. It’s not like you can really read off of S3, so why not compress then encrypt?
Kijana Woodard said,
August 26, 2007 @ 9:58 pm
I’ve read the other comments here. It depends on what you want to do. Other people could write software that does funky backups (versioning, advanced restore control, encryption etc). All this stuff sits on top of having the online disk drive.
Personally, I’d like to be able to share large files with others and host media for various my various websites.
For small businesses where various people work from home, this would be fantastic. We could have access to shared files without emailing them around or setting up remote networking.
Jack Boyce said,
August 30, 2007 @ 11:07 pm
Interesting comments. I see a couple of different product visions here:
(1) Be the invisible {versioning, offsite, incremental} backup tool that Just Works, and I hope I never need
(2) Be my Big Bad Drive in the Sky that I can read/write to from any machine, and which gives me a single consistent view of my data from anywhere
The things you’re planning to do in the near term (eliminating WebDAV dependency, delta-updates) serve both aims. In the longer run I suspect you’ll need to make a choice between (1) and (2).
Personally I would find both very useful, although I mainly use Jungle Disk for (1) today because of the absence of a syncing mechanism. Technically (2) is much more challenging I think, especially if you’re trying to make it work cross-platform, but it would be really cool if it worked well.
MikeB said,
August 31, 2007 @ 1:21 pm
>> First, on the client we are planning on adding native file system integration on all the platforms.
Jungle Disk » Blog Archive » Jungle Disk 1.43 Released said,
September 4, 2007 @ 9:18 am
[...] The next release will be focused on features as described in our recent roadmap post. [...]
Mark B said,
September 4, 2007 @ 12:31 pm
JD is simple and that is what makes it so good. Set it up and let it do its work in the background.
My top priority would be to have a function to restart files after they have been stopped half way through.
I support the suggestion for compression and the use of 7z.
I would not want a web interface as that would mean I would have to give away my encryption key and store online. This compromises the security of the solution.
Dana Seaman said,
September 8, 2007 @ 1:44 pm
I think the main focus should NOT be on replacing WebDAV but:
1. Versioning so I can save multiple copies of source code files in case I need to roll back.
2. Uploading only newer files. I currenly use SecondCopy as a front-end for JungleDisk to do this.
4. When you want to quickly update a single folder(can have subfolders) I find PcMagazine(ZdNet) WMatch can do this quickly and let you know beforehand what files are different. This can also be used to restore files.
Stu said,
September 16, 2007 @ 7:00 pm
Compression, for sure. CPU is cheap but network, especially to the US, is slow for much of the world.
Multiple backup jobs so I can specify different schedules for different things. Some things are important, some less so. Using ‘Backup cleanup’ is a real pain when everything is stuck in the one huge bucket. Specifying some jobs as auto-clean would also be good (although a full blown snapshot/timemachine type setup would be better).
Steve said,
September 19, 2007 @ 9:36 pm
I am looking forward to managing multiple client backups from a single account and an auto update function. At the moment, too much time is spent periodically checking each backup and updating to the latest version.
Otherwise fantastic and reliable product. Great work!
Valentin said,
September 25, 2007 @ 4:48 am
I have a very simple wish: a way to access (even if its read only) the data in the Jungle Disk Cache when offline.
I for one made the cache size so large that everything I ever uploaded (from this computer) should be still in the cache and it would be great to just access it there – even when offline.
Charlie K said,
October 8, 2007 @ 6:09 pm
I would like to suggest that a pause button that would allow you to suspend a backup if you need the bandwidth would be good.
Also, I think the UI could use some tweaking. It is sometimes hard to understand what the product is doing. I’d like to be able to see a list of files and directories that it is backing up.
Steven said,
October 15, 2007 @ 12:23 am
So does Jungle Disk support incremental backups now?
Jungle Dave said,
October 15, 2007 @ 4:42 am
Jungle Disk supports incremental backups at the file level – we’re working on support it in on the inter-file (block level) in the near future.
M Jungle B said,
January 12, 2009 @ 9:59 am
I support the compression suggestion…
Add an option (checkbox) to ‘compress all files’ for upload; then add the decompress option for the download. This will save several extra clicks and processes in which the user must monitor.
I work in the medical field and transferring patient studies with Jungle Disk has potential. The automatic compression will save the user time and money. Also, have you ever walked non-technical users through file compression… jeez.