Any hope of a true cloud-based future for Devonthink?

[Janov]

@rkaplan

When you get into the gigabyte data range Devonthink syncing does not work so well.

I have 6 databases syncing perfectly, all are larger than 3 Gb. I myself have never had any issues with any part of DT sync.

[BLUEFROG]

Welcome to the forums, @Janov! Glad to hear about your good experience!

[korm]

Just to note, “sync” and “share” (the OP’s request) are two different things that sometimes overlap quite a bit. But they are still different.

[OogieM]

@rkaplan

Why are you letting fans like me slip away rather than bringing Devonthink up to date?

From another point of view, if DEVONThink goes cloud only I will leave. Cloud is not really viable for many folks for a variety of reasons, security and access to the Internet being the primary ones. Try living in a rural area before you say cloud services are the bees knees.

[rkaplan]

@OogieM

From another point of view, if DEVONThink goes cloud only I will leave.

Nobody ever suggested that DEVONTHink go cloud only.

[CaptainEddie]

Amen! We are in rural upstate New York. We connect to the Internet via cellular broadband. Our local cable company is offering us a heck of a deal…over $19K to connect plus all other installation fees and charges for basic Internet. We can’t afford any cloud only services.

[OogieM]

@rkaplan

Nobody ever suggested that DEVONTHink go cloud only.

I believe you did with this statement.

@rkaplan

The sync feature is still a bit clunky and in any event does not really allow for true web publishing/sharing of data.

Any chance Devonthink will move to the cloud?

Cloud only is not an option. Even cloud preferred is not really good for me because of the limitations of access.

@rkaplan

But I also think we are beyond the point of dismissing the idea of the cloud as a primary data repository. There are countless data-critical organizations - both private and public - which do that today. Among other solutions, Amazon AWS and its competitors make it possible for even small organizations to store data in multiple cloud-based locations in a way that is more disaster-proof than local data. Nothing can ever be 100% secure whether it is local or in cloud.

It doesn’t matter how many copies of my data exist on servers widely separated to avoid issues with loss due to local disaster if I cannot access the data because I have no connection when I need to use the information! That’s what urban centric folks totally don’t get. Also folks in most of Europe, who have no idea of the distances involved in the American West. We have a single cell phone tower in our valley. The next one is 10 miles away. Now by chance we happen to have fiber Internet to the house, but I am not always at my house and wifi can’t reach into the pasture. We’re more connected than most, we have wifi access points in all the barns and outbuildings but that is rare for farming communities and families.

Cloud systems do not work in this environment.

[rkaplan]

@OogieM

I believe you did with this statement.

I made a request for my use case. I did not at all suggest that features should be removed for other users. The more options the better.

Cloud only is not an option. Even cloud preferred is not really good for me because of the limitations of access.`

Different people have different uses cases- that’s fine.

It doesn’t matter how many copies of my data exist on servers widely separated to avoid issues with loss due to local disaster if I cannot access the data because I have no connection when I need to use the information! That’s what urban centric folks totally don’t get.`

I get it. The current software works for you - wonderful. My point is that a good percentage of the computing world is moving to the Cloud so I think Devonthink may lose customers if they do not add that as a feature.

Adding cloud support would not hurt you at all. On the other hand, if lack of cloud support hurts Devonthink then that could impact their ability to support your local solution long-term. I think a better solution is to offer both.

[scottlougheed]

@korm

it is OK to keep those shareable documents in Dropbox, or OneDrive, or Box, and index them in your DEVONthink database as well as use, for example, Dropbox’s sharing features to share them to others.

I would say about 50% of my DT files live in a cloud service and are indexed, either because I need to guarantee access to them no matter what my equipment is (e.g., in the extraordinarily unlikely situation I am completely without one of my many devices with DT on it, or just because I want more flexibility in how I access them), or because they are shared with a collaborator.

I’m getting my cake and eating it too.

(that’s not to say that indexing doesn’t add a small layer of complexity, but if you’re careful it’s really trivial)

[rkaplan]

@scottlougheed

I’m getting my cake and eating it too.

(that’s not to say that indexing doesn’t add a small layer of complexity, but if you’re careful it’s really trivial)

You are just indexing meta data?

Or you are keeping copies in both Dropbox and Devonthink and updating them both?

[scottlougheed]

Files are stored in iCloud or Dropbox (depending on what they’re related to), indexed in DEVONthink. It’s important to note that DEVONthink creates full text indexes of indexed files. So within DEVONthink I can search metadata and contents of all the files in any of my databases, whether they are indexed or imported.

In my case, I also have DEVONthink syncing between multiple devices (I use iCloud for this) and I also have it sync the indexed files. This allows me to have access to indexed files DEVONthink To Go as well, which is important to me.

[korm]

@rkaplan

You are just indexing meta data?

Have you tried to experiment with indexing? Clearly you have use cases that might be solved with indexing, and it’s a lot easier to play with the feature than to speculate about it.

[cgrunenberg]

@scottlougheed

(that’s not to say that indexing doesn’t add a small layer of complexity, but if you’re careful it’s really trivial)

An upcoming release will simplify this a lot.

[rkaplan]

@korm

Have you tried to experiment with indexing? Clearly you have use cases that might be solved with indexing, and it’s a lot easier to play with the feature than to speculate about it.

No I have not tried it yet - I will read about it some more and give it a try - thanks.

[Cashew90]

I work on a multi-part article about Devonthink in general and my personal use of it in particular for my blog at the moment. This thread made me aware of the difference of storing files in a cloud and syncing it via a cloud. I obviously want to get it right in my article so could somebody explain to me what the difference is, please? Alternatively, pointing me to a place where the difference for Devonthink is explained would be great, too.

Thank you!

[BLUEFROG]

@Cashew90

I obviously want to get it right in my article so could somebody explain to me what the difference is, please? Alternatively, pointing me to a place where the difference for Devonthink is explained would be great, too.

It’s simple, really.

1. A DEVONthink database and its sync data are two separate things.
   A DEVONthink database is a package file with a .dtBase2 file extension.
   A DEVONthink sync location is a package or folder with a .dtCloud extension, created by our sync engine to contain the sync data.
2. Sync does not copy your databases to a sync location. It is raw, chunked, and optionally encrypted DEVONthink-specific data about your databases. DEVONthink or DEVONthink To Go use this information for importing and syncing the databases.
   (This is also why sync is not a proper primary backup.)
3. The mechanisms that drive cloud services are not compatible with the package file type of our databases (and some other applications’, as well). This is why you should never put your databases in a cloud-synced folder, and why we specifially disallow opening or syncing databases in these detected locations.

[ZeppoVA]

OK so now I am confused from Bluefrog’s explanation.

I have multiple databases synced via WebDAV - it all works great. The DBs have mostly managed and and some indexed files. But I tried using iCloud as the location for the sync location and each DB was synced on each Mac via iCloud sync location. This doubled the storage space required/used on each Mac as each Mac then required storage for the DBs as well as storage for the iCloud sync data in the bowels of the Mac for iCloud syncing.

(Boy the naming is tough to grab…)

Does syncing via iCloud as the sync location not require both the DB and a copy of the sync package on each Mac?

[jchiar]

I have 7K notes, 3.4Gb in size with 7Million Words and no issues with syncing that I see.

[BLUEFROG]

What “naming is tough to grab” ?

iCloud syncing is a two stage process. DTSync is done locally. After sync is finished, iCloud uploads to Apple’s servers, then to devices using your Apple ID. iCloud should clean up after itself locally, but we have no control over it and its behavior.

And you will ALWAYS have a local copy of the database.

[ZeppoVA]

Sorry - let me be clearer.

As far as I know, for iCloud keeps copies of everything it syncs on each Mac. Meaning that the working DT database is on each Mac in the Documents folder AND the syncstor document is downloaded from iCloud by the Mac in the local iCloud folder. The sync files include the actual files I put in or index with DEVONthink so if a DB is 1 GB in size, the Mac now has about 2 GB of disc space taken up between the database and the syncstor (synced via iCloud) Do I have that right?

The ”Naming is tough to grab” comment is about the nomenclature of database, syncstor, the package of files place in the syncstor location, etc. I just found it difficult to grasp what I needed to do when I set up syncing and still get confused about what to call each item when discuss it - case in point is my previous comment.

BTW my databases are about 50 GB total and sync fine and relatively quickly using a WebDAV set up on a low powered Synology NAS. They don’t all sync to the iOS devices.

I would like to use iCloud as the sync location but the storage space required on the laptop is effectively doubled because of iCloud’s syncing of the DEVONthink sync packages to the Mac. Is there a way to use iCloud as the syncstor location that does not double the storage needs on each Mac