Yeah, I'm not really sure what it's good for anymore, particularly since the cloud syncing is so weak.

Complete disagree. This is a thinking person's kit of software, exceedingly powerful if one bothers to take the time to learn it. Every update is a careful suite of improvements and once you've brought DEVONThink into your workflow, it simplifies and organises like nothing else out there.

You say that you can't find the data you just imported? Duh, it's in the Inbox, the folder choice at the very top of the sidebar. If you missed that simple effort, I can understand your challenges in understanding DEVONThink overall.

DEVONThink and DEVONAgent used together are the most powerful utilities for data gathering and organising on any platform. Bar none.

Yes, the UI design is getting long in the tooth and looks a bit dusty. But it's not that bad yet.

Seems to me you downloaded a copy, played around with it for half an hour, and wrote up your half-star review, burkanov. That low score alone suggests a chip on your shoulder or another agenda, not a measured analysis of a phenomenally crafted software utility.

@burkanov — You have a point that if you don't have a consistent filing system Devonthink can turn into a mess because it has so many options. It's also true that if I personally tried to keep everything in Devonthink it wouldn't be as useful. I used Devonthink as my primary information manager through an MA degree and it was incredibly helpful. Your claim of "unpredictability" is simply not true through. It does precisely what you tell it to do.

The reason why DT is a valuable tool is because it manages information from a wide variety of sources. It lets you browse the contents of filed information sources and compare them against others. For example, my database is full of academic papers. I can highlight citations and then later find one important sentence when I don't recall which PDF it's in or if I want to support a point I'm writing about but I don't recall all the citations I have collected. The Finder can't do this and other info managers can't do it with so many types of sources. It is not helpful—for me anyway—for active documents I'm working on like Word; for this I use Dropbox with cloud support and I can usually recall something based on the doc type and the file name.

Having cloud access to every single file on your computer is not productive—it's trendy. I don't want to work with large amounts of information on my iPhone. I do need mobile access to a lot of my files so I use a laptop.

The interface isn't eye candy like Evernote; but I don't expect this from a tool like this. If you've ever used Bento, DT looks great. It might not fit your personal workflow, but saying it belongs in the trash is misleading and not very fair.

My own approach as a heavy user of DT Pro is to use topical databases. I'm currently managing a hundred thousand or so documents among several topical databases. My main database, which focusses primarily on my interest in environmental science, technology and regulatory and policy issues comprises about 21,000 documents and about 24 million total words. It runs smartly (in more senses than one) on my MacBook Pro with 2 GB RAM.

I'm in process of building an associated database covering chemical analytical methodologies (both US and EU), statistical and other procedures and issues related to the evaluation of environmental data, sampling methodologies, etc. This may likely grow larger than my main database.

Splitting out the very specific methodological material not only keeps my main database speedier, but keeps the AI features and searches in my main database from being diluted by the specifics in the methodologies database. That's important.

As to organizational structure, personally I've never considered my organization hierarchical.

I approach organization as putting 'clusters' of related content into groups, as well as 'clustering' related groups.

Many of my documents are replicated into two or more groups. Likewise, I often replicate a group into two or more 'clusters' of groups. I will use smart groups to replicate documents for a particular purpose, or replicate results of a search into a new group of related items.

I rarely tag individual documents, as I rely on the combination of grouping, searches and the 'See Also' AI feature to identify material in the process of a research project. DT Pro provides other tricks such as Option-click on a term to see other documents that use that term, Command-/ to initiate a search on a text string, and 'See Selected Text' to provide a list of potentially related documents.

In the process of a project I'll often create rich text notes linking to documents of special interest, or 'mark' them temporarily (I usually clear such marks later) using label or state marking.

Disclosure; I'm the Evangelist for DEVONtechnologies. (12/5/2006, Version: 1.3b)