Synapsen IMHO has a horrible GUI. Zettelkasten has the advantage over Slipbox that it can export data in csv format (and does or will support XML in the future) so one can import data in other programs including DTP. This is important for me.
I try to figure out if i can use DTP as well but there is one impotsnt difference with programs as mentioned above and it is about keywords: i like to use a system where i can attach multiple keywords to specific notes and later do a search where i select some keywords from a list (which can in principal be very long). I find such a system dificult to implement in DT but i am wondering if anybody has any suggestions for this? OR uses some of the software mentioned above.
I recommend DevonNote, because it stores several filetypes and is easy to organize. For the keywords issue, click the Show Info button and keep that window parked in a corner or on the Dock. It will change as you move to different items. Store your keywords in the Comment window.
For the list of keywords, keep those on a separate DN item in plain text. You may keep it permanently open by using New Window and then selecting that item; and you may also minimize it to the Dock.
The only improvement on this system would be a automatic tagging, but I don’t know of a notes program that has that feature.
I see one major problem with doing searches. I would like to be able to use several criteria for a single search, like smart groups work.
Example: if i have a keyword “Spanish Civil War” and a keyword “Franco” in the comments of a file (among many other’keywords’) I want to combine those in a search (and specify that DTPO only searches the metadata). Some of the programs i mentioned earlier do exactly that: you specify for which keywords to look and voila.
Of course you can use multiple terms in a search of the Comment (only) field. An All Words search for “Franco” and “civil” would pick up your notes immediately.
I do a lot of writing and I almost never use keywords. If the content of your note mentions Franco and the Spanish Civil War DT Pro will instantly find your note whether you’ve used keywords, or not. I do All searches, which cover the name of the note, the content, the Comment field and the URL and Path.
Example: With more than 21,000 documents in my main database, if I do a database-wide search for “franco” (using ignore case) I get 27 hits in 63 milliseconds. If I now add “civil” and “war” to the query I’ll narrow the search. It so happens that my database contains only one reference to Generalissimo Franco, in the context of an analysis of the history and politics of the Middle East – observing that Basque separatists became more powerful in the post-Franco democratization of Spain. All of the other references to “Franco” in my database are about someone else.
Some tips about using and interpreting search results:
If I add “Francisco” to my query for “Franco” I get three results, The top one, in which the two words are contiguous, got the highest ranking, and that did mention the former Spanish dictator. the two lower-ranked items picked up “Francisco” from “San Francisco” and got much lower rankings. So although currently DT Pro doesn’t have the NEAR operator, rankings can often give one a good tip.
It can often be useful to replicate search results to a new group, and then do further ‘slicing and dicing’ by searches on that group. Although currently I cannot concurrently do multiple Phrase searches I can do that by searching for one phrase, replicate the results to a new group and then do a different Phrase search on those contents, and so on. Similarly, in this two-step process I can emulate the NOT operator.
If I’m searching notes for a particular class or project, I can narrow the search to the contents of a specific group that contains those notes.
If I’m reviewing references that are especially important, I can mark specific items using a Label. Now I can search very quickly for those notes that I really need to bone up on. Afterwards, I’ll probably search for those labeled items and - in a single step - remove all of the ‘important’ labels at once (select all of them, then change to ‘none’).
I don’t have anything against adding keywords to the Comment field of a document, but I don’t feel the need to do it often. In the old days before indexed content of databases, keywords were absolutely necessary to find anything. Bur in DT Pro I rely on organization (grouping), content searches (sometimes including smart groups, which I rarely keep permanently) and AI assistance such as See Also. As my research needs for my very large collections of references are continually changing, I would spend more time tagging (with a huge and growing list of tag terms) than writing if I tried to rely on tags. I simply work with DT Pro to almost totally eliminate the need for a priori tags.
But that’s just me. Other users are certainly free to disagree.
I was glad to read Bill’s thoughts on keywords–they match my own and my own use of DTP. My database is still relatively small, but I’ve found that searching DTP is easy, flexbile, and returns excellent results. Keywords may be something no longer even needed in digital research.
It would be nice, however, if DTP had a more unified search system where I could enter a search term, then simply use bolean operators to expand or limit the search. The current system involving those menu selections is a little awkward.
Bill, thanks very much for the answer. My need for keywords is mostly based on the idea that without assigning them i can not be sure DTP will find what i need.
I can take a quote from a book or article (about something very specific) and than i need to describe that proprerly in order to find it later. Example: i might have a bunch of quotes or clippings about the origins of the Spanish Civil War but if i do not put the phrase “origins of the SCW” somewhere (in the file or comments) a search will not bring all of them up because those exact words might not be in the quote/clipping (or they could be in various languages).
So the only way to know that all notes on a specific theme will be found is assigning a ‘keyword’ or so. (alternative is of course to group them correctly before). In the end with ‘keywords’ i feel i could control the outcome of the search as i want.
Having said this, one of the reasons i brought this up was to see if there are different approaches and how they could work with DTP. So this was very helpful.
Hi Gerben, if you are still looking how to implement something like a Luhmann-like-Zettelkasten my previous post might be helpful. It describes a way to use wiki-links and index files for a Luhmann-like-approach.
You could also use basic tagging by entering text lines like (as tags) in your notes and later search for the exact tag. An Application like iClip might be useful for this, because tagging is a matter of keystrokes.
Using wiki-links, custom file names (like 16.a23.a) and an index file, I can store some “hard links” to certain files. If I forget to tag or to set the file names, I can still find the document via DTs ai feature. I prefer DTs approach over synapsen, because the tool is much more flexible - if I want Luhmann-like links, I can use them, but the advanced AI digs up the rest.
Mark, thank you for your comment. In fact, it was your previous post, really very very helpful, that has started to make me think about a system and Zettelkasten-software as well. But i am definitely sticking to DTP.
I have some questions about your system. If i understand correctly:
each new file (clipping/note/whatever) gets a unique number which is also the name of the file
you then enter the number in an index file which is grouped according to keywords
To be clear: are the numbers completely random or is there a system in them? I ask because in your original post you said something about grouping files which confused me a bit: your groups were topical but how can one group files according to numbers in a linear order when they are attributed to files at random?
Don’t know if this makes sense? Anyway, i am experimenting at the moment but the ability to have links like this is indeed brilliant.
As for keywords: the ability to use an “add comments” script to add keywords to groups of files is very handy here.
Hi Gerben, glad that you liked the idea of using internal wiki-links this way. Since more and more stuff gets into the database however, I tend to take some shortcuts, for example:
In a perfect world: Yes. But the first thing I do with DTPO is capture items. They get into an INBOX folder, or directly into topic-related sub-groups. The ai-suggestions for the appropriate folders get better over time, the group panel is also helpful.
When I import lots of data, I often skip the naming. But once I know this file is important and I want to reference to it in a number of documents, it gets a number-based name, which stays the same even if I decide later on to move or to replicate the file, so that the links stay intact.
I only enter some items in that index file, when I want to be 100% sure that I look there first whenever I look for that topic. I found that an intelligent group structure and good cross-linking is more important here to find relevant items “in the neighbourhood”.
In principle, the numbering system is not important, but it should prevent identical file names. For my thesis, I have certain topics (with sub- and sub-sub-topic-groups), which are visible as groups.
3 Themes (that is an actual group name - the number keeps the sorting order, so that I remember the naming conventions)
with the subgroups
c new thematics
with a particular sub-group called
2 cognitive approaches
A new item for this group would have the “prefix” 3.c2 , for example. Once I have more than five items in a group, I try to set up a sub-group, with a new prefix. As you can see, I alternate numbers and letters, using the numbering system quite random, to help structuring the amount of data into topics. But the numbers are not hierarchical. “3” for “Themes” came up, because the group for the “themes” topic was the third top-level-group I created.
An exception to this convention is the naming of files related to bookends entries. These are numbered according to their bookends number (e.g. 070312-a), quotes get a number according to the first page of quotation (e.g. 070312-a-212 for p.212). I also subgroup these entries; the group name is "author - short title (even if I put it in 3.b2, for example). The basic idea is to see the book even in the group panel.
A side note: The more I work with DTPO, the more I learn to trust in the ai. I don’t like the idea of doing unnecessary things, so I only only name files with numbers the moment I have to, and I set internal links only when I want the connection to be visible. Over time I found that I could live without the index file for topics - there is a search function, after all, plus the documents are in groups. But I still use it from time to time, to enter files that relate to the topic but are filed somewhere else, or in a different group.
I am still experimenting with ways to speed up the transition from capturing / sorting to writing. Yesterday I tried this: First I browsed the groups in DTPro, then I replicated every interesting file into a “research” folder. Then I opened this folder, set the sorting order to “unsorted” and dragged the files around until they were in the order I’d like them to appear in the chapter. This way the quotes and chapters are visible during writing. I found this way much more useful than switching between Mellel and DTPO, but DT’s lack of footnote support in the current version makes it necessary to reformat everything later - painful but true. I am curious: How do you (plan to) use DTPro when it comes to actually getting things done in academia?
I use DT Pro to store all my research information. There is no other program that I’ve found that handles such a huge quantity of information so well. So it’s my ‘warehouse’ for information and research material for all projects, academic or otherwise. I organize things mostly hierarchically via groups or with tags in comments fields and smart groups. Mostly the former. Then I use DTs search functions to find things. For instance, for my dissertation, I would do searches for concepts or subjects of chapters, and DT would find EVERYTHING that contained something about that subject. Things I long had forgotten about or that I never would have thought of in the first place. It was extremely helpful.
I never use DT for writing or note-taking, though I know others do and are happy with it. I never could use it that way–too limited for my needs. So I use Scrivener for all writing and note-taking, project development, etc. I transfer any research material from DT Pro to Scrivener. Scr. does footnotes and has all the features I need for any type of writing and then some, so I no longer have to struggle to make another program work foro me (like DT).
For the few projects that require a word processor, I use Mellel. Usually only for projects that involve really complex footnotes/endnotes. I wrote my dissertation in Mellel then did the final formatting in Word to send to my advisors. Otherwise, I stick to Scr.
That’s how I use DT Pro in academia! I long ago gave up trying to have one program do everything. The programs I use are the best I’ve found for the uses I give them. DT Pro is the best I’ve found for storing and organizing vast amounts of information on a wide-range of subjects.
Thanks again for an elaborate answer, i understand how you work now.
One thing that keeps me wondering: the relation between your topical files and the Bookends-files. Do you link them among eachother according to their file names? How does that work when you have files which deal with the same topic, some being random clippings and thoughts, others being related to BE references.
As for my own system. I am still developing it and all these questions are part of that
So far this is what i have done to manage information for my PhD (but this is still very much changing):
my PhD is related to two broad distinct fields (Spanish Civil War and Jewish-Yiddish Studies) and i have created a separate folder for each “dimension”. within those folders i can make subgroups, for example for specific themes within Jewish Studies that relate to my research and also groups with biographical info (folder for each person of interest)
I have separate folders for each chapter of my PhD. I can fill those with replicates, archives of important websites, notes, clippings etcetera.
notes: this is what i am now trying to develop. Since last week i have begun to actively put some of my notes and also clippings from mailing lists, articles, websites into DTPO. I used Fastscript Lite to create a keyboard shortcut to send any selection (from whatever source/app) directly to the Incoming folder. Works very fast. I have still a separate group with all my previously made notes (mostly Word docs) that i would like to break into smaller clippings one day but for now i want to have them clearly visible in the db. I still prefer a sytem whereby i first put notes on an article in one document (i want to be able to read them as a summary of an article or book) and later break that document up in smaller pieces. I will probably end up arranging all my topical notes in the folder i created for each chapter.
One thing i am still not entirely sure about is this: do i use RTF-files for all notes or do i use a sheet with records. The reason behind that is this: i do an analysis of newspapers for which i keep an Excel sheet with 1000+ entries (about half with notes). I input these in Excel because i do some quantitative analysis. Luckily i can import the file into DTPO because it can import .csv files.
For me it would be ideal if records in a sheet, after import from Excel, could be linked to other files in the db. If this would be possible i woul be tempted to input new notes in sheets as well: i like to layout system of individual records which allows for a sort of note-card printing. And a DTP sheet can also be exported to a spreadsheet which has certain advantages for my research. The drawback is of course that each clip i send to my Incoming group would have to be turned into a record in a sheet, although they have to be renamed anyway so how much work it would be eactly i don’t know. But all of this is mostly thinking aloud since STP does not yet support links in sheets.
ALexandria, thanks as well! I have taken a look at Scrivener but for me organising my notes in DTP works best. Scrivener does look pretty good though…
Hi Gerben, I put all files that are related to an bookends entry into one folder; as a subgroup in the topics-group. For example, in the group “agenda-setting” I have a group “McCombs - setting the agenda” (book title). In this group is an rtf file called “070213-a”, which is also the number in bookends (I have set up a special number field in BE and I also mark these numbers on hardcopies). This is the “main” file for the book. sometimes I put some searchable PDFs of individual pages into this folder as well, but most of the time, typing is faster.
In this rtf-file, I enter notes, ideas and comments for this book. While I work with this book, I might create new files into the group, for example, “070113-a-81” for a quote from page 81. In the main file, I’d have something like “agenda-setting and frames: cf. 070113-a-81”, which is then wiki-linked. Using this method, I can link to the book from other topic-files (“cf. McCombs on agenda-setting: 070113-a”), but often I just replicate the whole group for this book or article where I need it.
@Alexandria: Thank you for your comment on scrivener: I like it mostly because of the footnotes feature. I hope that DT will have basic footnote support as well one day, because for footnote creation, the “see also”-button works wonders…
@Gerben: I did not have to use sheets so far, but my impression was that this feature is still on the agenda for improvement. Actually I try to avoid using them, because RTF text seems more flexible at the moment. I hope this will change.
Interesting to see how individual academic usage scenarios become… in the end, it is all about output, so I constantly have to remind myself not to “optimize” (procrastinate) too much.
All the best for your projects!
a really useful program - I will try it for using scripts. For tagging, maybe a freeware textexpander tool is handy: http://www.app4mac.com/rapidowrite.html. It works in Mail, safari, the finder and in Devonthink.