Go to top of the forum TheBrain Message Board RSS Feed
TheBrain Technologies Forum
Register New Posts
 
 
 


Reply
 
Author Comment
 
protein_function

Registered: Nov 4, 2009
Posts: 31
Reply with quote  #31 
I use http://www.zotero.org (not Mendeley, not EndNote anymore). 
protein_function

Registered: Nov 4, 2009
Posts: 31
Reply with quote  #32 

Here is how I import literature citations and link subordinate PDFs from Zotero into a Personal Brain (PB 5.5.2.4, Zotero 2.0.2, FF 3.6.8, WinXP)

(Preface) This is a great thread which is near and dear to my heart. I am a research scientist (pharma, genomics, biotech) and use the scientific literature and internet extensively.  I am also a hardcore PB and Zotero user and believe PB and Zotero are both the bomb.  In Zotero, I usually keep citations and PDF files to published articles.  I too have been wanting be able to use PB and my Zotero database together.  I do lots of research inside PB.  Like many in this thread, I ABSOLUTELY want and need to interface it with literature that I store in Zotero, especially for certain citations and projects.   Here is a work-flow that I discovered that works reasonably well for me (but suffers in that it requires too much foresight to set up the screen right etc.) .  This work-flow does not allow linking multiple references automatically and does not open Zotero or the citation in Zotero but goes directly to the PDF.

Work-flow to import literature citations and link subordinate PDFs from Zotero into a Personal Brain (v.5.5):

Open Zotero inside Firefox. Open Persona Brain.

I split screen with Zotero and PB each taking top or bottom half of screen (want access to both without obscuring the other).  This might work nice with a huge monitor or two monitors. 

Zotero: Select a reference to be exported to PB

Zotero: Right Click, Create Bibliography from Selected Item, Save to Clipboard, OK

Note: you may want to select a Bibliography format (before hitting OK above) that looks the way you want as a thought in PB. 

Zotero:  Expand the reference to show the nested underlying PDF file.  Highlight PDF.

Zotero: Show File (which opens the local folder where PDF is stored). 

PB: Paste thought into PB where you want it.

Zotero: Drag PDF file into PB either onto or beneath bibliographic thought that was just created. 

PB: clicking on thought or PDF opens the PDF file from the storage site. 

Note:  Zotero now allows remote and/or shared storage of libraries and perhaps this might also be incorporated into a different work-flow. 

I have been thinking about if and how this workflow or functionality could be improved with AHK Macros, PB 6, Mendelay (which is pretty interesting and improves rapidly - I need to look at this again) or however.  Any comments welcome.

rhodes

Registered: May 2, 2010
Posts: 373
Reply with quote  #33 

Thanks for posting.

Quote:
PB: clicking on thought or PDF opens the PDF file from the storage site. 

Although I don't use Zotero, you seem to have got it about as efficient as it can get. What don't you like about the way you're doing it now, or what would you prefer to be able to do?


__________________
Alan Rhodes
andreas

Registered: Sept 22, 2009
Posts: 87
Reply with quote  #34 
What I find extremely important is the time-factor: If something takes too much time (and your workflow with Zotero looks like it's time consuming) I will do it maybe for 2 weeks and then slowly stop doing it...

So I absolutely second anything involving automation.

Currently, I divide my reading workflow into 4 parts:

Getting information (daily):
-I set up an RSS Reader (FeedDemon) for all the journals I am interested in. Quickly browse for the important keywords or reactions (luckily, chemistry and biochemistry is mostly graphical information and I browse though reactions much more quickly than through text).
- Download pdf
- drag pdf into "collect papers" section of my brain. If no pdf is available yet (RSC journals), I copy the weblink into "waiting papers" (I will check back for pdf later).

Organizing information(dead time, e.g. after lunch):
- Open the pdf, scan through. Rename the pdf in the Brain: Year-Journal-Corr.Author. Make link to Corr. Author. Link pdf to keywords. Create thought icon of key idea/reaction.
This is the step that might be automatized.
- For most of the papers, this is it. I do not need to read the paper but I can come back whenever I need to.
- Tag the paper I really need to read.


Reading papers:
Actually read the paper, highlight info, make notes in the notes section.

Writing papers:
Only now, I need a tool like Zotero. Drag in the pdfs of the papers you want to reference into Zotero. Cite them in Word etc.

Some additional thoughts:
Chemistry is mostly graphical information. I cannot work with Zotero from the beginning since it is basically a list of written info. In the Brain, I have my pictures.

We have a shared Zotero DB for the group. I see this as a common base for pdf storage but encourage my PhD students to have their own reading workflow on top of it.

I have a lot of comments on scientists (gossip), companies, paper quality, etc. that I will never share with anyone. Having shared notes, sharing my literature Brain is out of the question. I just share the plain pdf, and Zotero does the trick beautifully.

Anyone mentioned that there are too many Journals out there?






protein_function

Registered: Nov 4, 2009
Posts: 31
Reply with quote  #35 
Quote:
Originally Posted by rhodes

Thanks for posting.

Quote:
PB: clicking on thought or PDF opens the PDF file from the storage site. 

Although I don't use Zotero, you seem to have got it about as efficient as it can get. What don't you like about the way you're doing it now, or what would you prefer to be able to do?



Quote:
Originally Posted by andreas

What I find extremely important is the time-factor: If something takes too much time (and your workflow with Zotero looks like it's time consuming) I will do it maybe for 2 weeks and then slowly stop doing it...

So I absolutely second anything involving automation.


rhodes, andreas and others, thank you for participating in this thread. 

I believe andreas answered rhodes question quite well.  If the workflow is too time consuming, cumbersome and intrusive for the objective, then it will not survive.  On the other hand, there is also a consideration of cost-benefit -- if the work-flow adds enough functionality and value to the data (in terms of usability etc), then it has a good chance of surviving. 

Staying on top of the scientific literature in ones field (I am a pharma biologist) is a daunting job (too many articles and journals, most of them not important to you, many demands on ones time). One needs mechanisms to tame this and make this more efficient (or effective), not less so.  So, the workflow should approach industrial strength and be as uncomplicated and fast as possible.  To be honest, though I have just started using the PB-Zotero integration workflow, I anticipate that my workflow is too tedious to perform for all but the most important articles (~5-10%) that I feel a need to capture in my own PB knowledge base. 

andreas

Registered: Sept 22, 2009
Posts: 87
Reply with quote  #36 
"there is also a consideration of cost-benefit -- if the work-flow adds enough functionality and value to the data (in terms of usability etc), then it has a good chance of surviving."

Absolutely. For me, the first scanning of the papers and the linking within the Brain feels like playing. It's a form of active procrastination and I like to do it during mental down time (e.g. Friday afternoon). I do not really read the whole content but at the end of the day I get the trends and the people involved. Plus, I have the info in the Brain.

The bigger the Brain gets the easier it is to put in new papers. Often, all the keywords and authors are already in there and I can link the new info easily to them. Plus, I often quickly check back on the earlier publications of an author by hovering over the thought icons...


If a student has a specific question (e.g. what reagent do I need to deprotect this group) I do find the answer in seconds and can print out the pdf or upload it into zotero. Since we have limited access to Scifinder (4 licences for the whole University...) often it is no option for me to go in there and refind the info. I have done the work once and it is in the Brain. But for this, I need ALL papers that touch my field in the Brain, not only the important 10%.
rhodes

Registered: May 2, 2010
Posts: 373
Reply with quote  #37 

Your replies provide interesting insights into your workflows. However, the only clue I get so far as to where the bottleneck lies is this:

Quote:
Create thought icon of key idea/reaction.
This is the step that might be automatized.

I still don't understand this very clearly. It seems you wish to create a thought with a special icon of some kind. Is this not possible already using a template thought (with icon) you can pin and recall when you need it?

I took a look at the Zotero video and it seems like a great idea but it's only a single-parent tree, unlike PB, which is multi-parent. This demands a more complex set of data relationships, meaning more work on your part. Unless you're running software which can think for itself, this process currently cannot be automated. Anyway, what advantage would there be in importing the outline of a Zotero database into your Brain?? You'd only end up with a duplicate, which needs pruning and syncing.

Zotero seems a great tool for accumulating mountains of citations but as far as I can see, leaves it all to the user to assimilate what it has collected, which gave me a rather uneasy feeling while watching the video, a bit like the sorcerer's apprentice. That's where you need PB to help you put it all into perspective. Do you really need all the stuff it collected? Less is definitely more. Even better would be a highly intelligent, natural-language application that could read, parse and sort it all for you.

Maybe if you could be much more exact about what you'd like to do, this would invite the rest of us to propose more specific suggestions and possibly get the developers interested too.

(Added) While I was preparing a reply, Andreas posted again, and his comment sums it up for me:

Quote:
The bigger the Brain gets the easier it is to put in new papers. Often, all the keywords and authors are already in there and I can link the new info easily to them.

That's it. Provided you have done the arduous groundwork of refining your design, and keyword format is very important here, it should become easier to add new data. Andreas' post mentions nothing about building new automatic features into PB, though I find that some of the macros which have been made available help me greatly and can be customized to make things go even quicker.


__________________
Alan Rhodes
andreas

Registered: Sept 22, 2009
Posts: 87
Reply with quote  #38 
Alan, I agree with you on Zotero. Wonderful tool for collecting, less helpful for organizing. But it is perfect for citing papers when writing papers, theses etc. Other tools like EndNote or Mendeley etc. exist. Zotero has open codes, though.

These programs can tag each paper with keywords that might serve as parent thoughts.

But the problem is that different people want to extract different things from a paper. Here is an example (see attached pdf). I would link this to the corresponding author, rename the thought and add a few parent keywords of my own. As thought icon, I would copy scheme 10. A different icon for each paper.

What I think would be helpful if it could be automated:
-Right-Click on a paper in Zotero (or other program) "import paper into Brain"
-Pop-up showing name of future thought, editable. Also showing possible parent thoughts that contain the name of the corresponding author and the keywords (add extra, or delete imported ones).
- Click OK and the pdf is imported into the Brain as a new thought containing the set parents.
- Maybe check if paper is already in the brain.

The process could be individualized by chosing a personalized "citation style" that would allow the naming of the though.


Attached Images
Name: brain.PNG, Views: 622, Size: 58.20 KB

 
Attached Files
pdf example.pdf (607.21 KB, 58 views)

rhodes

Registered: May 2, 2010
Posts: 373
Reply with quote  #39 

Thanks for the reply, pic and example. Allow me to make two comments:

Quote:
But the problem is that different people want to extract different things from a paper.

Only you would know what key words, phrases or paragraphs in this article need to be cross-linked to other Thoughts. PB cannot do this for you. Moreover, everyone's requirements are different. 

Quote:
-Right-Click on a paper in Zotero (or other program) "import paper into Brain"
-Pop-up showing name of future thought, editable. Also showing possible parent thoughts that contain the name of the corresponding author and the keywords (add extra, or delete imported ones).
- Click OK and the pdf is imported into the Brain as a new thought containing the set parents.

It sounds to me as if you are seeking a (Windows) shell extension or maybe a browser plugin that says "Import paper into Brain". The problem here is, where is this paper going to land? It could end up anywhere within the Brain unless you specify a location for it in advance. If you do that, you might as well DnD its favicon onto the Thought or into the Plex under the appropriate active Thought, which is probably why it's been implemented this way.

Two "automation" solutions to your problem come to mind.

The first is an entry form that can pop up over your browser with a number of fields into which you can copy and paste whatever details you want. The entry form needs to be customized to your requirements, but that needn't be a problem. Clicking OK on the form after it's filled in causes PB to be activated and the data to be pasted in at a preprepared location. PB doesn't have entry forms, preferring DnD, but I have seen some Windows macros that do and these are open-source so you can tinker with them to your heart's content.

The second is to highlight key words or phrases such as author's name, etc. in the PDF document, then paste them into Instant Search to see what already comes up. If you don't get any hits for one term you regard as important, you can create it and keep going. Don't link anything just yet.

When you have finished, open Reports and filter by "Date activated". This will display at the top of the list all the thoughts (keywords) you have just activated. Go down the list and Ctrl-click all the ones you will need to link, to add them to the selection box. Then import your PDF and link the contents of your selection box to it.


__________________
Alan Rhodes
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

TheBrain Mind Map & Mindmapping Software     Download PersonalBrain Mind Mapping Software
Powered by Website Toolbox - Create a Website Forum Hosting, Guestbook Hosting, or Website Chat Room for your website.