Dedoose Blog

BLOG

Dedoose Mixed Methods And Qualitative Research Software Version 1.2

10/5/2010

After weeks of some very difficult work our engineering team has finished our first major upgrade to Dedoose!   The biggest upgrade in the system is a component we call the SuperMegaGrid to replace Adobe's AdvancedDataGrid that we found just really couldn't handle the amount and type of data that we were throwing at it.  For instance with about 3,000 objects the previous data grid would come to a crawl on every computer we could find.  Now, with our new system we are able to get snappy performance out of our test collections of over 30,000 objects!   Since we got the opportunity to build this component from the ground up, our engineering team came up with a brilliant solution for easily getting to the data you're looking for through a filtering sub panel that automatically figures out how to create groups for items even if they are not defined as a list-based field.   You'll have to either check out the video's coming out later this week, or check it out for yourself.

Here is the full list of additions.

  • Replacement of Adobe's AdvancedDataGrid with Dedoose's SuperMegaGrid with far superior performance, filtering, grouping, column selection, and much more.
  • Overhaul of the "Select Data" workspace.  We replaced all the selectors with the new grid component and heavily optimized data selection resulting in a dramatic performance increase for selecting data.
  • Overhaul of Dynamic Descriptors and Descriptor Linking (thanks to Ian Beatty of  University of North Carolina at Greensboro for the feedback).   It was pointed out to us that the workflow for linking descriptors to resources as well as editing/adding dynamic descriptor values to resources needed major rework.  Thanks to your feedback we were able to devise a few adjustments to the system that now allow you to link and unlink descriptors to resources from within the document lists without actually needing to go into the document.  Just click on the Descriptors button that now shows in the document lister to bring up the "Linked Descriptors" popup.  This view has also been overhauled to allow editing dynamic descriptor values without having to unlink and re-link a descriptor.  Finally we heavily optimized descriptor linking and dynamic descriptor fields in general, so you should notice a dramatic speed increase here as well.
  • Support for auto descriptor linking on document import (thanks to Donald Jordan of University of the Pacific).   If you are having to import a lot of text documents and want to make the processes of linking these to descriptors automatic, export your descriptors, and in the first line of the txt file add "[dedoosedescriptorlinkid=xxxxxxx-xxxx-xxxxx-xxxxx-xxxxxxx]", where xxxx is the DedooseId from the exported spreadsheet.  This is useful if you are generating your text documents from another source through scripting.
  • Restricted batch document uploading to 20 files at time.  We noticed a few users attempting to upload a massive amount of documents in a single upload to the system, for various technical reasons we cannot at the current moment handle such a large batch of uploads at the same time.  So we've limited the uploads to a maximum of 20 files at a time to assure a satisfactory level of performance.  We hope to come back to this issue in the near future and be able to handle any number of files at once.
  • Finally there were a number of minor bug fixed resolved through the system, please let us know if you find something you would like to see improved.

We are about to update the site with an updated version of the User Guide and a entirely new set of training video's of Dedoose that should help people get acquainted with Dedoose much faster.  There is one general overview video and the rest are on 'how-to' use the Dedoose system most efficiently--No need for expensive training workshops with Dedoose!   If you have any comments or questions please feel free to comment here, on our facebook page (Dedoose), or shoot us an email to [email protected]

~ Eli Lieber

;
Back