Installing OAIS Software: DAITSS

On January 7, 2010, in Research, by Chris Prom

As I noted a few days ago, I have been trying to install and configure various pieces of software of use in and OAIS system, in an effort to discern whether it would be possible for an archivist to begin using them, even if they have very limited tech support. I spent a bit of additional time with RODA, and made some progress, which I’ll write about later. In the meantime, I thought I’d record my thoughts regarding the DAITSS software (Dark Archive in the Sunshine State). If you are not familiar with it, DAITSS is intended to run (as its name implies) as a non-public storage system, complaint with the general principles of the OAIS reference model. Like RODA and Archivematica, it includes (or at least allows for the integration of) various preservation actions, such as conversion of files from one format to another, using some of the same tools that are integrated into the other applications, such as ImageMagick.

Based on my testing, implementing DAITSS would be beyond the capability of most archivists. I found installation more difficult than with RODA, and gave up quite a bit earlier in the process, mainly because the installation documentation assumed a level of knowledge regarding server configuration that is far beyond that of any ‘power user’ or a typical archivists.

My detailed ranking and testing notes are after the break.


  • Cohesion and completeness of documentation: 2.The installation manual lists a short set of requirements/dependencies, but unlike the RODA documentation, no instructions are provided as to how to install the required software.  I found the software listed in the installation manual easy to install, based on the knowledge I had gained while configuring the server for RODA, but the documentation only takes you to the point where DAITSS is running as a background process.  In addition, one has to use the configuration manual to issue SQL commands to manually define contacts, accounts, projects, preservation rules, etc.
  • Required software dependencies: 1.It has fewer base dependencies than RODA, but it does not have a built in front end (either a client or web interface).  Therefore, to use it in any systematic way, another set of applications would need to be attached to it or developed.
  • Level of  technical knowledge and system admin authority required: 1.  Completing the basic installation was easy, following the instructions.  However, I ran into significant problems when trying to configure it for initial use.  You have to manually issue a long set of SQL statements, and I had to locate and install software to do that ( You could also use something like phpMyAdmin, but since I only installed mysql itself on the test machine, I did not have access to it .  One could install LAMP to achieve this, but again, you’d need to figure out on your own how to do it.  Similarly, while you can use basic unix file storage for the base storage engine, it would prefer that you use IBM’s Tivoli software, which appears to be a proprietary tool.  In any case, you need to manually define one or more storage locations by issuing SQL statements to add a row one of the DB tables.
  • Time required to install: 2.  Spent about three hours reading documentation and doing the base installation and configuration.
  • Flexibility of installation options: 2.  Unlike RODA, it all needs to be installed on one host machine. The mysql
  • Existence of support structure/user community: 1 There is a trac system for bugs and a mailing list.  However, there is no evidence that the software is being actively developed.  The major publication on it is dated 2007, and the last software release was in 2008.

Bottom line installation and configuration score: 5/20

Comments:  DAITSS would require very significant technical support to implement since it is basically a command line tool.  Furthermore, each project for which you would wish to ingest materials would require either that a long set of direct SQL statements be developed and executed (to create accounts, contacts, projects, preservation rules etc), or a web or client application to issue such commands would have to be written. Furthermore, it appears to be tailored much more closely to library than archival preservation notions.  It would not be useful to most practicing archivists, unless they are very skilled and/or have the resources to devote to large scale system development.

Tagged with: