The GRIN-Global database and Web interface were developed as a new standard, to allow easier transfer and comparison of seed bank accession data among the numerous, distinct national germplasm banks throughout the world. More interchangeable data means fewer ill-defined data fields (such as a list of varietal names). Apparently Jessica chose to use the Ti or Tc number (a single tobacco variety may have one or the other or neither) in place of common names. That is clearly more definitive, though less useful.
A separate problem is display of tabular data. The "Observations" data was cleanly presented in old GRIN using 3 columns (characteristic, value, source reference) and many rows. In GRIN-Global, we see the very same data transformed into a nearly unreadable display of 3 rows with a gazillion columns--so you have to horizontally scroll the enormously wide table, in order to see what's there. It also makes that content unprintable.
It may be that the apparently lost content (varietal names, etc.) is tucked away in obscure corners, but I was unable to easily locate a way to access it using the search engine (also newly standardized).
The pdf page of the original accession document may still be viewed, but the viewer that allowed paging forward or backward through the document is lacking.
New applications (new Web browser, new Microsoft Word, new forum structure, etc.) sometimes have a steep learning curve, even though they should be designed to be easier than their predecessors. Standardization of data fields often causes a loss of data. [I have two middle names, which could be easily written on a paper form, but are usually impossible to enter on a digital form.]
Constraining data improves standardization and transportability, but avoiding data loss requires more design effort (on the new data structure) and skill than is often the case.
Bob