Lock

You can find the server here

1.1. User interface, on-line help, choice of parameters and databases
very simple, not so many possibilities - only file window, email address (optional), choice of database and choice between core and pre-core superposition
1.2.Interactive versus E-mail results or notification
upload and also results interactive, you can get e-mail message about finishing the process of searching. You can not load directly the saved file, you have to copy whole .pdb file to the window or you can use PDB ID (but new structures don't have ID).
1.3. Documentation, access to literature references
Original paper in the .pdf format on the page
1.4. Confidentality
No information
1.5. Elapsed time between submission and results
it depends on chosen parameters and length of query structure,(pre-core superposition with smaller database and shorter protein is more rapid). It is among 5 and 45 minutes
1.6. How long are the results kept on the server ?
Authors announce 6 hours, but I was able to get back result even 4 days after the search

2.1. What information is reported for the hits?
contains list of matched structures sorted according to number of matched residues. To each matched structure, there is information about RMSD for matched residues and structural alignment in three forms (text, PDB and Chime) and pdb header
2.2. Measure(s) for goodness-of-fit and statistical significance?
number of matched residues and RMSD for them
2.3. Structure-based sequence alignment?
yes, in three forms and even text form is very good and useful (PDB and Chime format are the others).
2.4. PDB files with superimposed structures?
2.5. Interface with visualization software?
yes, if you have Rasmol or Chime
2.6. Hyperlinks to the other web-based resources (e.g., PDB)
links to Rasmol and Chime

3.1. Are there parameters the user can set to optimise the results?
you can choose between pre-core and core superposition, pre-core is quicker and should not be able to distinguish so well biologically meaningful and meaningless results.
3.2. Is there a choice of database? How much of the structural universe do they cover?
One can choose between six subsets of PDB generated by program Obstruct

Heringa J, Sommerfeldt H, Higgins D, Argos P. (1992). OBSTRUCT: a program to obtain largest cliques from a protein sequence set according to structural resolution and sequence similarity. Comput Appl Biosci 8(6):599-600

Max. Homology Resolution (A) Number of proteins

25% 1,8 201

40% 1,8 244

25% 2,0 312
40% 2,0 388

25% 2,5 474

40% 2,5 582
3.3. How often are the databases updated?
I was not able to find more information about subsets, but I guess from the size of databases, that the subsets are not updated at all since 1997