Kenneth R. Robinette wrote: > I don't understand the answer. That implies that no third party application > can make use of the certificates/keys contained within the Netscape/Mozilla > standard databases.
A third party application just needs to create its own NSS config directory. NSS will create copies of the databases in that config directory. If a third party wants to *share* the certificates/keys contained within the Netscape/Mozilla standard databases, they can't do this while Netscape/Mozilla is running. *Concurrent* access to NSS databases is not supported right now. > And by not testing on the most wide spread Windows based OS, you are almost > telling us developers to use the Microsoft browser and Microsoft crypto API > only! To add insult to injury, you also state that not only do you not > test, you don't even use the dll's on any system! I said the Mozilla client (or the PSM component) is using the NSS static libraries right now. All other NSS clients are using NSS dlls. We plan to make the Mozilla client use NSS dlls soon. I said the NSS team only test NSS on Windows NT 4.0 and 2000. (This is documented in our release notes, see, for example, http://www.mozilla.org/projects/security/pki/nss/release_notes_33.html.) This is not ideal, but just the Workstation/Professional Edition vs. Server Edition and single-processor vs. multiprocessor are already six Windows configurations we have to test on. The main reason we don't do NSS QA on Windows 98 is that we had a hard time getting our test harness (based on MKS Korn shell) to run on Windows 95/98. People test the Mozilla client on Windows 98 all the time. Netscape certainly tests the Netscape browser on Windows 98 and Me. Wan-Teh
