On Thu, 20 Mar 2003 [EMAIL PROTECTED] wrote: > On Thu, Mar 20, 2003 at 02:02:31PM -0500, Christopher Hicks wrote: > > On Thu, 20 Mar 2003, Zeljko Dimic wrote: > > > I don't think replacing transfer call with our own emulation would be > > > step in the right direction. By using emulation, we could be masking > > > bugs in our code that communicates to the registry, as well as possibly > > > registry problems. Our emulation may not be perfect, and we may mislead > > > you to write the code that would work within Horizon, but not in the > > > Live environment. For these reasons, I'd rather keep horizon as similiar > > > to operations of the Live system as possible. > > > > Providing certain domains which consistantly exhibit a given behaviour > > wouldn't prevent the rest of the domain space from functioning normally. > > > > If this is something that Tucows remains immovably opposed to there still > > seems to be a persuasive enough case to push for the various test > > registries (or at least one) to implement this. > > I'm not sure it even needs any other registry to be involved. What I'm > looking for it the ability to test my implemenation of the API between > my systems and OpenSRS' systems - and that's all. I don't want to care > whether or not OpenSRS sends my request off to some other test system - > that's their problem.
I only brought up having one of the registrars implment it if Tucows is steadfastly opposed to doing it themselves. > As things stand, it's possible to write code that kind of does stuff > with Horizon, and might well work on the Live system - but I don't > really know -- plus, depending on Verisign's system, my test-account > balance, and the phase of the moon, might fail my tests without having > changed[1]. I don't see how an emulation that might get slightly out of > sync with the live system is any worse. > > Remember, I'm not trying to write tests that prove that the entire > process works; I'm trying to write *regression* tests - i.e. I want to > know that my code does the same thing it did last time I tested it, > not that it now 'works' with whatever the test system is doing > today. I'm trying to reduce the number of variables in my testing to > one: my code. The test system as it stands doesn't give me that. Rigorous testing is a very good idea. It'd be nice if somebody added test cases into the official client code. Given that the last time I looked at it the UI handling was intimately intertwined with the communication to the server I'm not sure how this would work without running it through a web server, but Perl can do that. -- </chris> The death of democracy is not likely to be an assassination from ambush. It will be a slow extinction from apathy, indifference, and undernourishment. -Robert Maynard Hutchins, educator (1899-1977)
