Hi Oleg, > Both HttpCore and HttpAsync are being developed within the > HttpComponents project along with other components and I see no reason > for changing that. The trouble is that Jira requires a separate > "project" per a versioned artifact.
Thanks for the reminder. I knew there was a reason :-) > Therefore there is no way around > creating a new project in Jira for HttpAsync if we want to be able to > release separately from other components. It is still open for > discussion whether HttpAsync should have its own release cycle or > whether it should be released along with HttpClient. I am strongly in favor of having separate release cycles for HttpClient and HttpAsync. Here are some reasons: HttpClient has a more steady development cycle. I'm working on HttpAsync only in short, intensive bursts every few months. At least as far as coding is concerned. My job can require customer related travel on short notice. Since my development environment is not portable, I can't work on HttpAsync during the week at such times, and I might need the weekends to take care of other things. I'd hate to be the reason why a combined release is late. I expect HttpClient to reach both alpha and beta stage long before HttpAsync does. In particular the API freeze with the beta is an issue here. There is no point in freezing the HttpAsync API prematurely, nor in delaying the HttpClient API freeze because of HttpAsync. While the async interfaces may stabilize quite fast, the API includes more than just those. I expect HttpAsync to initially be buggier (or less stable). At the same time, it appeals to a much smaller community. I don't want to cut a new release for full HttpClient just to have some bugs fixed or features added for the HttpAsync users. cheers, Roland --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
