On Thu, Dec 18, 2008 at 8:06 AM, Joe Atzberger <ohiocore@gmail.com> wrote:
Brooke --
I don't even think we're talking about the same thing anymore. Versions will differ, and if you prefer an old version for its featureset or the specific implementation of a given feature, then stick with that version or please work to port the code over. No substantive feature has been deprecated capriciously. Indeed, bugs are sometimes introduced with new architectures and compromises between competing designs can alter requirements, interfaces or functionality. But this is normal to any developing project.
The point was already settled in the other thread that aspects of the HLT acquisitions implementation in question never made it to mainline Koha. As far as that goes, it was not "broken" or taken out: it was never there. So we're not talking about the same thing. I don't blame Katipo or Chris or HLT or anybody else, because managing customizations vs. mainline code can get complicated, especially using the version control systems of that time, with limited professional and community resources, etc, etc.
Just to clarify this bit. There was a version HLT were satisfied, it was in 1.0 ... it remained in a version they were satisfied up through and including 1.2.4. Then it got broken, I think around the time MARC storage in the db was added (lots of things got broken). HLT didnt upgrade to 2.0.0, but waited for 2.2.x Where acquisitions was fixed for them, but parts of that fix must not have made it into the code repository. This was before Liblime existed. So it's fair of them to say they never saw a working version. So 2 problems, 1 a feature got broken (it happens), and 2 the fixes didn't survive/make it into the 3.0.0 code base. Chris