On Wednesday, July 06, 2005, at 12:53 PM, Avi Yashar wrote: > Allie, I said 3.10, not 3.1. Please see Alexander Kunz's explanation. > I believe he said it better than I did.
Version numbering almost always exhibits a natural mathematical progression unless, the number defines a date or have some other similar meaning, or numbers are separated by non-numeric string. 3.10 is less than 3.9 3.1 is less than 3.9 3.10 cannot follow 3.9, claiming to be a later version. Not from how TB!'s version numbering has proceeded over the years. With all the sudden jumps or changes in TB! version numbering that's intended and not a typo, there has always been a mathematical increase in the numbers. If you've installed the latest MSI, open the readme file in the installation directory and follow the version numbering from TB!'s earliest versions. There has always been a mathematical progression in version numbering. Proposing that 3.10 follow 3.9 would certainly take them way off course with how they've been doing their numbering. > The point is that version numbers are not purely mathematical. This may indeed be the case for only a few apps. However, TB!'s version numbering has always followed a mathematical progression. > Indeed, some times they include a combination of numeric and > alphabetic characters. For example, 3.5b1 would be 3.5 Beta 1 and > 3.5c1 would be 3.5 Release Candidate 1. However, the numeric components follow a natural mathematic progression, especially when separated by 'periods'. x.y.z follows a single numeric progression. However, for x.y <string>z z follows its own numeric progression while x.y follows a numeric progression. The only way your proposal could work is if it were like: v3 beta/9 going to v3 beta/10 10 follows 9. Again, I'm speaking of TB!'s version numbering since its first release and not for apps in general. So, with their system, they can follow 3.5.36 with 3.51. > When beta testing is complete, the FCS (or production-level release) > would be 3.5. You aren't going backwards by dropping those extra > numbers (as would be the case mathematically) Right. > - it is just that those extra characters define a stage of > development. Yes. > So my point was only that 3.5 is not at all the same as 3.50. That is > true in mathematics, but it is not necessarily so with respect to > software versioning. With regards to TB!'s versioning system, it could indeed mean the same. >> This is as semantic a criticism as they come. You've just defined >> the meaning of semantic. > I don't think so. I am arguing for a logical and consistent system of > version numbering with TB. ... and then propose that *for TB!*, 3.10 could follow 3.9? -- -= Allie M.=- Using TB! v3.51 System Specs: http://specs.aimlink.name =-=-= ...Put on your seatbelt. I'm gonna try something new. ________________________________________________________ Current beta is 3.51 | 'Using TBBETA' information: http://www.silverstones.com/thebat/TBUDLInfo.html IMPORTANT: To register as a Beta tester, use this link first - http://www.ritlabs.com/en/partners/testers/
