On Tue, Apr 29, 2014 at 4:55 AM, Jim Jagielski <j...@jagunet.com> wrote: > > On Apr 28, 2014, at 2:44 PM, Brian LeRoux <b...@brian.io> wrote: > >> I feel the comments there are not really constructive or fair. Cordova >> changes too much? Sorry, static code means bitrot aka abandonware. > > True. But code that is "constantly" changing also means people > cannot standardize on it and they look for something more > "stable"... The trick is to find that happy medium. >
I know you're actively trolling, but I'll still comment on this. Some facts are important, and unlike you, I actually care about the facts: 1. The code isn't constantly changing. There's still a CLI-less workflow. You can use plugman to install plugins, or you can copy the Java classes manually and edit the config.xml so it works like the old project. This project may be more active than other projects, but my workflow for actually doing development on PhoneGap and Apache Cordova hasn't changed in almost four years. 2. The statements made above are absurd. It'd be amazing if Simon and I magically wrote the best code ever in PhoneGap 1.2 on Android, but I can tell you right now that version was garbage. That was code written back in November 2011. All minor versions before 2.x were done to ramp up to 2.0. We did a much better job ramping up to 3.0, but democracy and a bit of peer-pressure made sure that we changed a plugin API namespace at the last minute. That said, find and replace is trivial to do on all major IDEs and text editors, and isn't worth the rage that we've seen here. 3. We've kept 2.x support alive for six months before deprecating it. This was the last version before the pluggable module system. We have to move at the same pace as mobile, which luckily is slowing down, but is still ridiculously fast in comparison to server or desktop development. This means that we have to make sure that we're future-proof and that our code actually works with the latest SDKs while we try and add features and make mobile development actually easier. We've gone to great lengths to provide two methods for development, and while many people object to us providing both the CLI (which I recommend to our non-technical users currently) and plugman (which is easier than just copying over files), I think we're going to keep doing this for a long time to come, because of our users. Also, you have to be active when the decision is being made to have any impact in that decision. The namespace change was a terrible decision for so many reasons, since it was done to ease contributor pain at the cost of plugin developers. However, I was the only person in the discussion thread that pushed back against this, and I pushed against this as hard as I could. It's extremely hard for us to go back and reverse decisions once they've been made after a major release. Saying that your opinion matters a year after the fact is ridiculous, because not even my opinion mattered a year after the fact, other than the fact that I was there when it could have, and have it on the record.
