Hi Eric, thanks for taking the time to write this. It mirrors a lot of our internal discussions & I think it's a good thing to raise some of these points publically as an open project.
Before going into the points I'd like to clarify some of my personal thoughts. Notice that we don't have internal consensus inside the company either on some of these but hopefully once I explain these you'll understand my response more clearly. I'll first start with a statement of facts: - There are quite a few long time Codename One developers that keep the lights on here so we are doing something right - Growth for Codename One has been flat for a few years now, we are growing at a very small rate especially when compared to other tools - Most of these other tools have a lot of marketing $ and brand names behind them so we can't always do what they do to achieve growth - Part of the stagnation is Oracles mishandling of Java that hurt the overall community, this is reflected by the mass exodus of native Android developers to Kotlin I'd also like to state that the fact that we are a commercial company that needs to turn a profit is problematic in this market. Even though the features we charge for aren't available in any other tool the perception is still damaging. Unfortunately, I can't see a way around this without duplicity. As an engineer there is a strong desire to pull out my "engineering hammer" and fix the problem. E.g. if we just have this "one feature" people will flock to Codename One. That's wrong. With the exception of one missing feature (instant run on device/on device debugging) we pretty much exceed every other tool in the market. Personally I think we need to focus a lot on design and providing "ready made solutions" e.g. the Uber App and upcoming Facebook Clone. Our new emphasis on Kotlin also helps with communicating to the "down with Java" crowd. Of all the points you made I think point 2 is the most important. We need the experience of Codename One to be amazing out of the box... As to your points: 1. We would love to. The problem with that is that these pages make sense when there is a lot to show off. There isn't. The page would look worse than not having one at all right now. I would love to have an "Awesome Codename One" git page but it wouldn't have enough materials. As you saw when you tried to create a store it's a chicken/egg problem. The whole "if you build it they will come" mantra works only in movies. In reality a marketing person spends a huge amount of time on the phone talking to people and getting them to write things. That's what we did at Sun when we launched LWUIT. Our manager literally called up potential customers and convinced them to adopt our framework so we would have showcases for the launch. 2. I mostly agree with that. If you will look at our commit log just yesterday Steve added the CSS support into the designer tool which means it will ship by default with Codename One 4.2 which will come within a couple of weeks. He also made some changes to CSS based on my requests that finally make it competitive with the designer tool including things such as "live editing". I think we need to completely re-do the setup wizard and offer better app templates with better design. It's a matter of time/schedule. 3. That's great. Lets see the marketplace and the content you have there and then we can talk about how we can help promote this. I'd like to help in any way possible. 4. Steve wanted to create a Codename One desktop project, I'm not sure where he stands on that. We develop our own desktop apps (e.g. the new GUI builder & settings app) using Codename One so it's very possible. What I would like to do for localization is an "auto-localize" tool in the simulator. It's something that would probably be pretty easy to integrate. It would install a "fake" resource bundle that sees all the localize requests and gather the keys then suggest additions to the resource bundle. The localization data can be saved as properties under the res directory. Since properties are a standard file format in Java there would be no problem in editing this directly/converting this. It should be reasonably easy to write I just didn't get the chance to fit it into my priorities. You can submit a pull request but we are generally averse to adding big new tools. E.g. the CSS plugin from Steve spent a lot of time as a standalone tool because we didn't want to support something as big as that unless its ready for prime-time. If it's simple then it would be fine but I'd like it to fit with the future direction of the tooling. 5. I agree it makes sense to remove some things but these sort of things should be done very carefully. If you have something in mind file an issue. In terms of size the impact is negligible as the optimizers remove unused code. 6. I tried to do that a few times. We talked to large companies and tried to get case studies etc. but it never materialized. One of my friends used to do these interviews as a full time job. I understand it's a pretty difficult task normally as people are very averse to doing it. We don't need interviews with random developers. We need interviews with banks & hospitals that use Codename One but those guys are generally very secretive. 7. I agree with this. I've given this a lot of thought over the past few years. It's a big task though... The first step was breaking the GUI builder out of the designer tool but the goal is to ultimately retire that tool entirely. -- You received this message because you are subscribed to the Google Groups "CodenameOne Discussions" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. Visit this group at https://groups.google.com/group/codenameone-discussions. To view this discussion on the web visit https://groups.google.com/d/msgid/codenameone-discussions/9ac1741a-38e9-42d0-bf2d-25fcb2a2594e%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
