Re: [ANN] BubbleWrap 1.1.2 + stepping down

[Francis Chong]

Thank you for bringing this project and your time to guiding it for these months. It has been fun and certainly help a lot of fellow RubyMotion developers.

I have same feeling on the debugging tools on RubyMotion, and would be more confortable to use it on smaller pet projects.

On Thursday, July 19, 2012 1:43:18 PM UTC+8, Matt Aimonetti wrote:

I'm glad to announce the 1.1.2 maintenance release of BubbleWrap, here is the changelog: https://github.com/rubymotion/BubbleWrap/blob/master/CHANGELOG.md

And in case you don't know about BubbleWrap yet, here is the website: http://bubblewrap.io/

I would also like to welcome Dylan Markow, the 8th developer to join the development team.

I also have a personal announcement to make. This will probably the last release I'm actively involved with, at least for a little while. 

I started the BubbleWrap project to share some of the RubyMotion code I copied from project to project. 

It quickly became used by many which lead to great contributions, making this library RubyMotions's most popular community driven lib.

That said, I came to the conclusion that I don't have the required bandwidth to deal with all the RubyMotion bugs and lack of debugging tools and keep pushing BW further.

Just making sure our test suite runs properly has been challenging to say the least. 

The support we received from Hipbyte was quasi non-existent, I didn't receive any replies nor follow up for any of my bug reports or support tickets. 

I feel that RubyMotion needs to mature before I can invest more of my free time.

I still like RubyMotion but I'll only be using it for toy projects and quick prototypes. Because of this decision, it's hard for me to justify the time required to lead and support a project such as BubbleWrap.

It doesn't mean that the project is dead, we have 7 other active committers and many other contributors. 

Feel free to join them to make BubbleWrap even better.

Sincerely,

- Matt

[Christopher Pappas]

This is such a shame, particularly in the way that you phrased your stage-left.  

On Thursday, July 19, 2012 9:45:23 AM UTC-7, Veraticus wrote:

I've been using CoreData, networking and reachability, the appearance API, and CoreAnimation. While I'm not using OpenGL or CoreAudio, definitely my app uses some of the more complicated frameworks that Objective-C exposes. I will admit that integrating them has been difficult -- I just wrote a long blog post about how to get ShareKit working correctly, which is an enormous pain in RubyMotion -- but I still think it's easier than getting them working in Xcode, which requires header linking and compile flags and blah blah blah. So I still disagree that RubyMotion is only suitable for "simple" applications.

I can see why testing a wrapper like BubbleWrap would be so difficult, though, and how you'd run into annoying edge cases. Testing metaprogramming stuff is hard even outside of a new framework, and you're really pushing RubyMotion to its limits. I think I can safely speak for the community when I say BubbleWrap does some awesome stuff, and it'll be the worse for your lack of participation, Matt. Hopefully you'll come back to it one day.

And, to add to what Colin's said, debugging can definitely be a pain at times. I'm using a fair amount of Cocoapods, and it makes me really sad that stack traces don't go into them when they raise an exception. To get ShareKit working I had to add NSLog statements all over it and the Facebook-iOS-SDK to tell me what was happening where -- I really, really missed breakpoints when I was writing those. So, no, RubyMotion isn't perfect. But for me, the upsides still outweigh the downsides.

On Thu, Jul 19, 2012 at 11:11 AM, Matt Aimonetti <matt.aimonetti@livingsocial.com> wrote:

it sucks that you haven't experienced the level of support you'd consider ideal, Matt, I think RubyMotion is usable for production projects.

Support is only a small part of the issue. I do think you can push RM projects to prod and be fine, but it depends on the complexity of your app. In this specific case, BubbleWrap is more complex than your average iOS app. I didn't file a lot of bugs or weird behaviors I found, mainly because they were hard to debug and I indeed didn't feel I was getting the support needed to do that.

You are right, HipByte is a one man show (Laurent) and what he did is amazing. I'm sure it will eventually get better and more stable, but until then, I will stay on the cautious side of things for any important projects.

- Matt

On Thursday, July 19, 2012 9:02:49 AM UTC-7, Veraticus wrote:

All this said, RubyMotion (and HipByte) is essentially just Laurent and Eloy, isn't it? Considering it's just the two of them, the level of completeness and usability of the toolchain is astounding. I don't have a project in production yet, but I've been testing one for release with a limited number of beta testers; I've been successful in finding and debugging crashes with it. It's not super ideal, but it works.

Ultimately, while it sucks that you haven't experienced the level of support you'd consider ideal, Matt, I think RubyMotion is usable for production projects. Maybe I'll have changed my opinion in a month or so after I've released, though.

On Thu, Jul 19, 2012 at 10:37 AM, Matt Aimonetti <matt.aimonetti@livingsocial.com> wrote:

Interesting - *very much more* than $150 was invested in time and effort on BW and yet not willing to invest $150 for a lic.

Dennis, the reason is simple, it's bullshit. Laurent claims that  "Matt is not a paid customer" so that's ok not to give him support. (I didn't know that there was such thing as a BubbleWrap non paid customer) 

First of all, I have 2 licenses, one that was given to me by Laurent during the beta (for free) and another purchased by my employer (who paid money for it). If that wouldn't have been the case, I would have bought  a license myself. As you said, the time invested in BubbleWrap represents much more than the cost of the license.

So the whole paid customer excuse is absolutely nonsense, and even if it was true, nothing was communicated to me and no help was given on the project when we needed it.

Anytime someone pulls out the "toy projects and quick prototypes" stick and waves it around it tells me there are some unsaid motives at play here .

You're free to interpret my email the way you want. Look at Francis' comment after my post though, he's been one of the first active contributors to BubbleWrap and I think the first to join the team.

Ask James, another team member who wrote a big chunk of BW.

But at the end of the day, it's just my opinion, nothing else. Prove me wrong instead of just claim I had unsaid motives that are the reasons why I'm stepping down from a project I had fun working on.

- Matt

On Thursday, July 19, 2012 5:34:25 AM UTC-7, Dennis wrote:

Interesting - *very much more* than $150 was invested in time and effort on BW and yet not willing to invest $150 for a lic. So build up some serious community credibility to give a post like this more punch??? Anytime someone pulls out the "toy projects and quick prototypes" stick and waves it around it tells me there are some unsaid motives at play here ...

On Thursday, 19 July 2012 01:43:18 UTC-4, Matt Aimonetti wrote:

I'm glad to announce the 1.1.2 maintenance release of BubbleWrap, here is the changelog: https://github.com/rubymotion/BubbleWrap/blob/master/CHANGELOG.md

And in case you don't know about BubbleWrap yet, here is the website: http://bubblewrap.io/

I would also like to welcome Dylan Markow, the 8th developer to join the development team.

I also have a personal announcement to make. This will probably the last release I'm actively involved with, at least for a little while. 

I started the BubbleWrap project to share some of the RubyMotion code I copied from project to project. 

It quickly became used by many which lead to great contributions, making this library RubyMotions's most popular community driven lib.

That said, I came to the conclusion that I don't have the required bandwidth to deal with all the RubyMotion bugs and lack of debugging tools and keep pushing BW further.

Just making sure our test suite runs properly has been challenging to say the least. 

The support we received from Hipbyte was quasi non-existent, I didn't receive any replies nor follow up for any of my bug reports or support tickets. 

I feel that RubyMotion needs to mature before I can invest more of my free time.

I still like RubyMotion but I'll only be using it for toy projects and quick prototypes. Because of this decision, it's hard for me to justify the time required to lead and support a project such as BubbleWrap.

It doesn't mean that the project is dead, we have 7 other active committers and many other contributors. 

Feel free to join them to make BubbleWrap even better.

Sincerely,

- Matt

--
 
 
 

--