Hi there, First I want to thank you all for the personal effort and commitment you dedicate to this project, for me "it is priceless" gracias, thank you.
I don't think I have the knowledge to get into the underlying technical discussion; In my case, I had not even actively used Maven and ant until now, and of course I don't get to see the advantages and / or disadvantages of one or the other. I belong to the group of professionals who are "surpassed" migrating from Flex and that every morning we look at the Royale website to see if "there is a new sdk" 😝 I have carefully read this thread and that of the development group [1] and I am with Andrew. From my ignorance I understand that, if so much effort has been devoted to building ant + maven, on the CI Server, it is because it has been valued that "it is the best" (?), But if it has been possible to build with Maven successfully Why not do it this way and use time and enthusiasm in other challenges? How about version 1.0 😝? I reiterate my admiration for your "work" and I will try to create the best showcases of it. Thank you. I include the text in my mother letter ... Hola, Primero quiero agradeceros a todos la implicación y esfuerzo personal que le dedicáis a este proyecto, para mí “no tiene precio” 😊, gracias. No creo tener los conocimientos necesarios para entrar en la discusión técnica subyacente; en mi caso, ni siquiera había utilizado Maven y ant activamente hasta ahora, y por supuesto no llego a ver las ventajas y/o inconvenientes de uno u otro. Pertenezco al grupo de profesionales que estamos “sobrepasados” migrando desde Flex y que cada mañana miramos la web de Royale para ver si “hay nuevo sdk” 😝 He leído con atención este hilo y el del grupo de desarrollo [1] y estoy con Andrew. Desde mi ignorancia entiendo que, si se han dedicado tantos esfuerzos en la construcción ant+maven, en el Servidor CI, es porque se ha valorado que “es lo mejor” (¿?) pero, si se ha podido construir con Maven con éxito ¿por qué no hacerlo así y utilizar el tiempo y el entusiasmo en otros retos? ¿qué tal la versión 1.0 😝? Reitero mi admiración hacia vuestro “trabajo” e intentaré crear los mejores escaparates de él. Gracias. [1] http://apache-royale-development.20373.n8.nabble.com/Releasing-Finally-giving-up-td15323.html De: Andrew Wetmore <[email protected]> Enviado el: viernes, 27 de marzo de 2020 12:31 Para: [email protected] Asunto: Re: [Action Required] About how to release Apache Royale I do not have the technical knowledge to provide an informed evaluation at the build level, so I will try to look through the lens of outcomes: what are we doing builds for? Maybe right now a large proportion of Royale users are people who like to tweak the source code and do their own builds locally. I do not think that is a user group that can sustain this project or make a difference in the world of application development. Our target audience is people who want a stable build they can use to build their own applications. The faster we can get stable builds out, the faster we can become available to that audience. The faster our user community grows, the better Royale becomes: the community exposes bugs, requests or provides emulations, creates showcase-quality apps using Royale, encourages other people to give Royale a try.. I have trouble keeping up with assertions that seem at odds with each other: "Ant is the devil!" "No, Maven is evil!" And maybe I am missing something crucial, but I seem to be reading that it is possible to build a stable version of Royale that we can release via Maven, while still providing support for anyone who wants to build using ant. Is this so? Then why is that not the way to go? A On Wed, Mar 25, 2020 at 2:31 PM Carlos Rovira <[email protected]<mailto:[email protected]>> wrote: Hi all, if you care about this project, the PMC will need your help to know what's important to you about the "Release Process". Chris Dutz and I was working these days on release a new version. We finally give up due to the complexity of the process. The process involves to build in a CI Server with 13 steps and each one with some manual commands in CI server and in local machine. The process involves use Ant and Maven build systems, and that's what makes it so complicated. Maven is the the most widest and used build system for release as this statistics clearly states [1] and is perfectly prepared for CI servers and all the tooling needed nowadays for release process. Chris and I thinks that just using Maven can reduce the process to 2-3 steps and release without much problem, and even release often (each month or each 2 months). So we're asking to not require Ant to be part of the release process, since we think is not needed. (This does NOT means remove Ant, we are talking just about release, and people using Ant will be able to continue using it as a build system, so don't be afraid about Ant support in the future!). We're afraid for this project due to this problem, so for this reason is important that you express what's important from your point of view as an Apache Royale user. So please express yourself in this thread and let us know what path should we take. It's important. We want to hear from you! Thanks community :) [1] https://twitter.com/ChristoferDutz/status/1240219399551934473 -- http://about.me/carlosrovira -- Andrew Wetmore http://cottage14.blogspot.com/
