It's 2015 \o/
alpha-1.0+parrott has been finalized!
It had a recap late last year too. What this means though, in focus sheet terms, is that any loose ends were tied up, any known errors in the documentation were addressed, all samples and tests were verified to be running against the latest code, things like that. It also get's a github tag/release now, which is mentioned below.
First, though :
Happy Holidays and new year
This was a message I sent over the chat, but will relay parts here:
Happy new year everybody, hope it's a great one!
I look forward to forging ahead with the community as the year starts on alpha 2 onward for luxe, and the first focus sheet for snow (native sdk stuff).
Thanks btw, to everyone who has been testing and using snõwkit, I honestly appreciate the support during alpha and I can't wait to continue forward together and show you all the stuff in the pipeline.
2014 wrap up
Over the holiday period, I was pretty busy with an exciting update on my main project which kept me away from pushing stuff to the public repos, but that doesn't mean I wasn't working on snõwkit stuff. It just meant I could take a break from the focus and get distracted on more "fun" future stuff :p
I'll post about all that stuff independently, as it's quite in depth - but the first thing relates directly to UI.
I don't usually like to post about a library until it's available, mínt is not yet available, but it felt fitting to mention given the demand for a UI. I want to assure everyone that it's definitely a priority to me, especially when it comes to developing great games and apps.
To put it simply, mint is a game focused, minimally pure renderer agnostic, haxe UI library. It does all the logic, and outsources the rendering events to you, or the framework. For luxe, of course this would just work with everything else. What that all means in details and in practice will make more sense as I post info about it, but I want to also note that UI is a pain point for many developers - and getting it right out of the gate is imperative.
This also means it is being kept to a very tight focus (without sacrificing flexibility), as to prevent it being a bloated mess. There will be mínt specific posts in the near future, keep an eye out.
Closer toward the end of 2014, hxsw was added to the snõwkit libs page as well, which I spent some time on exploring as the backbone for shader handling in snow and luxe. I've made extensive use of the original C++ version back in previous engines, so I was excited to have it back in my toolset. I think I have a good workflow for it that will make it's way in during the alpha focus sheets for snow first, and then luxe.
I also have been working closely with my good friend Andreas on his tweening library delta. It's not ready yet, but it will be replacing the bundled Actuate in luxe, as it will be more tailored and optimized to the goals at hand. It's more inline with the targets and goals of snõwkit as a whole.
Of course your actuate dependency will just be from haxelib instead of bundled, so you shouldn't worry too much, but it means that more parts of luxe will be intrinsically tweenable in an optimal, strongly typed way. It also means that random blobs of code that aren't ever touched won't exist in the luxe code base, as it should be, so It's a good deal over all.
Now that alpha 1.0 has been wrapped up, you will notice that the docs page for luxe is sporting an automated version number:
This is directly tied to github releases, which is updated whenever we tag a milestone. If you were holding out on bleeding edge updates and only looking to upgrade to milestone builds, this would be the place to check.
During alpha, all releases will be marked as "pre-release" and as mentioned previously - will only be deployed via github directly. Once we reach beta (in a few focus sheets time), the releases will also be pushed to haxelib automatically.
One other thing that people have been asking about is more specific roadmaps. I am of two minds about trying to predict the future, as inevitably priorities change and things come and go, and I want the focus sheets to be directed on the most pressing needs at the time. This is especially true for the alpha time frame.
I will however try to post more about the things that are coming ahead, so that people can see where we are headed overall. This does mean that info about future stuff won't be given in any specific order, though, but at least you'll know that something will or will not make it into the core vs modules.
To keep this post on point, and the roadmaps easy to find, they will be posted separately in more detail soon, but here we can run over some quick points so you get some idea of the current landscape.
luxe is already listing it's second milestone on github for discussion and perusal. It's about Assets/Resources/Windowing specifically, and has a number of things that need addressing.
Here is the milestone list, and the discussion was opened so long. Be sure to chime in on this issue for any bugs or features relating to the topics, as once the focus sheet begins there won't be much time to toss your suggestions into the ring.
The alpha-2.0 focus sheet will be posted separately in full detail with it's name, as we did with alpha 1.0.
Another roadmap-like mention that I'll post in the coming week is the first named alpha for snow. This milestone will deal with the native sdk specifically, and will require some testing from us as a community to ensure the best possible results.
If you don't know exactly what the native sdk means - it's about extending the native features of snow through an sdk + api, in order to access external and third party code - without it being in the snow code base. An obvious example - iAds on iOS, or a third party jar library on Android.
There are quite a few things in branches that I have been working on for some time alongside the master branches on the repo - things like the new rendering backend for luxe (inside phoenix) which has been mentioned on the site a few times, there is also rewritten math classes (Vector, Matrix, Quaternion) that are A LOT more optimized in every way, with some concessions along the way. Other things include whole libraries, like mínt, for example.
There were some native additions in hxcpp recently in a user facing release on haxelib, which means a lot of stuff in the underlying layers of snow that I have been optimizing can surface into the master user base sooner than later. Originally it was slated for haxe 3.2, so it wasn't ideal to try and actively maintain. Since it's out now, it's easy to push up.
All in all, I'm excited for the things in the beginning half of this year - as we push through the alpha milestones together, and as we start to see more things being created in and around the community we can solidify the tools and frameworks to be the best they can be.
As a reminder, http://snowkit.org is your community. It's the place to share your projects, post project updates and dev logs, ask questions and get involved.
snowkit public chat
The snõwkit community chat has been a great place to get help and show off your work, but don't forget to post to the wider community for others to find and look through.
Since announcing, things have been steadily growing and I am ever grateful for everyone that takes the time to file issues, dig into bugs with me, and especially contributes code to the repos. Especially so, I love seeing what people are creating, and there are some really great things being made so far and it's so much fun to watch them being developed.
I believe the commitment from the community will be the reason any of this will make sense in the end - so really, thanks, and I look forward to the road ahead.
For interest sake,
As of now, there are around ~110 sign ups on snowkit.org, and 103 members on gitter, which is really great to see people finding their way here so early. The other day, luxe also passed 100 stars on github - if you want to and haven't yet, you could star the repo to show your support of the project.
to 2015, onward and upward!