World: r3wp
[!REBOL3]
older newer | first last |
Pekr 2-Jul-2010 [3605x2] | hah, ha! New milestone in R3 development seems to be close :-) I hope we get delayed modules too, so that gfx can be loaded on demand, which would be good for the X-less server - still one rebol version to use :-) |
Henrik - I saw small mention of some speed-up, due to more streamlined command interface (probably in opposite to DELECT). So - any chance we get updated 1000 cows demo or so, to see "the difference"? :-) | |
Henrik 2-Jul-2010 [3607x2] | we'll probably see that. I think he wants to work out all the bugs first. |
I've found that VirtualBox is a really bad testing platform. It slows down in places by a factor of at least 10, so I can't do any benchmarks. | |
Pekr 2-Jul-2010 [3609] | Is the API complete? Box, pen etc seems strange, there should be more, no? :-) |
Henrik 2-Jul-2010 [3610] | I don't think it's complete. He's just testing whether the whole thing fits together. |
Pekr 2-Jul-2010 [3611] | Good progress anyway. I noticed Twitter message mentions OpenGL. Well, I hope we get VID 3.4 with some nice bunch of styles first :-) |
Rebolek 2-Jul-2010 [3612] | I'll promise to update 1000 cows when the new gfx engine is ready :) |
Pekr 2-Jul-2010 [3613x2] | :-) I also like Cyphre's rotoobjects demo, or bounce one. Simply random draw shapes, with various levels of alpha, flying on the screen :-) |
btw - will DELECT still remain? New COMMAND type seems to remove the ability to have random order of parameters. And quite honestly - I don't like it. It might be good for very fast low level engines, where speed matters, but I still want the flexibility of random parameters. So - will DELECT remain for dielact builders, or will we have to fall-back to full parse in such a case? | |
Rebolek 2-Jul-2010 [3615] | Nobody said anything about removing DELECT, so I think it will stay. I never used it anyway, so I don't care. |
BrianH 2-Jul-2010 [3616x2] | DELECT could be used to convert a dialect block with unordered and/or parameters to commands into a command block with ordered, mandatory arguments. It does it already; all that would be necessary is to have if generate its results in a flat block rather than nested blocks. There would be no need to remove it - just change the output format. |
unordered and/or parameters -> unordered and/or optional parameters | |
Pekr 2-Jul-2010 [3618] | thanks. Then - let's request the change then :-) |
Gregg 2-Jul-2010 [3619x2] | IMO it's more critical to use to get a good GUI up & running with a broad set of styles. I agree Robert. Even if the set of styles is not broad, if we have a few good examples of how to write styles (and maybe some docs, which I am willing to help with), we can turn the community loose. |
We should also have a list of styles, and if a core GUI team member sketched out rough designs or inheritance/composition thoughts and prototypes, we stand a better chance of our work being used. | |
Robert 2-Jul-2010 [3621x3] | Correct, that's what we will do. |
And, at this time I will drive it cooridnated with everyone wanting to help via a SCRUM process. | |
And, I'm pretty sure it takes a couple of months and all the baseline stuff is lighyears ahead. | |
Henrik 2-Jul-2010 [3624] | http://www.rebol.net/r3blogs/0324.html Update on host kit. |
Steeve 2-Jul-2010 [3625] | Is that not the time to reveal some sources. I'll like to have a look. |
Graham 2-Jul-2010 [3626] | Sounds like the latest hostkit is just a few months away from release. |
BrianH 2-Jul-2010 [3627x2] | Well, delayed modules are going to have to wait until late next week. |
I suppose that's appropriate :) | |
Steeve 2-Jul-2010 [3629] | I don't expect something complete now, but it's nessessary to be accustomed when the code stay light. I don't want to discover the big one in one cast and feeling lost. |
Graham 2-Jul-2010 [3630] | You want to be acclimatised and not get a sudden culture shock! |
Steeve 2-Jul-2010 [3631] | exactly |
Graham 2-Jul-2010 [3632] | you have to be part of the gui developement team ... |
Steeve 2-Jul-2010 [3633] | why ? Is that a secret ? |
Graham 2-Jul-2010 [3634] | seems effectively so ... |
Rebolek 2-Jul-2010 [3635] | There's no secret, once there's something to release, it will be released. Right now, there's nothing right now. That will change soon. |
Graham 3-Jul-2010 [3636] | Is there a suite of tests for Database protocols ? I presume there was something for R2 ... can RT release those so we can use for testing? |
Robert 7-Jul-2010 [3637] | Ok... successful host-kit build! - Carl |
Pekr 7-Jul-2010 [3638] | Cool :-) |
Graham 7-Jul-2010 [3639x3] | what's the download link? |
Ok... successful host-kit build! about 1 hour ago via Socialite.app <<< that means Robert! | |
Carl uses the Twitter API | |
Pekr 7-Jul-2010 [3642] | I think that Carl really said that, but Robert just twitted about it :-) |
Henrik 7-Jul-2010 [3643] | correct |
Robert 7-Jul-2010 [3644] | I'm the ghost twitterer to help keep information flowing. |
Graham 7-Jul-2010 [3645x2] | And it's appreciated! :) |
note to Carl .. add a tweet button on Altme | |
Robert 7-Jul-2010 [3647x3] | So, Petr asked, what this message regarding the host-kit means: |
The components are there and work together, and more DRAW commands still need to be implemented. For this we priorized some task different to faster make it possible that we can start implementing DRAW commands in parallel to Carl's other work items. | |
There are still some questions to answer and quite a lot of little things to do. But IMO it's a big step ahead that the host-kit now builds. | |
Henrik 7-Jul-2010 [3650] | Internal release made of hostkit for A99. Still some work to do before a public release can be made. |
Maxim 7-Jul-2010 [3651] | funny, I see all I need so that I could start using it already. |
Henrik 7-Jul-2010 [3652] | still some compiler issues to solve. Cyphre is helping with that. |
Rebolek 7-Jul-2010 [3653] | It's just that ++ is more like -- in real life. |
Pekr 8-Jul-2010 [3654] | It is coming .... http://www.rebol.com/r3/docs/concepts/host-kit.html |
older newer | first last |