Release process changes

[Holgar]

To tell the truth, I'm completely new to such matters and heavily in doubt, if I could add here any useful statement. On the other hand, take me as a representative of the completely dull fan community

While finding myself here for a moment in some kind of crossfire, I tried to trace that back and found, that anything started here with a complete misinterpreting of this posting:
posting.php?mode=quote&f=20&p=41534
On the other hand: the link on Github pointed to the new version and the download section here had been updated, while the process of distribution started (0.40.0 dropped by here on Tuesday 30), For some newbie like me this is not distinguishable from a software release, so I felt a little disturbed pondering about the difference between a release and an *official* release.

So it's good to read the present discussion, which gives me a better insight into the processes here and therefore will help up my understanding in the future.

Curious as I am I do not really like the idea of being excluded from these proceedings with a new release on the horizon, but I think, it would really better things if there is a little more policy with this. It simply didn't look so good, and this was the only reason, why I called for action that way. Sadly to say, I'm not alone recovering my helpless remnants of written English, but also have to go deep into the linux side of life and reboot my head completely since it's also some time ago, when I really was used to commandlines and bash, So for the next future I'm bound to play the role of an onlooker. May be this changes over time

~

[Atahualpa]

[...]
While finding myself here for a moment in some kind of crossfire, I tried to trace that back and found, that anything started here with a complete misinterpreting of this posting:
posting.php?mode=quote&f=20&p=41534
[...]

I think we've already cleared this up. No worries.

Curious as I am I do not really like the idea of being excluded from these proceedings with a new release on the horizon, but I think, it would really better things if there is a little more policy with this. It simply didn't look so good, and this was the only reason, why I called for action that way. [...]

We don't want to completely exclude you from the release process. I'd rather take all internal discussion to the Team forums and post relevant information to the public release thread where there would be another discussion. (E.g. "Start of release phase", "Final videos are ready", "Change log is ready", or even "We got some problems, please test Windows 32 bit builds".)

[lysol]

Great discussion guys. Really interesting to read. I'm in favor of the second option with some kind of a closed forum section for release planning. As long as there are enough testers for each operating system that have access to the forum, I think it would be great.

A big problem with the current release threads is that they easily become 20+ pages long, which makes it really easy to miss important posts like Ace posting windows installs or testers confirming that the OSX packages work. Therefore, I think having a new, closed, release forum would make it more easy to have several threads about the release. Threads that shouldn't be spammed with comments here and there.

For example:

• Main release thread (the usual discussions where all is organized)
• Windows RC thread (with first post containing links to release package and the replies only commenting on if they work or not)
• OSX RC thread (same as above)
• Linux tar RC (+ apt?) thread (...)
• Video thread (Atahualpa posts as usual, people can comment here)

Having it this way, raevol can really easily see if all the binaries has been tested, and testers can really easily find the latest RC to test. The main release thread can then be used more efficiently, since the discussions are more focused on one topic.

[Atahualpa]

+1

[raevol]

A big problem with the current release threads is that they easily become 20+ pages long, which makes it really easy to miss important posts like Ace posting windows installs or testers confirming that the OSX packages work. Therefore, I think having a new, closed, release forum would make it more easy to have several threads about the release. Threads that shouldn't be spammed with comments here and there.

For example:

• Main release thread (the usual discussions where all is organized)
• Windows RC thread (with first post containing links to release package and the replies only commenting on if they work or not)
• OSX RC thread (same as above)
• Linux tar RC (+ apt?) thread (...)
• Video thread (Atahualpa posts as usual, people can comment here)

Having it this way, raevol can really easily see if all the binaries has been tested, and testers can really easily find the latest RC to test. The main release thread can then be used more efficiently, since the discussions are more focused on one topic.

This is a really insightful point. There's a lot of ways to skin the cat for this, but another idea would be to handle release tasks on the tracker. Then tasks can be assigned to people, marked with statuses, commented on, etc. Forum threads would work fine for this too, but just tossing this in as an idea.

[darkbasic]

I vote for option two, but please do not restrict access too much. It should also be stated somewhere that such section exists and that people could ask for access.

[SquireNed]

I vote for option two, but please do not restrict access too much. It should also be stated somewhere that such section exists and that people could ask for access.

The problem is that only a core few can have access by definition, or we wind up with the same problem because someone sends the video link to someone they shouldn't and then the metaphorical cat is out of the metaphorical bag.

[lysol]

I vote for option two, but please do not restrict access too much. It should also be stated somewhere that such section exists and that people could ask for access.

The problem is that only a core few can have access by definition, or we wind up with the same problem because someone sends the video link to someone they shouldn't and then the metaphorical cat is out of the metaphorical bag.

True of course. But if we invite someone to that hidden forum, we should of course make sure they know that they can't share stuff however they want before they get invited. The thing with the "leaked" video on reddit was that is was just "some random guy" that probably just browsed the forums one day and found a video he thought interesting. I don't think he intentionally wanted to be a dick.

[Atahualpa]

Our active member base is relatively small which should make it easy to select responsible users only. Everyone should be instructed to not share any delicate contents with other users (or with people on reddit ).

The following members could be given rights to the Team forums: active or even inactive developers, active team members, long-time members who have proven worthy , and willing testers (maybe restricted access only?).

As for the regular user, we would provide updates regarding our mysterious and top secret work (see below).

We don't want to completely exclude you from the release process. I'd rather take all internal discussion to the Team forums and post relevant information to the public release thread where there would be another discussion. (E.g. "Start of release phase", "Final videos are ready", "Change log is ready", or even "We got some problems, please test Windows 32 bit builds".)

[sjek]

would op for at a least read open forums as that one thing that keeps people interested allhought it can have it downsides sometimes.

edit: and case in point (not meant of hijacking), about hard implementation to rare reports / testers.
here's a scripted spell that trying to write and making to work at least in theory:

Spoiler: Show

Code: Select all

Begin radiustriggers

if player->getspell "vampire levitate"==0
	return
endif

short casting
float x
float y
float timer
short rotspe
short interval
short icount
short xstart
short ystart
float angle
float yre
float xre
short n
short radius
short xdire
short ydire

short placeonce

set rotspe to 90
set interval to 30
set radius to 500
set xstart to player->getpos x
set ystart to player->getpos y
set angle to dummyturner->getangle z

if angle<90
	set xdire to 1
elseif angle>270 
	set xdire to 1
else
	set xdire to -1
endif

if angle<180
	set ydire to 1
else
	set ydire to -1
endif 

if player->getspelleffects "vampire levitate" == 1
	set casting to 1 
	if placeonce==0
		dummyturner->position xstart ystart (player->getpos z) (player->getangle z)
		set placeonce to 1
	endif
endif

if player->getspelleffects "vampire levitate" == 0
	dummyturner->position 0 0 0 0
	set casting to 0
	set placeonce to 0
	set icount to 0
endif

if menumode==0
if casting==1
	set timer to timer+getsecondspassed
	dummyturner->rotate z rotspe

	if timer>n+1
		set n to n+1
	endif

	set yre to timer*2
	set xre to timer*2

	set x to xstart + n*xre*xdire
	set y to ystart + n*yre*ydire

	if angle > icount*interval
		placeitem "iron longsword" x y (player->getpos z) angle
		set icount to icount+1
	endif

	if (n*n*(xre*xre+yre*yre))>(radius*radius)
		set n to 1
		set yre to 1
		set xre to 1
		set timer to 0
	endif
endif
endif

End radiustriggers

it should start to summon swords in spiral going outward from player when certain spell is cast and active and makes excessive use of functions and calculus inside functions. still thinking after several iterations where the check is the proplem on which couple of years tagging along helps a alot. anyway still could learn at least some basic c++ as that can be some easy code change or then requiring more refactoring due to just the way how c++ works.
(edit: found and was many short variables as values made from float datatype, theoryproblem)

i think that problems will become more or less to this category althought i also think now that missing the point again with what kind of testing is needed to release versions but more to that direction.