Think Tank 2015

Important improvements are announced here.

Moderator: PPS-Leaders

Post Reply
User avatar
beni
Baron Vladimir Harkonnen
Posts: 949
Joined: Tue Oct 03, 2006 9:15 am
Location: Zurich
Contact:

Think Tank 2015

Post by beni » Tue Jan 20, 2015 3:55 pm

We just decided the date of the next official Orxonox Think Tank: Saturday, 7th February 2015

I really enjoyed the last think tank and developer meeting and I hope this year will be even better.

More Details:
Time: 15:00
Location: Supercomputing Systems AG, Technoparkstrasse 1, Trakt Edison, 5. Stock, 8005 Zürich

Let's gather the topics and issues we want to talk about during the think tank. It's going to be a relaxed atmosphere with food and drinks and also some gaming probably, but the purpose of the meeting is that we can discuss bigger decisions concerning Orxonox.

So please, go on, let's talk about what we gonna talk about ;)
"I'm Commander Shepard and this is my favorite forum on the internet."

User avatar
x3n
Baron Vladimir Harkonnen
Posts: 810
Joined: Mon Oct 30, 2006 5:40 pm
Contact:

Re: Think Tank 2015

Post by x3n » Tue Jan 20, 2015 8:27 pm

Location:
Technopark Zürich.
Image

We will gather around 15:00 in the lobby near the front entrance of Technopark, then move up to the SCS office together.
If you are late, that's no problem. Technopark is open at saturdays, so you can enter at your own. You find the SCS office in the 5th floor of the Edison wing. Use the bell next to the door.

Image
Fabian 'x3n' Landau, Orxonox developer

User avatar
x3n
Baron Vladimir Harkonnen
Posts: 810
Joined: Mon Oct 30, 2006 5:40 pm
Contact:

Re: Think Tank 2015

Post by x3n » Tue Jan 20, 2015 8:37 pm

Possible topic: Orxonox server (for forum, website, SVN, etc). What do we need, how do we organize it, who is responsible for what
Fabian 'x3n' Landau, Orxonox developer

User avatar
beni
Baron Vladimir Harkonnen
Posts: 949
Joined: Tue Oct 03, 2006 9:15 am
Location: Zurich
Contact:

Re: Think Tank 2015

Post by beni » Thu Jan 22, 2015 11:47 am

During the last PPS Convention several students have mentioned difficulties and issues. Maybe we can address those:
  • Lagging
  • AI
  • Code Complexity
  • Blender integration
  • Game Over conditions
Also I think several things are being worked on in almost every PPS. Maybe we can look at how these things actually progress and why it's so popular or deserves so much focus.

And a last point: I would like to pick up the discussion about content creation. We clearly create more code than content, which is fine, but I think we should talk about what we can do and how to get there.
"I'm Commander Shepard and this is my favorite forum on the internet."

flozi
Human Space Navy Sergeant
Posts: 13
Joined: Mon Oct 07, 2013 1:51 pm

Re: Think Tank 2015

Post by flozi » Sat Jan 24, 2015 10:42 am

I throw in one more thing to discuss:

How about restarting from scratch? I know this would be a huge step, but I think most of the mentioned problems could be easier addressed that way.
Also, starting from scratch is very exiting. As a student this would definitely motivate me more than taking over an aged codebase.

User avatar
x3n
Baron Vladimir Harkonnen
Posts: 810
Joined: Mon Oct 30, 2006 5:40 pm
Contact:

Re: Think Tank 2015

Post by x3n » Sat Jan 24, 2015 10:05 pm

More topics:
1) How can we improve the communication between PPS students and other Orxonox developers? We have lots of communication channels (forum, chat, mailinglists, tickets, wiki, etc) but often it is very difficult to know what's going on in this project because no one uses those channels.

2) How can we improve the activity in the Orxonox (developer) community? (more commits, more discussions, more development outside of the PPS)
Fabian 'x3n' Landau, Orxonox developer

smerkli
Human Space Navy Lieutenant
Posts: 20
Joined: Mon Mar 01, 2010 7:28 pm

Re: Think Tank 2015

Post by smerkli » Thu Feb 05, 2015 8:57 pm

Since I can't attend on Saturday due to craptastic planning on my side, I thought
I would at least state my inputs on the topics mentioned above here as a contribution.

Server:
- SVN permissions and accounts: possibly LDAP or simply move with
the old system since that worked so far.
- Data easy to move, but redirects nontrivial
- Same with Trac or new versioning system, maybe simply sticking with
what there is might be fine?

PPS:
- In terms of coding, I think the issue for more complex things like improving
the game performance require quite some hands-on coding experience that is
not taught in the ITET courses alone. What I think has been very popular
are tickets that are quite self-contained and are fairly local in the code,
such as creating game modes.
- AI seems like a very rich area, which I'd combine with scripting. Especially
for the advancement of a single player campaign, improvements to the camera
script controller and maybe dialogue integration seem like achievable PPS
projects that could enable more involving storytelling.

Content creation:
- A few recent tickets have made more things possible, such as the extension
of the space station modules with accurate collisionshapes. Combined with
the docking, this could now be made into levels where players travel from
one space station to another to get something or talk to someone, extending
on the current range of missions we have.
- While recording dialog for a story sounds fun, we would really need students
that are motivated to do so. Otherwise, I think just a textual dialog-popup
series would be fine to tell a story for now.

Restarting from scratch:
- I think restarting would in the long term not lead to a much cleaner
codebase, the whole model of students contributing over one semester is just
too open to keep code clean and well-structured. As we already discussed at
the previous Think Tank, we could introduce quality control and only accept
code that meets certain standards into the game's trunk. I think would be
more beneficial than a rewrite.
- Alternatively, we could also remove or refactor parts that are already present
and often present hurdles to people implementing new things. Examples for this
are for example the WeaponSystem which has a huge amount of features which
are not used a lot (yet) or the InputManager / InputHandler framework.

Communication PPS / other developers:
- We have done less advertising for the available channels in the recent PPS
iterations than before, however I think participation has never been very
high. This is likely due to many people looking at this as a PPS and
therefore a school thing, in which they generally don't participate with much
of their free time. I'm not sure how we could change that - the pizza evening
has worked well recently I think, maybe we could do this twice during the PPS,
one "get to know the developers" event and the usual pizza event.

Activity:
- I think the central issue there is that the multiplayer is not quite "there
yet". I've been trying to fix this by advertising a multiplayer debugging
ticket, but there were no takers so far. I think if we can get a gamemode
that is fun to play running in the multiplayer and actually start playing the
game more regularly, development motivation will also be higher.

Best,
Sandro

User avatar
x3n
Baron Vladimir Harkonnen
Posts: 810
Joined: Mon Oct 30, 2006 5:40 pm
Contact:

Re: Think Tank 2015

Post by x3n » Sat Feb 07, 2015 7:31 pm

Notes:

Thema: Content Creation
  • Bensch/Maxim: Kontakt mit ZHDK aufnehmen
  • Beni, Felix: Open-Source Content suchen
  • Editor: Steuerung wie in Blender? Editor komplett in Blender?
  • Quick-Reload des Levels
Thema: Code-Qualität
  • Doku: Im Code + Konzepte im Wiki. Aber woher wissen Studenten was es alles gibt?
  • Dev. Journals für Know-How Transfer?
  • Lernen damit umgehen - als positiv verkaufen
  • Fabian: würde Framework einführung machen, auch wieder Besuch in PPS
Thema: Kommunikation, Attraktivität / Aktivität
  • Wieder mehr Dev. Journals. Erfordert auch Reaktion von externen Devs auf die Posts
  • Es muss runherum mehr aktivität geben. Aktivität muss auch sichtbar sein (Webseite?)
  • Coding-Sessions
  • Attraktive Tickets, welche nicht nur im "eigenen Gärtchen" stattfinden
Fabian 'x3n' Landau, Orxonox developer

The Jo
General DuGalle
Posts: 121
Joined: Mon Mar 01, 2010 7:43 pm

Re: Think Tank 2015

Post by The Jo » Sun Feb 08, 2015 8:16 pm

Orxonox Thinktank 2015 - Protokoll
Anwesend: Fabian, Felix, Maxim, Johannes, Martin, Benjamin G.,Beni
  1. Video: Orxonox Gameplay Compilation
  2. Vorstellungsrunde
  3. Demo der letzten Änderungen
  4. Aktive Diskussion

Themenvorschläge:
  • Spielqualität/ Experience verbessern
  • AI: individuell pro Gametype
  • Code Qualität, Grösse, Alter, Komplexität
  • Blender integration: Workflow von Blender zu Orxonox
  • Game over conditions
  • Content creation vs. coding: zu wenig Content (Musik, Models, ...)
  • Manuell XML: besseres debugging, editor, helpers
  • PPS Tickets
  • Kommunikation: Wie stellt man die Kommunikation zwischen Ehemaligen und PPS-Studenten her?
  • Aktivität: Die Commits sind stetig gesunken. Wie macht man Orxonox attraktiver?
  • Server
Aktive Diskussion:
  1. Content Creation:
    Vision: Kooperation mit ZHDK
    braucht stabiles Game
    Mac werden benutzt! Orxonox muss auf Mac laufen.
    klarer Workflow (Wie modellen? Wie ins Spiel integrieren?)
    Wen spricht man an?
    Kommunikation
    Hat Potential. Interessant für ZHDK-Studenten.
    Wer nimmt die Aufgabe in die Hand? Evtl. Maxim
    Dieses Ziel als Tickets aufspalten
    Wieviel Vorbereitung braucht es?
    Andere Lösungen:
    Prozedurales Game Design, z.b. Skyboxgenerator
    Sprachsynthese
    Andere Open Source Weltraumspiele (Babylon 5 Mod) oder Contentquellen?
    Was für Quellen für solchen Content gibt es? Die Lizenz muss Debian-Compatibel sein. CC-BY, CC-BY-SA -> Felix und Beni wollen recherchieren.
  2. XML-Level-Editor:
    Erfahrung von Benjamin
    Level-Editor ist schwierig, eigenes Projekt für sich
    muss von Anfang an berücksichtigt werden – sonst muss man Content verwerfen.
    grosses Plus: bringt nicht Coder in die Community
    Unity3D
    Mögliche Umsetzung:
    Im Editor klöst nur die Aufgabe der Objektplazierung
    Probleme: Includes in XML via Lua
    Editor sollte ähnlich sein wie Blender, damit Studenten nicht neues System lernen müssen.
    Editor in Python als Blenderplugin? In Blender sieht es nie genau so wie im Spiel aus.
    Ticket: Level neu laden mit „F5“ -> spart Aufwand
  3. Code Qualität, Grösse, Alter, Komplexität :
    Hohe Fluktuation -> Wissen geht verloren
    Neue Studenten finden sich nicht zurecht. Was für Code-Lösungen gibt es schon?
    Selbst die Assistenten wissen es nicht
    Wie managen wir unser Wissen? -> Viel mehr Dokumentation!
    In der Praxis: Sowohl im Code dokumentieren, als auch Konzepte im Wiki haben.
    Wir haben wenig Doku – Wie kommt man mit diesem Zustand klar? Das ist die Relatität. Machen wir das Beste draus.
    Tutorials als Dokumentation?
  4. Attraktivität und Kommunikation:
    Es gibt genügend Studenten. Aber wie bleiben sie dran? Das zweite Mal ist besser.
    Artikel im Blitz
    Mit wenig Arbeit ein vorzeigbares Ziel erreichen. Z.B. gametype erstellen.
    Mehr Events
    Wie bringt man die PPS-Studenten und die Community ins Gespräch?
    Daily log: Jeder Student schreib kurzes Tagesfazit im Forum.
    Auswertung der Commit-Logs: Stetig abnehmend. Kick bei 2010.
    Es muss etwas laufen – man muss es sehen.
    Man braucht ein Ziel auf das man sich konzentriert.
    Einen Hackday/ Coding-Tag
    Man muss auch anspruchsvolle Aufgaben anbieten
Fail. Fail again. Fail better.

Post Reply

Who is online

Users browsing this forum: No registered users and 5 guests