Axel Kurki
Aseyakone
1
   |
Posted - 2013.01.17 12:31:00 -
[1] - Quote
On the "(feature X) wanted for (user group Y)" discussion: I kind of agree with Malcanis's thoughts (mentioned elsewhere) that any change designed to cater to a specific user group (e.g. newbies, high sec mission runners, etc.) will most likely benefit the most experienced user group the most, more than the intended user group. So one thing to ensure is to aim the features benefit as large a portion of EVE players as possible.
But, to my original purpose of writing this post, which is beating the dead, then undead, possibly repeatedly exorcised and raised horse that is the POS. It seems to me that there is no clear design goal with POS/starbase systems, and the current system is a collection of huge kludges no one really understands or wants to touch. Therefore, it is extremely hard to refactor and redesign the system, which leads into developers concentrating on developing excuses to keep from getting sticked to the steaming pile of manure ("we need to wait for art" / "art is expensive and cannot be done when the design is not finished"). Therefore, in order to get a replacement, there needs to be a clear road map on how to develop the replacement in small, bite-sized bits.
A possible step one would be establishing personal mobile habitation units - character-level structures. These would be much more limited than current corporate-level structures (i.e. POS/starbase), but would be designed from the ground up to be modular, for example some kind of a "lego brick" system or otherwise connected (see "building spaceships in Kerbal Space Program", replace fuel tanks with powergrid generating modules). The limitations mean that a personal starbase isn't as efficient as the POS (for example, slightly more expensive to run per PG/CPU unit produced) and its limited support for modules would mean that it isn't a competitor for a POS, but it would have its niche as a "one-man POS" for more casual players and newbies to get a taste for the system. Once the system could replicate all necessary POS functions, it could then be scaled up to corporate level and finally replace the POS system (assumed to be unmaintainable but necessary) with similar corporate-level structures.
There would be savings in making art assets and development time by starting with a subset of features. If the past is of any indicator, the initial step should be small so that there is no repeat of things like planetary interaction's vision of Civilization/SimCity and delivering FarmVille and then having the organization burned out and abandoning the entire feature. Rather plan small, doable and extendable, and commit to expand it at a sustainable pace. This needs to run parallel at all the necessary processes.
Example step one would be a central module which could be expanded by very basic modules (basic storage, defense and industry, such as an adapter version of a container, a turret and a single type of manufacturing module). This could be already very useful accross the board . If mobile enough, could be used as a "ninja" base in low/nosec ops. Regardless of system sec level, it might be used to extend to, say, a temporary single-slot laboratory. Game design provides the balancing. Programming provides the refactored (or new) back end. Art provides the design. Then, step two would add more module types, game design makes sure the design is expandable (and preferably, not exploitable), and programming expands the back end (and fixes exploits/bugs). Art expands upon the previous design to provide art for the new modules / effects. On the side there might be even room to expand the features of the system (say, design an access system). Once the system can replicate everything a POS can do, then would be the time to use the ready back end (which was designed to take into account this step, right?) to make art & balance corporation level structures. Which then would, finally, replace the POS.
Of course, it might not be very cost efficient to maintain parallel systems, but this is the reason why the "personal level" and "corporate level" structures require separate niches or at least different levels on the same scale. POS is seen as a necessary evil right now, but there is no reason why it couldn't be the fortunate good. If it can't be replaced in one fell swoop, then it would seem that the only alternative is to iteratively develop a parallel system until it can replace the old code. |