A Little Sidestep

Ooh I love to dance a little sidestep, now they see me now they don’t
I’ve come and gone and, ooh I love to sweep around the wide step
Cut a little swath and lead the people on
— The Best Little Whorehouse in Texas

I don't believe that the devs at CCP Games intend to mislead the EVE Online player community deliberately, but I'm sort of feeling that way after reading CCP Greyscale's recent post in the forums:

The massive overhaul to industry coming with the Crius expansion on July 22nd eliminates a limited number of slots in manufacturing and research facilities, changing instead to a scaling cost system - the more activity in a system, the higher the job cost. That means that players will no longer have to wait for their industry jobs to start, but they may pay a higher price for that availability.

Industry conducted in structures installed in a Player-Owned Starbase (POS) will receive bonuses for efficiency, providing an incentive to move research and manufacturing there, instead of using facilities in non-player character (NPC) stations. Previously, players installed multiple structures in their POS in order to have enough slots for research or manufacturing operations, but with the elimination of slots in Crius, there would be no reason to do this. In order to provide a suitable incentive to install multiple structures of the same type in a POS, CCP Ytterbium revealed in a dev blog published just last week that additional efficiency bonuses would be provided for doing so.

These additional efficiencies for stacking identical structures in a POS led me to the conclusion that industrialists would configure their operations around highly specialized production facilities, in order to maximize their profits. I anticipated that we would see a lot of large POS towers each producing single items at the lowest possible cost, especially in high sec.

But now, by suddenly reversing course and eliminating efficiency bonuses for stacking POS structures, CCP Greyscale has largely removed any advantage for industrial specialization. In fact, there is no apparent incentive to have more than one type of lab or assembly array in any industrial POS configuration.

Therefore, I anticipate that most players will choose instead to install the smallest POS tower they can, after Crius is released, to save fuel costs. The incentive to install a large tower no longer exists, if a smaller tower can accommodate all the different structures a player may need, while operating at one-fourth the fuel requirements. I expect to see a lot of small POS towers, fit as "general purpose" industry facilities, going up in high-sec space - and this is exactly what I plan to do on July 22nd.

POS Code: the Bane of CCP Games' Existence

To be fair to CCP Greyscale and the other developers at CCP Games who are working on the Crius industry revamp, the decision to reverse course on this design element appears unavoidable. CCP Games devs have asserted repeatedly that the current programming code for POS functions is extremely complex and difficult to modify. Under the existing code, at least an hour would need to pass after adding a new structure to a POS configuration before any adjustment in bonuses could take effect - too long a delay to satisfy CCP Greyscale and the Crius development team. According to a follow-up post in the forums, CCP Greyscale says:

Changes only taking effect on a long, silent cycle is a microcosm of everything that’s wrong with starbases, and we no longer operate in a mindset where that sort of functionality is considered acceptable.

The one-hour tick also poses major problems to any straightforward attempts to get around online/offline shenanigans. Yes, it’s relatively straightforward to specify solutions, but in practical terms it did not look likely that we were going to be able to fix this prior to release.

It’s also an additional source of complexity and confusion in a system that’s already got a lot of moving parts.
— EVE forums - Crius starbase feedback discussion

The labyrinthine nature of the current POS code presents a huge barrier to further game mechanics development for structures, and for this reason, it is on CCP Seagull's path for the future vision of EVE Online, which was revealed at Fanfest in May.

CCP Seagull revealed plans to tackle the thorny problem of legacy POS code at Fanfest 2014 in May.

CCP Seagull revealed plans to tackle the thorny problem of legacy POS code at Fanfest 2014 in May.

Until POS functionality is completely rewritten, CCP Games devs will always have to make compromises in their design decisions for structures. This latest change is only one example of the constraining effect of working with the old legacy code.

Raising an Eyebrow

Still, the late decision to remove this already announced feature puzzles me. All the industry changes were originally slated to be unveiled in the Kronos expansion, released June 3rd, but they were all delayed until the Crius expansion, to give the devs another six weeks to refine and perfect the mechanics. At the time, I applauded this choice. It's always better to release features that work, even if it means a slight delay, than to release broken mechanics and have to fix them on the fly, after the fact.

But this sudden zigzag on a key element, previously and publicly described in detail, chips away at my confidence that the Crius industry code is as well-considered as it should be. It makes me wonder: what other elements of Crius are at risk? Will we see any additional changes to the previously announced new functionality or mechanics, before July 22nd?

EVE Online players are generally paranoid - it's a required characteristic for surviving in New Eden. These kind of sudden reversals only serve to raise eyebrows in the player community, and increase doubts about the successful implementation of new features.

I hope my worries are unfounded. We'll find out soon enough.

Fly safe! o7