The Long and Morbid Tale of Sigil Games Online: Interview Edition

[Engels]

Ya, I worry too. With an interview like that, Schild may get some hep job offer somewhere and abandon us.

[LK]

This is why I'm gonna visit f13 forevah and evah and evah.  Getting first-hand reporting on a situation like this is 100x better than what I can expect from sites that regurgitate the news.

Excellent job and big kudos to schild.

[Lietgardis]

No scripting language, though.  Say what?!?  Surely there was no scripting language for a sane, understandable reason?  Like "we won't need a scripting language because content creation will be handled by our revoutionary new drag-and-drop interface.  Or maybe it's like Eve, and the bulk of the game is written in something pretty close to a scripting language so a seperate one wasn't needed.  No way did you have to code in C++ to get foozle 1 to walk from A to B and say "numbnuts"?!?

You don't need a scripting language if you have an extensible, data-driven infrastructure.  It does require a commitment to developing that infrastructure in a smart, maintainable way, though, so it requires more foresight than "whatever, the designers will code what they need themselves."

[Ixxit]

One thing I find interesting is that Dave Gilbertson (the ex modeler cum VP of Sigil) who is singled out by the ex Sigil employee as being one of major factors in the massive management failure is being retained by Sony to assume pretty much the same role.

From Smed's statement:

Dave Gilbertson will be the person directly responsible for the day-to-day management of both the Sigil Carlsbad office as well as Vanguard"

[Frelorn]

That was one hell of a read. I really feel bad for those people who put in a ton of work, only to get let go like that. Hopefully they will all find something better now.

[Morat20]

No scripting language, though.  Say what?!?  Surely there was no scripting language for a sane, understandable reason?  Like "we won't need a scripting language because content creation will be handled by our revoutionary new drag-and-drop interface.  Or maybe it's like Eve, and the bulk of the game is written in something pretty close to a scripting language so a seperate one wasn't needed.  No way did you have to code in C++ to get foozle 1 to walk from A to B and say "numbnuts"?!?

You don't need a scripting language if you have an extensible, data-driven infrastructure.  It does require a commitment to developing that infrastructure in a smart, maintainable way, though, so it requires more foresight than "whatever, the designers will code what they need themselves."

Jeez I'm in the wrong field. You guys get all the cool toys. If it wasn't for the fact that I have a family to support, I'd probably start pushing my resume around Austin in the hopes of getting my grubby little mitts on some of them.

Simond:

For example. this might recast the Shadow's of Luclin launch for EQ in a whole new light - instead of Evil Smed as The Man, forcing the poor, misunderstood artistes of Verant to ship an incomplete product; maybe it was Smed as a desperate manager, trying to shepard Brad's demented Vision Cult into releasing something...anything even halfway playable.

Interesting thought. I've always been of the opinion that there's a seriously sick management and design process at SOE -- I never could understand how the EQ2 team could be fixing a seriously sick game in a very professional way, while the SWG team was -- in essence -- taking a patient with a bad case of the flu and shooting him repeadedly.

[Mr_PeaCH]

This thread is worthless without pics (of giant stone penises).

[Nija]

[Jerrith]

Just a quick post, I'm at work, building atm...

As a former Sigil employee (gone as of mid-March this year), I'd say there are some points in there I disagree with.  Some of that is one person's opinion, not necessarily fact.

As for scripting, which would you rather use, something that requires you to type:

giveExperience(1000);

Or something that lets you:

click on a drop down, choose giveExperience, and then be presented with 4 radio buttons for tiny, low, standard, high, and a text field for level (with +/- buttons).

That's not exactly what it's like in Vanguard, but it's the general idea. 

[darrenl]

Well done good sir....well done.

D

[Yegolev]

I like typing things, it takes much less time and effort than click-to-victory UIs.  Not everyone agrees with me, though.

[Lum]

Ideally, you have both. See: NWN1

[Ubiq]

I like typing things, it takes much less time and effort than click-to-victory UIs.  Not everyone agrees with me, though.

In the example quoted above, though, you have to root through code to balance a quest.  This is less than optimal, especially if (when) you have to go on a large-scale balance pass, and ends up being how exploits or buggy quests sneak into the game.

[Raph]

I can't believe that this turned into a script vs datatable discussion. :)

Ideally, you data-drive everything that is data-drivable. This means everythign that relies on central balancing, everything with common datafields, everything that plain old looks like an Excel spreadsheet.

Some things, like quests, will be reduced in cool factor by compressing them into that. And that's where scripts can shine -- allowing flexibility and creativity in new behaviors.

[Yegolev]

I can't believe that this turned into a script vs datatable discussion. :)

It's one of my lovable charms.  I don't do the same sort of work that you (collective rednames) do, and for the sort of thing Ubiq and Raph describe then it is always nice to cut out the element of human error whenever possible, plus it's easier to work with and whatnot.  I assume.  In my case, a predefined menu just gets in the way unless whoever wrote the UI already knows what could be broken... in which case she should have just written an automated fix so I could keep surfing the net.  Being able to get to the guts of the machine is important for my work.  If I had to spend my days making sure rangers were overpowered in a fantasy MOG, I figure I would want a GUI as well.

[Ixxit]

Just a quick post, I'm at work, building atm...

As a former Sigil employee (gone as of mid-March this year), I'd say there are some points in there I disagree with.  Some of that is one person's opinion, not necessarily fact.

Another thing is that he/she  name names, and some of the attacks are personal, a conevience that anonymity grants.  While I understant that the employee doesn't want to burn any bridges by identifying themselves I think perhaps it would have been much more professional to talk in generalities (insofar as the people's names) like Nino did, regardless of whether or not Nino still works for SOE.  Whether he/she is right or wrong there is a stench of sour grapes.

[EDIT]  PS just to be clear Schild, you did a great job with your questions to ellicit such candid responses.  I think you should be a member of the White House press corps. 

[Baldrake]

Can anyone who's actually in the game describe what's fun about it (if anything?) After six months of polishing, is this going to be something worth trying out?

[Margalis]

No scripting language is absurd. If you want a graphical tool, make it work by producing script!

One QA person? LOL. Holy shit. It's even worse than I imagined. And not playing WOW. When one game does 20x better than its nearest competitor it is probably worth at least checking out.

Good job with the interview. Actual journalism - amazing.

[Pendan]

[I think perhaps it would have been much more professional to talk in generalities (insofar as the people's names)

Yea the managers should not only get enough money to buy houses after their mismanagement but also be able to keep their reputation intact. They deserve to have their cake and eat it too after all.

[Kitsune]

Can anyone who's actually in the game describe what's fun about it (if anything?) After six months of polishing, is this going to be something worth trying out?

I played it on a free trial after the release, and it boils down to 3D diku like every other fantasy MMOG out there so if that's not your tea of choice, skip it.  What it had going for it were some interesting mechanics (the diplomacy battles), interesting takes on classes (the psionicist for the crowd control-ish caster), and a big world that looked kinda pretty in places.  It felt fairly soulless, however; generic_fantasy_world_04.  At no point did I particularly care about any of the NPCs, cities, gods, or much of anything else in the world.

[Tcharels]

Well, with all respect to my old commrade Jerrith, while this is only one person's point of view, it is a very accurate point of view. The anonymous poster who did this interview.. definately biased, but.. very accurate. Yes, you can have both.

And I won't get into the scripts vs. data driven discussion, as Raph pretty much hit all the appropriate points.

Cheers.

[Simond]

By the way, if anyone wants some light entertainment after this drama, the Vanguard pre-launch forums appear to be back up.

[Hazard]

Well, it was as I  expected. Man I hate to have right sometimes. Poor sods that got fired, been there and hates the empty sinking feeling. Guys if you read this, I realy liked the game but the total ignore from guys like Brad about the issues like hitching and stutters got me to un-subribe. If they can't admit there are problems, then they aren't likely to fix the problems.

[Nebu]

Outstanding interview. 

What strikes me funny is just how similar this whole situation is to the Ivory Tower of academia.  You have people that are great at generating ideas leading hoards of people while knowing nothing about implementation.  I'm saddened by the waste of sheer talent due to bad decision making and implementation skills above them, but realize that this is an all too common scenario.  Ideas are easy.  The conversion of ideas to concept and concrete outcome, well... that's the hard part. 

I'm optimistic that there are some talented people from the Sigil team that will move on to bigger and better things.  I just hope that they don't have to deal with too much emotional stress in the meantime.  So much wasted talent/potential here.   

[Endie]

I like typing things, it takes much less time and effort than click-to-victory UIs.  Not everyone agrees with me, though.

In the example quoted above, though, you have to root through code to balance a quest.  This is less than optimal, especially if (when) you have to go on a large-scale balance pass, and ends up being how exploits or buggy quests sneak into the game.

Yeah, but you shouldn't have to root through code for apparently off-the-cuff numbers because you shouldn't be using literals in every quest.  It's facile, but

giveExperience(1000);

in your example, should be:

giveExperience(L_SCALAR_CONST * iThisLevel );

Or similar: you get the point: now you can find the level easily, and change the rate of advancement of the whole shebang with one altered constant.  And changing that might require ctrl-o, choose quest file, ctrl-f, but that's certainly no harder than (to vastly simplify the data dictionary involved)

select EXP_SCALAR, EXP_LEVEL from QUEST_REWARDS, QUESTS where EXP_SCALAR.QUEST_ID=EXP_LEVEL.ID and QUEST.FRIENDLY_TITLE="Whacking The Foozle"

and no more awkward than opening the quests interface, selecting the quest, opening the rewads section, pulling down the drop-down for the completion-step, opening the stage etc...  And that's what i referred to anyway in my "perhaps they had the other tools" bit.  Works, but too simple-simon.

Scrpting ftw.  I add scripting modules to most stuff I write, now, simply because I have the knack, and the class design on my library, down pat.

Edit: changed data-typing

[Engels]

Well, with all respect to my old commrade Jerrith, while this is only one person's point of view, it is a very accurate point of view. The anonymous poster who did this interview.. definately biased, but.. very accurate. Yes, you can have both.

I take it you were with Sigil for a while, before the axe fell?

[bhodi]

I'm continually amazed that game companies routinely fight the same sorts of management challenges as everyone else providing service in a computer field. I guess that I always figured the vision or the passion of the people involved would bring them together and smooth over rough spots, because, from what I can tell, there is a lot more emotion involved in game design than any other company I can think of.

Since there's so much emphasis on the product as a love-child instead of a solution provider, I'm frankly surprised the internals were that bad. After so many obviously boneheaded decisions coupled with dismal benchmarks, I'm shocked that word didn't get forwarded to people equipped to make necessary changes. A working man down in the trenches has a lot more incentive to speak up and rail against such obviously flawed decisions such as a woefully underweight QA staff when they have more of a personal stake in the finished product.

I'd also agree that offhand comments about house purchases with stock when speaking in front of 104 people in a parking lot after being notified of layoffs wins the Michael Scott award of the year from me.

[Soulslinger]

You know I've started thinking about the whole "Small group prototype MMO project" thing that Jeff will supposedly be heading up and remembered something very interesting. While I was onboard at SOE, there were a couple of small projects being worked on, mostly POC stuff, but kinda cool. One was called Tapestry and was going to be more of a virtual community ala Second Life... but on steroids. The other was more of an enviromental engine for a "living world". It would rain and rivers would swell. Leaves would fall from trees and pile up, creeks became rivers which became lakes and everything was affected by the climate. Snow would amass in drifts according to the wind etc etc. The demo I saw included rain puddling and washing out the sides of a dirt mound and some snow effect. This was over 2 years ago but I would not be surprised to see the environmental engine resurface. In terms of immersability, a game sporting that kind of environment control has incredible potential, imo. Then again, it could be a dead project and Jeff could be working on Epic Weapons for EQ3.

[Margalis]

QA team is a good litmus test for managerial competence. Bad managers almost without fail refuse to see the value of QA because QA does not product immediately tangible benefits or produce any new features. In fact QA slows down the introduction of new features.

[Tcharels]

I was. I'm about a year removed from it, but I still remember it clearly and know well what things were like, and that the interview was about as accurate a read on the situation as you could get.

I tried writing more, but, even after all this time, I find it hard to not sound a bit bitter myself, so, I'll have to leave it at that. I wish all the best for to my former colleagues and I hope they all land on their feet and soon. They are, for the most part, a great group of people.

Cheers.

[Endie]

...The other was more of an enviromental engine for a "living world". It would rain and rivers would swell. Leaves would fall from trees and pile up, creeks became rivers which became lakes and everything was affected by the climate. Snow would amass in drifts according to the wind etc etc. The demo I saw included rain puddling and washing out the sides of a dirt mound and some snow effect....

Man, I'd pay cash to play even Vanguard if it had that sort of immersion.

Well, probably.  For a while.

[Morat20]

I'm continually amazed that game companies routinely fight the same sorts of management challenges as everyone else providing service in a computer field. I guess that I always figured the vision or the passion of the people involved would bring them together and smooth over rough spots, because, from what I can tell, there is a lot more emotion involved in game design than any other company I can think of.

I'm not. If anything, I would expect game industries to have worse management problems -- at least for anything over a five-man shop.

There are a lot of game designers who still hold to the old-school hacker style of development. That is, they think they can pretty much just eyeball it and make it all fit together through sheer talent and force of will and deep understanding of the vision.

Which is, to be blunt, bullshit.

Requirements documents aren't sexy. They don't feel like red-hot talent. Iterative development, artifacts, work breakdown structures -- they're not fun shit to do, and most people who got into the games design business got into it to design games, not fuck around with "milestones" and "use cases" and all that crap.

The problem is -- if you're working on a multi-million dollar project with dozens of people, you sort of need that crap. (Tangent: Developing MMORPGs, I'd focus on a really tight evolutionary model -- lots of iterations, early cycles devoted to prototyping key elements, heavy on early abstraction to allow layering of features, and a serious up-front investment in core architecture. You don't need to know every nook, cranny, ability, or class anytime soon. What you need is a tight focus on getting a working engine going that'll handle your projected load, tools that allow the customization you're going to want, and a basic framework that allows you to design the actual 'game'. Initial game design would be pretty abstract and very flexible.)

You need artifacts, use-cases, milestones, and actual software engineering. You need solid management -- and Sigil apparently lacked professional management.

They might have been kick-ass designers, but they weren't good managers -- and once you're fucking VP of this, or Chief of that -- you need to be a good goddamn manager or so hands off you're just doing PR.

The games industry is slowing embracing professional standards (they're about a decade or so behind the software industry as a whole) for development. The MMORPG buisness seems to be lagging even them. With Sigil -- they seemed really prone to the 'rock-star' fallacy.

They had something like 20+ designers, 10+ coders and god knows how many artists, DBAs, etc. I don't care how much of a fucking rock-star you are or how awesome the fucking vision -- you need all that goddamn boring paperwork to make sure those 30+ people are working together and actually creating, instead of constantly fucking themselves and everyone else over and wasting money.

[Endie]

Requirements documents aren't sexy. They don't feel like red-hot talent. Iterative development, artifacts, work breakdown structures -- they're not fun shit to do, and most people who got into the games design business got into it to design games, not fuck around with "milestones" and "use cases" and all that crap.

The problem is -- if you're working on a multi-million dollar project with dozens of people, you sort of need that crap.

Dead right: that's why my evenings are spent doing a postgrad course towards my masters - M883: Requirements Gathering in Technology Management - for my masters.  It's not what I enjoy doing, but I know I need to do it better.

[Mandrel]

The one QA member is rather amusing, considering Sigil had openings for game testers posted on their website for over a year now.  I guess they knew they needed more, but no one got around to actually hiring them.

Another thing I never see discussion about VG is the lack of storyline or lore.  I don't think there was anyone who was the main writer for the game, and that the lore, what little of it there was, was written by any employee in their "free time".  I'm guessing this is just another symptom of the mis-management.

[Morat20]

Dead right: that's why my evenings are spent doing a postgrad course towards my masters - M883: Requirements Gathering in Technology Management - for my masters.  It's not what I enjoy doing, but I know I need to do it better.

I just slogged through (managed a B, which frankly was generous given my total lack of interest and corresponding lack of effort) my last graduate level project management course -- it was interesting comparing what we were learning to the mature processes that run fairly transparently around me -- and I learned an awful lot that might be useful personally (mostly that I am capable of doing my manager's job, and even his manager's job -- I just really don't want to). But still -- god is it boring. It's not bad if you're just using it as a template for personal development, but if you're actually leading a team -- it's got to be on paper, thought out for everyone, constantly updated and it's a ton of sheer work and very little fun-time.

If you're doing the software management, you're not going to have time to sling the code or design the cool quests or in fact do anything but manage. It's a lot of hard work, and about the only perk is when it gets done right and on time and on budget -- you can take some comfort in the fact that all those red-hot talented designers and coders only got there because you did a shit ton of heavy lifting behind the scenes.

Project managers rarely get named as influential. They're not rock stars. But fuck if they're not the keystone of whether your game is shit or not.

Of course, I'm just biased. My Master's is in Computer Science -- software management courses are more of a token gesture for me. I certainly don't want to do that heavy lifting. :)