Green Monster Games - Curt Schilling

F

Flight

Molten Core Raider
1,229
285
Cross posting from PotBS thread :

Wilfan said:
Good post.

Every big company gets this structure. Some argue that large companies can"t survive long-term without it. I"m sure that ex-Blizzard people would say the same thing about current Blizzard.
Click to expand...
The thing you can see with Blizzard is they are trying to develop methodologies, unlike the rest of the industry. Apart from anything else, you can tell that from their various staffing levels throughout their projects.


My opinion is that the industry needs to learn that whole new methodologies need developing and that the skill sets to do that does not exist within the industry.




Just the basic precepts are obviously not happening :

i) the old axiom - get it right first time; most especially at project inception and early definition stages; spend twice as much time now, save five times as much time later fixing it, plus have a solid product;

ii) specced with 10 years worth of resilience, reliability and scalability built in, with future xpacs in mind;

iii) project checks and controls across the board, for every level of employee;

iv) documentation - for example, split all software development into blocks of various sizes; make every block of code annotated and documented so that a new designer can come straight in and understand exactly what its doing and work with it. Possible to treat the whole development as a tree of blocks and have control over it from start to finish.



There is no quick fix, no way around the need for the development of methodologies for MMO development. It doesn"t exist.


38 Studios are in a truly unique position at the moment. Excellence throughout the company. They have the opportunity to make this happen. I hope they have the courage to learn to say, actually, maybe there is a better way to do things that we haven"t considered.

They should start work on this methodology; they should see it as a long term program. Create 10 or 20 Scholarships for University places based around developing this methodology - places that address every part of the company, to work alongside their company. Investing $500,000 in a project like this could be priceless.
 
D

Druixx_foh

shitlord
0
0
Flight said:
Cross posting from PotBS thread :



The thing you can see with Blizzard is they are trying to develop methodologies, unlike the rest of the industry. Apart from anything else, you can tell that from their various staffing levels throughout their projects.


My opinion is that the industry needs to learn that whole new methodologies need developing and that the skill sets to do that does not exist within the industry.




Just the basic precepts are obviously not happening :

i) the old axiom - get it right first time; most especially at project inception and early definition stages; spend twice as much time now, save five times as much time later fixing it, plus have a solid product;

ii) specced with 10 years worth of resilience, reliability and scalability built in, with future xpacs in mind;

iii) project checks and controls across the board, for every level of employee;

iv) documentation - for example, split all software development into blocks of various sizes; make every block of code annotated and documented so that a new designer can come straight in and understand exactly what its doing and work with it. Possible to treat the whole development as a tree of blocks and have control over it from start to finish.



There is no quick fix, no way around the need for the development of methodologies for MMO development. It doesn"t exist.


38 Studios are in a truly unique position at the moment. Excellence throughout the company. They have the opportunity to make this happen. I hope they have the courage to learn to say, actually, maybe there is a better way to do things that we haven"t considered.

They should start work on this methodology; they should see it as a long term program. Create 10 or 20 Scholarships for University places based around developing this methodology - places that address every part of the company, to work alongside their company. Investing $500,000 in a project like this could be priceless.
Click to expand...
This man needs to create a game. I believe a company founded on these principles, even with very marginal game IP, could eventaully go head to head with Blizzard. Preach on, brother.
 
W

Wilfan_foh

shitlord
0
0
Moved from Tabula Rasa thread:

Flight said:
I"m pretty much done eulogising about this. People who aren"t in MMO companies can see the problem - they see it in one crap product after another. People inside the dev companies all say, hey this is how it is, we have the skills and people, just keep throwing money at it.
Click to expand...
This is common to the whole software field, from the small, departmental Java/VB applications to boxed software products to multi-million dollar enterprise software packages.

But let"s take a look at your proposals. They"re not new and were advocated and tried plenty of times.

i) the old axiom - get it right first time; most especially at project inception and early definition stages

Problem is, you don"t know what you"re doing at project inception. You have a vague fuzzy picture in your mind. In this "vision" (har, har, har), everything works fine, until you start implementing it. Then nasty details you never knew about and whole areas you didn"t examine in detail wreck your plan.

Now, you can spend more effort and time elaborating your vision. But you can"t get out of the fuzzy realm if you don"t implement and test. That"s why Brooks (and others) advocates throwaway prototypes (beancounters tend to force you to promote your prototype into finished product, so "throwawayness" is crucial).

Slightly on-topic: while TR threw away their fantasy-themed version 1, they didn"t learn much from it. They changed to sci-fi setting and gun combat (and who knows what else) and whatever they learned before didn"t apply anymore. Plus, I"m pretty sure they threw away only art assets and basically reskinned the game while keeping the engine intact.

ii) specced with 10 years worth of resilience, reliability and scalability built in, with future xpacs in mind;
Click to expand...
How can you predict 10 years in advance? And if you can, why didn"t you implement all those things already? Blizz would be very happy if they were to know that 25 man raids are the future back in the friends-and-family beta...
And how do you actually "spec" that, apart from putting "This MMO must have with 10 years worth of resilience, reliability and scalability built in, with future xpacs in mind" in the design docs? This is just passing the bucket.

iii) project checks and controls across the board, for every level of employee;
Click to expand...
Elaborate, please. Milestones, bug/issue trackers, statistics applied to testing/QA and timesheets are widespread. So are micromanaging bosses, most of which don"t know one bit what they are actually managing.

iv) documentation - for example, split all software development into blocks of various sizes; make every block of code annotated and documented so that a new designer can come straight in and understand exactly what its doing and work with it. Possible to treat the whole development as a tree of blocks and have control over it from start to finish.
Click to expand...
Functional, structural or object-oriented decomposition is nothing new. The problem is, your last statement does not follow from it. You"ll inevitably decompose incorrectly to get full control and you"ll often find out you need several decompositions to cover all aspects you want to "control". And you"ll still have to build the big picture from the smaller blocks - either you, the reader, or the poor writer each time something changes.

Then there"s the thorny issue of documentation. You"ll either have to lose most details or your documentation will be as incomprehensible as the code. Detailed documentation that doesn"t (directly) describe the code is more reasonable goal, but different aspects of software (such as game mechanic, performance, installation, library user guide etc.), though it"s a bitch to keep in sync. Just technical writers won"t do, you"ll have to involve the implementors in it.

I"ll skip the issue of useless documentation as we all know it too well (an example:ALBPM Process API- this is what ~$50k/CPU buys you).
 
N

Noah EQ2_foh

shitlord
0
0
TBH Most of the items mentioned by Flight in his posts seem to be the "basics". Nothing revolutionary or ground breaking by any means. I"m willing to bet most companies designing games already try to plan for these things.

Words on a page are just that, words. Most people that post seem to act like how the IBM commercials portray Corporate Culture with "buzz words" and "innovation meetings" but at the end of the day.... no one has done anything.

Wanna be impressive? Post a rough example of HOW you would put in 10 years of scalability. What system would you use specifically to ensure "getting it right".

I personally am tired of general bland posts that the only thing changed in them from someone"s post in 03 is verbage.

It is a lot easier to give feedback than actually impliment a successful change for the better. Most don"t have the balls to stick their necks out like that.
 
OneofOne

OneofOne

Silver Baronet of the Realm
6,628
8,104
After the first few novel-sized posts I just assumed he was looking for a job and trying to impress Curt.
 
S

Sunnyd_foh

shitlord
0
0
OneofOne said:
After the first few novel-sized posts I just assumed he was looking for a job and trying to impress Curt.
Click to expand...
Glad I wasn"t the only one! Thought perhaps I was getting cantankerous in my old age.
 
F

Flight

Molten Core Raider
1,229
285
OneofOne said:
After the first few novel-sized posts I just assumed he was looking for a job and trying to impress Curt.
Click to expand...
I"ve made clear I live in the UK, but there are a number of people who have been involved in this thread who Curt should consider giving interviews to; if Wilfan works in the areas he is talking about he is certainly one of them. Actually, I live in Liverpool - after a year of involvement with Tom Hicks, as a sports fan, I"m not sure I ever want to work with anyone else from the Baseball world



Noah EQ2 said:
TBH Most of the items mentioned by Flight in his posts seem to be the "basics". Nothing revolutionary or ground breaking by any means. I"m willing to bet most companies designing games already try to plan for these things.
Click to expand...
Noah, I agree absolutely that we are talking about getting the basics right. It absolutely is the basics. Its a fact that in an industry that has evolved the way the MMO industry has its a massive step to get them to do the foundations right. It needs re-inventing. It demands culture change.

You"ve got a well documented history of playing Sony games; there are pictures all over the internet of you with various of their devs, so its understandable you repeatedly stick up for them.


How many of the basics have Sony got right in the last decade ?


i) They put out an EQ1 expansion that was designed around a level cap increase - but pulled the increase until the next expansion. They left the content balanced for the level cap increase. A whole expansion destroyed due to one poor, massive decision.

ii) They churned out half completed expansions every six months, with no thought for the state of their game or their long term reputation - it was all about a fast buck.

iii) They released SWG which was totally dependant on player made armor and weapons, with a totally broken armor system.

iv) They not only introduced the NGE to SWG, they did it a few days after releasing an expansion for the game. They let the customer base buy the expansion, before they told them about the massive changes they were making to the game.

v) EQ2 is an example of how to do exactly the opposite of everything we are discussing. Curt made a post knocking one of the VG devs for later talking about the mess the company was in. He said the guy should have had the guts to face up to the senior management and tell them the state the game was in.

The original EQ2 dev team did this, because the game was in such a state, half way through development. Sony sacked / replaced / reallocated them. The entire initial development of EQ2 was a total mess. Even the devs are on record as saying it was so poor at release they never played it. It has never recovered. They have gone from one expansion to the next, not knowing if its a group game, a solo game, a raid game. Zero continuity.

Now they"ve released RoK which totally misses the group content - its solo then raid. They threw in a few dozen AAs for everyone, at the last minute, which was a huge bonus to some and next to useless for others, such as those where each AA tree depends on having a different weapon type equipped. Its constantly reacting and tacking things on, instead of having planned ahead.

I disagree that they plan ahead, anywhere near enough.



I"ve given specifics on what I see are the problems and how to fix them


PROBLEMS

1. Lack of skills within the industry

2. Managers are too insecure to admit their lack of knowledge and ability

3. Time scales are driven by techies - software and hardware techies are both the same type of creature - they want to play with the latest shiny toys and with the latest hardware and software capabilities. They will do this to the nth degree if they are allowed, in totally inefficient fashion.

4. The techs and creative folk don"t want anything to change. They want to be left alone to be creative, control time scales and play with shiny new gear.


These are just the initial, high level problems. They lead on and snowball to lots of others as companies develop.



SOLUTIONS

There are no easy quick fixes. I foresee a 5 year time scale for the development and growth of a brand new company, which is on top of every aspect of the development of MMOs".

1. Initially recruit from outside the industry in areas where more can be learned - particularly QA. Use the people to pass on their skill sets and grow the staff around them.

2. Education, education, education. Top to bottom. Massive investment in staff personal development. Managers have to know the areas they are supposed to be managing, so they understand the work their staff are doing and can"t be bullshitted.

3. Scholarships. Recruit people with a 5 year development plan, to include a degree, as necessary. 10-20 places. Being totally serious, recruit a large number of those people from this forum.

Recruit managers and QA folk and sponsor them through a software development or game development Degree or Masters Degree. Recruit software people and have a small number of them do QA or Managerial qualifications, as necessary.


For a small investment of $500k - $1mill you could have a spine of people prepared to work in this industry in a way hardly anyone at all is at this moment. This industry is going to grow exponentially over the next few years, particularly with emerging markets in the Middle and Far East, into a multi billion dollar industry. The few companies who survive and do anything of note are going to have to do things radically differently to the way they are done now.
 
G

gnomad_foh

shitlord
0
0
Flight said:
***massive SNIP of good information***

PROBLEMS

1. Lack of skills within the industry

2. Managers are too insecure to admit their lack of knowledge and ability

3. Time scales are driven by techies - software and hardware techies are both the same type of creature - they want to play with the latest shiny toys and with the latest hardware and software capabilities. They will do this to the nth degree if they are allowed, in totally inefficient fashion.

4. The techs and creative folk don"t want anything to change. They want to be left alone to be creative, control time scales and play with shiny new gear.


These are just the initial, high level problems. They lead on and snowball to lots of others as companies develop.



SOLUTIONS

There are no easy quick fixes. I foresee a 5 year time scale for the development and growth of a brand new company, which is on top of every aspect of the development of MMOs".

1. Initially recruit from outside the industry in areas where more can be learned - particularly QA. Use the people to pass on their skill sets and grow the staff around them.

2. Education, education, education. Top to bottom. Massive investment in staff personal development. Managers have to know the areas they are supposed to be managing, so they understand the work their staff are doing and can"t be bullshitted.

3. Scholarships. Recruit people with a 5 year development plan, to include a degree, as necessary. 10-20 places. Being totally serious, recruit a large number of those people from this forum.

Recruit managers and QA folk and sponsor them through a software development or game development Degree or Masters Degree. Recruit software people and have a small number of them do QA or Managerial qualifications, as necessary.


For a small investment of $500k - $1mill you could have a spine of people prepared to work in this industry in a way hardly anyone at all is at this moment. This industry is going to grow exponentially over the next few years, particularly with emerging markets in the Middle and Far East, into a multi billion dollar industry. The few companies who survive and do anything of note are going to have to do things radically differently to the way they are done now.
Click to expand...
Whole heartedly agree with you Flight.

You have nailed it quite well and this is speaking from a background of over 30 years in the nuclear industry doing Nuclear QA. There is a radical difference in the 2 fields but the ideas and methods stay the same (course if we screw up then we create radioactive wastelands for 30,000 years )

The software industry especially the MMORPG genre is completly inbred and myopic. Of course this occurs across many fields of endeavor and is not really limited to the genre.

I knew 2 NASA engineers that used to have a computer company here in Virginia that was one of the first Apple dealers in the Tidewater area. Well they have been out of business for over 20 years since they were techno experts but knew nothing about business. As long as most gaming companies keep pursuing the current type of development with the new shiney toys and tech taking precedence they will continue to fail.

If you want a gaming software business to succeed you need business savy people running it not a bunch of techno geeks and gamers. People can say what they want about the "suits" but it has worked for Blizz, EA, and a few others. The rest are flash in the pan one night stands never to be heard from again.
 
N

Ngruk_foh

shitlord
0
0
gnomad said:
If you want a gaming software business to succeed you need business savy people running it not a bunch of techno geeks and gamers. People can say what they want about the "suits" but it has worked for Blizz, EA, and a few others. The rest are flash in the pan one night stands never to be heard from again.
Click to expand...
You bring in the best people in the world to run the business.

Those people shield the developers from every possible distraction, create a truly incredible environment, allow them to make the game.

Allow them to do what you pay them to do, while creating passion and energy in the team for the products, and each other.

Understanding that nothing is more important than what they do. None of it matters if they don"t get it right, every time. Screw ups are fine, but progress from the screw ups is essential.

You also never allow the PR/Marketing aspects of your company or product to drive development, budget, timelines.
 
Z

Zarcath

Silver Squire
96
54
gnomad said:
If you want a gaming software business to succeed you need business savy people running it not a bunch of techno geeks and gamers. People can say what they want about the "suits" but it has worked for Blizz, EA, and a few others. The rest are flash in the pan one night stands never to be heard from again.
Click to expand...
I only attended GDC once, and it was after being in the industry. I didn"t like it.

Groups basically broke down like this.

A) PR/Marketing people
B) Actual Dev
C) Spectators

Real devs kept to themselves. They"re nerds, they have bad social skills, they"re innately distrustful of others and are unwilling to talk about anything beyond idle chit chat. I was trailing behind a group of devs being led around by a suit, and the devs were whining about "meeting people" and "do they really have to go?", and the suit was saying "YES, make an effort to talk to people".

It"s understandable why devs aren"t willing to be more friendly with people outside of their company. It"s very rare that a dev can speak openly or have an opinion without fearing for their job. I think that creates a bad environment for fresh ideas or positive feedback. Bad ideas will fester internally without any outside opinion (NGE, VG, TR).

I like that Curt and others from the company are free to come here just to shoot the shit. More companies should make that effort. Brad only showed up to spin his newest vision and Smedley rarely ventures here unless it"s for a PR reason. It"s important to establish an open and respectful relationship with the customer base, even if we"re not the target market, it builds a positive image for the company and the people that work there.
 
F

Fog_foh

shitlord
0
0
Zarcath said:
It"s understandable why devs aren"t willing to be more friendly with people outside of their company. It"s very rare that a dev can speak openly or have an opinion without fearing for their job.
Click to expand...
This seems like an extremely silly blanket statement, unless you specify a company. Surely you don"t think that every game development shop is some faceless, regimented corporate monolith?

(Disclaimer: I know nothing about the gaming industry but something about the software industry. The majority of companies I"ve seen are more open than that.)
 
Z

Zarcath

Silver Squire
96
54
Fog said:
This seems like an extremely silly blanket statement, unless you specify a company. Surely you don"t think that every game development shop is some faceless, regimented corporate monolith?

(Disclaimer: I know nothing about the gaming industry but something about the software industry. The majority of companies I"ve seen are more open than that.)
Click to expand...
Rather silly question to ask, given what we"ve seen on these boards.

Due to the nature of the industry, in general, devs don"t like to bad mouth other devs because the industry is largely made up of the same people. If you ever want to work in the industry again you don"t up and out call people idiots even if it"s warranted. And the available market for experienced devs is (was) somewhat small, and devs don"t necessarily like to help other devs out in fear that they"ll soon be replaced by the very person they"re helping. Also when a dev speaks it"s perceived as gospel to the public and PR has to say they spoke out of turn.

In short, MMO companies treat their employees as liabilities.

I don"t know how the work environment is fairing now, after the EA fiasco and scuttlebutt over unions. But before, devs would constantly worry about job security.
 
F

Flight

Molten Core Raider
1,229
285
Wilfan said:
Fantastic post
Click to expand...
I have to start Wilfan by saying I love this post. Though I might appear to be "disagreeing" with you, I"m not. We share a lot of common ground in our beliefs and I see this as brainstorming with you, with us approaching the same issues from different angles.

Lets remember, though, that I am talking specifically about the MMO industry. Yes there will be common ground with other large software companies, but there are many differences.


Wilfan said:
But let"s take a look at your proposals. They"re not new and were advocated and tried plenty of times.

i) the old axiom - get it right first time; most especially at project inception and early definition stages
Click to expand...
I totally disagree. I don"t believe any of the high profile MMOs put in nearly enough time ahead of "getting stuck in". In every sense, the industry needs to learn that the more time you put into planning, the more time you save exponentially later - most particularly at inception. The evidence of that is rampant. And not just into planning requirements, but into contingency planning.

There is not nearly enough "whatif ?, what if?, what if?" and too much, "that sounds cool, lets get the staff working on it straight away". "What if" should be constantly on the mind of the people driving the project.


Wilfan said:
Problem is, you don"t know what you"re doing at project inception.

Now, you can spend more effort and time elaborating your vision. But you can"t get out of the fuzzy realm if you don"t implement and test. That"s why Brooks (and others) advocates throwaway prototypes (beancounters tend to force you to promote your prototype into finished product, so "throwawayness" is crucial).
Click to expand...
I agree with you on throwaway and prototypes. When I talk about getting it right first time I don"t mean you will methodically work things through and get what you want first time you try it. I mean don"t allow mistakes or, for example, lapses in documentation to go through when you implement something, which will entail spending 5 times as much effort later to work through where the problem is and fix it, than it would to do it properly.


Wilfan said:
How can you predict 10 years in advance? And if you can, why didn"t you implement all those things already? Blizz would be very happy if they were to know that 25 man raids are the future back in the friends-and-family beta...
And how do you actually "spec" that, apart from putting "This MMO must have with 10 years worth of resilience, reliability and scalability built in, with future xpacs in mind" in the design docs? This is just passing the bucket.
Click to expand...
You don"t have to make precise design decisions 10 years in advance, but the systems you introduce should certainly have 10 years of scalability built into them.


Lets look at three examples of where this didn"t happen and ask ourselves if this is acceptable.

9 years ago, with the introduction of EQ, Sony made one of the biggest mistakes in MMO history with the introduction of one spell.

All together - Complete Heal.

How much forward planning there ? Zero. I will give you this - it turned into something they didn"t envision. They didn"t plan for expansions or further success. We"ll let that go, because it was so long ago - as long as they learned their lesson.

Uh-oh, 9 years later and guess what. They are still making the same mistakes over and over. In one expansion, RoK, we see numerous examples. We have had the afore mentioned AA increase. Zero forward thinking when they were introduced and when they increased the number of them.

An even better example - the level 70/80 arrow debacle. The damage system innate in the bow/arrow system does not scale past 70. So even at level 80, the best damage bows are level 70, rather than the ones introduced in the new lvl 80 expansion. (Is this fixed yet ?)



Every design decision should be coded with 10 years of scalability built in. If a fireball does 10 damage at lvl 1 and 50 damage at lvl 50, then the code should allow it to have the potential to do much more.

If an AA system is introduced, with a planned initial allocation of 50 AA, the system should be designed and coded so it still makes sense if it is expanded to more.

Mudflation is not going away. It needs planning for. This is what I am talking about when I say ten years of scalability.


Wilfan said:
Elaborate, please. Milestones, bug/issue trackers, statistics applied to testing/QA and timesheets are widespread. So are micromanaging bosses, most of which don"t know one bit what they are actually managing.

Functional, structural or object-oriented decomposition is nothing new....

Then there"s the thorny issue of documentation.....
Click to expand...
I want to convey something here, triple underlined. I"m not for coming down on an organization, crushing everyone with endless bureaucracy. My entire management style and theories are based upon looking after staff, fulfilling them in their work and making them totally happy in their work and their environment. Thats how I believe you get the best out of people individually and collectively.

Lets consider implementation, not brainstorming or theory work, when it comes to a devs workload. Devs and creative people need directing and they need boundaries. To allow that to happen in a reasonable fashion, the people who are responsible for allocating work loads and schedules need to have a good understanding of the work in question. I would suggest there is an extremely small number of those people and the ones who do exist could benefit massively from personal development.


This feeds into documentation and design methodologies. This is going to be a lengthy process. The Scholarships, personal development and education I have suggested would all feed developing the software development methodologies. The development of the documentation is going to fail, unless you have the managers I have outlined, who understand, can work with, can communicate with both the "techies" and the "suits".


On software methodology, the MMO industry needs something wholly new, which will borrow from existing theories and principles, but would need tailoring to both the intial development and the longevity and scalability. Personally, my pet theory is, to use a bad analogy, like a software version of the OSI 7 layer model. using a hierarchy of building blocks, with standards for each layer, to include how they plug into the next layer.



Wilfan, work with me here. Dispute where I"ve got it wrong, but consider where I am hitting the spot, build on it and make suggestions.
As I"ve said, the problem is that the industry is driven by a relatively small number of individuals who are not the best they can be. The entire industry is under-skilled, under-managed and out of control. The tools do not exist to provide control, even if sufficient skills are introduced. I"m suggesting developing those tools and introducing the skills. How do we make it happen ?
 
F

Flight

Molten Core Raider
1,229
285
gnomad said:
Whole heartedly agree with you Flight.

You have nailed it quite well and this is speaking from a background of over 30 years in the nuclear industry doing Nuclear QA. There is a radical difference in the 2 fields but the ideas and methods stay the same (course if we screw up then we create radioactive wastelands for 30,000 years )
Click to expand...
Any suggestions on how things could be improved in the MMO industry, Gnomad ?
 
B

Bellstian_foh

shitlord
0
0
I"ve always wondered if it"s possible to have a truly scalable MMO engine. Graphics that look like WoW on low settings with high performance, and a high setting like EQ2 or VG. I know nothing about how game engines work...but is it feasible?
 
You must log in or register to reply here.