Vehicular Combat

John McLear (Hector Busamante)

Over the past week, while traveling in NYC, I took a chance to revise the Gear file based on some work another player had done (he turned some Traits--general super-powers--into "gear" and had some questions). The chapter has been revised and streamlined--the general intent made more clear, illustrated with a couple more examples, and rationalized across multiple pages a bit better.

There is still a "gap" in the JAGS rules here: Vehicular Combat.

Vehicular Combat: What We Know
We have tried before to write a cohesive vehicular combat rules-set. Here's what we know right now:

1. Vehicles have a DRIVE Score. Most vehicles have a set of DP (a small amount) set aside for something called DRIVE (it was called CORE) in the original rules. If you "shoot this out" the vehicle stops. This is necessary because a car has a few hundred DP based on weight but a single shot from a handgun can stop it (and we know that, for example, destroying the tires is a lot easier than crushing the whole car). Hitting Drive usually requires a called shot that hits by 4+. This makes vehicles fairly easy to stop (if unarmored) by decent shots.
2. Vehicles provide Cover. If you declare you are shooting at someone in a vehicle they get some cover. This is generally 3pts for a standard car and maybe 2pts for a motorcycle (and maybe 4pts for someone in a car "crouched down and hiding." These basic cover rules assume that even if your shot penetrates the car easily--which, our research, suggests will happen for any civilian vehicle and nearly any firearm--there are complexities around hitting targets you can't see well that provide defense.
3. A moving vehicle provides a set of to-hit modifier negatives to both shooters in it and attackers. Generally firing from a non-mounted gun gives a -2 in addition to some other speed modifiers. I think the general speed modifiers for most vehicular combat will clock in around -3 / -4. For L2 attackers a "drive by" is pretty random.

What Else Do We Know?

Here are some things I think we know.

• A vehicle moves based on its speed--not the driver's Initiative. Faster vehicles move first (higher Init) and every vehicle must move before you can check for collision. As an Advanced Rule, a vehicle can move "in reaction" to another vehicle's movement even if it is slower--that is, if I move my vehicle counter on top of yours and say "my PC jumps from my car to yours!" you can say, "Not so fast, I take my reaction move to push my vehicle ahead--see, there's still space." If my vehicle is faster, which it must be, since I moved first, I will, by the end of this, still have closed the space-gap--but not, yet--all the way.
• Having a vehicle move basically straight and not accelerate costs the driver 0 REA. However, any kind of combat movement, acceleration, turning, or deceleration, costs the driver 5 REA. If this is not paid (the driver is Stunned, Dazed, Unconscious, or uses their REA for other things) the vehicle is "out of control" and may crash. We do not have concrete rules for this--but a skill roll is probably involved with some significant negatives based on situation.
• A Vehicular Dodge is allowed to stop an attack that hits DRIVE. This is the primary purpose of Combat Vehicle Operations skill. It is all important: if your Space-Wing Fighter Jock gets shot by the enemy fighter craft, (a) the enemy is almost ALWAYS shooing at DRIVE (your engines) (b) their shots are either 20mm cannons or "space 20mm cannons." They do a lot of damage. If you are hit, your vehicle will EXPLODE and you will probably DIE. You have a large battery of SP (because you are a super fighter jock) who spends some of those (instead of the craft's DP) to make sure that you don't get blown up. These Success Points are essentially your "DP" for vehicular combat. NOTE: a vehicular heavy weapon hitting the body of the vehicle may fuck it up pretty badly anyway as they can do 100's of points of damage.

Things We Do Not Know

There are some things we are, currently, not clear on.

1. When a vehicle gets shot up without a called shot (i.e. at the driver or at DRIVE) how do you determine if, for example, characters were hit? If I fire an RPG at a normal car, if it detonates in the passenger compartment it'll likely kill everyone inside. If it hits the back bumper it'll annihilate it but that's all. Sailing ships, for example, would take all kinds of cannon hits without sinking: apparently what got most of them was not the raw damage from the hull being breached but rather fire.
• There seem to need to be some rules for "blow-through" (the vehicle is hit but the damage doesn't hit anything vital and it's cosmetic, you lose the radio, etc.)
• There needs to be a way for a character with defensive SPs to spend them to mitigate hits with vehicular weapons that would kill the whole party: there has to be a reason to get into a vehicle. Your gaming experience should not become totally fragile.
2. Are there any specific rules for specific maneuvers? If we were playing with counters, how do I move the counter? How do I handle 3D combat? It appears to me that, without resulting to a battle-map, most maneuvers are rolls against vehicle ops skill to accomplish something. Failure may or may not result in a crash (usually not, actually, our current thinking is that crashes are usually the result of either (a) low / no skill and a missed roll or (b) a high-skill level character specifically "risking a crash" to accomplish something harder.
3.  What are the needed "standard dramas"?
• Racing Dramas: We have some good ones. We've tested them and they work.
• Battle Ship Exchange Of Fire Dramas: For fleets or long range combat where maneuver is more strategic (facing of ship rather than exact location--lack of "dodging" and more about being hard to track with guns, and so on) do we have special rules?
• Dog Fighting: How do we handle situations where a more maneuverable ship tries to out-maneuver several attackers. NOTE: because of the way SP points work, if you have a fairly small number of them, larger numbers of attackers add up fast. Is there some kind of dog-fight drama that simply "prevents being shot at?"
4. How are vehicles used in "ordinary combat"? For example, what is my roll to run over someone in a car? If I use the car as a weapon with some regularity ... do I pay AP for it? If I am buying "ship's guns" with AP's--ultra long range cannons that (presumably) do a ton of damage ... but never get any closer to "the action" than orbit or a port ... are there rules for that? How do those cost in terms of APs?
5. How do I charge APs for certain kinds of vehicles?
• Mad Max's car: almost no one has a car. He has a great one. How many AP is that? How many AP is a cycle?
• Mal's Space Ship: does a PC pay AP to be "the captain" who owns the ship? Or is that a non-AP convention?  Do PCs pay APs for "better personal fighter-craft?" How about "ugrades for their assigned mecha?" (indications are "yes"--but do we have any guidelines as to how much?)
6. It appears from research that "vehicular heavy weapons" are quite capable of targeting individual humans (and that is my experience as well as an anti-vehicular wire-guided missile launcher in the army: I hit targets down range that were not much bigger than a playing card sized area. It could have been a person). Furthermore: current fire-control systems seem to almost guarantee a hit if you can line up the targeting system on the target. This makes movement and cover all important in modern vehicular (tank) battles. Do we have some set of rules that (a) covers vehicles and rolls to take cover but also (b) covers on-foot PCs and the need to take cover? If we do not have good rules for how cover protects, facing a military heavy weapon is simply a death sentence and however "realisitc" that is or not, it's not good for the game.
7. How do different kinds of weapon-mounts work: fixed guns vs. turrets? Indications are that turret guns allow fire in situations where fixed guns don't. Slower moving heavy guns have some kind of modifier that lighter machine guns do not, and so on. Maybe those "dog-fight dramas" allow you to evade fixed guns but not turrets?

There's probably more but ... that's what I got now. 

Have-Not Game

Cyber Eye (Santiago)

We hit a major milestone in our 2-year Have-Not game last night when we exited a major series of "dungeons" which had taken us maybe eight or so play-sessions to close out. The adventure was a culmination of a number of threads bringing us closer to the probable conflict between us and ... the rest of the world. At this point, having gotten around 1 level per "dungeon" (with a few extras) we are around 104 AP characters--immensely powerful having come from 8 AP to start with and well above almost anything else in the world.

The game was never meant to be completely open ended--we knew it would end and that we'd move on and play something else: it's pretty clear we are on the back-nine of the game-world. Our next game will likely be a super-hero game designed to stress-test the existing rules and while I'm looking forward to that it's bitter-sweet to see the current game which I've greatly enjoyed winding down.

Our general mode of play (these days, anyway) is to have pretty lengthy games--but I think our more regular pattern (over the past, say, 10 years) is to start games--have them run a few sessions--and continue them if they "catch fire" or dump them for something else, if not. Our last three games, a moderate-length Psionics game (we were students at a school for telepaths where some of the students were a-moral children of the conspiracy that ran the world), a lengthy Cyborgs game (we were cybernetically enhanced US Marshalls living in an age where "the singularity," or, more appropriately, several possible "singularities" were about to happen), and then Have-Not (2 years+) seem to be beating that spread.

I'm not sure if this is because our limited play-time (2.5 hours per week) makes it harder to start new games or because we're getting better at hitting winners. I think it's the latter--but a high barrier to entry suggests that could be responsible for the track record. I'm not sure how to objectively test this (although it feels like while the Psi game had some weak points, the Marshalls game and this one are certainly home runs).

How Do You Make A Car Invisible?

Officer Law (Hector Busamante)

One of the sub-systems that we're working on right now is Gear (equipment, devices, etc.). The rules for gear are complicated for two reasons:

1. You do not always pay AP for gear. In no game I can imagine would you pay AP for things like normal clothes, a toothbrush, etc. That's (reasonably) common sense--but in a lot of games you won't even pay Archetype Points for things like weapons and armor. For, for example, a modern-day Cops game the PCs would have very few AP (if any) and things they buy would generally represent unusual native abilities (such as an exceptional physique or extreme detective skills). The gear would just be "issued" to them in-game.
2. When you ARE paying points for gear there are still a few ways it can go:
• You can take a Trait like Powered Flight and turn it into gear (a "flight ring"). This is an example where being equipement is usually a (small) disadvantage (or, if you make a very limited flight-rig that's dangerous and difficult to control, a substantially larger disadvantage).
• You can take some gear you did not pay points for (like a normal car which "anyone" can get in most games for no AP) and turn it into a tricked out vehicle. That's another implementation of the Mod-Point rules. In this case, while you do pay character points (MPs) for the modification, you do not pay them for the initial piece of gear (assuming it's something "anyone" in the game could have).

Which Brings Us To The Question

So as I look over the Gear rules, I find that to a certain extent they work "okay." Ability modification rules are always complex (take, for example, a theoretical simple set of rules that simply reduces the cost of an ability by a set % amount per defect added: how do you handle the case where the %-total reaches 100% (free) or above 100% (you get points for having a defective ability)).

Our current rules work as follows for modifying a non-combat power: you pay (for enhancements) or get (for defects) a certain number of Mod Points (MP). So being a "simple hands-free device" is -1 MP (a defect). That "-1 MP" is per 8 AP of the Ability. 

So if I have 1 to 8 AP spent on Flight and I turn it into a "Flight Ring" (magical, thought-controlled, small and convenient) I get 1 MP "back." If I have, say, 16 AP spent on Flight and I turn it into a Flight Ring I get 2 MP back.

I can use those MP to buy other small advantages or I can turn them into Damage Points if I want to.

That, conceptually, isn't so hard.

Okay--but what if I want to make a car invisible (I have a cloaking device for my automobile!). Firstly, the rules, as written right now, don't address this (or do only barely). Secondly, the answer isn't that easy even going by "what I think should be the case."

Let's See

1. We presume the "basic car" is 0 AP in cost. In most games you don't pay Archetype Points for a normal car. In a Mad Max post-apocalypse game where the guy with the car is a big deal? Maybe. But in most games you don't. For purposes of this question, let's assume you don't.
2. There are rules for "Awesome Upgrade" (which invisibility certainly is) but these are generic and  are best applied for stuff that we don't and are not going to have rules for. In the case of Invisibility we do know how it effect combat. We know some things about entering and leaving "cloak" (8 REA Long Action).
3. There should be some basic rules for "modification cost to ability that does not directly effect the character." Let's say that "doesn't effect you is half price for the power: so if I want to put Armor on the car to make it tougher for 4 AP I get 8 AP worth of armor on the device. This would be represented, by the way, as -16 MP cost (for people keeping score). That rule isn't clearly listed--but let's say it was. Okay ...
4. But Invisibility is a TAP-cost power: it's cost is based on a percentage of your Total AP Points. Is that still right for the car (which is 0 AP points) or is it based on your character (which is, let's imagine, 32 AP)?

What are the right answers for these points?

1. We already have some rules for modifying "0 AP gear." Firstly we tend to "Assume" it cost 8 AP for purposes of doing the Mod Point math. That is, if some enhancement is listed as +4 MP (which is 1 AP for the character to spend) that's for 8 AP worth of an ability (so to put that enhancement on a 24 AP Trait would cost 12 MP or 3 AP if you wanted to pay AP for it). For a 0pt piece of gear we assume it's 4 MP. This is a simple rule (a 747 jetliner is the same "AP cost" as a Ford Focus--which probably isn't quite right--but hey).
2. While it isn't 100% exactly how invisibility will impact vehicular combat (even from what we know about it) we can make some good guesses: it is hard to "shoot out the engine" of a vehicle you can't see--called shots become far more risky or impossible. That can be a big deal when shooting at vehicles where a lot of "hits" won't disable anything critical. So, yeah, we should probably try to use as much of the power description as possible.
3. The cost for having a trait only work on a device is probably heavily dependent on the device. For example if you take the Trait "Flight" and declare it "only works on my clothes" (which you then wear and fly around in) that's like having the Flight power yourself. If you take Flight and declare it works on a camera so it can hover around and take pictures ... that's not like you having flight. Big difference. It also matters whether you can be expected to be inside the device or not. That's a big deal. Having an invulnerable clock-radio that will ignore any attack that hits it on your bed-side is cool. Having an indestructible trench coat that does the same thing is much, much better.
4. We could do something like say the TAP cost for the vehicle is based on it's assumed cost (8 AP) but there is a point raised: does it cost fewer AP for a low AP character to have a cloaked car than a high AP character? Maybe it should? Especially if the character can, say, fire out the window while remaining mostly invisible (the reason Invisibility is TAP costed is that it presumes the advantages will be more severe the harder the character hits, etc.). So there is a case that the cost for the car being cloaked should be based on the cost of the owner.

So, you know--that's something we're thinking about.

Proofing The Opening Chapter

Gravity Control: Hector Busamante

I'm having my mother read the opening chapter of JAGS Revised Archetypes. Firstly she is a competent editor and has edited at least one other published book. Secondly she is not a gamer in any sense and, although she has, actually gamed with us (story below) once or twice, does not natively "get it." If she can understand the chapter I figure I'm in good shape.

I ran a game for both parents and an experienced game friend several years ago when Hurricane Wilma knocked out the power for several days. With very little to do and housing a friend from an hour north who was staying by himself, we figured it might be a way to pass the time. It was a success.

The game I ran was one where the characters were "Space Rangers" who were new on assignment and wound up with a disabled ship in "pirate space." They, keeping their identities secret, wound up on a hidden pirate space station (large) trying to beg, borrow, or steal parts necessary to repair their ship. The only visible supply of parts were in the hands of a local crime boss (to an extent, everyone on the station was a crime boss) and the parts could be had for "bounty hunting."

The PCs could've chosen to try to steal the parts (I made it clear that was possible but risky--they were heavily guarded) but decided to run down the bounties. It turned out the targets were two lovers on the run (Romeo and Juliet) and in the end the characters had to decide whether or not to bring them in (to get off the station), let them go and try something else, etc.

They hit on an idea of taking some of the bodies they'd accumulated and having them surgically altered to appear as the targets. It was an idea I hadn't thought of and allowed a win-win where the PCs got their bounty and the lovers got to escape--the (very short, tightly run) was deemed a success--and fun.

Observations
I have observed (and leveraged this) that new players really like price lists. I created the game with a short list of gear that they could choose from (not knowing what would happen--but knowing something would) and let them take a few minutes to make trade offs. This seems to be a way to generate interest and energy early on. I think AD&D really benefited from it's (often obscure) price list. I think a lot of people miss this.

I also wanted to give the players strong parameters (you are on this station, there are these things around) but no strong guidance. Even the core decision to attack the crime boss or try to play the game was explicitly left open (of course bounty hunting was so attractive I doubted they'd gravitate to the riskier solution). The big decision (what to do with the lovers) was left completely open. There was no easy answer (I figured they would not send the two kids back to face death--but I thought they might do it).

For making characters, I think it's clear that the full number of options is overload for new players. I gave them basic character templates and let them pick from a set of equal cost "packages." This is "simplification for new players" but it illustrates a point that should never be forgotten: the more work you do to make everything in play relevant to the game the tighter the experience will be.

I didn't put in cyber-hacking because I knew that there wouldn't be cyber-hacking in the 4-6 hours of play we were going to do in a blacked-out house. Now, this rule can (and has been) taken and run far too far with: if I didn't put cyber-hacking in JAGS because "JAGS isn't about that" it'd be a huge mistake. In a game that is designed to, maybe, be run for 2 years with the same characters and players you are necessarily going to want to do things differently than in a 4 hour game for players who have never played before and may never play again.

For those players they expect--and are reasonable in it--that whatever formative decisions they make, they will be relevant. When making a character for a game that may run hundreds of hours those expectations are often going to be different. A player who understands both role-playing and the dynamic of their specific group well will know how much agency they can expect to have when introducing elements into the game (i.e. a cyber-hacker space ranger will, in a long enough game, find places with a cyber-net and will generate situations where the hacking is likely to be useful even if the first pirate space station they crash on isn't one).

This is also something behind the thinking of costs in JAGS: I find that having an "8 point" slot where you can put "one really good thing" (8pts), "two pretty good things" (4pts each), or "four okay things" (2pt costs) is easily grasped and that the ease of understanding and responding makes for a good dynamic during play.

Immortality ...

Dr. Death (Domain Control: Death) by Hector Busament

Let there be no mistake: we're still churning through JAGS Revised with a vengeance. There are all kinds of powers, combinations of powers, new powers, and expansions on old powers that we need to nail down. I'm very pleased with the product but it's not finished.

Let's talk for a moment about immortality.

Immortality
In JAGS there are a few powers that confer--or come down around--making you immortal. These fall into several categories:

1. The very cheap. Things like Won't Die Of Old Age, Ignore "Death" Results (but you can still be killed by extreme damage, decapitation, etc.), and various kinds of hard-to-kill (but still normal to knock out) are pretty cheap. Taking random "death results" out of play isn't that expensive and most PCs don't die of old age anyway. There is Soul Jar which lets you be immortal so long as your enemies don't find your vulnerable hidden soul--which, of course, they will start looking for as soon as they know you've done this. It's cheap--but it has ramifications.
2. The Medium Cheap. There are forms of Immortality where, for example, you can be "put back together" if someone is there to reassemble the parts. There is a form of for-real immortality where you come back in a season. These are around 4 AP or so. Inexpensive compared to, well, not having to worry about dying ... exactly.
3. Expensive. In the TAP-Cost range is Immortality where you come back in a day or so--or even "nearly instantly." This costs, scientifically speaking, Yo Momma. There is also Invulnerable which is kinda like Immortal--and Very High Defenses (hundreds or even thousands of points of Armor and DP) which can simulate this for most purposes. These are also around 80-95% of your points.

Some Observations

The first observation I want to make is that immortality is, in my experience, usually corresponds to an otherwise expensive character. Greek gods, for example, are a lot of points. Invulnerable attackers tend to have pretty powerful attacks--not always but, like, Juggernaut is no wimp. JAGS addresses this by assuming he's a LOT more AP than the PCs (so he can spend like 90% of his power on being invulnerable and then have his 10% remaining to trash the PCs). 

JAGS does handle this (Scale Number, for example is slightly more elegant than just saying he's 1000 AP or something but the fact is that these abilities fit into a category similar to abilities like "mimic / absorb the whole team's powers" that are hard to model in a point-based game system.

Our Thoughts

So here are a few of our thoughts.

1. The General Mode of Play. Our mode of play is, on observation (although this is not so explicit) that so long as the PCs are generally within "bounds" death is unlikely. That is: there are few unannounced death-traps. There are not "random encounters" that include things likely to create a TPK. There GM usually "plays fair" to an extent around PC death and suicidal actions are generally double checked to make sure the player is taking on the risk / understands the situation*. And so on--in a "level based" model I would say that it's usually fairly clear where you are going from a Level-1 encounter to a Level-5 encounter and the GM just doesn't "spring that on you."
2. Being unable to be taken out of a fight (invulnerable, immortal and back pretty fast) is worth a lot. It means you generally can't lose so the cost has to make sure you usually don't win (if you are "equal points") because 'balance.' Balance is a tricky concept at the best of times but clearly in combat if you can spend AP's to win a battle you can also spend APs to "never lose" and that models pretty expensively.
1. BUT: what if Immortal, back-real-fast was only, say 50% of your points? Or maybe 70%? We generally don't think PCs will spend more than 50-60% of their AP on defensive abilities in the most extreme case and we know from modeling that if you spend 70% of your points on something that does NOT apply to combat mechanics (i.e. you can be killed and you re-form about an hour later) you will lose almost all your "equal-level" fights. In other words: maybe the current costs of around 80-90 AP are too high?
2. ALSO: For Invulnerable we are stating that if you spend X% of your points (90? 95?) on, say, Armor, at some break point (the TAP cost for various kinds of Invulnerable) you get EVERYTHING (you go from, say 90 Armor to 9000 Armor or -infinite- Armor). This is an interesting statement: we DON'T say that for Strength, for example--why not? Because not-losing is palatable but always winning (I hit for -infinite- damage) seems more problematic to us.
3. We do not directly address the game-play ramifications of immortality and we have only limited exposure to them in practice. Being able to take suicidal risks (the immortal guy loads up with explosives and runs into the enemy midst) is something that is outside of the simulator and is more in the realm of game play. Being able to insult the Emperor and get away with it because you can't be executed (and, if you "return to your home realm," can't be captured) may have impacts on play beyond the game mechanics. We think that many of these impacts are negative as they remove drama (or even just the appearance of drama) so they should be carefully considered.

The Big One: When Do You Come Back?

Our biggest question, though, is around when the PC comes back. Firstly, do we go in-game (days, minutes, Rounds, months?) or do we go meta-game (sessions, hours of play, etc.). It is not hard to switch back and forth for a group that has a good routine down but being prescriptive about it could have problems (our play sessions today are 2.5 hours--but in high school a "play session" could last 48 hours). 

Secondly, for games with any sense of pacing (which is, to our experience, most of them) coming back "in a month" or a season is like not coming back at all. An entire campaign could last a season. Our current game has run, real time, around 2 years and I'm not sure we've covered a year of in-game time.

The obvious answer is to have some kind of "intent" mapping. If the power says "You come back in a season" that could be "you come back in two sessions" if the GM agrees that makes sense. That's okay. But do you really want to not play for two sessions? I mean, if you were dead-dead you'd come back as soon as you could make a character and get re-integrated. If play time is the thing, what do we say to someone who says "I'm paying AP for the 'privilege' of less play time." (yeah, but you get to keep your investment in the character--so there is that). 

The Other Big Question: When Is It Appropriate?

When am I allowed to buy Immortality--even the cheap kind? In some games like "Dungeon and Dragon style games" that might not be right for the "feel of it." How do we codify that?

JAGS Continues ...

You are probably going "What the heck is going on with JAGS!? Where's the next update." I realize many of you can't make it through the day without regular updates so I apologize for being sparse. Okay, not really--but it helps me to keep this blog updated once in a while :)

Let's see:

1. We are creating a SLEW of characters using abilities that haven't been used before (of which there are many) and trying higher point builds (128 AP--supers characters). This is informative.
2. We are proofing the book. This is depressing (not so bad--but, man, there are a lot of mistakes in there and I'm not just talking about typos).
3. We are still collating artwork. 

For each single character sketch (of which there are a lot) we are actually doing their AP Traits. So we have a strong guy and he's Built plus some other stuff. We have a super-heroine with gravity control and give her various capabilities -- doing the stats--which should get printed next to the picture in the final book. As far as I know statting out the artwork is pretty unusual for a game book.

Something We Learned

So the simulator tests Resisted Attacks--remember: an RA is like Blind-Target, poison gas, mind attacks, etc. It has an Intensity score that is compared to your DP (plus some other stuff) and then there's a Resistance Roll and several levels of effect. So if I use some kind of Mind Control and my Intensity dominates your DP (you're like a normal guy and I'm like a master villain) I get a Critical or Catastrophic effect and can order you to kill yourself or something. If you're more my peer then maybe I get a Standard or Major effect and it's not so bad.

Well, in the Simulator we were testing VERY simple characters--all attack and defense and DP. Not so much in stuff like Super Hacking or whatever. It wasn't relevant. But what we discovered when making a bunch of characters is that the real, interesting characters had a LOT fewer DP than our test characters. We also discovered that the less powerful Resisted Attacks (compare "Tear Gas" to "Nerve Gas") had way too high Intensity so that even if they required a Catastrophic Effect to take you out, their proportionally higher Intensities would get that.

So we are re-tooling our Resisted Attack charts to both reduce the Intensity for most attacks a bit and to make the "weaker attacks" have even lower Intensities than they used to.

What Else?

Well, we did get some interesting looks into the powers and have revamped several of them "on contact" with a real character. Maybe I'll post some of the characters we're making here ...

Cover for Domain Control chapter by David Chow