This Article Has Been Delayed

I recently thought that someone should write an article critically examining the gaming industry and its habitual issues with delayed products and false expectations. It should, I thought, be an article that takes an unflinching and unbiased perspective, measuring carefully a gentle balance between the starry-eyed desires of consumers with the realistic constraints placed upon developers. Granted, I was pretty sauced at the time and I see now how that would be a really really boring article. No, what we need is a poorly researched, sharply biased piece more derived from intuition and imagination than fact – you know, the sort of thing youÂ'd see at CBS! See, I can be topical!

[now, when I say topical, I donÂ't mean that I can be applied to the skin like a salve or balm. But, I digress – dramatically.]

IÂ'm well aware that we have a few developers perusing our site on a basis that I might term at least occasional, so a word to you gentle souls before I call you scum and roll you through the dirt. I love you guys like brothers. Brothers I never really knew, and who might have given me wedgies in front of your pimple faced friends when we were kids, but brothers all the same. You guys are out there on the frontlines every day doing whatever magical incantations and shaman shaking dances it takes to make computers perform feats fantastic, and for that I thank you. But, it is the kind of thanks IÂ'd give to a chef cooking my dinner while I wait insatiably hungry following a month of eating carpet lint, which is to say that itÂ's a genuine thanks but one tempered with my desire to kill you and take whatever youÂ've got finished.

Perhaps a part of the issue I have stems from my inability to relate to the stress placed upon you guys. IÂ've never had anyone anxiously await the product of my work. No one ever messages me (and, I hasten to point out via my Freudian typo, no one ever massages me either) while I write my articles demanding a firm release date, except maybe Certis, and then only because if I write something it means he doesnÂ't have to. Lazy is not an adequate substitute for anticipation in this example. So I canÂ't say what the demands are for the spotlight you guys work under.

But, as I said this article isnÂ't about being reasonable and patient. IÂ'm a gamer and a citizen of the net, so reason and patience are foreign and offensive to me. In the long run I think I speak for everybody when I say, shove your well considered delays and give us the damn games!

Not to be too on the nose about the whole thing, but IÂ've seen weathermen with better forecasting skills than most developers when announcing a new project. I realize there are unforeseen difficulties, but IÂ'm guessing this isnÂ't the first time most of you have put together a game, so when you show up and say your new game Awesome Texture Five Thousand X-3 Xtreme will be out by the end of November, I donÂ't think itÂ's too much to ask that you not be wrong by half a decade. IÂ'd far more respect a shrug of the shoulders and a soft chuckle when asked about release than picking a date out of a hat. The fact is, head shakingly sad as it is, when you tell gamers stuff like release dates it affects their lives. They leap to their feet, skid across the kitchen in their Conker slippers, and circle September 30th 2003 in big black Sharpie surrounded by a halo of asterisks and stars. IÂ'm looking at you Valve! And, donÂ't think I donÂ't also see you, Bungie, in the back of the class passing notes to Blizzard.

And IÂ'm looking at you guys because when your games get delayed all you see is the volcanic ejaculations of message board vitriol, but you donÂ't see the quiet shakes in the dark because you were supposed to be playing Half-Life 2 and instead youÂ're playing Everquest again, and that means you probably need to be back on the reuptake inhibitors, and damn you just got this job, but we all know how itÂ's going to be when the guild goes on a raid to kill the Sleeper but youÂ're supposed to be up at seven to clean the deep fryers and this is your third can of Bawls so thatÂ's just not going to happen, and Sally doesnÂ't call anymore except to ask if youÂ'll please for the love of all thatÂ's holy send her Indigo Girls cds back, but don't deliver them in person until you get a shower and new clothes, but you donÂ't even want to admit to yourself that you absolutely love the song Gallileo and listen to it while trading Spider Silks in the bazaar, and how you so wish you were playing Half-Life 2 instead Â"….

*gasp*

So, you see, itÂ's bad to delay games, because thatÂ's how crazy stalkers are born. The net doesnÂ't need more crazy stalkers; itÂ's pretty much at capacity.

Now, I know what you guys would say, Â"oh, weÂ're just trying to make the best game possible, and we have to make sure the product is of highest quality, and blah blah blah.Â" Well, IÂ'm a problem solver, and I have the solution. Instead of delaying the game, have it done when you said you would, and have it complete at that point. See, I can be reasonable. If that means you guys need to work twenty-four hours a day under constant threat of violence to soft squishy parts, itÂ's a risk IÂ'm willing to take. I think itÂ'll be a breath of fresh air.

Ok, I think IÂ've slipped into farce somewhere back in the third or fourth paragraph, and thatÂ's really only tolerable for a very short time. We all love games, and thereÂ's a reason we gamers donÂ't just rise up against you. ItÂ's because if it were left to us to make the games instead of people with training and at least a hint of functional knowledge, weÂ'd all be playing a lot of very bad games that released on time. Truth is, we gamers are angry monkeys in the cage. We can throw all the feces we want at the Plexiglas, but itÂ's ultimately a pretty futile gesture. So, if you could show us just a little humanity and trim up those delays weÂ'd surely appreciate it.

Now, can I please have my Half-Life 2, because IÂ'm pretty sure my EQ icon winked at me the other day.

- Elysium

Comments

First of all, I have the solution to your EQ problem. Uninstall it and burn any cd's you have that make it easy to reinstall it on your system!

As far as delay's are concerend. This sort of thing will only get better when publishers realize that sending out bogus release dates that can never be reached is detrimental to a games success once it is finally released. That will never happen unless we sheep... I mean gamers begin to control the market with our buying power and start showing developers what we think about being teased with false promises.

Of course that is as likely to happen as a cow flying to the moon, we are pretty much screwed.

edit: because I think to fast and write too slow.

Damn straight!

EDIT: Oh, and way to make it through a game delay article without mentioning 3d Realms. Congratulations, it's an internet first!

Wow: Bungie, Blizzard, Feces throwing Monkeys, Plexiglass (Mmmm plexiglass), and Half-life 2 all in one article - Congrats!

Gaald wrote:

As far as delay's are concerend. This sort of thing will only get better when publishers realize that sending out bogus release dates that can never be reached is detrimental to a games success once it is finally released. That will never happen unless we sheep... I mean gamers begin to control the market with our buying power and start showing developers what we think about being teased with false promises.

Shouldn't the invisible hand do that for me? I don't want to have to actually do anything or forsake buying a particular game; the market needs to correct for me whilst I remain comfortable in robe and slippers.

This sort of thing will only get better when publishers realize that sending out bogus release dates that can never be reached is detrimental to a games success once it is finally released.

Eh, the developers are as responsible for this as the publishers. And we're not talking about the game business only here, most software projects are finished later than estimated, cost more than estimated and often enough have less features than originally intended. It's the planning phase were the first mistakes happen. Some parts are always underestimated, some schedules are way too optimistic, never taking unforeseen problems into consideration. If we were living in an ideal system, there wouldn't be such a thing as crunchtime.

You know, delays don't really bother me much anymore. I have a really hard time getting excited unless I see a date advertised somewhere besides a rumor mill/ebgames.com and evidence that they're actually that close to being done. Most of the time, I just forget about the game until it comes out, then pick up whatever good game comes out this week. Halo 2 is the only thing I'm really anticipating like that, simply because I've seen it in action and they've been pretty consistent on the release date for a few months now. Anything less and I just can't seem to get worked up over it, I'm too used to developer's horrible planning.

Also, I'm a software developer and I know how hard it is to come up with an accurate schedule on a pretty mundane product, let alone a rapidly moving target like a video game. I can't imagine what it'd take to come up with an accurate schedule. So any dates that come out from developers and publishers I treat like I treat our development schedules, with a grain of salt.

Wow. I Sig'd you and you mentioned reuptake inhibitors all in one article! You're so dreamy, Ely!

Developers need money from the publishers. Publishers don't want to give their money out unless they can get a certain return on their money over time (usually highly unrealistic return on their money it should be mentioned). Therefore, the only way developers can get the publishers' money is to promise (read: lie) a return on investment that will get them the money. The actual development time is irrelevant. You want to promise enough features to get the publisher interested and promise them in the time they unrealistically want, if you want to get your hands on the money. The developer that does this the best is the developer that gets the money.

Once the developer has the money; spends the money; and can show some fruits of their labour, it is much easier for them to go back to the publisher and ask for more money and renegotiate the time/features required for the end product.

The problem with the customers begins when the publisher is convinced that the product will be actually completed. It is the publisher that tells the consumer when the product will be done and it is the publisher that does all the hyping of the product. Unfortunately, the publisher is not realistic about this process and inevitably "promises" the ship date that the developers gave them last (even though the developers pushed the date back more than once by now the publisher didn't seem to have drawn any conclusions from this).

The whole process is one evil vicious cycle. Publishers get burnt by giving money to people who lied to them about what can be done in what time frame, so they pucker up and make it more difficult to convince them to hand over their cash. Which in turn forces new developers to lie even more about features/deadlines. Which in turn makes promised ship dates slip even more.

To make the problem more difficult, some of the new technologies or genres are new to the development world and bring a whole new host of problems with them. Problems that cannot be identified prior to the development process. Last but not least is the problem of quality assurance in the software development industry. Software developers are supposed to be engineers, yet unlike engineers do not have an industry governing body to regulate who is allowed to practice the trade and who is not. They are also not thought sufficiently many engineering quality controls, as those controls do not translate well to the software development and the true science of software development quality controls is only just emerging. This results in most developers being incapable of trully foreseeing the scope of a project at its onset. Not because they are stupid, but because the project is so overwhelming that no unprepared human being could trully grasp it. As an analogy, imagine asking your local car mechanic to build a Formula 1 car. I'm sure he would know how or could overcome obstacles, but I guarantee you that he would underestimate the time required by at least 100%. Oh and while we are on the topic, a little factoid: the best and most organized project estimators, employing the best possible engineering techniques, will err anywhere from 50% to 200% in their time estimates (according to many years of research from NASA).

As far as HL2 goes... I am starting to believe in a conspiracy theory. Let's just say that it was not in Valve's financial interest to release the game last year. The code theft may have not been their direct doing, but it's amazing how convenient it was and how much time it bought them when it comes to their dealings with Vivendi. It basically gave them an unlimited "delay until we are done" card, which is a very powerful bargaining tool against the publisher. Come to think of it, I have HL2 executable on my hard drive as I write this, yet the official release date on all gaming websites is TBA. Makes you wonder, doesn't it?

Anatomy Of A Software Project:

1. Management wants Project X done in "Y" (where Y = "Complelely delusional short period of time, about as realistic as asking a person to stop an ocean tide with a teacup")

2. Development insists "Y" is not nearly enough time.

3. Management ignores the fact that no project of this size has ever been completed in anything less than "2.5Y", but says, "hey, you guys are good, you can do better!"

4. Development "negotiates" management up to a timeframe of "2Y". Management insists anything longer is completely unacceptable, and feel like they've been good managers because they've doubled their originally planned timeframe. Development has now agreed to complete a project of a certain size much faster as anyone else has ever accomplished.

5. Development needs to find a way to meet this ridiculous schedule. Pre-production basically doesn't happen. Planning is done on napkins at lunch, and coding begins immediately.

6. Management wants an update on progress. Development has tons of code, most features showing progress, although no efforts have yet been made to ensure smooth integratation of modules will happen. Development will jump over that bridge when they get there.

7. Original timeline is about 3/4ths passed. Lack of planning and organization hurts development. Code is thrown out, and other code takes lots of re-tooling to match up with other modules. Some modules plain do not line up with each other, and require rewriting. The lack of a solid specifications means modules don't exactly work with each other like they need to. Premature integration of code reveals problems and creates some of its own. Coding standards are non-existant.

8. Time approaches "2Y", and management wants to see their product. The project is in complete disarray. Management sets a new deadline, a few months after "2Y". Development begins hacking things together to try and get some features to a "finished" state. New code is constantly being written, without time to concern itself with integration issues with the rest of the project.

9. A competing product releases with Feature Z. Marketing insists the project must match Feature Z or else be dead on arrival. Adding Feature Z invovles ripping the project open, and performing some open-heart surgery to install the new functionality. The latest deadline passes. Management is pissed. Coding standards and a specification materialize extremely late. Existing code is re-worked again. Feature Z flat-out breaks functionality elsewhere. The broken features are ripped out and begun again from scratch.

10. Management places another deadline. Burned out team members get hired away by headhunters. Work slows as replacement members are brought aboard and up to speed. Jack down the hall insists his networking code is fine, and won't let anyone else near it. The rest of the team worries about the program itself, figuring they'll worry about networking at the end. Despite a slow pace, pieces of the project start to come together.

11. Last couple deadlines passed without completion. Management sees the project as out of control. Questions arise about possible dropping of the project. Regardless, progress has been smooth since the team stopped trying to squeeze into an insane deadline. Project starts to show signs of life.

12. Team is confident that its project is approaching completion. They inform management that a releaseable product is coming soon. The team looks to integrate Jack's networking code into the rest of the project, only to find Jack's code is not well self-contained, and tied itself to old modules that have been scrapped and replaced. Efforts to "polish" the program halts as networking code becomes the sole priority. Management and marketing plan a release very soon, on the strength of the optimism just expressed by the development team.

13. Efforts to write new networking code go well, though most existing work had to be scrapped. Management is informed that they'll have to slide the release a little longer. Management is furious and feels like they're being yanked around by those "engineer guys".

14. Project finally comes together to a releaseable state. QA has a boatload of issues for development to fix, but management will wait no longer.

15. Software is finally released, in about "3.5Y" time. Management considers the development a runaway failure, as having taken 3.5 times longer than they originally planned. Development is amazed that they only exceeded the "best possible" development time by that much. Customers finally receive the product after having been told a string of "release dates" between "2Y" and "3.5Y"

Amen, brother. Enough with the sketchy release days and inevitable delays!

...and damn you just got this job, but we all know how it's going to be when the guild goes on a raid to kill the Sleeper but you're supposed to be up at seven to clean the deep fryers and this is your third can of Bawls...

New job isn't everything you'd hoped it would be, eh?

Sketchy release dates and delays are inevitable, imo, and will continue to be so until the industry matures significantly. Nearly every software project is exploring uncharting territory. I'd rather they take the extra time to get their product into a shippable state instead getting my hands on the game a couple of months earlier.

Lol, nice one Elysium, as usual.

I don't really have anything to add at this point, except to say that delays are usually the programmer's fault The artists and designers can meet their deadlines

5. Development needs to find a way to meet this ridiculous schedule. Pre-production basically doesn't happen. Planning is done on napkins at lunch, and coding begins immediately.

This is the key problem, at least in my world.
Actually, I'm impressed, you plan on napkins.

And you left out that most of the "coding" that begins immediately is an exhaustive search of existing code from which you can cut and paste working functions, classes, snippets in a vain attempt to meet the agreed upon unrealistic time frame.

And this is where a lot of the problems in QA come in, because you're using code that you "know" works, yet somehow, doesn't work in the new codebase.

One thing, though. Coding is not a science. It's not engineering. Most developers I know are closer to artists and artisans than scientists and engineers. We spend as much effort and brainpower making our part of the app "beautiful" as the artists. It's just that there are fewer people that can appreciate the beauty, and only a mere fraction of those that would appreciate it will ever actually get the chance to see it.

I would argue that coding is a sort of beast unto itself, which is why trying to apply the old labels to it never seem to fit.

That said, I think there's plenty coding can learn from engineering. Also, I don't think engineering is ignorant to "beauty" either (not your claim, of course, but rather just an extension of one of your thoughts).

But it comes down to a bunch of pointless semantical arguments. Coding is what it is.

Great stuff, Legion, but I think you missed a step (which we'll call it 4a): Since Development has negotiated to do the work in 2Y, they don't really need all those extra people, so 1/4 to 1/2 of all resources are assigned to another project.

duckilama wrote:

One thing, though. Coding is not a science. It's not engineering. Most developers I know are closer to artists and artisans than scientists and engineers. We spend as much effort and brainpower making our part of the app "beautiful" as the artists. It's just that there are fewer people that can appreciate the beauty, and only a mere fraction of those that would appreciate it will ever actually get the chance to see it.

While I agree that we developers exhibit the same tendencies as artists and authors, I'm not so sure about people's inability to appreciate it. I think a "beautiful" application is not only built out of elegant, well-designed and documented code, but it is also robust, easy-to-use and easily maintained. If developers were allowed to build such an application, these features would be visible on the bottom line and in customer satisfaction levels.

Then again, maybe we're all crazy. I mean, who ever really "got" cubism?

You forgot to mention that after a game is finally released, after numerous delays, it will be partly broken. The developer will announce plans to release a patch that fixes several fatal flaws and adds functionality almost immediately after release.
By announcing a patch so fast, wouldn't it indicate they knew the problems existed before they released the game, thereby blowing the "we'll release it when it's ready" excuses for the original delays out of the water?

I worked for a period of time as a game developer. I was not overly impressed with how the 'system' worked. Now, by no means do I profess to be an expert at developer-publisher relationships. I went back to the simple yet satisfying life of an engineer long before the train-wreck that is game development could claim my sanity. I do agree with much of what MoonDragon and Legion said above.

Publishers have unrealistic expectations and developers want to make them happy ... after all, the publisher is the sugar daddy. You do see some developers with good track records for getting games out the door in a timely fashion. I think Raven is one that never gets enough credit for consistently generating quality games in a timely fashion. Of course, they also work mostly with established technology and don't have the problems some other developers may face.

One thing that game developers seem to do a piss poor job at is project scheduling. Developing a game is not much different than a construction project. You have certain tasks that you need to complete and other tasks that rely on these tasks before they can be started. The construction industry has scheduling down to a science. If there is a delay, you know it well in advance because your project schedule has slipped. Aside from technology issues (which admittedly can be a big unknown), I don't see why game developers can't to a better job of managing their projects.

My 2 cents ...