Today I'm converting text from the book into the game (see previous post). That should give an accurate idea of how long the whole book will take. From there I'll be able to give a daily "percent done" update for puzzles, cut-scenes, secondary dialog and primary dialog. When those reach 100% the game goes for beta testing.

Breakthrough! Over the last few weeks I've posted my concerns that the dialog is going to take too long. Well I think I cracked the problem. I created some code that will let me basically take text straight from the book. Click on a person, and this triggers a description or conversation from the book. I will still need to add puzzle-specific stuff, but now I will only need to add what is really needed: the foundation, the mass of detail, is already there.

But how to make it work? How does that make sense in terms of a game? It means adjusting the user's role a little, but I think it strengthens the game.

PREVIOUSLY: the role of the user was kept a mystery, hinted at, but not revealed until later games. NOW the immediate role is clearer, but in a way that should strengthen the mystery. Basically, you (the user) have godlike powers. But this is not like other games. In most games, "godlike powers" means "destroy stuff!" But in real life (with the occasional exception) gods and spirits don't behave that way. Gods and spirits have unlimited knowledge, yes, but they influence events BEHIND THE SCENES. Never directly. Gods and spirits plant ideas in your mind, but they almost never reveal themselves directly. Look at Christ, or a djinn, or almost any real world gods. That's how they operate. That's why faith is needed. That's why unbelievers scoff and say there is no proof. I spent half my life as a fundamentalist Christian, and I think most games about gods just don't "get it."

Besides, I think it's cool to tell the user "you have godlike powers!"

The only down sides are, (1) Victor Hugo wrote a great deal about some people (e.g. about Bishop Myriel) and far less about others. And, (2) maybe people won't like Victor Hugo's style. But if people don't like Victor Hugo, what are they doing playing a game called "Les Miserables: the game of THE BOOK" ? :)

This is a test image for the new graphics idea. I think you can see that the art has a lot of potential, but it also needs a lot of tidying up. And it looks a lot worse when the pixels are doubled (its native resolution is too small to be practical, except maybe on mobile phones, and that market s too short lived to be practical.)

I did some calculations, and the extra effort is not justified at this stage. Which is a pity - if I was starting again I'd give it serious thought.

The character on the left is a frame from a video of my son, climbing. I thought video would be the real problem, but it was easier than I thought. Yet even here it would take some tidying up.

Well there you have it. The art that might have been.

Fixed the erratic walking problem. Hurrah!

After much experimenting, the "new art" idea turned out to be a dead end. So the art is not changing again. However, I am still determined to find a solution to the dialog problem. The game as it is now designed requires at least a month's extra time to write dialog that will probably never be used. far more dialog than will ever be used. Meanwhile, Les Miserables has plenty of material that I'm not making good use of. I need to somehow rethink the dialog part of the game so I make better use of the book and don't waste time on stuff nobody cares about. It's a tough nut to crack, but the beauty of making my own game is that I'm totally free to make changes like this, even at this late stage.

Back on the subject of new art, in case anyone is interested, my idea was to use classic illustrations and paintings exactly ad they are, in their full detailed glory. The story would have been reworked to minimize the need for animation, and essential animations would be simple video. It looked great! But the devil, as always, was in the details. Every stage involved some little tweaking that was very labor intensive, and by the end of the experiments it became clear that this would not save any time at all. And while some benefits were clear (much of it would have looked beautiful), there were also drawbacks (low resolution, uncompromising copyright issues, etc.) Still, it was an interesting idea.

Update: I think the real problem is complexity. The game is fundamentally complex. Ironically, it became more complex to compensate for its simplicity! Minimalist art can be a little boring, so I added dozens of extra people in most scenes, to make them look more interesting. But that multiplies the background code (leading to endless little fixes), and multiplies the interactions (what if person X interacts with object Y?). Not only does the extra work take extra time, but the extra stuff is not very interesting for me, so probably won't be too interesting for anyone else.

The only way I can get away with removing the redundant characters is to make the art more interesting. That is easily done - I now have a collection of great art that I didn't have when this project began. But it raises three big questions. In order of importance:

1. Would this save more time than it takes? I did some experiments, and I reckon I can replace all the graphics (again!) in just three weeks. Four, tops. (Why so quick? See next point.) Will it save three weeks of work? In theory, no. I have budgeted the remaining work and there is no spare flab. HOWEVER, in real life EVERYTHING is taking longer because of the complexity I mentioned above. So the remaining two months of the current schedule will almost certainly expand into three months. By simplifying the game I save that month.

2. The only way to make the art look much better while being simpler (and being quicker to produce) is to do this game at low resolution. I have plenty of art that looks great at low res, but looks rubbish at higher resolutions. Will people accept a low res game? My gut feeling is that they won't care as long as it looks good. Low res gives much better file sizes, so that's the selling point. A few years ago hgh res was everything, but in thjese days of ubiquitous high res, and mobile phones and Youtube, I think people are more accepting of low res. As long as it isn't positioned in direct competition with identical high res games.

3. Would people miss the minimalist white style? The reaction has been positive, but not "this is the best thing ever." For example, the teaser trailer made zero difference to the number of people signing up to be told when the game was released. The new at would have looked MUCH better, though that's mainly because the teaser would be viewed at low res.

Making this game is such a learning curve!

I am in a major quandary. My heart wants to make some more major changes, but my head says "it's too late, you've gone too far already."

Oh dear, another long blog!

Development has been taking longer than planned over the past few weeks, and I keep finding little problems that take me firther and further from where I originally wanted to be. Here are two more examples from yesterday and today. The first one was a problem with changing scenes, and came as a result of the code now being at an advanced (and complex) state. That took six hours to fix: basically a whole day wasted.

The second problem is still not fixed, and has more fundamental implications. Remember how, many months ago, I had to stop using a main character because AGS adds a dark outline to characters? The chap who makes AGS agreed that it was a genuine bug, and recommended changing to the next version. Unfortunately the next version cannot be made to run in Linux, so that's a no-go. So my only choice was to remove the main character from the game. Now I'm getting deja vu. There seems to be a similar class of problem, but this time with characters walking.

Another user spotted this problem years ago, and it was added to the "to do" list but (so far) it looks like it was never fixed. So far my efforts to find a workaround have come to nothing. It is looking like I may have to re-write a lot of the game so that important characters stay still.

These are typical of the issues I'm facing every day now. I'm pushing AGS much further than it was ever intended to go. AGS was designed for light hearted adventures and pixel art, and I'm trying to use it for heavyweight adventured with smooth art, a lot of scaling, crowds of people, and so on.

Another issue is with the general direction of the project. For the very first release I naturally want the first story to stand up on its own as something special, and this means spending along time on puzzles, dialog, etc. But long term that is not what I want, and not what I planned for. For me, the stories and art are simply a McGuffin, an excuse to explore amazing ideas.

Over the past few years I have adapted and changed my plans in response to what users seem to want: nice art, some action, a story that is very close to the book. Each change has been very small and appeared very sensible, but one by one they add up to create a project that takes longer and longer to produce, and is is less and less what I care about.

I'll tell you what my dream game would look like:

• Low resolution pixel graphics.
• Minimal animation.
• Thousands and thousands of scenes covering all known concepts and places.
• Each story would be complete in maybe three or four hours, and there would be hundreds of them. This ease of use amnd vast catalog would be a major selling point.
• Another major selling point would be long conversations with people. These would arise naturally: the game would be easy to code, so any spare time will naturlaly go to the dialog.

One by one the original goals have been compromised.

• Art: Low resolution pixel art seemed like a sales killer, so I went for higher res sketchy art instead, Sketchy art was also a sales killer, so I finally went for smooth art and a minimalist style instead. This is just as quick to produce (hurrah!) so didn't seem like such a bad idea. However...
• Smooth detailed art leads people to expect smooth, detailed animations. Even minimalist animations can take a long time because every line has to look right.
• The new art meant I had to cut out the "thousands of scenes" idea. And the memory leak in Sludge meant I had to move to AGS, requiring a further streamlining of the game, focusing purely on Les Miserables and not on anything else.
• The smaller game world meant I could no longer rely on exploration and serendipity to provide the entertainment. Instead I had to spend more time on Les Miserables itself, and the personalities involved. This takes much longer and frankly doesn't interest me. What I love about Les Miserables is crime, punishment, redemption, poverty, the class system, questioning right and wrong, conflicts between two sides that can both be seen as right (in real life Victor Hugo fought AGAINST the barricades, remember), God, fate, hedonism, desparation, what makes a nation state, etc. The other stuff, the romance, the chase, the gun battles, really don't interest me at all.

All of these changes have seemed quite sensible on their own. I know that people would rather read a romance than a treatise. People would rather see a beautiful painting than flat pixels. But the cost is very high: a single story takes much longer to produce. It's much less rewarding for me, as developer. Higher demands (coupled with the fundamentally unusual concept) mean that I constantly have to overcome technical bugs, which slows things down further. And looming in the future is the possibility that the final stage, secondary dialog, will take longer to get right.

The current plan:
Put in as many hours as possible, try to have something that pretty much works ready some time in May (good grief!" May begins in 3 days!). Continue to add dialog and polish, and respond to testers. Have it complete in early July. More testing and polishing, on sale mid July.

Realistic scenario?
To do this properly, it may take linger and longer and longer. I'm currently playing the game through, making "minor" changes. This could easily take until May 1st. The next step is adding the precise details to all 185 events. At six events a night that's May used up. It could take the same amount of time to add the essential core dialog. That's the end of June. Maybe another month to add basic non-essential dialog: end of July. Another month for changes suggested by testers: end of August. Another month for polish: end of September. That's another two months on top of the already stretched schedule. And what if other tasks end up taking a long time as well? The game could end up being a whole year later than the original date of December 15th 2007.

What my heart wants to do:
I want to rip out all the uncompleted stuff that takes forever. And either do without it, or replace it with something more focused.

I'm not stupid enough to completely redesign the game at this point. I very much like the core code, and the story engine that I blogged about a few weeks ago. Those are the core strengths of this project. And the story (based on 185 events from the book) is sound. I don't see any problems with those. What I do see problems with, and this is the cause of the expected delays, is take a long, hard look at the complexity. Look at the stuff that adds time to the development process but does not add a great benefit to the finished product.

I'm going to spend the rest of today doing a cost benefit analysis, looking at:

• where the development time is going and where it is LIKELY to go,
• what are the actual (or likely) benefits to the end user
• are there any changes that would end up saving time and improving the balance?

Progress report. I'm currently working therough the first complete play through of the game, making changes as I go. Naively, I thought this would take just one evening: just click, click, click 185 times and the game is done. But I'd hugely underestimated the time that "minor" changes can take. Two examples from yesterday:

Example one. Click on Javert to make him notice Fantine. In order to have dramatic tension this needs to include the "snow down the dress" incident. No, I'm not going to animate a snowball fight (that would take at least a week), but I did need to create a new view of Fantine plus a new guy (the snowball thrower).

Example two. Click on a street kid and on some junk to get him thinking about bending wire. Originally I planned to just use the box of metal junk in the Montreuil factory. But when playing through, this did not make sense. So I needed to re-read the book, and find an alternative. That took half an hour.

Multiply this up by the 185 events in the game, and you can see how a "quick play through" is actually taking two or three weeks. The next stage is to go through again and add cut scene details. Not the big animations (thouse are already done) but making sure that everyone is in the right place. There are nearly three hundred characters, and most of them don't move (it takes a month to create a full set of animations for a single character). So the cut scenes need to be carefully designed so that whenever someone enters or exits, the camera just happens to be looking the other way, but in a way that the user does not notice. All of this stakes time.

After the run through (now) and the cut scenes (next week) comes adding the basic dialog. This will be the minimum functionality, enough to make it clear what is happening, but without any rich Les Miserables goodness. That is the stage when it will be playable by someone else.

The earliest date for releasing test copies is May 15th. The latest date (based on current experience) is May 30th. Expect more details as the dates get closer.

I need ideas for how to get help with dialog?

General tidying is taking a lot longer than expected. I'm looking ahead at the schedule, and there's one obvious solution: farm out the secondary dialog. I had a month set aside for going through Les Miserables (the book) and adapting it into incidental comments, suitable for use in random conversations with strangers.

This is the kind of task that any reasonably educated person could do - I certainly don't have special talent in that area. But any payment would have to wait until after the game is sold, and would not be very high. Something in the order of a month's hard work for five hundred dollars, to be completed by the start of July but no payment until one year later.

I'm sure there are educated folks in India who would consider that good pay, but how to guarantee quality?? (And how could they be assured of payment?) My ideal solution would be to find some Les Miserables fans who'd do a few hundred lines each for the fun of it and see the money as just a nice bonus. But why would they accept such a tight deadline?

If anyone has an idea how to get dialog writers under these conditions, I'm very interested!

Stil working away. The characters are beginning to develop personalities at last. I'm seeing some of them in situ for the first time, and it comes as quite a pleasant surprise. I think Myriel (the bishop) works best so far. Which is probably as it should be, as the story begins with him and he is Valjean's spiritual anchor.

Still working through the game, event by event, making sure all the right characters are in place, they look right, and so on. This morning, for example I combined the "normal character" and "bystander" code so all characters can be treated the same. I also changed some backgrounds (outlines characters and soft backgrounds don't always work I discover). And created another character (the "well dressed man" bystander was far too small), and did various other tweaks. I'm hesitant to upload more screenshots until this stage is complete, as the difference between a "good" screen and a "bad" screen is a fine line.

Merrily testing and refining away. Got the "give me a clue" button working, made it so that the moving people stop when you press the space bar (otherwise they can be hard to click), added a farmer's wife character, moved a lot of code out of rooms to a single text file where it's much easier to find and use, decided to put Bahorel in the library in the early days, and so on. Just another day...

Another day, another five minute change that ends up taking five hours. :)

Some of the characters look too grey, so I need to give them a slight hint of color. In Sludge that was easy. You just create a light map with that color at the place you want. but in AGS it's hideously complicated. For one thing any tint is global or nothing. (there is a limited ability to tint different areas of the screen, but you can't change color and brightness at the same time). Worse, the tinting command involves three colors AND saturation (which is surely redundant?) AND luminance. And experiments show they don't behave as you expect: changing blue changes green instead [edit, this may have beena problem in my coding, but I'll try and blame someone else anyway], setting luminance to zero or 100 both result in white, changing the red from zero to 100 results in pure white until 8% then it suddenly jumps to pure red. luminance and saturation seem to be the wrong way round. And so on. Hoo boy, getting subtle colors out of this will be a LONG job.

I'm not really complaining, as AGS is free and pretty much anything is possible with AGS once you get your head round it, but for tinting maps, Sludge was infinitely better.

EDIT: Now I'm happier! I made a little function that lets you change a character's color using the keyboard. It's quite fun - you choose a characters, and WHOOSH! They're all black, or bright yellow, or whataver color you want. Users won't be able to access it (it doesn't really fit into the theme of the book) but I love it. :)

So busy today I almost forgot to blog!
Been busy making sure the early puzzles/events work smoothly.

You know, looking back on the development process, the one BIG, BIG thing that I did not (and proabbly could not) plan for is/was the ENORMOUS amount of time needed to learn anything new. Right back at the start I just calculated how long each task should take, assuming I did it right first time. But of course the first time you do something it has to be done three or four times, or else done extremely slowly. Still, the good thing about doing something with no experience is, when you finally get it right then you DO have experience, so the next time you do it IS fast.

The work now is becoming much more pleasant, because at least I'm dealing with actual characters on screen and not with just lines of code. I'm getting closer and closer to the experience that an actual user would have. Which is reassuring.

I just had a great idea!

One of the MOST time consuming and LEAST enjoyable parts of making this game is trying to second guess what users will think. For every "correct" action I need to anticipate five or ten "incorrect" actions and then create some kind of response. So it's five or ten times as much work, but half of it is wasted because who knows what users will REALLY want to click on?

So I made one of those executive decisions you've all come to know and love. Instead of spending a whole month (or more) on trying to second guess the user, I'll send the unfinished game for testing and ask testers to tell me who they click on during each puzzle. That way I can make sure I write dialog where it is needed, and not where it isn't.

Result: the test version is out sooner (middle, not end, of May), and I spend more time on what matters.

I was just reading a review of The Lost Crown. This is another British one-man developer, but he's a little further down the road than me (it's at least his third game) so I thought it was interesting to draw some comparisons. I think it illustrates what I'm trying to do different with this game.

• His game looks nicer, in a traditional sense - proper 3D characters, the familiar style of art that every game uses. But my game will stand out as more distinctive. Which if course could be a very good or a very bad thing. :)
• Similarly, his game uses the familiar tools of inventories, conversation trees, visible avatar, etc. I suppose if nine tenths of adventure games use these methods then they must work. But I've never been one to follow the well beaten path.
• His story sounds very polished - he is able to deliver genuine emotion. No doubt he is a very talented guy, but his experience in making previous games will help. And of course using familiar tools (see previous points) will also help. I am hoping that I can achieve a similar emotional connection through heavy use of Victor Hugo's ideas, but of course only time will tell. And as I said, he has the benefit of more experience.
• He appears to deliver one game every two years. That is very good going for a one man band. Of course I aim for one every six months, but the style is completely different.
• He requires a 3D acceleration card and reasonable spec in the computer. Nothing like the requirements of big name games, but this shuts out the non-gamers that I hope to reach: the kind of person with a five year old computer who has no idea if they have a graphics card or how much RAM they own.
• Judging by the games that are sold successfully, his kind of story is exactly what people want: A new character, a new problem, the problem is solved, the audience is entertained. End of transaction. But I find it hard to care. I want something more lasting. I want stories that teach me something about the real world. I want stories that are so profound that people are still talking about them hundreds of years later.

I guess it comes down to the whole reason for making my game: nobody makes the kind of game I want to play so I' making it myself.

Wow, sometimes coding evena "simple" line is like wading through treacle. I just spent four hours trying to track down a strange crash, finally traced the offending line, and tracked down references to it on the message boards. Guess what: I managed to conveniently combine (1) an undocumented feature (so the code was doing something fundamentally important that is never mentioned in the documentation) with (2) a known bug.

Aaaarrgh!

I suppose it's a good sign, since it means my own code was not fundamentally broken, but that's another day wasted...

It's taken me a couple of days longer than planned to get the final parts of jumping between games working properly. But this shouldn't affect the overall schedule, as I had originally planned to do that last of all, after beta testing. I changed my mind and did it this week instead because (a) I like to get unpredictable stuff out of the way, and (b) the idea of joining stories is the core of this game, so I want it to be very plain to users, even when there is (at first) only one story.

One trivial point is that the cross game code involves writing the essential variables to a text file in between games. So anyone can open the file and see the name of every single event in the game! It's like a built in walkthrough. But the text file doesn't have the events in the right order, so I don't know what effect that will have.

I got a shock this morning on another board (a technical forum). I'd asked a question, didn't get a reply, so added a reply to my own question. Then I DID get a reply, so I added a humorous thankyou. A moderator then jumped in and said (quote) "this is a forum and not a chat room. Don't do that again."

I have never before seen a forum that forbids ANY off topic comments or friendly banter. Is this common? Have I just led a sheltered life?

It seems to me that a small amount of friendly banter is useful on a technical forum:

• Banter lets the other guy judge whether or not you are wasting their time.
• It reduces the risk of the other person becoming bored or offended (these are serious risks when posting anonymously).
• It allows emotion to come across: many questions, even technical ones, are merely covers for a deeper underlying issue.

What is your experience? How many serious forums demand total and complete on-topicness?


PS the same forum, which shall remain nameless, also has a rule that forbids the same person posting twice in succession, and generally tries to reduce the number of posts, which leads to people editing earlier posts rather than adding updates. This is clearly useful to people who read a finished thread (as threads are shorter), but makes the thread more confusing while it is ongoing, because each post may cover multiple angles, and earlier posts can change without warning.

The really interesting thing is that nine tenths of the output of that forum is intended for comedy purposes, but the tone of the all moderators is completely humorless.

PPS I would have posted this question on the "other topics" section of the board in question, but I'm beginning to think that they don't speak my language.

The jumping between games seems to work well. I still have to finalize how many of the variables to save: there are hundreds (potentially thousands) of variables remembering exactly what point every character is in different conversations. Saving these things between games is inelegant - it takes up far more space than it needs to, and is so complicated that hard-to-find bugs become likely. And to be honest, I don't think most users would even notice if all secondary conversations are occasionally reset to zero. But I'll need to think about it some more.

In other news, I tweaked the story code to make it more focused on the conflict at any moment. In an adventure game (or ANY game) it's essential that the user always knows what they're supposed to do. And in a novel it's essential that you don't go too far before reminding the reader of the current dilemma. Truly great authors like Hugo and Tolstoy can afford to spend several chapters describing history and whatever crosses their mind, but we mere mortals have to use cruder means (like conflict) to keep the user's attention.

By the end of today I should have ELEVEN different games working! OK, nine of them only have one screen and some links, for testing purposes. This pretty much fixes ten of the titles of future games, since they are all linked from locations in France. These are not the next nine games (I don't have a direct link to , but they are games I have definite plans to do in the future. When they are finally done, Les Miserables will magically link to them. They are:

1. War and Peace
2. Journey to the Centre of the Earth
3. Origins of the Gods (i.e. the Theogeny by Hesiod)
4. Treasure Island
5. Julius Caesar
6. Henry V (I couldn't think what else would link to northern France, and I plan to do the major Shakespeare plays anyway since they're nice and short!)
7. Fairy Tales
8. Around the World in Eighty Days
9. The Divine Comedy
10. The Descent of Man

That last one is a bit out of left field, and won't be an early game. But would love to make games based on some classic non-fiction texts (or yiou can call it fiction if you're a fundamentalist Christian!) To me the history of mankind's struggle is far more exciting than any made up story.

I'm looking at how to connect different stories. I need to work this out now, so that when the second story comes out it will just plug into the first one, so you don't have to download the first one again - you'll just find out that the first game is magically bigger than before! It's important to me that people should be aware of the big picture, right from the start. The lure of this game is not that it contains Les Miserables, but that something as substantial as Les Miserables is just a small part of this epic.

When people download the first game, if they try to explore too far away, like going toward the hills or the sea, they'll come up with a message that says "this road leads to Rome" or "this path leads to Dante's forest" or "this boat takes you to the Greek islands" and a note that that other story wasn't found on the computer, but to check EnterTheStory.com for the latest stories. I'll have to work on the exact wording, as I don't want people to think the game is broken or has missing parts. But I do want people to look forward to the next part, when they can eventually continue down that street or across that hill and see what they find.

Just revised some of the main pages on EnterTheStory.com. I'm not a great writer, and tend to ramble incoherently, but the index, about and gameplay pages are at least better than they were. They still need more work of course.

On the coding front, I found another set of messy code that I was able to tame, and that feels good. Coding is very much like entering a room full of insane monkeys, wrestling with them for months, and training them to stand in a line and serve tea, open doors, fetch my pipe and slippers, etc.

Unimportant but slightly interesting piece of news this morning (to me, anyway). Anyone who's explored the "other" link on EnerTheStory.com knows that I'm interested in comic books, particularly the Fantastic Four. Not because of what they are (utter rubbish), but because of what they once were, and could be again. All good fiction is simply a MacGuffin for exploring ideas that apply in the real world. The Fantastic Four was originally an experiment in "what would happen if superheroes were real" and they were designed in such a way (a family, explorers, publicly known) that almost any possible story could be explored in their book. Unfortunately they became so popular in the mid 1960s that their creators stopped experimenting, and started milking the brand. Since 1968 the Fantastic Four have been stagnant, cliched superheroes, mostly utter rubbish, and sales have declined continuously. I do not like the current FF, except for a faint nostalgia and hope that maybe one day they could be good again.

So I keep an eye on the comic news, just in case anything changes. This morning an item caught my eye: http://www.comicbookresources.com/?page=article&id=15944

It's a Fantastic Four mini-series where the team enter famous novels and interact with the stories. And the author even makes a big deals about the novels being out of copyright (that's part of his story). And one of the stories they enter is Dante's Divine Comedy. Sounds familiar? My two worlds (Enter The Story and Fantastic Four) are coming together!

From the sounds of it, it's being played for laughs, and it doesn't stick very close to the originals (I don't recall any superhero battles in Jane Austen's novels, but I'm not really an Austen fan so I could be wrong). So once again it's going against what I'm trying to do. I want to get the actual power of the novels themselves, I don't want to just make use of the characters or update them. I want to use the novels as a way to explore reality, not as a way to escape from it. It's a completely different mind set.

But still an interesting coincidence.

Actually, those with long memories may recall that I plan to feature the Fantastic Four in this game eventually, but they will be completely unrecognizable to anyone except to the few people who "get" the original concept. No danger of copyright or trademark violation.

I did it! I did it! I actually played the game!! Not all the way through, but far enough that I'm pretty sure it will play all the way through except for the inevitable minor and easily fixed bugs.

As I wrote yesterday, it still needs dialog, clues, minor cut scenes, and hundreds of small changes, but for the first time it is actually a game and not a proof of concept!

This would probably be a good time to work out how long it actually takes to play all the way through. I really have no idea. But there are 170 events. I know exactly where to click for each one, and that takes maybe 20 seconds per event. So it's probably reasonable to guess that a hardcore gamer would solve each "puzzle" in one minute, and a normal person might average three minutes per event, so it will probably have 8 hours of gameplay for a typical user. That's shorter than a big name game, but I have to start somewhere. I think that's reasonable for 15 bucks, and of course what really matters is if those 8 hours are rewarding.

Anyway, the point is, it's now a game!

Hopefully by this time tomorrow I should be able to play the game all the way through from start to finish. (Crosses fingers, toes, etc.) Of course the dialog, clues and cutscenes won't be al finished, so most of the time it will be a question of click in the right place, see a message "test mode: task XYZ , event N completed)" then repeat 168 times. But it's a major milestone.

The last few days have been intensely boring, going over old code. And also horribly slow and unrewarding, since it required rethinking things I've already thought through before. So it's good to be through the desert and seeing real results on screen once again. At last!

Juts uploaded a different photo for my profile. My kids think the other one was too creepy.

Made it over another psychological hurdle.

A few of weeks ago I completed the list of plot points (who does what, what happens next, what needs to be in place, what clues should be available). There were 33 plot points, which made it easy to see the outline of the story (the barricades were one plot point, the rescue of Cosette was another, and so on). But this also meant the individual plot points could become very complicated. It was clear that, as I refined and polished them they would become unfeasibly messy. For example, something as "simple" as the barricades sequence contains dozens, perhaps hundreds of individual actions that all inter-relate. The simple code would then become a tangled mess, and bugs would be everywhere and would appear as fast as I could eliminate them, like cutting of the hydra's head and having two more replace it. That was a few weeks ago.

So at that time I replaced the 33 plot points with 200 events. Every event was very clearly defined, and written in almost natural language, so it was far better code. But the down side is that I had to convert the old code, which isn't as simple as I thought, because the thought processes behind each style of code are very different. And the complexity of the old style allowed me to think it was "nearly" finished when it wasn't. And having two hundred sections of code is just daunting... I'd spend three hours on it (effectively a whole day), have a sore arm and strained eyes, and only have maybe five items done. So it was hard to be motivated. So over the last couple of days I spent more time on may web sites than on coding. But I did finally do enough coding to break the back of the problem. I've done enough work on the 200 events that they no longer appear daunting, and I can trot through them at a fair pace, maybe just ten minutes each.

I will still need to go through them again to give names to all the clues, etc., but that part requires less thought and can often be left until a later stage, so it isn't scary any more. I'm back on track. :)

I had a day off from working on the game today. A day off! I know. Shocking. Back to normal tomorrow.

Not much to add. Still converting the 33 "plotpoints" into 200 "events."

Putting Dante next changes the natural order of which stories come after that. So here is a revised schedule for the next few years. Obviously it's subject to change! But I need to plan at least two years ahead for financial reasons.

• By APR 15 2008: all events are complete ("events" replace the old concept of "plotpoints" - see recent blogs about new code)
• By APR 30 2008: the game is fully playable, with temporary dialog.
• By MAY 31 2008: final dialog in place, sent to beta testers. Note that this is a couple of weeks later than originally planned, as the dialog really needs more time than I originally allowed.
• By JUNE 30 2008: second beta released. This is the version I am happy with, having polished it for a month based on beta testers' comments.
• By JULY 31 2008: beta testers should be more or less happy (I hope!) so the game is released. Yes I know this is later than June, but I always said I would not release the game until beta testers were happy. So I think this is the most realistic date. This time last year I was hoping for December 15th 2007, but 3 months were added for changing the art and another 3 months for changing the engine.
• By AUG 31 2008: Dante's Inferno is fully planned. With a shorter book, and the benefit of everything I've learned doing Les Miserables, this should be a much quicker task.
• By SEP 30 2008: all scenes should be complete. Expect the first screenshots. By this point I hope to be selling 20 copies of Les Miserables a month. I have no idea what will actually happen, but this seems like a realistic goal based on other downloadable games I've seen.
• By OCT 31 2008: all sprites should be complete. Expect daily pictures of demons, angels, giants, etc.
• By NOV 30 2008: Dante should be fully playable, with basic dialog.
• By DEC 30 2008: Dante sent to beta testers.
• By JAN 31 2009: Dante is released, exactly six months after Les Miserables. I hope to be selling 25 copies of the combined game each month at this point. Which story comes next? Thinking about it, the natural third story is Shakespeare's Julius Caesar. Why? because (1) it introduces another major theme of the game: Shakespeare! (2) it's short (3) it's well known (3) it has many connections with Dante (through Virgil, shades/spirits, Brutus, etc.). It's also the natural place to see the origin of Peri, who we probably meet in person in Dante.
• By JULY 31 2009 Julius Caesar should be released.

What comes next? Well we may as well tie up all the divine loose ends once and for all. I'm very tempted to make the next game Hesiod's Origin of the Gods. What a title! So many games allude to these stories, but I don't know of any game that has presented them directly. Let's go right to the source! This will finally reveal the big story behind Peri, the meaning of life, and all that stuff. Or rather, it will reveal whole new levels of story on a larger and deeper level. I have the next five stories roughed out based on that (I've mentioned most of them before: Fairy tales, Journey to the Center of the Earth, the Time Machine, Esoterica, and probably Around the World in 80-Days). This blog entry is long enough so I'll leave it there.

Sometimes it hits me how DANGEROUS this project is. For me, anyway. I'm not just creating my first game, I'm not just relying on a genre and business model that has never been tried (classic serious novels, world that builds in instalments), I'm also breaking all the rules of traditional gameplay. Most games appeal to primitive brain functions like fighting, racing, or nurturing (I mean primitive in a biological sense, not a moral sense). A few games appeal to higher brain functions for puzzle solving. Adventure games fall into this category - the inventories and conversation trees are basically puzzles dealing with simple objects or simple concepts, and the enjoyment comes from solving the puzzles, with secondary enjoyment from the entertaining characters. But my game swaps this around putting the characters first and the puzzles second. My game goes for broke, and tries (I must be crazy!) to appeal to higher concepts, in the same way that great literature appeals to the highest ideals and abstract principles.

Yesterday I just had to stop and ask myself "what was I thinking?? An adventure game without inventory or conversation trees or even verbs? Expecting people to be interested in dialog about God and faith and poverty instead? Am I mad??? Who would play it????"

But then, the same question could be asked about any literary novel. To many people, "literary" just means "boring." Long chapters discussing the merits of political theory or religious tradition, then chapter after chapter with no action at all? And the hero ends up lonely and misunderstood then dies? Who would read a book like that? Where are the car chases? Where are the murders? What kind of an action hero is that wimp Marius? And of all the revolutions to choose, why choose 1832, when almost ANY other revolution would have been more interesting?

So I'm taking quite a risk! Anyone who approaches this game expecting something like other games, will be disappointed. Or at least, will probably finish the game very quickly. The game's only merit is in its attempt to cover bigger topics. The deeper stuff MUST be entertaining or the game will fail.

Take a concrete example, from yesterday. I was adding dialog about Napoleon's campaigns. My first attempt ended up looking like a list of names and dates. How boring! It's so easy to be dull. But this is Napoleon! The antichrist! (According to the opening chapter of War and Peace). The one man army who conquered Europe! The focal point of modern history (well, that's one way of looking at him anyway). One of the most amazing people who ever lived, a man whose achievements have never been equalled in modern times! Yet it's so easy to make him sound dull.

Another example (one I mentioned a few days ago) is fence posts. One of the scenes in the game has several fence posts. If a character looks at a fence post, isn't that the most boring event imaginable? Yet fence posts represent land boundaries. Land ownership is the basis of almost all wars, it's the basis of almost all wealth, it's the basis of slavery and poverty and hopes and dreams and nostalgic idylls of country life... even a fence post can be amazing if presented in the right way. But it's not easy.

So I'll probably spend a couple more weeks than planned just playing through the game and changing it, finding the weakest ideas and changing them to be the strongest. This game stands or falls on its ideas.

Today I'm converting some old dialog across. I originally wrote it for Perry to talk to people (so you would choose from various topics). But now that Perry is a character who won't appear in the flesh until the second release, so the game is more like the book - you don't know what a person will say until you talk to them. In one way this is a bad thing, because you have less choice of topics. But in another way it's a good thing because you're more likely to stumble across the unexpected. The idea of choice in games is often misleading - if presented with a choice of "talk about a convent" or "talk about slang" why would you choose either? You don't know if a toppic is interesting until after you've heard it. Authors know this. They don't create list of topics like an encyclopedia, they create intersting characters and surprise you.
.
Actually there IS a strong element of dialog choice in the game, because there are so many characters to choose from, and they don't all respond in the same way. Javert and Fantine, for example, are unlikely to think the same on anything.

Part two of today's blog: the short term quandary:

I said recently that I had over 4,000 monologs to write. Obviously that's not practical, so I spent the day organizing them into groups that could use similar responses. Which was good, because I was able to focus on topics that are important to Les Miserables: justice, poverty, faith, etc.

But now I'm left with a problem: I have about 200 boring topics left. Like when you click on a person then click on a wall, or click on a piece of grass. I can't just have the person say "it's a wall" because the person you click on might not be in the same room. I need plausible dialog that anyone might say if they suddenly and inexplicably thought about a wall or a piece of grass.

These things are not important to the game, but they need to be there to add richness, and to be decoys from the puzzle objects. Should I really go to the trouble of writing dialog just in case the user decides to look at the wall or the grass? Or should I just write some default statement like "I don't know why I suddenly thought about that"? I mean, how often are users going to click on the unimportant stuff?

Ironically, the unimportant stuff is much harder to write than the important stuff. If a character thinks about Javert, or a church, or even a door, it's easy to think what they might say. But if the same character thinks about a wall or a piece of grass, what would they say? It's a tricky problem.

Two quandaries today, one long term, one short term. I'll blog each topic seperately. First quandary: which story should come after Les Miserables?

This decision has to be finalized now, because I will soon be writing the final dialog, and it should naturally lead to the next story (without twisting the original novel too much). For some time I have planned to create War and Peace next. This was why:

1. It is conceptually very similar to Les Miserables: a 19th century national epic.
2. Through Napoleon, one story leads to the other, and they are geographically similar (I could reuse some graphics!)
3. It reinforces the selling point that "Enter The Story" is about BIG stories.

But on the down side...

1. It's a very long story (and I'm still learning) so the second game would almost certainly be late. Nobody would believe the core concept of "a new story every six months." It would also have serious financial implications.
2. Its similarity may be a big weakness, not a strength. What if people don't like nineteenth century novels?
3. The link between War and Peace and Les Miserables is problematic, because the second story is set BEFORE the first.
4. No visitor to this site has ever shown the slightest interest in War and Peace. Of course now I'll get a ton of emails saying "it's my favorite novel!" :)

So I'm strongly minded to make jump straight to the third story, Dante's Divine Comedy. And leave War and Peace til later. Why Dante?

1. It's short and easy to make, so would be ready by the end of this year, 2008.
2. "Dante's Inferno" has instant appeal to regular gamers.
3. It would be relatively easy to have one story lead into the other.
4. It would jump right into the big back story that links all the other stories (War and Peace would only add more teasers).
5. It is a radically different style of story, giving a beter idea of the range of topics to be covered in Enter The Story.

I'll have to plan exactly how Les Miserables merges with Dante, but the more I think about it the more I like the idea.

Very short blog. The power to our village is turned off most of the day today.