Friday Facts #105 - The Grey Zone

[Peter34]

One benefit of Steam, potentially huge, is visibility. Findability. Discoverability.

Also, with Factorio already having a very enthusiastic player base, it's quite likely that existing owners will flock to Steam, once it arrives, and give it very positive ratings.

[Gammro]

I don't think releasing on steam is going to be a benefit to the game all that much. All to frequently, I see big game launches on steam that look like they have had millions of dollars of investment into developing the game/story/graphics/engine etc.. but have low numbers of purchases even after weeks, it's sad in some cases. There are tens of thousands of games on steam, and I don't see steam's userbase growing that much. The game being bought directly from the creators is likely going to be the best option for the continuation of it's design period. Take a look at minecraft pc, they never partnered with any steams, and they did just well, plus not having a middle man taking some percent of the purchase.

Minecraft did not need to partner with steam because it's popularity blew up fairly early into development in a time when there weren't as much indie titles known to the mainstream. They reached a critical amount of users before they even had the chance to get on steam(as the platform was different back then), so I don't think it's a very good comparison.

In general, comparing to other successful indie games is hard because we only hear about the success stories and not of the failures. Because nobody has heard of those games. Steam doesn't guarantee financial success, but it is another outlet to advertise. And without any form of advertisement, you're guaranteed to fail.

Steam is also far from stagnant in it's userbase, some headlines from the past years:
2010: Steam Surpasses 30 Million Account Mark
2013: Steam boasts 65 million accounts, 30 percent growth since last year
sept. 2014: Steam Reaches 100 Million Users And 3,700 Games
feb. 2015: Steam has over 125 million active users, 8.9M concurrent

It's hard notice the growth in number of accounts without actively looking for it, but steam is still growing faster and faster. Of course, there's also a huge growth in games on the platform in the last few years. So it might be harder to actually get found.

[Peter34]

If it's of interest to Wube:

I first was told about 7 Days to Die by a friend, back in 2013, probably the last half of 2013. I then actually tried it (coop MP) at his place in something like April 2014, which was my first experience actually playing a voxel-based game (I knew about voxels from having seen Minecraft footage on TV, but had never really thought about their usage in a more serious kind of simulation game), and I fell instantly in love with it. I came to like the survival genre, especially when paired with crafting and building. I quickly found, in dev material, an overt promise that the game simulation would eventually be broadened sufficiently to let me play as one of my favourite fictional character, MacGyver. That thought hooked me even more.

Apart from being interested in 7D2D, I also took a more general interest in the survival game genre, especially non-horror surrival (I tried Robinson's Requiem back in the 1990s, but never got very far with it), and so every few months I'd browse some of the game categories in the TvTropes wiki, especially ones to do with games with a survival element.

That's how I came across Factorio.

Probably in the summer of 2014, but back then it still wasn't my inclination to use YouTube to watch footage of games I was curious about, so it was some time later, late October 2014 I'm sure, before I actually bought it. That got me instantly hooked. It's a fun game, and the main difference between that and 7D2D (which I still like a lot) is that Factorio retains a game-like element if you disable enemy attacks, whereas in 7D2D if you disable zombie spawning then the only reason to play is to utilize the opportunity to learn the game, to improve your understanding of the game's survival, building and crafting game mechanics, which can then be utilized later when playing for real (i.e. with zombie spawning enabled).

That's three ways where I might discover games: Being told by friends with whom I share certain tastes. TvTropes, mainly its game genre categorization system. And YouTube videos, mostly if there's a youtube gamer I'm already following who tries a new game (which actually hasn't happened to me yet, although it's close with Empyrion: Galactic Survival which I've seen Kage848 play).

I think for me it's very much about looking for games that cater to my somewhat niche interests. I've come to dislike twitch game, and I long ago stopped liking games focused only on combat (because such games tend to spend very little time and effort simulating things other than combat - that is they don't simulate broadly). I also prefer character skill over player skill, e.g. the lack of interventive control in battles in Crusader Kings 2 vs the direct control of company or battallion movements in games such as Total War. That's why I'm excited about skills eventaully being added to 7D2D, and why I'm unhappy about them being unlikely to be added to Factorio.

[Drury]

I don't think releasing on steam is going to be a benefit to the game all that much. All to frequently, I see big game launches on steam that look like they have had millions of dollars of investment into developing the game/story/graphics/engine etc.. but have low numbers of purchases even after weeks, it's sad in some cases. There are tens of thousands of games on steam, and I don't see steam's userbase growing that much. The game being bought directly from the creators is likely going to be the best option for the continuation of it's design period. Take a look at minecraft pc, they never partnered with any steams, and they did just well, plus not having a middle man taking some percent of the purchase.

Well as I said earlier, you can make a game about rocks and have it sell millions if you pitch it well. In fact, I believe this exact scenario has already happened, not even kidding here.

There's this indieapocalpse sentiment going on these days. Tales about how you go on Steam, there is no guarantee you make it through to your customers. It's a good way to scare small devs to try harder. And big devs not to forget about proper marketing. But in the end, contrary to popular belief, there is a formula for punching through the fog of shovelware. Factorio is set on a good course.

As you said, they already have their publishing method. People are already buying the game. It's not much, but there is the word of mouth going on. Is the game any good? Draw your own conclusion. That is a solid base for Factorio success. A lot of games that fail on Steam don't have any of that. They just kind of assume they can attract people with good gameplay. Without even providing a demo.

And hey, if the game fails to attract attention on Steam, no big loss there. They'd have to screw up much harder than that to close business, I'm assuming. Whatever the case, that is a nonsense. Factorio is like Titanic. You'd have to do something crazy like go full Steam ahead into glaciers to sink it. So let's not do that.

[slpwnd]

Thanks for the input guys, quite some valuable material there. Based on the discussion in the previous FFF (and also based on the amount of work we were running behind) we decided to skip the planned Steam November release and wait till January. So we have like 3 months for preparations and polish.

It is great to see that this discussion is taking a course of reason-based opinions to the point. Please continue if there is more you have to say.

As for distributing the game on Steam. The way I see it (and it was mentioned already) is that Steam gives us (potential) visibility. That means the opportunity to interact with players who don't watch youtube videos, are not active in the gaming community and haven't had a friend tell them about Factorio. Yet, still Factorio might be the game they will really enjoy. So our task here is to make the game presentation convey the right information to these people.

The trailer is the first step. I really like the idea mentioned by Drury of having a short gameplay video. We will put one together or maybe ask some of the Youtubers doing lots of Factorio videos to try to come up with one.

As for the presskit / textual game presentation. Yes, we don't have a "marketing / presentation specialist". I will see if it is possible to find someone in the given time. The challenge would be to find someone with experience in (PC) games I guess. However there is a good chance we will have to do this ourselves. In that case any recommendations are appreciated.

[ratchetfreak]

Now about modding:

I've already had my rant on this forum about it but try to see things from the modder's point of view when adding new features.

For example the combinators, what is the first thought a modder has when seeing those? I want to create my own combinators to do X.

To do that a modder needs access to the circuit network (being able to read the signals on circuit and output to a network).

Another is creating a electricity producer. Now the only options are a solar which is tied to the solar cycle, an accumulator which won't do what you want or a boiler which requires feeding it heated water. What is preventing you from adding a "electricity-producer" entity with a priority and an internal buffer to be filled by the mod code?

[Drury]

Thanks for the input guys, quite some valuable material there. Based on the discussion in the previous FFF (and also based on the amount of work we were running behind) we decided to skip the planned Steam November release and wait till January. So we have like 3 months for preparations and polish.

It is great to see that this discussion is taking a course of reason-based opinions to the point. Please continue if there is more you have to say.

As for distributing the game on Steam. The way I see it (and it was mentioned already) is that Steam gives us (potential) visibility. That means the opportunity to interact with players who don't watch youtube videos, are not active in the gaming community and haven't had a friend tell them about Factorio. Yet, still Factorio might be the game they will really enjoy. So our task here is to make the game presentation convey the right information to these people.

The trailer is the first step. I really like the idea mentioned by Drury of having a short gameplay video. We will put one together or maybe ask some of the Youtubers doing lots of Factorio videos to try to come up with one.

As for the presskit / textual game presentation. Yes, we don't have a "marketing / presentation specialist". I will see if it is possible to find someone in the given time. The challenge would be to find someone with experience in (PC) games I guess. However there is a good chance we will have to do this ourselves. In that case any recommendations are appreciated.

The best marketing you may get would be out of partnering with a publisher, but I'm not sure if that is what you want. I take it you prize your independence too much to tie yourself up like that.

There are marketing managers out there you could hire, as mentioned. I know exactly none, unfortunately.

Last but not least, simply getting in touch with the indie developer community. Personally. They're weird but generally harmless. Most of all, they have experience with releasing games.

Speaking of which, will you attend http://www.gdsession.com/ ? Never been there, but there was Vávra apparently, so that's a seal of quality.

[Boogieman14]

That TotalBiscuit video mentions Simon Callaghan as someone who knows how to market a game. Takes a little bit of google-fu to find him inbetween the horseracing trainers and musicians, but it seems http://www.johnnyatom.com/index.html is his website. Might be worth shooting him a message.

[slpwnd]

Speaking of which, will you attend http://www.gdsession.com/ ? Never been there, but there was Vávra apparently, so that's a seal of quality.

Me and kovarex were there last year. We even had a short Factorio presentation on Friday before the conference in the bar (but it was loud, overcrowded and with poor technical support). Unfortunately this year I will be in the UK at the time of the conference. But for sure there will be someone from the office.

[cpy]

I hope factorio will have great steam launch but as far as i go, i only use factorio web to get new releases. This is one of really good indie games out there and i don't need it on auto update on steam to ruin something i made or made with friends. I just can't wait for space.

[Drury]

I hope factorio will have great steam launch but as far as i go, i only use factorio web to get new releases. This is one of really good indie games out there and i don't need it on auto update on steam to ruin something i made or made with friends. I just can't wait for space.

Good thing you mentioned that.

Devs - when Factorio gets on Steam, you can have multiple versions of the game available at the same time. I don't have a game on Steam, but from my (user's) perspective, these show up in the BETAS tab in game's properties. You can have all previous Factorio versions available from there and have the same functionality as downloading DRM-free from the website without any automatic updates taking place. It's even more convenient as you just pick from drop-down menu, let it automatically download the previous version and that's that. Euro Truck 2 does this. More games need to do this.

Some games like Skullgirls opt for having multiple instances of a game available as a separate product altogether that is given to all owners of the parent product automatically, but that seems like a clunky solution to me, Now I have effectively two games in my library, Skullgirls and Skullgirls Beta. They both update as they normally would (regular Skullgirls practically never, but there's nothing preventing it).

[jockeril]

can't read through 4 pages X20 comments on each right now but first thing I noticed is a text mistake:

* Platforms: PC only - Microsoft Windows, Max OSX, Linux

It says <max OSX> - is that some new OS I haven't heard of (not possible) or a typo ?

I also think that Factorio is not a tower defense game... or a very good (and interesting) one !

One final word - I think it's a good idea to invest in a professional press relations person to design the PR page (and advise on release timing) - it's a one-time deal and it can be a turning point or a distant memory... and like -root says: GO BIG OR GO HOME !
(or just ask #notch how it's done... )

good luck to WUBE & #Factorio on steam

[Paranoimia]

auto-generating the Lua API documentation

This is by far the most exciting (to me) issue mentioned in the Friday Fact; when might we actually see it in a roadmap?

I wanted to get into modding, but the current state of the documentation made it very frustrating. The only decent way to learn how things worked was to look at existing mods, but that doesn't help when you want to try doing something that they don't do.

I had actually tried making a tool to auto-document existing mods (to help understand and learn from the objects they use; which is why this issue is of interest to me), but only had limited success due to the format of mod data. Would be nice to hear more about this, so I know if I need to give it another shot or not

[kovarex]

auto-generating the Lua API documentation

This is by far the most exciting (to me) issue mentioned in the Friday Fact; when might we actually see it in a roadmap?

I wanted to get into modding, but the current state of the documentation made it very frustrating. The only decent way to learn how things worked was to look at existing mods, but that doesn't help when you want to try doing something that they don't do.

I had actually tried making a tool to auto-document existing mods (to help understand and learn from the objects they use; which is why this issue is of interest to me), but only had limited success due to the format of mod data. Would be nice to hear more about this, so I know if I need to give it another shot or not

Did you take a look at this?
https://forums.factorio.com/wiki/inde ... ua_objects

[DaveMcW]

Half the pages in the wiki are still using the 0.11 property names.

[Oxyd]

Half the pages in the wiki are still using the 0.11 property names.

I just browsed through it and it seems that most of it has been updated to the new names.

Nevertheless, this is one of the motivations behind generating the docs automatically: The script automatically tracks the relationship between our internal C++ names and external Lua names. So, were this script available before 0.12, updating the docs for the new names would've been a matter of simply re-running it.

[voyta]

Looking forward the Factorio Saturday Facts

[Brambor]

Yeah, no chance they will get the FFF in time. FFF still could be at Friday, but the #106 :/

[Paranoimia]

Did you take a look at this?
https://forums.factorio.com/wiki/inde ... ua_objects

Yes, that was part of the "very frustrating" comment. A wiki is not a good place for documentation.

From the perspective of someone trying to get into modding (ie. knows nothing about the API, like me), having to go to each and every single page to learn the little tidbits of information that are inside is very tedious. Information should be concentrated together in relevant chunks, instead of spread out over the number of wiki pages there currently are. Formatting and readability is another issue; a wiki can make it difficult to display the information in a good way. And manually updating every page can be very time-consuming.

What I was looking for wasn't even in the wiki at all, or at least I wasn't able to find it. I was trying to add an extra tab on the inventory menu to create a "quick-find" section for commonly crafted items. I had to resort to looking at an existing mod (thanks, bob) to find out they are called item-groups, how to have them actually show up, and how to have items show up in them. After all that, I ran into other issues that were vague and confusing, so I just stopped trying. This was why I thought of making a tool to document mods to learn from them; it was better than learning from the API documentation itself.

Searching the wiki now for item-groups, I can find the Group page, but I wouldn't have been able to understand what Groups were just from the information on the page.

I don't want to sound too harsh here, so I'll provide some context for my dislike for the documentation. Documentation is a bit of a touchy subject for me right now.. you'll see why

I'm a software developer myself. I'm used to using libraries and APIs that rely on their documentation to attract and teach developers how to use it. Here are a few examples of docs that I've worked with (and find easy to work with):

• * http://backbonejs.org/
  * https://developer.mozilla.org/en-US/doc ... /Reference
  * http://yui.github.io/yuidoc/api/

If it can't teach me how to use it, then why should I put that much more effort into wading through the frustration to figure it out? There are alternative libraries I could use.

Another thing is that I just went through this process myself. I just got a new job, and our first task was fixing all of the documentation for our public API. Yup, that's right: I just spent two months doing the exact thing you mentioned as a side-comment. And it was great; it'll save me a lot of effort down the road. We have a build process that automatically lints, tests, and documents our code. People can produce more with our service, and I can spend less time supporting developers having a hard time with our docs. Win-win.

Guess I'll wrap this up.. for a game with a core aim to be very moddable, the documentation makes it difficult to want to mod it. I really hope the documentation gets the attention that its been lacking.


Edit: I just read the latest FF. Think of the EEI level for modders as well

[ssilk]

Yes, the wiki is a but frustrating.

In the end what only counts is the source code. How many times did I read docs, that just lied? That had wrong parameter description, or that implied a way of usage, that was never implemented? Countless.
When I come to that point, I look into the code. What does he tell me?
I can see for example: don't ever never use that function. or I see: it can solve the problem much more elegant,if using it in this context. An automated process to generate the docs out of the code helps a lot, but such docs tend to be taken out of comments, not code. Comments in code lies.

What counts in the end is running code. Not text. Otherwise it's like a salesman that tells the customer, how cool a product is. But the truth reveals, only when the customer opens the box and uses it. Or it's the same as reading about how cool sailing (put in your favorite hobby) and doing it.

Or (to bring in meta-discussion) static knowledge vs. life.