Web Presence Improvements


A while ago, I posted a survey here for Thrive fans to tell us where our outreach is working and where it isn’t. Rather than analyse it in that thread, I’ve moved it here to reduce clutter (and place it in the more appropriate section, #outreach). If you want more detailed statistics, here’s the Google Sheet with the full data. I don’t have time to go through it in with a fine tooth comb, but if anyone else finds anything interesting, please share.

My analysis here won’t be anywhere near as thorough as our last survey, but we did still get some useful insights so it’d be a waste not to act on them. We had 155 responses in total, though I think some were duplicates since they contain exactly the same information. Percentages when quoted are relative to number of responses to each question, which varied.

  1. The website and our subreddit were the most used web presences, but the subreddit (78) actually outstripped the website (69). Next was YouTube (60), and the community forum (56) and development forum (53) were closely matched just behind. Given the places this survey was posted, it’s not surprising to see sites such as ModDB (7) and the Steam group (9) so far down in comparison. My hypothesis is that most of our followers are passively watching the subreddit and don’t engage much with anything but major news, which is why we have more than 4000 subscribers but very little activity.

  2. Unsurprisingly, the least well known web presence was Bandcamp (121), but I mostly put it in there on a whim to see whether anyone but me even cared about it. Our other small presences were next, such as DeviantArt (99) and Steam (93). Interestingly, Discord was also pretty unknown at 86, which suggests the people using the Discord are a small but loud subset of people interested in Thrive. Worryingly, 16 people didn’t know we had a website. This shows a significant ineffectiveness in the way we communicate via social media.

  3. The next question asked how well each person felt they understood what’s happening in development. The split was: 34 (22.7%) clear, 59 (39.3%) adequate, 45 (30%) vague and 12 (8%) little to none. I think we need to do a better job of communicating the day-to-day workings of the project to our fanbase.

  4. This question asked what notified people of the latest Devblog, i.e. which news source they generally see first. Most were either Facebook, the subreddit or the community forums. Most other answers were unsurprising (randomly checking website, YouTube, accidentally misinterpreting the question as what they wanted us to improve about the Devblogs, etc.) but a couple expressed surprise that we had Devblogs.

  5. Which is backed up by the next question. 20 people didn’t know we had Devblogs when asked how much of the latest one they’d read. An encouraging number (66) said they’d read all of it, although just how much they engaged with is still unknown.

  6. In a vote for which Devblog was everyone’s favourite, every Devblog except #4 had at least one supporter, which is nice. Most popular was #11, the most recent, which gathered 17 votes. I added this question to see which formats we should aim for in future Devblogs - other popular posts were #1, #3, #5, #7 and #10. These were all either release announcements or went into more detail on a subject, such as the survey analysis or collaboration with Species: ALRE. I like making these types of detailed Devblogs, and the statistics show they’re generally more popular than minor development updates without accompanying releases.

  7. 55 people were surprised we had livestreams, compared to 35 who knew but didn’t watch them, and 58 who watched any amount of them. I would say we should communicate them better, but we should probably think about better structure for them first.

  8. The next question asked what topics people would want to see disucssed in a podcast/livestream. Some of the more interesting answers: the engine, the history of Thrive with issues and future solutions, cellular-multicellular transition, features of upcoming versions, how to speed up development, agents/bacteria, ways to support us, game controls, who we are as people, outreach plans, auto-evo and finances/promotion. Some of these have been addressed in writing elsewhere, but if we were to have an organised vocal discussion it might better express our thoughts and find possible solutions.

  9. Next question was on improvements to the website. This is a topic I want to discuss more in future in this thread. Three points appeared multiple times so should probably be addressed: better links to pages such as the gallery, some sort of checklist/roadmap, and more frequent updates. As much as I myself would like a roadmap, our efforts to create one in the past have always gone awry because the features we plan for don’t match the features people work on. If we were to have one, it would mean a major shift in the work ethos of the team, so this probably warrants a more detailed discussion elsewhere. Otherwise, people seemed pretty happy with the website.

  10. Then was the question on improvements to the community forums. We have a few interesting responses here. Some suggested shrinking the number of subforums in the interest of ease of navigation, others said they’re fine as they are. Various others of note: adding more images, recently updated threads on mobile, and removing the less…erm…tactful creative threads which verge on fanfiction. All of these are in my opinion valid points and should be considered.

  11. A similar question for the dev forums. Here, the need for cleaning up and reorganising was more apparent, linked to difficulty navigating. Both these questions also revealed that some people want better navigation from the website, with the suggestion that the Get Involved page become a forum link. I don’t think that exact format would work, but the sentiment is valuable and a reorganisation of navigation links across sites should be considered.

  12. Few people had much to say in regards to general opinions on web presences. Those that did suggested, among other: they should be more interlinked, find a single channel to spread news, update dead sites, discuss more on the dev forum instead of Slack, more activity in general, Tweet occasional funny bugs we encounter as Dwarf Fortress developers do, and post memes on the subreddit. The last of these has already been taken care of, and some of the others might be unreasonable given the team structure, but they’re still food for thought.

  13. The next question asked what an individual would show first to a friend if they wanted to generate an interest in Thrive. The answers here were surprisingly mixed. The most popular seemed to be some variation on the website, which is encouraging since it’s supposed to be the introduction to all things Thrive. Also frequent were the latest trailer and the game itself. Some people felt they should show off the concept first, others more concrete progress on the development forums or the Wiki. Others suggested the community forums, since in its current state the game is insufficient and the community is more likely to draw people in. Further answers included: the origins and history of the project to solidify faith in the dedication of the team, concept art, the music (:D), our Devblogs, and this video. Respondents were generally divided into two camps: those who wanted to show the grand scale picture and those who wanted to keep expectations realistic.

  14. Something I’ve been thinking about a lot, the next question asked how people felt towards concept art at the front of the website. 89 people (64.5%) said it would engage newcomers, 35 (25.4%) said they’d prefer seeing actual development progress, and 14 (10.1%) entered their own response. You can trawl through the Google Sheet to see all of them, but here are some responses I found interesting: realistic concept art of near-future development is fine, a balance of both, more screenshots than concept art, and concept art with warnings that it isn’t in the game yet.

  15. The next question ended in a slight stalemate. It asked whether people would prefer blog post or video updates and the split was almost even (58 videos, 67 blog posts). In retrospect, I should have added an option for both/either, since this was a frequent response for those who entered their own answer. Those remaining divided the options by type of news, suggesting small updates would work better as videos and vice versa.

  16. The split in opinions on what length the news should be was similar: 62 said long updates every month, 71 said short updates every week. I realise every week is almost impossible for us, but we could still post updates far more frequently than we do. Alternative answers were somewhat contradictory - some said we can be flexible and update whenever we have something update with, others said they wanted updates the same time every single month, no exceptions.

  17. Respondents were overwhelmingly in favour of detailed updates from developers on major game concepts such as CPA, whatever form they took. Only 2 people (1.4%) said they weren’t. This answers @Seregon’s question from one of our livestreams.

  18. Finally, I added a catch-all question for suggestions on outreach. It gathered some insightful ideas. Some of the more interesting were: indulging questions asked by the community instead of directing them to a previous discussion, more conversation on the development forums and less on Slack, more Q&A sessions, more surveys, smaller releases to get more feedback, visit conventions (somehow I doubt this will happen, but it’s fun to speculate about it), better livestream structure, and weekly contests for things.

So what do we do about all of this? Well, for a start, you lot can see all the data now, so feel free to discuss the above below with your own opinions on opinions. Ultimately we didn’t find anything Earth-shattering here, but I feel it was worthwhile nonetheless.

Over the next few weeks, I’ll go through the website, community forums and development forums and start making minor changes based on some of the suggestions. Most prominent is an adjustment to the navigation links, which I’ll try and standardise across all three.

I also want some kind of recent updates feed, which I’ve trialled on Slack using RSS. Its main issue was inability to distinguish types of post on the community forum, so I may have to omit it entirely. The development forums, YouTube and such should certainly be combined into a feed though.

I’d also like to ask whether anyone has a topic they’d be up for writing a Devblog on. @Seregon expressed interest in a detailed explanation of the CPA system, either as a video or blog post. @hhyyrylainen, since so many people seem interested in the engine, would you be up for writing one on its structure and justifications for decisions about it? And of course, anyone else who feels they’re an expert on something current in development could make something. The bottom line from this and other surveys is the need for more frequent updates and communication, and items like these describing design decisions in detail could be exactly the thing we need. Also, @Naggorath would you be up for a discussion on the possibility of having video updates as you suggested elsewhere?

We should also have a talk about organising this forum. So far there’s little organisation and navigation consideration to speak of. What could make it better? Bear in mind some concepts might be limited by the capabilities of Discourse. We can have that discussion in this thread instead.


I think it would be cool / I would be up for being on a podcast discussing those suggested topics in 8.
Many of those topics would really benefit from having @crovea or @Moopli join the podcast.

I think we could setup a trello board where people could move tasks around as they do/plan to do them. And then put a preview of that on the website. This would fix the biggest issue I currently have with the roadmad thread/wiki page which is that I (or most other devs, too?) can’t add stuff or mark that they made progress on something. Unreal Engine does this type of thing: https://trello.com/b/gHooNW9I/ue4-roadmap but with like “TODO”, “IN PROGRESS”, “DONE” lists with tasks in them.

I can certainly try. But recently I’ve been thinking about the issue again, as I’ve been waiting for jjonj to review my changes for the past 3 weeks. And I think such cases would take way less time if we used a premade engine, so I’d preferably try to get some input on the reasoning. But I could certainly at least list all the points that have been made.

I could also just write a technical overview of all the things going on and what’s created in what language, as I think a few people have asked that.

Post-0.3.3 Forum Management

With premade engine you mean things like unity, or are you talking about something else?


Yes, that was what I was referring to


I intended to work through a load of minor site improvements today. I’m not quite sure how, but my afternoon just kind of disappeared in a flurry of frustrated perfectionism, so I didn’t get as much done as I wanted. I spent far too long on it, but we now have a couple of feeds on the website home page showing the latest Devblog and forum posts (partly inspired by @Narotiza’s launcher mockup). Tomorrow I’ll restructure the entire front page to make them fit better.

Also, what started as an attempt to standardise header links ended in a complete overhaul of the community forum design:

I’m still on the fence. I feel like there’s an ideal design lying somewhere between the two extremes, but I haven’t found it after several hours of tinkering. The feedback thread is here. The old design is more or less available to return to at any time, but I’m hoping someone can come up with something better than either. Be warned, though: there are never as many customisation options as you need to make your dream a reality.

Anyway, here’s a list of things I’m going to do over the next couple of days. If anyone has any thoughts, let me know in advance.

  • Work on the dev forum cosmetics
  • Add more prominent links to the gallery and gameplay overview on the website
  • Update the gameplay overview for 0.3.3
  • Add more concept art to the website
  • Look into starting a Trello board as @hhyyrylainen suggested


I’ve also made some changes to the website, reorganising things and using a sidebar to put important items closer to the top.



It’s looking great so far! I especially like the sidebar. Though, I think the download button should be higher up so you don’t have to scroll as much to find it



I’ve had a thought. We’ll hopefully be ready to release 0.3.4 in a few days, and it doesn’t really have many trailer-worthy features to speak of. The programmers have worked hard on several things, but none of them have any clear visual changes. A trailer would likely be either too short or disappointingly drawn-out.

I’m wondering, perhaps we could use this as an opportunity to make a dev diary style video as you suggested? Instead of a conventional trailer, you (or someone else) could explain the launcher, the process system updates, the Lua fixes and upcoming possible engine change. It would notify people of the release without relying on a forced cinematic approach.


Sure, I think it might be cool to do something similar to what Montie Olsen here did (I forgot what his username here is)
He made these two videos:


Maybe something similar (A lot shorter though) or get in touch with him. I think he presents the project in a nice way. Also if we are gonna do some kind of voice-over, then we need someone else to do it, because I am not that good at that kinda stuff.

So yeah, I am just gonna go through the launcher tomorrow. I imagine I can use fraps to film (If not, then suggestions for other recording tools are welcome)


I definitely recommend OBS for recording, it’s completely free and has no watermark at all.


I can most certainly do a shorter version if you want. I have all the assets ready and such, and I do need to update my channel!


Awesome if you write a draft for the voiceover (Presuming that’s what we are going to do) and eventually record it, then I can record some footage and edit it all together. Should we talk more in slack and coordinate it?


Sounds good!