Announcing a new Construct 3 suggestions platform (21H2)

1 favourites
  • 2 posts
From the Asset Store
Casino? money? who knows? but the target is the same!
  • Want to submit a suggestion for Construct 3? Start here.

    Background

    From 2017-2021, we ran a public suggestions platform for Construct 3. This worked well for a while. However over the years it collected so many suggestions it represented a truly insurmountable amount of work for our small team. Even just administrating all the submissions became a lot of work. Further, for some older ideas it was no longer clear that they were still necessary or that lots of people still wanted them. This also led to some disappointment as ideas could sometimes seemingly get lost in the pile.

    Consequently we've decided to make some changes to try to improve the process. We don't have to keep things this way forever - we can always make more changes in future if necessary. This post outlines the new approach, and some updated guidelines to help make sure submissions are useful and compelling.

    Half-year phases

    We are planning to start from scratch with a new and empty suggestions platform about twice a year. This is intended to make sure suggestions are still relevant, and reflect changes to your priorities over time. Each phase will be named with the year (e.g. 21 for 2021) followed by H1 or H2 for the first or second halves of the year. For example we are starting the 21H2 platform, meaning the second half of 2021.

    The old suggestion platforms will still be left behind, but archived and no longer actively reviewed. This provides a historical view of past suggestions and votes, and also allows for referencing previous submissions, such as if you wish to resubmit an idea when a new platform is set up.

    Limited voting

    While we know some users would love to vote on lots and lots of ideas, we intend to force users to think carefully about what their top priorities really are. As a small team with limited resources, it's simply impossible to address a thousand ideas, and we want to focus our energies on the few ideas that will make the biggest difference to the most people. Therefore we are limiting each user to 10 votes only. You can also apply up to 3 votes to a single idea if it's really important to you.

    We are also introducing a policy that ideas with zero votes will not be considered at all. This is intended to make sure we get a smaller number of carefully considered ideas, rather than thousands of ideas that are difficult and time-consuming to review, let alone implement. Unfortunately due to the way the suggestions platform works we don't appear to be able to prevent idea submission once you've used up all your votes, but this policy is intended to discourage something like one user posting 100 suggestions, which contributes to generating an insurmountable amount of work. Further if a suggestion cannot attract even a single vote from anyone, it seems unlikely to be worth consideration anyway.

    Posting suggestions

    Please register an account on the suggestions platform using the same username you have on the forum, if you don't have one already.

    Please search to see if your suggestion has already been posted before submitting a new one. Duplicate suggestions will needlessly split votes and make your idea look less popular than it really is.

    In order to make sure suggestions are useful and realistic, we are also introducing some stricter guidelines about posting suggestions. With bug reports we have required information to make sure we can help you and avoid clogging up the issue tracker with loads of reports that are impossible to do anything about. Similarly we are now adding requirements which must be met in order for us to consider suggestions. Every suggestion must include the following:

    1. A comprehensive description: this should be at least a paragraph and detail precisely how your suggestion would actually look and work. Screenshots of mock-ups can also be helpful. If the description is merely a sentence saying something vague like "make a feature to handle enemy AI", it will be declined as it's impossible to act on a suggestion that vague. Also this should not refer to other software, e.g. "make a feature like software XYZ" or "make an official version of third-party addon XYZ": it's not always clear to us how other things really work or that it's even a good idea to implement them the same way in Construct, so similarly these types of suggestions will be declined. Make sure you describe how the entire feature actually works in the context of Construct.
    2. It solves a real problem: ideally the suggestion should make something possible that was previously impossible. Otherwise ideally it makes something that was previously very difficult significantly easier. If it's not clear that either is the case - especially if it's adding a second way to do something that's already possible - it will likely be declined.
    3. It covers alternatives and workarounds and explains why they are not sufficient and why the proposed feature is necessary. If it seems reasonable that you could use some other combination of features, even if it takes a bit of extra work, the suggestion will likely be declined.
    4. It includes the reasoning why the idea is important. It should answer the question: "With our limited resources, why should we implement this idea and not another one?"

    Please only describe one suggestion per post. Posting multiple suggestions is difficult to manage: it can mix very easy and very difficult ideas; it's not clear which people are voting on; and we only have one status to apply for the entire post. If you describe multiple features anyway, we will either decline the suggestion, or update its status according to any one suggestion in the post at our discretion.

    Note this time around we have removed the 'Minor suggestions' category. Users are often not aware of the technical complexities of their ideas and so can't easily judge if something is really a minor suggestion or not. We even struggle with this ourselves: as described in the blog post The unexpected complications of minor features, seemingly trivial suggestions can end up running in to unforeseen complications and spiralling in to weeks of difficult work. Therefore we aren't attempting to have any way to judge if an idea might be quicker or easier.

    Responses to suggestions

    Please note we do not guarantee that Scirra will respond to suggestions. As described above our policy is suggestions with zero votes will not be considered at all, but in addition to that we do not guarantee that we will respond to suggestions with any number of votes. The reason for this is that this alone would be a huge amount of work for our small team with already limited resources. Even providing a short response can involve a lot of work, involving things like assessing the technical feasibility across a large and complex codebase. Further, staff comments can easily expand in to extended discussions about what the suggestion really meant or the merits of different approaches, taking up even more time (especially across hundreds of suggestions). Therefore our policy is that we do not guarantee any responses at all.

    We will make a best-effort attempt to post official responses to highly-voted ideas, subject to available time and resources. However we are not stating a specific vote count at which we will provide a suggestion: we've already seen ballot-stuffing happening on the old platform, and it seems likely any specific threshold would encourage ballot-stuffing in order to force an official response.

    We do not guarantee implementation of features

    Even if an idea is the top voted one, that does not guarantee we will implement it. Voted-on features are highly subject to hype and imagination ignoring the real-world constraints of engineering and practicality. There are also several other ways we prioritise what to work on, including what we see happening with support requests, bug reports, forum posts, common questions and problems, opportunities created by new technologies, usage statistics, our own business goals, and more. The aim here is just to collect feedback, and the suggestions platform is just another one of several ways that we get feedback from users.

    Notes on feasibility

    Working on a large and complex software project involves many complicated technical considerations, most of which are probably not obvious to users. Here are some further reasons suggestions may be declined.

    • They're simply too massive a change to the product to be feasible. For example adding support to "make 3D games like Crysis" could amount to an entirely new product. There are huge risks to taking on a project like this - blindly jumping in to it could easily ruin the company.
    • They are not technically feasible. For example a user may ask for a plugin to allow X-ray vision through the phone camera. Unfortunately that's simply not possible. In other cases it may be just about possible, but infeasible due to the sheer complexity or amount of work involved. This isn't always obvious to someone who doesn't have to do the actual engineering involved.
    • It breaks backwards compatibility. For example a user may suggest a completely new and different way for the Families feature to work. Even if it's a brilliant idea, it would likely break thousands of existing projects that already rely on the feature working the way it already does. Users whose projects break between Construct versions rightly get very upset, so we strive to ensure compatibility between Construct updates. Ideas which would obviously break this are not something we can really do in practice. We can in some cases add a parallel feature and try to phase out the old one, but this is technically complicated, and confusing to users who wonder why there are two options and which one they're meant to use.

    Another factor in reviewing suggestions is how much work it involves. Some ideas are useful and can be added relatively easily in a day or less (although sometimes even small changes can become unexpectedly complicated, as mentioned previously). However other ideas will clearly involve months of work and have long-term maintenance implications too. We will be much more stringent when reviewing such large ideas, since they are far more costly and risky to work on.

    Try to vote wisely. If you sink all your votes in to lots of huge engineering projects, it's unlikely you'll see anything happen with your voted-on ideas. If you try to focus your votes on things that at least appear to be smaller and easier, you are more likely to see them implemented. Once a suggestion is shipped you get your votes back too, so you can then move them on to your next preference.

    The suggestions platform

    If you're happy with all the above, and accept the requirements and understand our policies such as when and how we respond, you can find the suggestion platform here: Construct 3 suggestions 21H2

    Happy suggesting!

    Archived suggestions platforms

    Note archived suggestions platforms are no longer actively monitored or reviewed.

    construct3.ideas.aha.io - archived suggestions posted 2017-2021

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads
  • Is it possible to feature some ideas here on the Construct.net site?

    I think visiting the suggestion platform is a bit "out-of-the-way" for most people (or those like me, at least..). I usually visit Construct.net weekly to check for updates, but visiting aha.io is not a priority for me.

    This matters because ideas now has limited time to get noticed, or they will be archived. Not because an idea is old does that mean it is now obsolete. Maybe there are good things in the pile!

    It would be more easier if there is some sort of "idea leaderboard" here on the site, which shows the ideas on top of the list at the moment, and links to aha.io for voting. This way, more traffic goes to the votes, and you can get more accurate data about what the users want (since more people are "actually" voting)

    I found that Aha.io offers a developer API, maybe you can use it to list the top 20 ideas on a new page here on Construct.net.

    Thank you.

Jump to:
Active Users
There are 1 visitors browsing this topic (0 users and 1 guests)