Open source version is dead

Sim9

Hello,

I used Bug Genie on two of my past projects and really enjoyed it. I went to check the website to see about getting the latest version and it looks pretty different now. The .org version of the site is completely gone, and all the links to github are super buried. On the pricing page, it's not clear at all what is commercial features now and what is free (some mention of modules, but not a full list of what's free and what's paid).

Even on the github release notes, it sounds like updates for modules is no longer available to open source community. If I can't even get security updates, it hardly makes sense for me to use this for my next project since it's on a public facing server.

Anyway, seems like the future direction of this project is commercial. No judgements here, I get priorities can shift over time, but that direction doesn't really fit my needs. FWIW, thanks for making one of the best open source trackers and good luck in future, myself I'll probably be heading back to Trac.

zegenie

Sim9 thanks for taking the time to write up your thoughts about the direction of the project - I'm glad you're sharing it with us. I'm also glad to tell you that your assumptions aren't correct, even if I can totally understand it from what you're seeing.

The reason the .org website is gone is because we didn't have the capacity to maintain them both, so we went with the hosted website, and are working on putting the community components in a more prominent position. In the meantime, we figured we could set up a good and functioning forum so that our users wouldn't think everything was dead. We moved the github source links to the global footer so it shouldn't be hard to find, as well as putting a global banner on the "download" / "register" page for people looking for the github downloads. It hasn't been our intention to hide it, but simply work with what we had until we had our website reworked.

We're trying out different ways to support the project financially, but at our core (and our code) we're not just committed to being open source, that's simply what we are, and it's not changing. We're a very small team working with what we've got. If you (or anyone else reading this) wants to join us and help shape the way forward, we're all ears and open arms.

Sim9

Thank you, I'm glad to hear I was wrong and thank you for the invite. I try to contribute back any changes I make to open source software I use -- I haven't made many changes on my Bug Genie instance, although a few years ago I did privately contribute a security disclosure I found to this project which got quickly patched. 🙂

Regarding the website, maybe it might be effective to borrow some of the user flow of Mantis' home page? There are some nice things about their design:

It "feels" like an open source project page as opposed to a commercial page.
A lot of this is also that you appear to get everything in the open source download. TBH, if I'm looking for a free project to do a particular thing and I see 'pricing' link anywhere above the fold, I bail out and go to the next thing on the search results.
The critical links (Demo, Download, Hosting) are all immediately obvious
It doesn't feel like they are pushing one of the routes stronger than another.

Compared to Mantis, there's also some things Genie does well with its current design and hopefully can be kept:

Feature walk-through
Interesting screenshots
Mantis stealthily upsells for the mobile version, which seems a bit less honest since people aren't going to realize that until they've already gotten an install going. Meanwhile, if you can find the Get page (I managed to once I found out it was in the footer) Genie is more up-front about this. Although, it seems kind of weird to monetize it at all in 2019 when mobile-compatible software is more or less expected for any project.

Sidenote, there's no screenshots of the source control integration. That feature sounds really neat, so it would be cool to include that somewhere (it's also not hooked up for the official instance).

I'm by no means an expert in open source, but it seems like the projects that get the most adoption provide the full offering out-of-box. Either making a commercial project or an open source project is both good, but mixing commercial components and open-source components together ends up satisfying neither group of users, I think. If you're looking for open source stuff, you want the whole offering and you want to be able to self-host it. Even stuff like autoupdating or installing modules easily, the most popular open source projects like Wordpress offer this as standard.

Although I do think the non-code stuff you provide like optional hosting, training, support is a pretty fair way to monetize. Maybe even things like feature bounties, or sponsorship from QA companies who use TBG and are looking to find new customers? And bespoke/commissioned development for those esoteric company-specific features or custom theme development seems like it could be a good offering.

zegenie

Thanks for the thorough reply. I think you're making some very valid points about "feel" and perception, however that doesn't mean we have to look unprofessional - an open source project isn't about a "homemade look" but providing a clear overview and at the same time get the message across to our users that you get the full offering also if you're choosing the open source download edition.

As much as we'd like to think otherwise, self-hosted instances aren't nearly as popular as they were before - and trying to provide a true open source alternative to our closed competitors means we have to look at ways to finance our hosted service. We can't provide the same level of service as the major players (like Wordpress, JIRA, etc) can do for free, but we can definitely find ways to enhance our offering with paid services without cannibalizing our open source offering.

The mobile look is an example of the wrong way to do it. Providing an outstanding experience out of the box - whether on mobile / desktop or any other form factor - should be the default in 2019. I see that the mobile UI is still a bullet point in our comparison chart, but it shouldn't be for long as that's not how we want it. Your suggestions about bounties, sponsorships, etc, are all good suggestions and things we already provide for anyone asking. I'm very interested in continuing this conversation with the goal of finding a nice middle ground. Anyone else wanting to chime in feel free to do so, we're really interested in your feedback.

Sim9

That's great to hear regarding the mobile version!

Yeah, I think it's 100% fair to charge people for cloud hosted instances. There's very clear compute resources involved, and it makes sense for people to pay what they use given the cost of each additional cloud instance.

But the overhead for each self hosted instance should be pretty minimal. For instance, I don't think things like autoupdating on self-hosted instances should ideally be a premium feature. Having a secure ecosystem with all the latest patches by default is definitely something that I think is a worthy goal. This should be something that can be hosted very efficiently and so doesn't need to become a premium feature. To make it more efficient:

The version check API component of autoupdating could be hosted behind cloudflare, in which case it could be scaled essentially for free.
The file download could be hosted via github as it is currently, or if not possible, then via backblaze. Assuming 50,000 self hosted instances are all pulling the latest version (4.48MB) once a month, you'd be seeing about 219GB of traffic, or about $2.19/mo in bandwidth charges on B2. (This could be made even more efficient by just distributing only the changed files, but at low scale, full download is probably fine)