Echoes into the abyss...
(first of all, your question: How many arms and legs does the typical human have? ... won't accept "two" nor "2" ... I didn't realize the typical human has 4 arms, for example.)
I am sitting on the other side of the tracks from the previous posts, but I keep running into the same hold pattern as the rest. I had an employee several years ago (i.e. circa probably 2010 who worked on creating an ecosystem of the WyUpdate paradigm... subsequently never pushed to production simply for the lack of perceived renewable editions of this product.
On one hand, I'm kinda glad we didn't buy in so tightly to it (i.e. I suppose sitting at 2018, in hindsight, perception has become reality)... on the otherhand, I wish we proved to be completely wrong.
As it stands, we've gone through iterations of all the standard variations of installers like NSIS (Nullsoft), InstallShield, Spoon(+Home Grown Solution), Third-Party consult (epic failure), (etc, etc etc...), and now currently InstallAware (with all sorts of congealed "stuff" being employed just to work around it, and every other previous answer's issues if you know what I mean).
However, I've been in need of a reliable, consistent, usable product exactly like what your product has purported itself to be since 2004... The major battles I have with any iteration of Installer product is the vast difference between initial setup\configuration and maintenance releases (sound familiar?), aside from the semantics of simply being able to accommodate them, when it comes to the ability of future staff being able to keep pace with the learning curve necessary to further development. Sadly, high turn over is just a fact of life where I work as well an absolute necessity for security and quality assurance over non-nominal end-users (Government work at a University)... which just exacerbates the point.
Perception.
Even if this product were 99% reliable for "when it was built"...Even if this product were 99% "future-proofed"...Even if this product were 99% "feature-complete"...
The perception of the product deteriorates (and in such, deteriorates the perception of any other product, service, etc. of the company and the company itself as the public face of those products, services, etc.) in a non-linear fashion over nothing but "stale"-time. "Stale"-time becomes simply the public-facing time between Major releases with "sign of life" Minor releases.. in most cases even if only compiled as an active "coming soon" page within the product's stable release "site". What you need is a living testament in good faith to end-users
If you were to simply addend your posts and products (think of, say, Windows 7 ProgressBar) with "always updated" iterations (i.e. Windows XP-10 ProgressBar or even just Windows 10 ProgressBar), you give a better perception of a "living" company. Recompile your Panel2006 and make it Panel2018...
Post some other tangential products your staff has inevitably made as replacements and workarounds for the lack of proper stock (\free\available\etc) <things>.
Link through general collaborative space for (if not handled in the above case) what your staff has come across where they were made to at least pause for a moment and reflect upon the "I wish I just had" <thing> but inevitably relented to some other mental avenue for lack of time\money\deadlines\etc. At least then many of the above collaborators could feel more vested in your release products rather than wanton for sign you even still exist, as it were.
Even further into an idealistic view, take your last stable release (and eventually every future stable releases) "as is" and open up a couple sections of the source which most predominantly contain end-user bugs, issues, "feature requests", etc. and let them see if they can fill in the gaps themselves in the meantime. They'll still buy in to the future (literally) while satiating those who come to you as:
1) developers wanting or needing functionality you may never roadmap,
2) contributors to future editions who are able to make their cost-benefit-analysis -time- more bountiful than you've admitted in several responses above to being able to justify with the overall source-revenue of this product,
3) end-users who are ultimately always hamstring'd by the inability to compile their way around a program/need and instead tend to find themselves trying to cobble together disparately sourced combinations of <stuff> at each step of their own process to accomplish each <step> they come across (again, those "opened" sections can at least be provided an "as is" "plug-in" comprised of the above user submissions as well as any "approved" modifications of your own),
4) Prospective new funding sources for you who are going to run the gamut between wanting simple, clear, intuitive final products and extensible, robust, full featured, configurable and evolving integration software (all of which currently are just as likely right now to see the last release date and move on; if not, the next step is generically to find legitimacy within the periphery of the same site leading them to this support forum rife with unanswered premises and the above posts highlighting genuine concern over the future of the products themselves).
None of which are going to take kindly to simply being ignored (or perceive flippantly standoffish responses promising things which "clearly" never materialize, hardened and disillusioned programmers can attest : see Perl6, history)... or for the end-user explanation read the opening and closing sentiments here: https://www.evanmiller.org/a-review-of-perl-6.html) no matter how exceptional your one-on-one support may be.
Ripping a quote from the link above is as eloquently and aptly put as any potentially meaningful drivel I've managed up to this point:"The explanation, of course, is banal: if youre a geek, writing correct computer code is a lot easier than reprogramming peoples perceptions."
I get that (hopefully) the lot of you are tending towards being more software-development-centric vs software-use-centric (me too)... but you've put yourself out to the world as the latter now and there's no turning back so clip that faux-tie on your tuxedo shirt and at least pander to the capitalist end-user crowd... (so we can get back to our dungeons and code monkey-around in the shadows like we all want)
Ars Longa, Vita Brevis.