As I mentioned in previous installments of this series, I’m now mostly focused on tying up loose ends. Part of that effort has involved long‑overdue updates to the website over the past few weeks - reorganizing its structure, improving the flow of the text, and refreshing many of the screenshots to give the software a cleaner, more contemporary look.
With BMT Micro now able to process Chinese payments and a new partnership in place with a Chinese reseller, I also reopened the floodgates to China. The last time I attempted this, the Winstep server was instantly overwhelmed by the sheer volume of update requests pouring in from Chinese users.
I can now confirm that the free version of Nexus has over 280,000 active users in China, with a confident estimate of roughly 400,000 total installations. Because these systems had not been updated in over a year, they all attempted to download the full 40 MB installer instead of the lighter 16 MB update package. Worse, the Winstep Updater’s repeated failures to reach the website - due to the previous firewall block - caused most clients to retry in the same concentrated 10‑minute window. The resulting surge completely locked up the server, and once the real numbers became clear, the reason was obvious. I had no choice but to block China again.
No single bare‑metal server can handle a load of this magnitude, so reopening China required a multi‑pronged solution:
• moving update files to a CDN (Content Delivery Network),
• rewriting the Winstep Update Manager so Chinese downloads are redirected to the CDN instead of the Winstep server, and
• ensuring that future connection failures would not cluster retries into the same narrow time window.
Instead of every request hitting one machine, the load is now distributed globally - which dramatically improves reliability and absorbs traffic spikes.
Once all of that was in place, the final step was to reopen China gradually rather than all at once. The firewall contained 55 separate Chinese IP range blocks, and on the 16th of April I started removing them one by one over the course of 24 hours to control the flow of incoming requests.
As each block was lifted, users in that range immediately downloaded a small 2 MB update to the Winstep Update Manager, which then directed them to retrieve the much larger 40 MB application update from the CDN instead of the Winstep server.
This worked and prevented the Winstep server from becoming overloaded again.
Now I have a different problem: serving hundreds of thousands of updates and more than 10 TB of data through the CDN is not cheap. And while having nearly half a million users in China sounds like a massive opportunity, it’s also a massive liability if it doesn’t translate into revenue.
When I originally developed the free version of Nexus, it had to compete directly with well established docks like ObjectDock and RocketDock. To stand out, I added features that probably should have remained exclusive to the paid editions - and that strategy worked. Nexus became the de facto standard for docks.
But success brings its own challenges: for many users, the free version is already “good enough.”
The “Buy me a beer” toaster I added to the latest free version has produced some results -better than nothing, at least - but overall the response has been discouragingly weak. It’s clear that passive support alone isn’t enough.
Different markets also respond differently, and what works in one region doesn’t necessarily translate to another. As someone put it rather eloquently, relying on love to generate revenue is an inefficient strategy.
That said, I want to be absolutely clear about one thing: I will never remove features from the free version or bundle it with adware. I’ve been approached multiple times over the years with offers to monetize Nexus that way, and I’ve always refused. Reputation takes years to build and seconds to destroy - and that’s not a line I’m ever willing to cross.
So the focus now is on improving how the free version communicates value.
The same person who offered that insight also suggested adding subtle prompts inside the free version when users attempt to access particularly valuable features. We both agree that we shouldn’t push or annoy users; instead, we should guide them toward the version that best fits their needs. Taking it step by step is essential.
Going forward, premium features - such as Create New Dock, Add Grid Stack, and others - rather than being completely absent will now become visible in the interface and in the right click context menus, marked with a small lock icon. Users will naturally discover them as they explore. When they try to access one, the application will provide a clear, contextual explanation along with the option to upgrade.
The goal isn’t to push or annoy, but to guide - helping users understand what’s possible and what they gain by upgrading.
Finally, in terms of actual feature improvements to the applications themselves, I’m convinced that the single change with the greatest impact will be the re-introduction of the Winstep Theme Builder - even for users of the free version.
The Nexus skinning engine supports capabilities no other dock offers, such as overlays and advanced compositing, but right now, most users never fully tap into that power. Giving users a dedicated tool to harness that power - rather than simply copying existing RocketDock skins - changes that completely.
For the first time in years, users will be able to create, customize, and share complete themes without needing external tools, arcane knowledge, or manual file editing. It empowers creative users, lowers the barrier to entry for new designers, and encourages a community‑driven pipeline of fresh content.
A strong theme ecosystem keeps the software visually fresh, reduces support overhead, and naturally drives interest in more advanced features.
In short, the Theme Builder doesn’t just improve the product - it strengthens the entire Winstep platform.
A natural progression from this would be finally opening the Winstep Themes Gallery to the public, but that depends on reaching a level of sustainability where supporting that ecosystem becomes viable.
Attachment:
Screenshot 2026-04-21 065401.png [ 104.85 KiB | Viewed 202 times ]
Attachment:
fb1ccbf4-5949-4699-8704-51c1470b2ba2.png [ 1.22 MiB | Viewed 202 times ]
Posted: Wed Apr 22, 2026 3:58 am 
Hope "The Thing that must not be Named" 
will be comprehensive, not just an initial release for Nexus? Might encourage people to look beyond Nexus...
Window border might need a little work? Worried about the use of Big Tech AI - it's all going to be used in further training AI, and goodness knows what else happens to your data. 
