Yes, I will continue to stay on top of security issues, performance
issues, compatibility issues, bugs, and anything else that adversely
affects the quality or usability of TurboVNC.
My message below is a call to action, not a letter of resignation. I am
not quitting, and TurboVNC is still very much an active project.
However, moving forward, the pace of new feature development will depend
on the pace of funding. I have enough guaranteed yearly funding to
cover project maintenance and maybe even a few minor features. However,
that funding is insufficient to keep the project moving forward at a
relevant pace, and it is certainly insufficient to keep abreast of major
changes such as Wayland. (It is also insufficient, in and of itself, to
pay me a living wage.) Basically, if I had to rely on the General Fund
to implement major new features, it would take years to implement them,
and few people would care by the time I finished.
The reality of independently maintaining a project such as this is that
it's a volume game. The more users TurboVNC has, the more likely it is
that users will proactively find and report issues, and the more likely
it is that organizations will fund new features. Fixing those issues
and addressing those feature limitations makes TurboVNC more attractive,
which brings in more users. Rinse and repeat. If TurboVNC goes into
"maintenance mode", then the user base will likely dwindle to the point
that it lacks the critical mass needed to keep the project relevant.
In a broader sense, this is also about the struggle between open source
and proprietary development models. When TurboVNC came on the scene
more than 20 years ago, it was the fastest Linux remote desktop solution
in the world, the only one fast enough to stream OpenGL-rendered frames
from VirtualGL at "interactive" frame rates. Now you can throw a rock
and hit three other Linux remote desktop solutions that are fast enough
to handle the output of VirtualGL, although two of them are probably at
least partially closed-source and the third is probably a b**** to set
up. As far as I know, I am the only one trying to crowd-fund
enterprise-quality remote display software, because there isn't enough
money in that business model to support even one developer, much less
entire teams for marketing, sales, support, R&D, IT, etc.
I have dealt with similar struggles vis-a-vis libjpeg-turbo. For nearly
the first decade of its existence, libjpeg-turbo was a loss leader,
essentially subsidized by funded development on VirtualGL and TurboVNC.
However, that came to a crashing halt in 2018, when I didn't make enough
money to cover my rent, health insurance, and other basic needs. I put
out a similar call to action for libjpeg-turbo, and in the years since,
its funding has become a lot more sustainable. However, now VirtualGL
and TurboVNC funding is the problem.
DRC
On 2/11/25 1:26 AM, Kimmo wrote:
Hi DRC
Good call. Thanks for all the hard work you have put into development
at your own cost. To me this software has been enterprise quality
without enterprise-level license fees.
Appreciate the bug fixes. What are your thoughts about the security of
TurboVNC with this decision? Do you have plans to keep the current
features up-to-date if any issues arise regarding security?
Best regards
Kimmo
sunnuntai 9. helmikuuta 2025 klo 20.57.00 UTC+2 DRC kirjoitti:
I'll get right to the point.
I have been developing TurboVNC, libjpeg-turbo, and VirtualGL
full-time
and independently for the past 16 years, the majority of my 29-year
career. I have managed to scrape by with funded development and
support
contracts on these software projects, as well as some patronage
(mostly
from corporations that rely upon the software.) However, the
amount of
money I make through this work is not great. My gross income is
something like half of the median for the U.S., and my net income
is in
an even lower percentile because of self-employment taxes and other
business expenses that corporate employees do not have to pay.
Prior to
2024, my income had been relatively flat since 2012 (except for 2018,
which was a really lean year.) That was already bad enough,
because it
meant that I was not able to keep up with inflation. (In fact,
were it
not for marrying someone during the Pandemic who is smarter about
real
estate than I am, I would have had to quit this game years ago.)
2024 was another really lean year (not quite as bad as 2018 but
close),
and a big reason is that funded development on TurboVNC and VirtualGL
has dried up. None of the features in TurboVNC 3.2 received any
funded
development at all. I was able to use the TurboVNC General Fund,
provided mostly by Santos and Crimson Vista, to pay for some of them,
but most of the work on this new release was pro bono. TurboVNC
3.2 is
a "Hail Mary pass" of sorts, a desperate attempt to maintain
relevance
against a rising tide of competitors who have deep pockets and whole
teams of people who do what I do. Unlike my competitors, I don't
charge
money for TurboVNC. I also don't wrap an open source project within a
paid closed-source product and reserve the best features for the paid
version. TurboVNC is TurboVNC. Thanks largely to libjpeg-turbo,
TurboVNC's performance is no longer unique in the industry, but
many of
its features (the Session Manager, for instance) are unique in the
open
source world. Unlike most open source projects, TurboVNC is
maintained
with enterprise-level quality control and version management
processes.
Being independent allows me to develop TurboVNC in a
platform-agnostic
manner, free of any one organization's agenda.
TurboVNC has helped a lot of people, and I want it to continue
helping
people. I strongly believe that it should remain free-- in terms of
cost, in terms of license, and in terms of agenda. However,
developing
this software is not free. It costs me money, both in terms of actual
expenses and in terms of opportunity cost from continuing to do this
rather than make 3-5x as much working for a corporation. Bug fixes
have
been and always will be free to end users (paid for with the General
Fund.) However, I simply can't afford to continue developing major
new
features for free. All major new feature development from this
point on
will have to be funded. That means that TurboVNC 3.3 and all future
major releases will live or die on the strength of that funding. If
there isn't enough funding for TurboVNC 3.3, then there won't be a
TurboVNC 3.3.
How can you help?
- If you represent a corporation or other organization that needs
additional functionality in TurboVNC, then contact me for a free
estimate. For inspiration, here is a list of features, requested by
various members of the TurboVNC community, that are in need of
funding:
https://github.com/TurboVNC/turbovnc/issues?q=state%3Aopen%20label%3A%22funding%20needed%22
- If you are an individual TurboVNC user, then please consider
sponsoring our project through GitHub:
https://github.com/sponsors/TurboVNC. If everyone who used TurboVNC
contributed even $5/month, then I would have no funding problems
at all.
Every dollar received translates directly into time spent on TurboVNC
development, maintenance, and support.
DRC
--
You received this message because you are subscribed to the Google
Groups "TurboVNC User Discussion/Support" group.
To unsubscribe from this group and stop receiving emails from it, send
an email to turbovnc-users+unsubscr...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/turbovnc-users/76b396b8-12ca-483a-b99f-a97b28e865b8n%40googlegroups.com
<https://groups.google.com/d/msgid/turbovnc-users/76b396b8-12ca-483a-b99f-a97b28e865b8n%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
You received this message because you are subscribed to the Google Groups "TurboVNC
User Discussion/Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to turbovnc-users+unsubscr...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/turbovnc-users/39f18f48-41e9-4a98-9156-62021d1e308c%40virtualgl.org.
