Hi,

I have written an authoritative DNS server since 2005.  This february 16th
it will have the last Open Source release at version 1.8.  The Open Source 
development was a great prototype (for me), but I feel that asking for 
donations is not going to make me a lot of money so I intend to port it to 
Microsoft Windows (and perhaps Mac OS) in the next two years.  

I also intend to keep this part non-open source, and you may be able to buy
that port in a microsoft store.  This is just part of a greater plan to 
eventually enter the firewall market as a cloud based layer seven firewall.  
Many systems already exist doing this, but I'm hoping my approach will 
eventually get me a minute market share enough to pay some bills.

Now to my question(s):

1. Does the LibreSSL port to windows work?  If so, great!  That will easen
the porting work.  

2. How hard would it be to port the imsg framework to Windows?  I understand
there is descriptor passing involved which windows doesn't know.  But I'm
confident that an alternative can be found.  Does a windows port to imsg
already exist?

3. Just out of the blue, is there Windows efforts for pledge and unveil?  I
don't intend to port them but leave them be just like the Linux port that is
already working.

Please, don't feel annoyed that I'm porting to Windows.  It is just an effort
to gain a larger marketshare of people that could use this as a product.  After
nearly 20 years I have finally a chance to make some money.  Something I never
had before.  Also version 1.8 will always be around, it will never go away.
And in a few years I do intend to release version 1.9 (without windows port),

I'm a firm believer that the Open Source model benefits the cream of the crop
(the people with skills on top), but it doesn't benefit everyone.  I'm not
a hotshot programmer, I'm mediocre at best.  This is why I want to adopt an
"open core" business model.  This may be selling out to some.  So what.

Also the days of closed source are almost finished.  People with enough ML/AI
power can devise decompilers that are able to make a fine human understandable
code (in C) of a binary.  I have seen screenshots of C decompilers that label
variables var0, var1, var2, var3 etc etc.  So non-coherent.  But with a bit of
AI the var1, var2, varN... can be rearranged to something more understandable.
This also means that open source will win, but its significance will not be
so obvious anymore.  So I give my "closed source" part a few years before they
are decompiled back into source.  Hopefully enough time to make a bit of money.

Thank you for your help along the way for the last 19 years!  And who knows
you can always fork my open source version and continue development for all.
It would be nice to see what you're doing with it and even participate but my
priority for the next two years is re-education as a social worker and when I
can to work on this windows port, so that I have more options to make money
in 2026 and beyond (before I reach retirement age in 20 odd years).

-peter

Please reply with CC to me since I'm not on tech@ and misc@ lists for the time
being.

Reply via email to