"quitting havenco", not quite!

[Ryan Lackey]

Two people appear to have forwarded communications from a
semi-private chat system through two mailing lists to cypherpunks.
("gale" at this stage is semi-private, and this is the first time
anything has been forwarded from it to public lists.)  These people
are no longer on my christmas card list, or whatever.

I suppose I should now explain:

1) HavenCo is now making sufficient money and is stable enough to be
"boring".  This is a *good* thing.  Over the past year we've completed
most of the remaining engineering, and some of our other staff are
taking over more and more of the operations, first and second level
support, and administration.  Nothing really breaks anymore; it's easy
to add new switches and routers without changing the underlying
architecture; it's all operations.

2) I'm hiring engineering and sales managers in the next 3-6 months to
take over more day to day responsibilities.  This is independent of
anything else.

3) I'm working on sofware to automate more operations, and doing
various conferences throughout 2002.  We're also looking at other
sites; when we actually move forward on one of those, it would be a good
test for the software/hardware systems.  (some of these have been
deployed recently.  I can now do an entire new customer install
without ever even seeing the customer's hardware; comes pre-configured
from the vendor with PXE and it just gets plugged in, racked, and 
auto-configures).  This means regardless of anything else I will be
spending more time elsewhere in the world.  It is silly that I can do
everything over ssh from wherever in the world and yet still sit in an
office a few feet above the racks.  If something breaks, we just flash
to a hot spare machine remotely and then order a replacement.  Worst
case, someone pulls some drive carriers out of the front and puts them
in a different machine; not exactly rocket science.

4) I'm working on a couple of side projects which may be
ready in 2002.  These include some book projects,
an offshore open-source development framework with quality control,
some papers to bring a couple of fields together (crypto to p2p,
traffic analysis to finance), a comprehensive survey of Gerald Bull
and engineering analysis of the supergun, and some ecash and 
tamper-resistance related projects.  I hope to have "abstract" level
work on all of them completed, and then finish one at a time when I
have time.  Right now, I'm finishing off the completely automated ISP,
and 2 papers.

5) When such things are ready to deploy, I will either switch to just
doing board things for HavenCo, or will stay on as CTO but have less
day to day responsibilities.  Some of them might be done by HavenCo,
which would mean shifting from colo operations to that project.  Some
of them are pretty minor; they can be "trialed" without distracting
from my day to day job.  Most of them would want to be separate legal
entities from HavenCo, but might involve the same people.  The exact
date of any transition is not fixed; I suspect it will be a gradual
process from mid-2002 until mid-2003.  

6) Even if I shift to working on something else, it's entirely
possible I'd live on Sealand 1-2 weeks/month anyway.  It's actually a
decent place to get work done, just lacking in entertainment (which
may be related)

7) The process of setting up additional secure colocation centers
around the world is pretty slow -- even if we can do most of the
technical things in 4 weeks, political negotiations with telcos,
governments, etc. take a long time.  They are also not something *I*
am directly involved in.  Also, various fiber projects elsewhere in
the world have been delayed by the dotcom collapse, so the options are
a bit more constrained as to where we can go.  We can "cheat" by
coloing with other people in those markets, with a common management
interface and maybe network -- specialized offshore colo providers,
telcos, or major colo providers who happen to be there.

Motivations for this include:

(I started thinking about this a lot more at HAL2001 in Holland, not
entirely sober I guess...)

1) I have worked 7-day weeks for the last 18 months.  I've had several
trips to conferences, and 5 days of real vacation.  I have slept on
concrete floors and desks because we brought over a bunch of computer
gear before furnishings.  I have missed Burning Man twice, a bunch of
interesting conferences, been woken up at 0300 and on dark boat rides,
climbed around on ladders 300' above the ground attaching lights,
given press interviews with impacted wisdom teeth and passed out at a
pharmacy afterward waiting for a prescription for antibiotics, etc.  I miss 
{girls, alcohol, sushi, ...}.  SFBA really is one of the nicer parts
of the world, and I've still not yet had a decent vacation in asia,
australia, etc.  It isn't that much fun to have a successful company
if you can't actually enjoy the results (I have 5TB of mp3s, but
haven't gone to a live concert since...1996)

2) That I really want to work on a software project.  I have this
lingering need to do something intellectually challenging -- HavenCo
is certainly a feat of *something*, but it's not a huge scientific or
technical challenge -- more an engineering, political, marketing, or
integration challenge.  Which is nice, but I'd like to do something
which contributes to the scientific knowledge of mankind; I've
taken advantage of amazing scientific advances made by people over
thousands of years, but not really contributed much myself.

3) It's been 24 months since we started, and I'd like to something 
different at some point, just to see if it can be done successfully
twice.  And I can self-finance it this time, which is a definite
plus. (having gone from about USD 30k in debt when I started HavenCo)

4) Doing sales and engineering at the same time is quite stressful.
It really would be best to have two distinct people doing each role.

5) Good people are cheap to hire now :) (jobs at havenco.com, not me please)

6) It would be nice to work on something which isn't "man vs. mankind"
but actually "man vs. nature".  If the world's governments became
reasonable overnight, and people stopped acts of terrorism, theft, or
whatever, there would be no real need for HavenCo.  Yet, there would
still be motivation for doing things which actually contribute to
man's knowledge, ability to conquer natural environments or other
challenges, etc.  HavenCo is a necessary precondition for a lot of
those things, but is not the goal in and of itself.

7) The current market is actually a decent time to set up new
projects; there's no competition since no one else is funded, there
are much more reasonable expectations (HavenCo hasn't IPO'd for
billions of dollars, but we're profitable about 6 months after we
started selling stuff, and growing at a reasonable rate).  Plus, now
that HavenCo is up, it's entirely possible for an offshore-centric
startup to get started with no real expenses; operate unincorporated
or incorporate on some small caribbean island, accept payment in
online currencies, have everyone communicate via PGP or ssh to the
box, and colo initially one but then several boxes with us.  Many
of our customers operate exactly like that.  For a sufficiently 
interesting project, we might even invest a box or two in it; so you 
could have zero startup or monthly costs during development.  

8) I can.
-- 
Ryan Lackey [RL7618 RL5931-RIPE]	ryan at havenco.com
CTO and Co-founder, HavenCo Ltd.	+44 7970 633 277 
the free world just milliseconds away	http://www.havenco.com/
OpenPGP 4096: B8B8 3D95 F940 9760 C64B  DE90 07AD BE07 D2E0 301F