Scott Raney
2018-06-12 01:12:58 UTC
I've got proxyfor.me updated to the latest versions of the frameworks
and the bugs fixed and am setting it up to do a shadow of a local
government for the next round of beta testing. Any of you have any
experience or suggestions for how/where to do this?
Some of the issues/constraints:
1) It's of course got to be an English-speaking area (I've got a
design for wiring in automatic translation, but that will have to wait
for some future version).
2) One of the things I plan to test is the proposal of allowing people
to collect their name off of the public voter rolls to do
authentication (I'm planning to allow people outside the area to
participate, but to keep the votes separated (registered in the
locality vs. other)). In Colorado that database is available for $50
(it's *much* more expensive in most other states, and many of them
require you to contact *each county* to build a statewide list).
Colorado also has an easy-to-use web site that will allow people to
look up their registration. I'm thinking to key off of the voterID
field, which is something no one would know or be able to spoof unless
it's their own registration (they use their name, zip, and birthday on
the State's site to look up their voterID, and the birthday isn't in
the public database). Unless of course they bought the database
themselves: I'm thinking to check for fraud via that by recording some
sort of device fingerprint to make sure that no more than a few
account creations come from any one machine. There are several
libraries out there that do this, and if a machine is flagged we then
use that to block entire IP address ranges. Given the massive number
of attacks against matchism.org from Russia and Ukraine I've had to
deal with (they've even cracked all of the CAPTCHA routines!) I'm
tempted to just ban those entire countries from accessing proxyfor.me
until we're ready to replace their governments with new ones that
aren't so tolerant toward, or even directly engaging in, this sort of
antisocial behavior.
3) The locality has got to have good public information sites,
including all the handouts that the local council members use to make
their decisions.
4) Choosing a combined city+county is appealing because ultimately
we'll get to dealing with the revenue side (budgeting), and the
majority of that comes from property taxes which are set by counties.
Surprisingly there are only a couple dozen examples of this in the US,
so I may have to work around this somehow eventually.
5) Even so, I'm going to have to extensively prune the list of
proposals: As is the case for the US congress the majority of things
voted on at the City/County level are non-substantive (like
recognizing national hot-dog day!) and there's no need to waste The
People's time on that crap.
I'm mostly looking at Denver (around 700K population), which meets all
the requirements above, but would like to consider others even if they
don't. The tradeoffs seem to be that big cities have a lot more varied
topics, but most of that relates to managing concessions (Denver city
council spends almost half of its time dealing with the airport and
the various sports and entertainment venues they own). That's boring
and IMHO not something the government should be micromanaging anyway
(i.e., the management itself should be contracted out). Boulder
(population 100K) would actually be almost ideal because of the
manageable size combined with a high participation rate (more people
typically show up for Boulder city council meetings than in Denver
despite the huge population difference, and Boulder has the most
awesome access to documents of any city I've seen), but I frequently
have to deal with the City of Boulder for various reasons (mostly
related to our rental properties) and so am somewhat reluctant to
paint a target on my back...
Any other factors or places I should consider?
Regards,
Scott
and the bugs fixed and am setting it up to do a shadow of a local
government for the next round of beta testing. Any of you have any
experience or suggestions for how/where to do this?
Some of the issues/constraints:
1) It's of course got to be an English-speaking area (I've got a
design for wiring in automatic translation, but that will have to wait
for some future version).
2) One of the things I plan to test is the proposal of allowing people
to collect their name off of the public voter rolls to do
authentication (I'm planning to allow people outside the area to
participate, but to keep the votes separated (registered in the
locality vs. other)). In Colorado that database is available for $50
(it's *much* more expensive in most other states, and many of them
require you to contact *each county* to build a statewide list).
Colorado also has an easy-to-use web site that will allow people to
look up their registration. I'm thinking to key off of the voterID
field, which is something no one would know or be able to spoof unless
it's their own registration (they use their name, zip, and birthday on
the State's site to look up their voterID, and the birthday isn't in
the public database). Unless of course they bought the database
themselves: I'm thinking to check for fraud via that by recording some
sort of device fingerprint to make sure that no more than a few
account creations come from any one machine. There are several
libraries out there that do this, and if a machine is flagged we then
use that to block entire IP address ranges. Given the massive number
of attacks against matchism.org from Russia and Ukraine I've had to
deal with (they've even cracked all of the CAPTCHA routines!) I'm
tempted to just ban those entire countries from accessing proxyfor.me
until we're ready to replace their governments with new ones that
aren't so tolerant toward, or even directly engaging in, this sort of
antisocial behavior.
3) The locality has got to have good public information sites,
including all the handouts that the local council members use to make
their decisions.
4) Choosing a combined city+county is appealing because ultimately
we'll get to dealing with the revenue side (budgeting), and the
majority of that comes from property taxes which are set by counties.
Surprisingly there are only a couple dozen examples of this in the US,
so I may have to work around this somehow eventually.
5) Even so, I'm going to have to extensively prune the list of
proposals: As is the case for the US congress the majority of things
voted on at the City/County level are non-substantive (like
recognizing national hot-dog day!) and there's no need to waste The
People's time on that crap.
I'm mostly looking at Denver (around 700K population), which meets all
the requirements above, but would like to consider others even if they
don't. The tradeoffs seem to be that big cities have a lot more varied
topics, but most of that relates to managing concessions (Denver city
council spends almost half of its time dealing with the airport and
the various sports and entertainment venues they own). That's boring
and IMHO not something the government should be micromanaging anyway
(i.e., the management itself should be contracted out). Boulder
(population 100K) would actually be almost ideal because of the
manageable size combined with a high participation rate (more people
typically show up for Boulder city council meetings than in Denver
despite the huge population difference, and Boulder has the most
awesome access to documents of any city I've seen), but I frequently
have to deal with the City of Boulder for various reasons (mostly
related to our rental properties) and so am somewhat reluctant to
paint a target on my back...
Any other factors or places I should consider?
Regards,
Scott