Originally posted by Gawdzilla
Inspired by an exchange with Xtraeme, I humbly offer
The purpose would be a research organization that uses standard scientific rigors to examine reports of signs of ETI on this planet (loosely defined,
its brief could include the solar system.)
There's usually only one opportunity to get something like this right. I know many people that participated in SETI that eventually got frustrated
with the software (or in some cases due to how intensive the CPU usage was it would crash hardware). These people don't return.
At the same time if the application doesn't provide some form of entertainment (ie/ the screen-saver mode for SETI) people aren't going to download
Also restricting it to a search for extraterrestrial intelligence assumes the application can, or should, only test one hypothesis. I envision it to
be much broader. UFOs are such a bizarre phenomenon in the sense that the hypothesis set
massive. I don't rule out macro-scale quantum manifestations, something as bizarre as inter-dimensional clipping, perhaps evidence of quantum
retrocausality, or strangest of all evidence of wholly or partially transcendent entities (intelligent or otherwise).
For people that weren't a part of the other conversation
I think this
provides a good high-level overview of the problem:
It's one thing to make a small tool that does automatic identification of airplanes, weather balloons, low magnitude planet / star identification,
etc and also notifies other human beings in the area of the sighting.
It's another to hook it up so as it progresses through the on-device automated identification, it not only notifies humans, it distributes
observational parameters to remote peripherals. Next would be the in-depth analysis step. With decent image processing software / audio decomposition
/ overlaying other data you could probably automate ~50-60% of the harder identifications. To decrease CPU load it would be best to break it up in to
distributable tasks using BOINC
. That still leaves ~40-50% that's manual. This also has to be distributed in an
efficient manner. My best solution is mimicking Amazon's Mechanical Turk
With a fully realized system identifications could probably happen within say 10-20 minutes. Anything outside of that would be a genuine unknown.
To copy and paste some of my ideas from this thread
Originally posted by nablator
The most difficult problem in ufology is methodology. Applying the scientific method to ufology is very difficult. What is the best way to make
progress, and not fall into the many traps that so many ufologists have fallen into?
- How to get uncontaminated data. Statistic analysis has failed due to 1. the huge number of sightings that cannot be confirmed or dismissed as
hoaxes. Discriminating between the real unidentified, the misidentifications and delusions is nearly impossible. 2. the socio-psychologic aspect of
You nailed the problem on the head. However I think this is solvable. Bluntly, the problem is the mechanism used to capture the report.
My current area of interest is computational and distributed tracking of UAPs using widely available technology (i.e./ GPS-enabled camera phones /w
tilt-sensors). I'm currently developing a distributed
mobile UFO detection, analysis, identification, and reporting platform
for the iPhone that I later intend to port to the RIM BlackBerry Storm
and Palm PRE.
The idea is simple, use the iPhone location services to notify other users within a certain radius of a persons UAP sighting to distribute the
workload of recording numerous angles of photographic data and to rapidly deploy people with additional sensory equipment to the location.
Not only would this software package increase public awareness it maximizes the number of people recording data (more first-hand reports); decreases
the lag between a sighting and the time it takes experts with equipment to analyze the event; and gives investigative reporters access to
up-to-the-second sightings which may have the happy side-effect of a professional camera crew recording a UAP live, up-close and personal.
One thing I'd like the application to do is attempt to automatically identify various obvious UAPs. For instance if a person is observing the night
sky and they see a bright object that isn't moving on the horizon I would want this application to use the accelerometer and the GPS device to
determine heading (ie/ have the person walk in the direction of the object thus determining N-E-S-W orientation) and then using the tilt-sensor
(oriented with the horizon transposed to the position of the object in the night-sky) to calculate azimuthal position.
With that kind of data I can quickly search a starmap database to quickly locate night-sky objects with a negative magnitude. Likewise I can query
against meteorological databases to check for weather balloon releases to try and auto-rule out obvious explainable sightings.
I've tried to contact some of the bigger names in the field, with the hopes that I can get a little funding, so I can properly execute on the
- How to verify a hypothesis. Fortean phenomena are not subject to repeated testing.
This is another reason I like the idea of such an application. If I open-source it and use a platform-independent programming language it can be
connected to any kind of sensory equipment. Thus as we start to formulate falsifiable hypotheses we can test them with all the devices deployed in the
field using this software package.
[edit on 13-4-2009 by Xtraeme]