IRC chat today! Join us January 31, 11am California time, 19:00GMT, for our weekly setiQuest IRC chat at irc://irc.freenode.net/#setiquest
Or in a browser: http://webchat.freenode.net/?channels=setiquest
Lets make an agenda. I'll start with my suggestions, please feel free to reply and add your own.
I want to work more on the google+ project, but I need a few specific guidelines about how you want it to work for SETI. Maybe we can talk about it as a group today.
4. Status of the Galaxy Zoo/ SETI Live App - will there be a beta test
Good question. I'll ask them.
No Beta test - they do this internally. So, you'll have to wait till March.
A couple weeks ago Michael brought up the topic of goals. What is the goal of the IRC meeting? What is the goal of setiQuest? We should talk about goals.
Here is a nice article about the ATA and SETI.
Jrseti, you are correct, people will see RFI signals and think they have discovered SETI. False announcements or people just getting over-excited are not good for the project. My personal mantra is that every signal in the setiQuest data is RFI unless I can prove otherwise. The proving is the difficult part. I think the solution is in educating users. Show them that there are an incredible number of signals in the data and make sure they know that this isn't going to be easy.
Don't flame too hard, I am simply thinking aloud. I asked myself what it is that I get from the Seti Quest, and the answer came back that I get access one hell of a Radio Telescope ( true I don't get to steer it). The data that you put up on the Web is exactly what you get to work with. Although your primary interest is Seti, that isn't all that is buried in the data that you provide. If you change your format to only supplying the data that simply is adjacent to what you consider an interesting signal, you maybe throwing away data that I might find interesting.
I also started thinking on why it is that there is so little activity on the site, and came to the conclusion that unless you are a Sigblips look alike, there is very little you can do with the data. You join up thinking you could be the one that detects the “Take me to your leader” and all you get is a pile of random numbers and get told well it might be in there somewhere. No wonder they rapidly loose heart and rapidly vanish from the scene.
So what is my suggestion. Well for one thing I find frustrating is that the effective bandwidth for getting access to the data is abysmal – due to other calls on my broadband, and my limited pension I can only download one of your files a month. I would be really happy if I could go onto the Seti Institute Shop and buy a volume of say Crab nebula DVD discs. I would be quite happy to pay over the odds for it, as this would give me access to signals almost immediately – without having to wait 6 or so months before I can start a project. So I thought if the Seti Institute could make a small profit on the data, why not go one step further and supply an ATA Radio telescope console. My vision here was maybe something like a large capacity USB memory stick, which could act as the storage requirement for one volume of data, a set of discs and a layman's GUI that the average person could download his current data volume (e.g. the Crab nebula) and some nice buttons that will allow them to explore the data, maybe the data discs could provide something like a script that would allow instant steerage to interesting objects. The same mechanism could be used to publish the 'coordinates' of any interesting signals discovered by the users on the forum.
From my experiments one can sacrifice computation speed by doing the silly approach of putting the display in the loop. I find it absolutely fascinating to watch the signal grow out of the noise as each iteration is displayed – and wouldn't be surprised if most people found the same, given the choice of rapidly changing stream of numbers to a more slowly evolving graph – I know what I would choose. I well remember the excitement I got watching the 0329 pulsar shape slowly emerging from the noise, and thinking that I am actually seeing a rotating neutron star.
Obviously the SETI staff don't have time to develop such a console – that's not what they are paid for. However providing we can get a requirement specification document written I would think that we could find enough volunteers to build it. The GUI must have a straight forward API so that further tools can easily be developed and simply integrated into the console; the only problem being the authors willingness to let the Seti Institute sell their work for a slight profit. You never know if you could sell enough of these, you could employ someone to interface more regularly on the forum.
Just a thought. Hope it is more use than all of my other contributions.
I agree it would be useful if participants of your calibre could purchase and receive bulk observation data on hard media.
You mention that you don't "get to steer" the telescope. This is a big missing piece towards generating a thriving setiQuest community. The setiQuest community should have allocated to it a sliver of telescope time.
An idea I support, which has been presented in various forms in the past, and implemented once incompetently in the form of Cloudant, is to not distribute the data, but rather for most users to act on the data in a central location by submitting algorithms.
A social website might be developed (no, not setiQuest in its current form), with a chat area, an area where proposals could be submitted, and area where proposals could be voted on, and an area where everyone could see the results of the experiments (executed proposals) of others in the form of tabular data and graphics and comment on them. Proposals, that involve telescope pointing and data capture might have difficulty bubbling to the top due to the severely constrained resource. But, the community would decide. Proposals, that involve running algorithms against data (sandboxing details and languages to be worked out) would most likely execute in a timely fashion, especially at first. The proposals would be stacked in a queue, kind of like job submission in an old IBM mainframe computer.
It is important their be no constraints on signal types. Note that compamp files and current GZ plans are same-old-same-old Fourier transform data output.
PS Dave, please sent to my home email address (rob @ ackrman dot net) your mailing address in England and a description of your main research computer (e.g., OS type).
Personally I don't feel the desire to steer the telescope. The reason why is because I don't have a good reason to point it in any particular direction. I've found thousands of signals in the setiQuest data and trying to follow them all up seems pointless, if not impossible. Now if I had access to multi-beam data and I found something interesting then that would be different. In such a case I still wouldn't need to steer the telescope myself, announcing my findings would probably get the ATA operators interested enough to investigate on their own. If this happened a lot then a board as you suggested could be setup to manage this but if that was necessary we'd likely have a bigger problem, a technical problem.
All that said, being able to steer the telescope may be a good marketing tool to get people interested in the project. You know, people want to be in the drivers seat and feel in control. The Discovery / SETI Live press release even made an issue of steering the telescope:
Can you point to any evidence on the setiQuest forum or elsewhere of anyone wanting to be in control of the telescope? Seems more like a convenient way to dismiss a perfectly legitimate concern. What Rob was talking about was the need to be able to do even a tiny bit of follow-ups for any meaningful transactions to occur. There's a big difference between "severely constrained" and "not happening at all".
Yes, thank you Anders, that is correct - the desire would be for an experienced SETI Institute observer to "control" and "steer" the telescope. The setiQuest community would establish the observation parameters (e.g., RA/Dec., frequency, ...). Then, after the scheduled observation, the recorded data would be made publically available in a timely manner.
If I were to submit a proposal to the community today, it would be to reobserve BD144586 to see if more "Herpy Dots" occur. The community might hold up my proposal, if there were other observing proposals in the queue, arguing that more datasets should be searched first for the absence of "Herpy Dots."
The real-time multi-beam .compamp files Jrseti mentioned could solve the bandwidth overload problem you're experiencing. The .compamp files consist of about 92 seconds of 4-bit quadrature samples at a 711 Hz width. The files are 71 KB in size and about half that if compression is used. The .archive-compamp files mentioned are about 16X wider and larger.
Yes, this is a tiny sliver of bandwidth compared to the 8.7 MHz width of the old setiQuest data files but the real-time and multi-beam qualities are a major advantage. I'd prefer real-time access to multiple beams of the full 104 MHz beamformer output but that isn't going to happen anytime soon for a number of reasons. What seems promising from the IRC chat is that this .compamp data initiative looks like it will happen. I'm hopeful but we'll need to wait and see.
One thing worries me about the proposed 'compamp' files (what is a compamp? I have never met the phrase before); are we losing the full data files? The point I was trying to make in my previous posting was not that I can't steer the telescope, indeed like you I have no desire to. The point I was trying to make was that someone is pointing one hell of a telescope at interesting astronomical objects, and giving me access to all of the data. This contains a lot more interesting stuff than simply SETI signals. I have enjoyed extracting pulsar signatures, investigating low frequency oscillations from the Crab pulsar, investigating the effectiveness of Wavelet transforms on the telescope signals. If all that is now going to be supplied is a small slither of signal surrounding areas of potential SETI detected signals, the effectiveness of being able to use the data for non-Seti applications will vanish.
I know that SETI Quest primary motivation is the search for ET, and not to provide Dave Robinson some interesting astronomical data to play with, however the ability to see real astronomical information buried in the noise, which you might be about to filter out of the data stream is really quite worrying.
I agree Dave ... furthermore, ET does not necessarily send signals that have highest signal-to-noise in Fourier transform space, therefore, the "compamp" files may filter ET too. The original idea was not to constrain setiQuest searchers to simply follow what is mostly being done automatically by SonATA, but rather to provide an experimental space where a discovery might be made using an alternative method.
(what is a compamp? I have never met the phrase before)
"compamp" is short for "Complex Amplitude" It is the raw, but bit-reduced, output of the Fourier transform not detected (squared to power)
The .compamp sample data isn't the output of the Fourier transform any more than the original 8.7 MHz setiQuest data was. It's just a bunch of packetized time-domain quadrature samples of a narrower bandwidth.
I suspect the SETI Institute's meaning and use of the .compamp format has changed over time. Originally it might of been frequency-domain complex amplitude values but it isn't anymore.
It is narrow enough, in context, for me to refer to even the independent stream as Fourier transformed data (through the channelizer and DX). I must be careful now, because if we have future interactions, there might be confusion between us if you were to refer to an independent 1Hz bin as a "time-series." Suppose compamp files surrounding a candidate signal were placed side-by-side, which is what had been previously suggested by Jill in the chat to search for "wider-bandwidth" events, would you refer to the compamp files as a set (two or more) as Fourier transformed data? Just wondering...
I would refer to the two "compamp files surrounding a candidate signal placed side-by-side" as two narrow bandwidth files of time-domain quadrature samples that overlap slightly in frequency.
When people see "Fourier transformed data" they immediately think frequency-domain samples. The Channelizer/DX may be using the Fourier transform to accomplish its filtering and resampling but the output is quadrature time-domain samples. I'm just trying to use terminology that avoids time / frequency domain confusion.
A .compamp file is generated by the DX in SonATA when it detects something interesting. It is a packetized file format that has 92 seconds of 4-bit quadrature samples at a 711.1111 Hz sample rate. The files are small (72 KB) and are generated for multiple beams and polarizations. The .compamp files will be exported to the cloud to feed the GZ SETI Live project. The infrastucture is mostly in-place and the setiQuest project "might" get access to this data.
Note that there is an .archive-compamp file format that is a wider version (16 subbands) of the basic .compamp file.
Dave, I agree with you that I'm not happy about the downsizing of the setiQuest data. I'd like to have access to the multiple beams and polarizations of the full 104 MHz bandwidth. But unfortunately this isn't going to happen anytime soon for a number of reasons (a lack of resources, narrow Internet pipe out of Hat Creek, ...). I'll keep hoping though.
On the bright side here are a list of benefits of a .compamp file stream:
Think of all the new analysis possibilities ...
[11:00am] _Jens_: Hello everybody
[11:00am] gagan: jrseti_: i wish to enroll again....can i?
[11:01am] sigblips: Greetings Earthlings!
[11:01am] jrseti_: gagan - yes, I'll email you today and we can discuss.
[11:01am] Michael_Mo: Greetz
[11:02am] jrseti_: Hello all: It is 11am California time
[11:02am] leash: Hello.
[11:02am] jrseti_: The agenda is built from the discussion at http://setiquest.org/forum/topic/community-meeting-2012-01-31
[11:02am] gliese581c: Title: Community Meeting 2012-01-31 | setiQuest (at setiquest.org)
[11:02am] janebird joined the chat room.
[11:02am] jrseti_: Agenda:
[11:02am] jrseti_: 1) Goal of the IRC chat and SETIQuest? making data be a more central theme for setiQuest.
[11:02am] jrseti_: 2) Website issues, spam, any design suggestions or topics?
[11:02am] jrseti_: 3) Google+
[11:02am] jrseti_: 4) The poll. let's make a new one. Anyone want to write a blog about the results of the current one?
[11:02am] jrseti_: 5) Galaxy Zoo / BETA?
[11:02am] jrseti_: Anything else?
[11:03am] gagan: jrseti_:ok.....
[11:03am] Michael_Mo: Let's roll!
[11:03am] jrseti_: OK ---
[11:03am] JIll joined the chat room.
[11:04am] JIll: hi
[11:04am] jrseti_: Last week we discussed: What are our goals.
[11:04am] jrseti_: The discussion pointed to the fact that the data and the analysis of data seems to be the big attractor
[11:04am] jrseti_: Do I have that right?
[11:05am] Michael_Mo: I think so as a consensus of what we were attracted to, but no bottom line on the GOAL of SQ
[11:05am] JIll: does it cost us anything in resources to keep SonATA code published on github?
[11:05am] sigblips: Data is important to me. What is important to other people?
[11:05am] jrseti_: Jill: no
[11:06am] Michael_Mo: I think, Jill, that the point was that the SonATA code was not as interesting as the ATA data
[11:06am] JIll: discovering a signal - anyway we can
[11:06am] jrseti_: We should keep the SonATA source code open sources and on GitHub
[11:06am] JIll: Micheal - just verifying that there's no cost to continuing to make it available openly
[11:07am] Michael_Mo: Okay. that's fine, but the goal of SQ is still needed
[11:07am] Michael_Mo: "Any way we can" is not a plan, of course! :)
[11:08am] _Jens_: I guess everybody here wants to contribute to SETI in a way he or she can. So far only people with certain special knowledge can contribute, but that'll change soon I guess.
[11:08am] _Jens_: So the goal for SQ is to make that possible
[11:08am] jrseti_: Jill: What is the "official" reason setiQuest was started? The thing you say in speeches?
[11:09am] Michael_Mo: Yes, that was our original broad, excellent, goal. The question is whether we have any current role in that. Would our time be better spent holding a bake sale for Galaxy zoo or UC Berkeley, for example?
[11:10am] JIll: it's part of the wish - to empower Earthlings everywhere to become active participants in the ultimate search for cosmic company
[11:10am] jrseti_: Jill must be looking it up
[11:11am] sigblips: Maybe we should ask what do we want from people? A goal might pop out traversing that route.
[11:11am] Michael_Mo: Well-said, siggy
[11:11am] Michael_Mo: (may I call you siggy?)
[11:11am] sigblips: Sure.
[11:12am] jrseti_: One thing we wanted was people to help or review or understand our source code. That did not happen.
[11:12am] JIll: i sent a sugestion of what we might ask for help with this morning when Jon asked for agenda items
[11:12am] sigblips: Sigmund and the Sea Monster was a favorite TV show of mine as a kid.
[11:12am] jrseti_: Jill: Did not see the item
[11:12am] Michael_Mo: So, with x billion humans out there, what do we want from them. The lack of coding is sad, but not the end of the world.
[11:13am] JIll: jrseti - posted it to forum
[11:13am] jrseti_: where
[11:14am] JIll: i think we need to make our request for code help more specific - the SonATA code is pretty intimidating
[11:14am] JIll: jrseti- in response to your suggested agenda - i thought
[11:14am] jrseti_: I have asked. There is a group of open source project suggestions on the WiKi (that probably need to be revamped)
[11:15am] jrseti_: Jill: forum post did not "post"
[11:15am] JIll: argh..
[11:15am] JIll: let me tyoe off line for a moment
[11:15am] jrseti_: It has been expressed more than several times in the IRC chat - coding is hard and people are busy.
[11:16am] Michael_Mo: and most open source projects are based on code that the coders may use themselves, not so applicable here.
[11:16am] jrseti_: Maybe that is correct.
[11:16am] sigblips: Yes that is an issue with SonATA.
[11:17am] Michael_Mo: But we've all agreed it would be a miracle if even a half dozen coders showed up.
[11:17am] Michael_Mo: So, what about the other eight billion humans.
[11:17am] gagan left the chat room. (Quit: Page closed)
[11:17am] Michael_Mo: sorry, seven billion
[11:18am] jrseti_: It comes back to data, i think. If we can devise a way for more than a handful to help us analyze the data
[11:18am] jrseti_: In some form or another - don't know what that is
[11:19am] jrseti_: The goal is to allow the world to help with the search. Analyzing the data is about the only way.
[11:19am] Michael_Mo: Right. But is there support for SQ executing on it
[11:19am] jrseti_: FYI: I posted a new project idea at http://setiquest.org/wiki/index.php/Main_Page#Open_Source_Projects
[11:19am] gliese581c: Title: Main Page - setiquest wiki (at setiquest.org)
[11:20am] JIll: analyzing data is only part of it - the entire framework of how you interpret the analysis and respond to it is just as important
[11:20am] jrseti_: Michael: support is minimal. So what we do has to be done with minimum support
[11:20am] jrseti_: Suggestion:
[11:20am] JIll: here's what i wrote this morning (sort of)
[11:20am] sigblips: You should add http://setiquest.info to the main setiQuest page too. Maybe in one of the menus and/or the main page.
[11:20am] gliese581c: Title: SetiQuest Info Service (at setiquest.info)
[11:20am] JIll: I was suggesting that we make an art gallery out of some of the signals ‘siggy’ is preparing for GZ as test dataset. Then invite people to create different data visualizations of the data that might make signals more visible – the sort of thing that David Robinson did, but schemes cannot be that complex. We’d evaluate them for detectability as well as compare them to CPU requirments for current renderer.
[11:21am] jrseti_: Sigblips: I willonce I get the kinks worked out. Should be soon
[11:21am] Michael_Mo: I like that idea, Jill
[11:22am] Michael_Mo: But, as with the other ideas, such as my own game layout, or even Galaxy Zoo, then what?
[11:22am] jrseti_: I think we should post all the signals we see, with the compamp data. That will attract attention and give us an opportunity to educate them on RFI and how we spot true signals.
[11:22am] sigblips: It's a good idea. People like looking at pictures. But transforming that eyeball gazing into something useful for the project won't be easy.
[11:22am] jrseti_: With not much support needed, it can be automated
[11:23am] jrseti_: Sigblips: it is a start. then maybe someone will be interested to help us develop something around it
[11:23am] jrseti_: One usefulness that will come out of it is educating people about the signals
[11:24am] sigblips: Maybe we can make it a game. For example: Here is a compamp file with some RFI in it. Now what is it?
[11:24am] Michael_Mo: Alien or Cell Phone? Cute
[11:24am] JIll: but as Micheal keeps saying - there's not much support for developing that infrastructure. it is getting developed NOW for SETI Live -- so i think setiQuest should become a channel for infusing improvements into SETI Live
[11:25am] jrseti_: Jill: True, but we can't put 100% into SETI Live because if SETI Live "dies", we have nothing to fall back on.
[11:25am] JIll: we went to a lot of trouble to get GZ funded to develop SETI Live - and we used them because they have already demonstrated that they are the gold standard for citizen science.
[11:26am] JIll: i humbly submit if GZ can't make SETI Live a success - no other effort will, and we should then chalk it up to a failed experiment - machines can do SETI in a systematic way that humans can't
[11:26am] Michael_Mo: Nothing wrong with that. They seem like cool guys who know what they are doing. The question remains whether SetiQuest has any useful function in light of that.
[11:27am] jrseti_: Michael: yes, that is correct. If we decide 100% to go with GZ, what about setiQyest?
[11:27am] JIll: yes - i think we become an arena where improvements to SETI Live can be developed and vetted before being implemented by GZ
[11:28am] jrseti_: Jill: Can you explain that a bit?
[11:28am] Michael_Mo: So... you think Galaxy Zoo developers will go for that? It would be unusual behavior.
[11:29am] JIll: by its nature GZ has to be fairly hostile to 'suggestions' of improvements. they do their development in a standard way and maintain their projects with small staff
[11:29am] Michael_Mo: Of course. That's how I behave too.
[11:30am] JIll: jrseti - suppose David Robinson posted to GZ his ideas about better signal processing? what would they / could they do about that? we can try to grow a community that filters and develops incremental improvements tha GZ can then swallow
[11:31am] jrseti_: Jill: That is fair enough. A big part of that is data. Posting data, educating people about the data, educating them about the process.
[11:32am] sigblips: Maybe education should be one of setiQuest's goals?
[11:32am] jrseti_: And we have the means now to automate posting data to the server.
[11:32am] jrseti_: I am thinking about posting the "*.archive-compamp" and waterfalls, not the data we post for GZ
[11:33am] jrseti_: We need to educate more people before we will start to get much good feedback
[11:34am] JIll: education has always been one of setiQuest goals
[11:34am] jrseti_: If we post new data we can generate some forum topics about the data. Along the way will come education
[11:34am] JIll: there are a bunch of tutorial materials on the site - i'm using some of that to write copy for SETI Live
[11:35am] jrseti_: So, back to goals for setiQuest - anything decided yet?
[11:35am] sigblips: How many .archive-compamp files are you envisioning? What prompts them being posted? When ever SonATA sees something interesting?
[11:35am] _Jens_: Who's the target audience for this education?
[11:36am] sigblips: I think it has to be aimed at different levels. Not everyone has the same background.
[11:36am] Michael_Mo: Or cares to develop it.
[11:36am] jrseti_: these files are created by SonATA whenever a signal is detected. The compamp data is saved along with 8 subchannels on either side. Not that many, but depends on how noisy things are.
[11:37am] jrseti_: So, these will be more than just a lot of static
[11:37am] Michael_Mo: That might be more interesting than just plain noise
[11:38am] JIll: there will be lots of static in this - usually only one signal in one subchannel, but 16 surrounding subchannels are saved.
[11:38am] jrseti_: Jens: Good question - I would say the people that are interested in helping with the search, but have no time or resources to write code.
[11:39am] JIll: note also these are bands that SonATA can deal with - not the crowded bands being sent to SETI Live
[11:39am] jrseti_: Jill: Yeah - probably just want to use the one subchannel?
[11:39am] JIll: if you do that you preclude possiblity of recognizing broader patterns
[11:40am] sigblips: Any signal wider than 533 Hz won't fit in a single subband. Basically that's anything wider than a narrowband signal. Or a fast drifting narrowband signal.
[11:41am] JIll: having said that - we don't examine the archive for broader patterns - maybe something setiQuest could do
[11:41am] Michael_Mo: Aliens who cannot compress their signal better than 533 Hz do not deserve our notice.
[11:41am] Michael_Mo: (kidding)
[11:41am] sigblips: That would mean a low baud rate.
[11:43am] Michael_Mo: yes
[11:43am] JIll: but such patterns might help us recognize the source of the interference and schedule around it.
[11:44am] JIll: there is a huge amount we can learn from our archive about our signal environment - we don't have the internal resources to mine that data.
[11:45am] JIll: OTOH we don't have the internal resources to respond to all claims of detection of ETI signals, and i'd like to do better than seti@home which just ignores its email
[11:46am] sigblips: They must get a flood of email.
[11:46am] JIll: so a self-organized and self-policing group of setiQuest volunteers who take on this functionality might be good idea.
[11:46am] jrseti_: We need to be clear about this in setiQuest.org. Be upfront, and try to educate. No email - we use the forum, maybe others will help
[11:46am] jrseti_: Jill: yes
[11:47am] Michael_Mo: Yes. Agreed. I offered up a scheme for that. At the end, however, there must be a commitment to look at at least one signal at at least some time, or it's just playing
[11:47am] JIll: also SETI Live (GZ) finally hired a data 'scientist' to work with the volunteers and their efforts to understand the sources of signals - we should leverage that too
[11:47am] jrseti_: yes
[11:48am] JIll: Micheal: please remind me of your scheme
[11:48am] Michael_Mo: Basically a feedback loop, where "hits" would get vetted by other examiners.
[11:49am] Michael_Mo: THe "success" of a hit would raise both its "value" and the reputation valiue of its finders.
[11:50am] JIll: so the difference between the setiQuest data in AWS now, and this proposed archive of stored .compamp data is that there will be at least one signal in each dataset? is this the best way to find new kinds of signals?
[11:50am] Michael_Mo: An algorithm would evaluate a weighted valiue for a suspected hit, and when the value came up to a sufficiently high level, it would get attention at the top
[11:51am] JIll: it could be - we realize that some filter has to be invoked - not many folks have worked through the raw data we supplied the way sigblips has.
[11:51am] Michael_Mo: Well, that is a new factor. More than a "hit" needs to be found. But still a "this is what we are looking for" could be the target
[11:51am] Michael_Mo: or it could be run in the reverse method: "these are NOT the droids you are looking for."
[11:51am] sigblips: Jill: It would definitely be a faster way of finding signals. Let SonATA do the hard work, that is what it's good at.
[11:52am] Michael_Mo: But yes, definitely needs a filter Or filters. But the data do not care how they get looked at
[11:53am] JIll: but by definition, SonATA has already done the hard work and found those signals - what is the value added?
[11:53am] sigblips: ID? A second opinion?
[11:54am] jrseti_: Education and maybe seeing signals within that data that SonATA did not see?
[11:54am] Michael_Mo: Either better finding or better sorting. If not, shut the program down right now.
[11:54am] janebird: Most of the candidates that Sonata is finding and archiving are not visible to the eye.
[11:54am] jrseti_: A second opinion - that is a good one. Maybe SonATA occasionally detects signals that are not there?
[11:54am] JIll: ultimately i think the value added will be the datamining that finds any correlations among discrete signals over the spectrum, across space, and time that we are now blind to
[11:55am] jrseti_: We are not making much progress on he agenda 9.9
[11:56am] sigblips: True, but this is really good stuff to be talking about.
[11:56am] jrseti_: yes, much needed
[11:56am] JIll: if we have set the thresholds correctly we should be working where we have a 50% probability of missing a signal
[11:56am] jrseti_: Jane has a good point.
[11:58am] JIll: perhaps this analysis can help us find a better set of threshold values - with RFI in the picture they always have to be set empirically rather than theoretically and we've not had the luxury of lots of test time prior to observations.
[11:59am] JIll: so post-analysis of observational results should inform of of how well we are doing.
[11:59am] sigblips: Or you could set the thresholds so that you get a lot more false signals and then let the SETI Live army sort it out.
[12:00pm] JIll: just the sort of scientific/statistical analysis the project leader should be doing - hmm...
[12:00pm] Michael_Mo: Hmmmm
[12:01pm] JIll: sigblips: with that approach we end up following up on every observation and observing efficiency crashes - though it certainly will do that at the start of SETI Live.
[12:02pm] jrseti_: I have asked the SETI Live folks if there will be a BETA test. Have not heard back yet
[12:02pm] JIll: time to start driving. thanks. bye
[12:03pm] jrseti_: Hey all - time is up. Good discussion! We can post further thoughts to http://setiquest.org/forum/topic/community-meeting-2012-01-31
[12:03pm] gliese581c: Title: Community Meeting 2012-01-31 | setiQuest (at setiquest.org)
[12:03pm] leash left the chat room. (Quit: Page closed)
[12:03pm] jrseti_: We'll get to the other agenda items next time.
[12:03pm] sigblips: This was a really good meeting.
[12:03pm] JIll: no reply yet - my guess is yes, but limited to volunteers they've worked with before. they are taking site mock ups to the floor of the Adler for look and feel checks
[12:04pm] JIll left the chat room. (Quit: Page closed)
[12:04pm] jrseti_: In the mean time, it may be good for some of us to come up with one or 2 sentences that defines a goal for setiQuest - based on today's discussion
[12:04pm] jrseti_: Watch http://setiquest.org/forum/topic/community-meeting-2012-01-31
[12:04pm] gliese581c: Title: Community Meeting 2012-01-31 | setiQuest (at setiquest.org)
[12:05pm] jrseti_: bye!
[12:05pm] sigblips: OK, I'll post this transcript to the forum.
[12:05pm] sigblips: bye
[12:05pm] jrseti_: great
This was the best IRC chat we ever had. And the topic was ... data!
[11:26am] JIll: i humbly submit if GZ can't make SETI Live a success - no other effort will, and we should then chalk it up to a failed experiment - machines can do SETI in a systematic way that humans can't
If this project does not become a success, please remember that you made a strategic decision to ignore the advice of myself, Karsten Wade and many others. Hence any lack of success can be ascribed to the failure of the scientist's decision-making capabilities - not to the underlying premise that SETI can be done more effectively and more efficiently by involving outside contributors.
You seem to have very high expectations for GZ's ability to be the sole authority on what constitutes good science, but I certainly do not, and I doubt a lot of bona fide scientists do. It's not exactly in line with the ideal that science should be capable of withstanding scrutiny from other researchers and even the public.
However, the criteria of success seems so low it will probably not be a problem. The media will print a couple of routine "wow, aliens!" articles, and that should be enough to keep the buzz going and the money aflowing.
I concur with Anders, but must admit to some bias after having been rejected by GZ as "unqualified."