Is it a 'real' or a 'CM'?

Discuss your experiences with and ideas about Stardust@home here.

Moderators: Stardust@home Team, DustMods

jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Is it a 'real' or a 'CM'?

Post by jsmaje »

Many questions in these forums depend on whether the movie referred to is 'real' or a 'calibration'.
There was an assumption early in phase-one that you couldn't display CMs, which was never true - see here.

In fact, all you need to do to know what type you're dealing with, even with phase-two movies, is R-click on the image, go to its Properties (I'm talking IE here; not sure about other browsers), and see whether the 'Address/URL' field includes either "... / real /..." (i.e. it's a real movie), or "... /test_train/ ..." (i.e. it's a CM).

I hope this isn't giving away any state secrets, but if so, then the state needs to up its act!
DustSabre
Posts: 63
Joined: Sat Dec 09, 2006 4:51 pm

Post by DustSabre »

Wow, you're right. With just a quick right-click, you can tell in a second whether any given movie is a calibration or a real movie. It doesn't say what sort of calibration movie it is, but that doesn't really matter a whole lot. I wonder if you can even figure that out by looking at the webpage code.

Does this constitute a new way to cheat? Opinions, anyone?
the moon
Posts: 177
Joined: Thu Aug 03, 2006 12:34 am

Post by the moon »

Hmm I never knew about that. That will be useful for me when I see an inclusion and I don't want to click it but I do anyway thinking it might be a CM. Now I can tell for sure and don't have to fill my events list with things I already know aren't tracks.

And yes it is also a way to cheat, but theres only about 500 people left still dusting. If any of them are still only concerned about their score and not finding tracks, then... I don't know, that'd be weird.
fjgiie
DustMod
Posts: 1253
Joined: Sat May 20, 2006 8:47 am
Location: Hampton, SC, US

The Dirty Dozen

Post by fjgiie »

The Dirty Dozen

Boy, I wish the “Ch” word had not been brought up.
Here is a little test that I would like for us to take. The answers will be given later.

What from this list do you consider Ch’ting ?

1. Using a screen magnifier. IE 7 has a good one.
2. Using strong reading glasses to magnify the tracks.
3. Breaking into the Stardust site and raising your score.
4. Breaking into the Stardust site and changing the code so it is worthless.
5. Right clicking the movie to check if it is a real movie before clicking a track.
6. Using java script and HTML to write a code that will tell you CM or real movie.
7. Using java script and HTML to write a program that will click the tracks for you while you sleep.
8. Keeping a list of CM’s that you have trouble with.
9. Working through the night just to get an advantage on fellow dusters.
10. Studying the forum posts and other literature so as to gain an advantage on fellow dusters.
11. Spending more time on CM’s than real movies.
12. Searching for dust at times you should be taking your family out.
13. Baker’s question: Using the events viewer to check if a movie is a CM or real.
Nikita
DustMod
Posts: 994
Joined: Wed May 17, 2006 8:33 pm
Location: Indiana, USA

Post by Nikita »

Well, #12 is clearly wrong. Of course, burning dinner because of dusting isn't wrong...unless you are my family! :lol:

Seriously, it really comes down to this: if you need to right click to find out if it is a calibration movie to get a higher score, well, I say, so be it. A few people doing it won't affect the total project. Science and understanding are the ultimate goal. But for the brave hearts willing to risk all and click on that thing, risking their score to say "Well, this might be something..." I say "Hurrah!"

John and all, I'm not saying you are wrong for figuring it out, in fact, it is the creative and imaginative that lead the way for new things. But with so few dusters, I challenge you to stick your neck out and go for it! After all. like fjgiie pointed out, any computer program can be made to find the CM's.

We can do one of two things: make the already overburdened Stardust@home team do something more and make them mask the properties so we don't skew the scores of a project we have already put so much time into or decide that we can analize these without needing to know if they are real or not.

So what do you say...are you in?!? :P
From dust we come
marcia
Posts: 51
Joined: Sat Aug 05, 2006 1:04 pm
Location: bergen, netherlands

Post by marcia »

I agree with you, Nikita: stick your neck out and go for it, take the challenge.

But what I'm asking myself since the beginning of phase2 in connection with this problem, is if the team has looked for a rather long time to the CM's themselves? For if you have good eyes and a logic brain, and some experience, you can see rather good what CM's are, without to right-click or what kind of smart computer-things at all (and I don't believe I am the only one who can, t.i. most of the time, for I make my mistakes). And I know it's 'on purpose' to make the CM's difficult to detect, but then the CM's has to be, in my opinion, not to distinguish at all (or a very, very little bit :wink:) from the real movies. That's IMHO the only way to create a useful detection threshold.

Marcia
DustSabre
Posts: 63
Joined: Sat Dec 09, 2006 4:51 pm

Post by DustSabre »

1. Using a screen magnifier. IE 7 has a good one.
2. Using strong reading glasses to magnify the tracks.
3. Breaking into the Stardust site and raising your score.
4. Breaking into the Stardust site and changing the code so it is worthless.
5. Right clicking the movie to check if it is a real movie before clicking a track.
6. Using java script and HTML to write a code that will tell you CM or real movie.
7. Using java script and HTML to write a program that will click the tracks for you while you sleep.
8. Keeping a list of CM’s that you have trouble with.
9. Working through the night just to get an advantage on fellow dusters.
10. Studying the forum posts and other literature so as to gain an advantage on fellow dusters.
11. Spending more time on CM’s than real movies.
12. Searching for dust at times you should be taking your family out.
13. Baker’s question: Using the events viewer to check if a movie is a CM or real.
Well, if you ask me, only 3, 4, and 11 are real cheating. All others are legitimate, inasmuch as the stardust team hasn't specifically made it impossible to do them, and insofar as they don't seem to effect the project. I think that right-clicking to learn whether a movie is a CM or real ought to be allowed for anyone who wants to do it. It does slow you down somewhat, so I doubt it will become a habit with anyone, rather it will probably be used when you think that something might be an artifact, but you're afraid that you're looking at a CM and so you click on it anyway. That sounds fair enough to me. All you can know is that it is or isn't a calibration movie; you don't know which type, so you're still sticking your neck out when you click it one way or the other.
elainekeefe
Posts: 190
Joined: Fri Aug 04, 2006 4:38 am
Location: Massachusetts, USA

Post by elainekeefe »

I admit I checked out that right-clicking just to see if it were true, which, of course, it was. However, I hardly see the point. Even if I know it's a CM, I'm still going to click on the wrong dot, LOL. Also, checking so you don't click on an inclusion just because you think it might be a CM doesn't work either, since inclusions have suddenly, in Phase 2, become desirable. Actually, if someone can tell me exactly what it is we're looking for, I'd much appreciate it. The longer I work on Phase 2, the more confused I get. :lol:
solaris
Posts: 4
Joined: Mon Aug 13, 2007 9:22 am
Location: Ontario, Canada

Post by solaris »

elainekeefe wrote: The longer I work on Phase 2, the more confused I get. :lol:
Ditto. :)
the moon
Posts: 177
Joined: Thu Aug 03, 2006 12:34 am

Post by the moon »

elainekeefe wrote:Also, checking so you don't click on an inclusion just because you think it might be a CM doesn't work either, since inclusions have suddenly, in Phase 2, become desirable. Actually, if someone can tell me exactly what it is we're looking for, I'd much appreciate it. The longer I work on Phase 2, the more confused I get. :lol:
I still refuse to click on inclusions, but it's up to you now to decided what to look for. You've been doing this a long time, you have all the same information as everyone else. Only click on stuff that you personally think could be a track. Everyone else will do the same and the best candidates will rise to the top thru democracy. Not a perfect system but better then a dictatorship where everyone only clicks on what the team says they should click on.
jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Post by jsmaje »

It seems from the all the postings above that I have indeed revealed some sort of 'state secret'.
I can't imagine why - it's rather obvious (even though I only sussed it when writing my animated jpg thingy).

Firstly, I agree with DustSabre that "... it does slow you down somewhat, so I doubt it will become a habit with anyone."
Secondly, also with Nikita: "...stick your neck out and go for it!"
And in fact I personally don't make a habit of it. The challenge of deciding for myself is part of the satisfaction I gain from the project.

Thirdly, though, as the moon says: "That will be useful for me when I see an inclusion and I don't want to click it but I do anyway thinking it might be a CM".
So, if the team think this simple ruse might compromise the project in any significant way, then it's surely up to them to do something about it.

John

PS: I reckon fjgiie's 3,4 & 12 must be the worst sins (we've all done 12, but is there in fact any evidence of anyone having committed 3 or 4? )
Nikita
DustMod
Posts: 994
Joined: Wed May 17, 2006 8:33 pm
Location: Indiana, USA

Post by Nikita »

John,

Indeed, it does seem like a "state secret", especially to those of us who don't know enough to have figured it out!
I didn't think you used it all the time any way, you've been at this for too long. Who would cheat this whole time? It would have to get soo boring!
Hmmmm, #12... do you think the system was shut down only to get us reunited with our families before starting phase II?
From dust we come
jsmaje
Posts: 616
Joined: Tue Aug 15, 2006 8:39 am
Location: Manchester UK

Post by jsmaje »

Nikita wrote:...I didn't think you used it all the time any way...
No, I never did this prior to the break because (a) I never found the need, and (b) I never thought of it!

It wasn't until writing the animation program that I needed to know how to cater for both reals and CMs. I hadn't realised until then that their urls were different, but R-clicking gave the appropriate addresses (also evident from the SD@H html source; in IE, click View > Source).

In fact, on any web page you can simply R-click > Properties to see the url of an image. This is so basic that I can't think the team weren't aware of it. They eventually dealt with other CM give-aways in phase I (the address bar contents, jpg pixellation, etc.), so if this is now perceived to be another problem they will have to somehow anonymise the movie urls.

Meanwhile, I'll try and bite my lip while deciding if it's a track or an inclusion!
And as another example of a simple procedure to make life easier (of which I confess to having been unaware) may I recommend Wolter's F11.

John
fjgiie
DustMod
Posts: 1253
Joined: Sat May 20, 2006 8:47 am
Location: Hampton, SC, US

Right Clicking

Post by fjgiie »

If I may, I will reduce that list of thirteen down to three questions. 1 and 2 are not cheating. 3 and 4 are not cheating either, but criminal offences that can be prosecuted by law.

5, 6, and 11 are the only questions and I would like to express my ideas about these two really, because 5 and 6 are the same thing.
5. Right clicking the movie to check if it is a real movie before clicking a track.
6. Using java script and HTML to write a code that will tell you CM or real movie.
I do not get a vote on this one but if I did, I would vote to leave it as it is now. If that feature were taken away there may be tracks that would not get clicked on. The idea is to find real tracks. If a duster is worried about clicking on a clear CM, he/she/it may not click. By knowing that the movie is real, there is not that problem.

11. Spending more time on CM’s than real movies. This may get to be a very controversial subject. The idea of rushing through the real movies with a quick no-track so as to get to a calibration movie and score points was the orgional problem. That would be wrong and not help find real tracks. We do need to give real movies our best look because that's the only place a track can be found. The CM's have been certified to not contain real tracks.

As long as the times are within reason though, this question should not be a problem. I know I have trouble trying to click right in the center of a calibration track and it takes me a little longer. The mouse-pad has sticky syrup from eating and the mouse does not slide easily. It just takes a little longer to click a track than click no-track.

Like I said, just my own thoughts -
DustSabre
Posts: 63
Joined: Sat Dec 09, 2006 4:51 pm

Post by DustSabre »

That is true, insofar as it's possible to fastforward through the movies from calibration to calibration. However, you would need to manipulate the time spent on the real vs. calibration so as make it look like you weren't cheating. The only practical way to do that would be to use a computer program that pauses on each real movie to make it look like it's being studied for tracks, when it really isn't.
Post Reply