vertically flipped (upside down) movies and trajectories

[Ulrich Stern]

Hi John,

Ctrax appears to do some amount of vertical (upside down) flipping -- for both movies and trajectories -- and this flipping depends on the input video format!  Here what I found:
* compressed movies (M-JPEG, Microsoft Video 1): open non-flipped in Ctrax, y values in MAT-file are flipped
* uncompressed movies (created w/ VirtualDub): open flipped in Ctrax, y values in MAT-file are non-flipped

While there are easy workarounds (e.g., flip movie with VirtualDub etc.), it may be good to add this as a low-priority ticket.  (I looked into this after my MATLAB script to analyze trajectories produced strange results after I changed from uncompressed to compressed movies.)

My guess is (have not looked at code) that the reader for uncompressed movies incorrectly flips, and then another flip was added to correct the reader flip before the MAT-file is written.

Thanks,

Uli

[John Soong]

Might this have some impact on analysis? I'm trying to break acceleration and velocity into x and y components (to measure curvature) using the magnitude of velocity and acceleration multiplied by either the sin or cos of orientation angle (in addition to the problem of not ahving figured out where 0 degree orientation is, i.e. whether it is North or East).

But the combination of ambiguous flipping + ambiguous phase shift of orientation angle shouldn't affect calculations of curvature, right? (I'm using the relation K = || r'(t) x r''(t) || / ||(r't(3))^3||, where x is the cross product.)

2012/6/11 Ulrich Stern <ulrich...@gmail.com>

--
You received this message because you are subscribed to the Google Groups "ctrax" group.
To view this discussion on the web visit https://groups.google.com/d/msg/ctrax/-/rvdDRyKICnAJ.
To post to this group, send email to ct...@googlegroups.com.
To unsubscribe from this group, send email to ctrax+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/ctrax?hl=en.

--

John Riemann Soong Lüwen

Biochem, Physics UVA '13

"Every difference of opinion is not a difference of principle. We have called by different names brethren of the same principle. We are all Republicans, we are all Federalists." -- Thomas Jefferson

 

We want structures that serve people, not people serving structures. (l'Odéon, mai '68) 

[Ulrich Stern]

John Bender & John Soong,

In the compressed movie case (where the y values in MAT-files are flipped), using FixErrors becomes strange since the trajectories do not match the movie in the background.  While one could create an extra flipped movie as a workaround, this may make the flipping issue "low to medium" priority.

I have not used much of the Ctrax MATLAB code, so I will let others answer on impact on analysis.

Thanks,

Uli

On Monday, June 11, 2012 1:42:45 PM UTC-4, jrs5fg wrote:

Might this have some impact on analysis? I'm trying to break acceleration and velocity into x and y components (to measure curvature) using the magnitude of velocity and acceleration multiplied by either the sin or cos of orientation angle (in addition to the problem of not ahving figured out where 0 degree orientation is, i.e. whether it is North or East).

But the combination of ambiguous flipping + ambiguous phase shift of orientation angle shouldn't affect calculations of curvature, right? (I'm using the relation K = || r'(t) x r''(t) || / ||(r't(3))^3||, where x is the cross product.)

2012/6/11 Ulrich Stern <ulrich...@gmail.com>

Hi John,

Ctrax appears to do some amount of vertical (upside down) flipping -- for both movies and trajectories -- and this flipping depends on the input video format!  Here what I found:
* compressed movies (M-JPEG, Microsoft Video 1): open non-flipped in Ctrax, y values in MAT-file are flipped
* uncompressed movies (created w/ VirtualDub): open flipped in Ctrax, y values in MAT-file are non-flipped

While there are easy workarounds (e.g., flip movie with VirtualDub etc.), it may be good to add this as a low-priority ticket.  (I looked into this after my MATLAB script to analyze trajectories produced strange results after I changed from uncompressed to compressed movies.)

My guess is (have not looked at code) that the reader for uncompressed movies incorrectly flips, and then another flip was added to correct the reader flip before the MAT-file is written.

Thanks,

Uli

--
You received this message because you are subscribed to the Google Groups "ctrax" group.
To view this discussion on the web visit https://groups.google.com/d/msg/ctrax/-/rvdDRyKICnAJ.
To post to this group, send email to ct...@googlegroups.com.

To unsubscribe from this group, send email to ctrax+unsubscribe@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/ctrax?hl=en.

[Brian Tietz]

All,

Just wanted to add in that we've seen this same issue using .avi files recorded on a Casio camera. We're looking at just flipping trx.y in the .mat files, but videodub looks like a good workaround as well.

Brian

To view this discussion on the web visit https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.

To post to this group, send email to ct...@googlegroups.com.

To unsubscribe from this group, send email to ctrax+un...@googlegroups.com.

[John Bender]

VirtualDub, right? Only Roy calls it VideoDub...

This is actually a reasonably hard problem to solve correctly. Ctrax
uses a third-party library to read compressed AVIs, and Matlab has its
own methods, so if they interpret the same data differently, there's
very little we can do about it. Does this happen in every compressed
AVI?

One pretty easy workaround might be to add a checkbox in FIxErrors
that would let you flip the movie for display purposes. Does that
sound like it would resolve everyone's complaints? Or does the
FixErrors auto-tracking track upside-down in compressed AVIs, as well?

JB

>>>> ctrax+un...@googlegroups.com.
>>>> For more options, visit this group at
>>>> http://groups.google.com/group/ctrax?hl=en.
>>>
>>>
>>>
>>>
>>> --
>>> John Riemann Soong Lüwen
>>> Biochem, Physics UVA '13
>>>
>>> "Every difference of opinion is not a difference of principle. We have
>>> called by different names brethren of the same principle. We are all
>>> Republicans, we are all Federalists." -- Thomas Jefferson
>>>
>>> We want structures that serve people, not people serving structures.
>>> (l'Odéon, mai '68)
>>>
>>>
>>>
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ctrax" group.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.
>>
>> To post to this group, send email to ct...@googlegroups.com.
>> To unsubscribe from this group, send email to
>> ctrax+un...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/ctrax?hl=en.
>
>

> --
> You received this message because you are subscribed to the Google Groups
> "ctrax" group.

[Ulrich Stern]

Re: compressed AVIs, the third-party libraries used by Ctrax appear correct; compressed movies are displayed correct (non-flipped) in Ctrax.

I think there are two bugs in Ctrax:
1. the reader for uncompressed movies (which is non-third-party code as far as I recall) flips.  As a result, uncompressed movies are displayed flipped in Ctrax.  But I have tried only uncompressed movies created by VirtualDub, so I am not sure whether this applies in general to uncompressed movies.
2. the trajectories are likely always (compressed and uncompressed) flipped before they written to the MAT-files, but this problem is hidden in case of uncompressed movies because of 1.  (The two flips cancel each other for uncompressed movies.)

But this is only my guesses based on the behavior I saw.  Something else may be causing it.

Thanks,

Uli 

>>>> ctrax+unsubscribe@googlegroups.com.

>>>> For more options, visit this group at
>>>> http://groups.google.com/group/ctrax?hl=en.
>>>
>>>
>>>
>>>
>>> --
>>> John Riemann Soong Lüwen
>>> Biochem, Physics UVA '13
>>>
>>> "Every difference of opinion is not a difference of principle. We have
>>> called by different names brethren of the same principle. We are all
>>> Republicans, we are all Federalists." -- Thomas Jefferson
>>>
>>> We want structures that serve people, not people serving structures.
>>> (l'Odéon, mai '68)
>>>
>>>
>>>
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ctrax" group.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.
>>
>> To post to this group, send email to ct...@googlegroups.com.
>> To unsubscribe from this group, send email to

>> ctrax+unsubscribe@googlegroups.com.

>> For more options, visit this group at
>> http://groups.google.com/group/ctrax?hl=en.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ctrax" group.
> To post to this group, send email to ct...@googlegroups.com.
> To unsubscribe from this group, send email to

> ctrax+unsubscribe@googlegroups.com.

[John Soong]

Could you add an option to flip the Y-values when exporting to a MATLAB file as well? That way if a .mat file is coming out flipped, we can recognise that and re-load the movie and re-export it.

2012/6/13 John Bender <johna...@gmail.com>

[John Bender]

I understand why that sounds appealing, but I think it might be too
complicated in practice. Many users automate or batch-process their
movies, so this option would have to be configurable per movie at many
points within Ctrax. A checkbox in FixErrors would be an easy,
non-destructive fix that doesn't require any re-exporting.

JB

[John Soong]

Perhaps there are simple scripts to flipt trx values in MAT files if needed? I would write one myself, but I am not familiar with the limits of the coordinate system. For a 640x480 video, do the valid x values range from 0 to 639?

2012/6/13 John Bender <johna...@gmail.com>

[John Bender]

Either that or 1 to 640. It's easy to try it and see which. But if it's a vertical flip, you want the y-coordinate.

I still think it's better to flip the video display than the data model, though. It's probably as simple as applying flipud() to the image as it's read from disk.

JB

[John Bender]

A flip up/down checkbox for FixErrors is available in the 0.2.5
release of the Matlab toolboxes. Please let us know if it gives rise
to any further tracking errors (I didn't test auto-tracking with the
video flipped).

JB

[Ulrich Stern]

Thanks for adding the new checkbox.  Will try it soon.

I was wondering whether there is a reason to not fix the underlying issues in Ctrax?  I looked a little more into what I posted earlier on this:

1. the reader for uncompressed movies (which is non-third-party code as far as I recall) flips.  As a result, uncompressed movies are displayed flipped in Ctrax.  But I have tried only uncompressed movies created by VirtualDub, so I am not sure whether this applies in general to uncompressed movies.

When creating the uncompressed movie that opens flipped in Ctrax, I had used the recommended "Luminance only" in VirtualDub, and VirtualDub gives "Y8" as FOURCC.  Most movie players except VirtualDub appear to not open this format; SMPlayer does open it, and the movie appears non-flipped there.  So Ctrax seems to incorrectly flip Y8 movies.

I also created an uncompressed movie in color using Ctrax.  (The FOURCC is likely blank for this movie.)  Interestingly, it opens correct (non-flipped) in Ctrax.  Unlike the Y8 movie, it also opens, e.g., in Windows Media Player (non-flipped).

2. the trajectories are likely always (compressed and uncompressed) flipped before they written to the MAT-files, but this problem is hidden in case of uncompressed movies because of 1.  (The two flips cancel each other for uncompressed movies.)

I did a quick check of the coordinate systems in both MATLAB and Ctrax, and they appear flipped -- MATLAB has y=0 (or 1) at the top of an image, while Ctrax has y=0 at the bottom.  This would explain 2.  But this was a quick check only, so I may have missed something.

Thanks,

Uli

>> >> >>>> ctrax+unsubscribe@googlegroups.com.

>> >> >>>> For more options, visit this group at
>> >> >>>> http://groups.google.com/group/ctrax?hl=en.
>> >> >>>
>> >> >>>
>> >> >>>
>> >> >>>
>> >> >>> --
>> >> >>> John Riemann Soong Lüwen
>> >> >>> Biochem, Physics UVA '13
>> >> >>>
>> >> >>> "Every difference of opinion is not a difference of principle. We
>> >> >>> have
>> >> >>> called by different names brethren of the same principle. We are
>> >> >>> all
>> >> >>> Republicans, we are all Federalists." -- Thomas Jefferson
>> >> >>>
>> >> >>> We want structures that serve people, not people serving
>> >> >>> structures.
>> >> >>> (l'Odéon, mai '68)
>> >> >>>
>> >> >>>
>> >> >>>
>> >> >>>
>> >> >> --
>> >> >> You received this message because you are subscribed to the Google
>> >> >> Groups
>> >> >> "ctrax" group.
>> >> >> To view this discussion on the web visit
>> >> >> https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.
>> >> >>
>> >> >> To post to this group, send email to ct...@googlegroups.com.
>> >> >> To unsubscribe from this group, send email to

>> >> >> ctrax+unsubscribe@googlegroups.com.

>> >> >> For more options, visit this group at
>> >> >> http://groups.google.com/group/ctrax?hl=en.
>> >> >
>> >> >
>> >> > --
>> >> > You received this message because you are subscribed to the Google
>> >> > Groups
>> >> > "ctrax" group.
>> >> > To post to this group, send email to ct...@googlegroups.com.
>> >> > To unsubscribe from this group, send email to

>> >> > ctrax+unsubscribe@googlegroups.com.

>> >> > For more options, visit this group at
>> >> > http://groups.google.com/group/ctrax?hl=en.
>> >>
>> >> --
>> >> You received this message because you are subscribed to the Google
>> >> Groups
>> >> "ctrax" group.
>> >> To post to this group, send email to ct...@googlegroups.com.
>> >> To unsubscribe from this group, send email to

>> >> ctrax+unsubscribe@googlegroups.com.

>> >> For more options, visit this group at
>> >> http://groups.google.com/group/ctrax?hl=en.
>> >>
>> >
>> >
>> >
>> > --
>> > John Riemann Soong Lüwen
>> > Biochem, Physics UVA '13
>> >
>> > "Every difference of opinion is not a difference of principle. We have
>> > called by different names brethren of the same principle. We are all
>> > Republicans, we are all Federalists." -- Thomas Jefferson
>> >
>> > We want structures that serve people, not people serving structures.
>> > (l'Odéon, mai '68)
>> >
>> >
>> >
>> >
>> > --
>> > You received this message because you are subscribed to the Google
>> > Groups
>> > "ctrax" group.
>> > To post to this group, send email to ct...@googlegroups.com.
>> > To unsubscribe from this group, send email to

>> > ctrax+unsubscribe@googlegroups.com.

>> > For more options, visit this group at
>> > http://groups.google.com/group/ctrax?hl=en.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ctrax" group.
>> To post to this group, send email to ct...@googlegroups.com.
>> To unsubscribe from this group, send email to

>> ctrax+unsubscribe@googlegroups.com.

>> For more options, visit this group at
>> http://groups.google.com/group/ctrax?hl=en.
>>
>
>
>
> --
> John Riemann Soong Lüwen
> Biochem, Physics UVA '13
>
> "Every difference of opinion is not a difference of principle. We have
> called by different names brethren of the same principle. We are all
> Republicans, we are all Federalists." -- Thomas Jefferson
>
> We want structures that serve people, not people serving structures.
> (l'Odéon, mai '68)
>
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "ctrax" group.
> To post to this group, send email to ct...@googlegroups.com.
> To unsubscribe from this group, send email to

> ctrax+unsubscribe@googlegroups.com.

[Ulrich Stern]

PS: fixing underlying issues is not high priority IMO; I am mainly curious.

[John Bender]

That's the reason for not fixing them. :)

The underlying issues are reasonably complicated because of the
plethora of cameras and codecs, the multiple platforms we support, and
the interaction with Matlab on those platforms. I'm also not sure
there's a standard for the encoding order in the first place. It seems
like a big investment of time (fixing and debugging) for a small
payoff, when the alternative (the checkbox) works already.

JB

>>> >> >> >>>> ctrax+un...@googlegroups.com.

>>> >> >> >>>> For more options, visit this group at
>>> >> >> >>>> http://groups.google.com/group/ctrax?hl=en.
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>> --
>>> >> >> >>> John Riemann Soong Lüwen
>>> >> >> >>> Biochem, Physics UVA '13
>>> >> >> >>>
>>> >> >> >>> "Every difference of opinion is not a difference of principle.
>>> >> >> >>> We
>>> >> >> >>> have
>>> >> >> >>> called by different names brethren of the same principle. We
>>> >> >> >>> are
>>> >> >> >>> all
>>> >> >> >>> Republicans, we are all Federalists." -- Thomas Jefferson
>>> >> >> >>>
>>> >> >> >>> We want structures that serve people, not people serving
>>> >> >> >>> structures.
>>> >> >> >>> (l'Odéon, mai '68)
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >> --
>>> >> >> >> You received this message because you are subscribed to the
>>> >> >> >> Google
>>> >> >> >> Groups
>>> >> >> >> "ctrax" group.
>>> >> >> >> To view this discussion on the web visit
>>> >> >> >> https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.
>>> >> >> >>
>>> >> >> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> >> >> To unsubscribe from this group, send email to

>>> >> >> >> ctrax+un...@googlegroups.com.

>>> >> >> >> For more options, visit this group at
>>> >> >> >> http://groups.google.com/group/ctrax?hl=en.
>>> >> >> >
>>> >> >> >
>>> >> >> > --
>>> >> >> > You received this message because you are subscribed to the
>>> >> >> > Google
>>> >> >> > Groups
>>> >> >> > "ctrax" group.
>>> >> >> > To post to this group, send email to ct...@googlegroups.com.
>>> >> >> > To unsubscribe from this group, send email to

>>> >> >> > ctrax+un...@googlegroups.com.

>>> >> >> > For more options, visit this group at
>>> >> >> > http://groups.google.com/group/ctrax?hl=en.
>>> >> >>
>>> >> >> --
>>> >> >> You received this message because you are subscribed to the Google
>>> >> >> Groups
>>> >> >> "ctrax" group.
>>> >> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> >> To unsubscribe from this group, send email to

>>> >> >> ctrax+un...@googlegroups.com.

>>> >> >> For more options, visit this group at
>>> >> >> http://groups.google.com/group/ctrax?hl=en.
>>> >> >>
>>> >> >
>>> >> >
>>> >> >
>>> >> > --
>>> >> > John Riemann Soong Lüwen
>>> >> > Biochem, Physics UVA '13
>>> >> >
>>> >> > "Every difference of opinion is not a difference of principle. We
>>> >> > have
>>> >> > called by different names brethren of the same principle. We are all
>>> >> > Republicans, we are all Federalists." -- Thomas Jefferson
>>> >> >
>>> >> > We want structures that serve people, not people serving structures.
>>> >> > (l'Odéon, mai '68)
>>> >> >
>>> >> >
>>> >> >
>>> >> >
>>> >> > --
>>> >> > You received this message because you are subscribed to the Google
>>> >> > Groups
>>> >> > "ctrax" group.
>>> >> > To post to this group, send email to ct...@googlegroups.com.
>>> >> > To unsubscribe from this group, send email to

>>> >> > ctrax+un...@googlegroups.com.

>>> >> > For more options, visit this group at
>>> >> > http://groups.google.com/group/ctrax?hl=en.
>>> >>
>>> >> --
>>> >> You received this message because you are subscribed to the Google
>>> >> Groups
>>> >> "ctrax" group.
>>> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> To unsubscribe from this group, send email to

>>> >> ctrax+un...@googlegroups.com.

>>> >> For more options, visit this group at
>>> >> http://groups.google.com/group/ctrax?hl=en.
>>> >>
>>> >
>>> >
>>> >
>>> > --
>>> > John Riemann Soong Lüwen
>>> > Biochem, Physics UVA '13
>>> >
>>> > "Every difference of opinion is not a difference of principle. We have
>>> > called by different names brethren of the same principle. We are all
>>> > Republicans, we are all Federalists." -- Thomas Jefferson
>>> >
>>> > We want structures that serve people, not people serving structures.
>>> > (l'Odéon, mai '68)
>>> >
>>> >
>>> >
>>> >
>>> > --
>>> > You received this message because you are subscribed to the Google
>>> > Groups
>>> > "ctrax" group.
>>> > To post to this group, send email to ct...@googlegroups.com.
>>> > To unsubscribe from this group, send email to

>>> > ctrax+un...@googlegroups.com.

>>> > For more options, visit this group at
>>> > http://groups.google.com/group/ctrax?hl=en.
>
> --
> You received this message because you are subscribed to the Google Groups
> "ctrax" group.

> To view this discussion on the web visit

> https://groups.google.com/d/msg/ctrax/-/Q7ARNKBKPJsJ.

>
> To post to this group, send email to ct...@googlegroups.com.
> To unsubscribe from this group, send email to

> ctrax+un...@googlegroups.com.

[Ulrich Stern]

Makes sense.  It was easy to add the equivalent of the FixErrors checkbox to my analysis script.  And John Soong's suggestion of a "flip MAT-file" script becomes an attractive alternative in case checkboxes will be required in other places...

Thanks,

Uli

>>> >> >> >>>> ctrax+unsubscribe@googlegroups.com.

>>> >> >> >>>> For more options, visit this group at
>>> >> >> >>>> http://groups.google.com/group/ctrax?hl=en.
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>> --
>>> >> >> >>> John Riemann Soong Lüwen
>>> >> >> >>> Biochem, Physics UVA '13
>>> >> >> >>>
>>> >> >> >>> "Every difference of opinion is not a difference of principle.
>>> >> >> >>> We
>>> >> >> >>> have
>>> >> >> >>> called by different names brethren of the same principle. We
>>> >> >> >>> are
>>> >> >> >>> all
>>> >> >> >>> Republicans, we are all Federalists." -- Thomas Jefferson
>>> >> >> >>>
>>> >> >> >>> We want structures that serve people, not people serving
>>> >> >> >>> structures.
>>> >> >> >>> (l'Odéon, mai '68)
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >>>
>>> >> >> >> --
>>> >> >> >> You received this message because you are subscribed to the
>>> >> >> >> Google
>>> >> >> >> Groups
>>> >> >> >> "ctrax" group.
>>> >> >> >> To view this discussion on the web visit
>>> >> >> >> https://groups.google.com/d/msg/ctrax/-/HHzqGyoVyqsJ.
>>> >> >> >>
>>> >> >> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> >> >> To unsubscribe from this group, send email to

>>> >> >> >> ctrax+unsubscribe@googlegroups.com.

>>> >> >> >> For more options, visit this group at
>>> >> >> >> http://groups.google.com/group/ctrax?hl=en.
>>> >> >> >
>>> >> >> >
>>> >> >> > --
>>> >> >> > You received this message because you are subscribed to the
>>> >> >> > Google
>>> >> >> > Groups
>>> >> >> > "ctrax" group.
>>> >> >> > To post to this group, send email to ct...@googlegroups.com.
>>> >> >> > To unsubscribe from this group, send email to

>>> >> >> > ctrax+unsubscribe@googlegroups.com.

>>> >> >> > For more options, visit this group at
>>> >> >> > http://groups.google.com/group/ctrax?hl=en.
>>> >> >>
>>> >> >> --
>>> >> >> You received this message because you are subscribed to the Google
>>> >> >> Groups
>>> >> >> "ctrax" group.
>>> >> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> >> To unsubscribe from this group, send email to

>>> >> >> ctrax+unsubscribe@googlegroups.com.

>>> >> >> For more options, visit this group at
>>> >> >> http://groups.google.com/group/ctrax?hl=en.
>>> >> >>
>>> >> >
>>> >> >
>>> >> >
>>> >> > --
>>> >> > John Riemann Soong Lüwen
>>> >> > Biochem, Physics UVA '13
>>> >> >
>>> >> > "Every difference of opinion is not a difference of principle. We
>>> >> > have
>>> >> > called by different names brethren of the same principle. We are all
>>> >> > Republicans, we are all Federalists." -- Thomas Jefferson
>>> >> >
>>> >> > We want structures that serve people, not people serving structures.
>>> >> > (l'Odéon, mai '68)
>>> >> >
>>> >> >
>>> >> >
>>> >> >
>>> >> > --
>>> >> > You received this message because you are subscribed to the Google
>>> >> > Groups
>>> >> > "ctrax" group.
>>> >> > To post to this group, send email to ct...@googlegroups.com.
>>> >> > To unsubscribe from this group, send email to

>>> >> > ctrax+unsubscribe@googlegroups.com.

>>> >> > For more options, visit this group at
>>> >> > http://groups.google.com/group/ctrax?hl=en.
>>> >>
>>> >> --
>>> >> You received this message because you are subscribed to the Google
>>> >> Groups
>>> >> "ctrax" group.
>>> >> To post to this group, send email to ct...@googlegroups.com.
>>> >> To unsubscribe from this group, send email to

>>> >> ctrax+unsubscribe@googlegroups.com.

>>> >> For more options, visit this group at
>>> >> http://groups.google.com/group/ctrax?hl=en.
>>> >>
>>> >
>>> >
>>> >
>>> > --
>>> > John Riemann Soong Lüwen
>>> > Biochem, Physics UVA '13
>>> >
>>> > "Every difference of opinion is not a difference of principle. We have
>>> > called by different names brethren of the same principle. We are all
>>> > Republicans, we are all Federalists." -- Thomas Jefferson
>>> >
>>> > We want structures that serve people, not people serving structures.
>>> > (l'Odéon, mai '68)
>>> >
>>> >
>>> >
>>> >
>>> > --
>>> > You received this message because you are subscribed to the Google
>>> > Groups
>>> > "ctrax" group.
>>> > To post to this group, send email to ct...@googlegroups.com.
>>> > To unsubscribe from this group, send email to

>>> > ctrax+unsubscribe@googlegroups.com.

>>> > For more options, visit this group at
>>> > http://groups.google.com/group/ctrax?hl=en.
>
> --
> You received this message because you are subscribed to the Google Groups
> "ctrax" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/ctrax/-/Q7ARNKBKPJsJ.
>
> To post to this group, send email to ct...@googlegroups.com.
> To unsubscribe from this group, send email to

> ctrax+unsubscribe@googlegroups.com.