patchGNU Octave - Patches: patch #8472, Data file with measurement data of...

 
 

patch #8472: Data file with measurement data of a penny

Submitted by:  Thomas Weber <tweber>
Submitted on:  Sun Jun 1 14:33:43 2014  
 
Category: NonePriority: 5 - Normal
Status: DonePrivacy: Public
Assigned to: NoneOpen/Closed: Closed

Add a New Comment (Rich MarkupRich Markup):
   

You are not logged in

Please log in, so followups can be emailed to you.

 

(Jump to the original submission Jump to the original submission)

Tue Aug 19 21:33:43 2014, comment #14:

The penny.mat file was added to the build system with this changeset: http://hg.savannah.gnu.org/hgweb/octave/rev/c8240a60dd01

Rik <rik5>
Project Administrator
Tue Aug 19 16:59:42 2014, comment #13:

That seems fine to me. If you'll take care of that part, I'll fix the run-octave script and push my changes to savannah.

Thanks.

John W. Eaton <jwe>
Project Administrator
Tue Aug 19 16:44:15 2014, comment #12:

One thought I had was to subdivide the examples directory into

examples/code
examples/data

where the data directory would be installed into $octdatadir.

Rik <rik5>
Project Administrator
Tue Aug 19 16:32:00 2014, comment #11:

Where should this file go in the sources? etc/data? Some new data directory? I'd like a directory that contains only auxiliary data files like penny.mat so I can append that to the search directory in the run-octave script. All the files in this directory should be installed in the new $octdatadir directory. Is that OK?

John W. Eaton <jwe>
Project Administrator
Mon Aug 18 17:08:44 2014, comment #10:

I attach an mbox with the mail exchange. It is just one message, but the discussion was done using full quotes, so the mail exchange is complete.

(file #31912)

Thomas Weber <tweber>
Tue Aug 5 15:06:42 2014, comment #9:

That's good news. We just want to have something on hand showing the conversation, but the particular format isn't important. Upload whatever is most convenient for you.

Rik <rik5>
Project Administrator
Tue Aug 5 14:02:54 2014, comment #8:

I got confirmation from NIST:
"Yes, the information you requested is in the public domain as stated in the "Use of NIST Information" section of our website. You can cite the data as: Intelligent Systems Division, NIST."

Should I upload the mbox file here or just a PDF showing the mail exchange?

Thomas Weber <tweber>
Fri Jun 20 15:05:56 2014, comment #7:

The NIST standard disclaimer seems to cover software. Is this really software or is it data? They seem to have a different possible license for Standard Reference Data.

On the other hand, it doesn't seem like this applies very well either because the Standard Reference Data Act is talking about fundamental physical and chemical data.

I don't want this to be painful for them, but I wonder if they can release it in to the public domain? Or whether they can release it under the same terms as they did before (to the Mathworks), whatever those were.

There seems to be some precedence for the public domain. If I look at their Disclaimer page it says

This is covering the web pages and anything presented there, but it sounds like anything that they don't have to maintain copyright on, such as Standard Reference Data, is generally released to the public domain.

Rik <rik5>
Project Administrator
Fri Jun 20 09:20:07 2014, comment #6:

I will ask the contact person at NIST about publishing the email. Would a confirmation that the data is covered by http://www.nist.gov/public_affairs/disclaimer.cfm be sufficient?

Thomas Weber <tweber>
Tue Jun 17 19:17:39 2014, comment #5:

I had the submitted file run on Matlab and it does produce the correct image, so this is a bug with Octave's implementation of contour.

It's probably this one https://savannah.gnu.org/bugs/?func=detailitem&item_id=39370

Rik <rik5>
Project Administrator
Tue Jun 17 17:00:27 2014, comment #4:

I converted the xls with

and I can confirm that these data are equal to these in penny.mat from Thomas and equal to the data returned from Matlab 2013a.

When plotting with contour (fltk toolkit) I also see some lines (as described by rik) which changes when I move the cursor over the figure. In my assessment the penny data is correct but contour shows a bug.

Andreas Weber <andy1978>
Project Member
Tue Jun 17 15:45:21 2014, comment #3:

I talked to jwe about adding penny.mat. Part of his response:

"If there is no URL we can point to, then I think we should include the email that gives permission or ask for some statement from this person at NIST"

If none of the three methods above are possible then maybe we have to archive off the e-mail exchange and use that.

Rik <rik5>
Project Administrator
Mon Jun 16 11:35:13 2014, comment #2:

I don't have access to Matlab. I attach the original .xls file I
received from NIST.

(file #31564)

Thomas Weber <tweber>
Fri Jun 13 18:20:01 2014, comment #1:

I tried the penny.mat file that you uploaded with the sample code from
http://www.mathworks.com/products/matlab/examples.html?file=/products/demos/shipping/matlab/penny.html. There is a difference for the first example which uses contour to draw the penny. In Octave there are 3 line segments that seem to go across the face of the penny at different angles.

I see two possibilities: 1) The conversion from the Excel file to Octave format left in some errant points, or 2) Octave's contour routine has some problems (we already know it does have some). Do you have access to Matlab? We could check whether your file loads and displays correctly in Matlab. If it does then the problem lies with Octave.

Rik <rik5>
Project Administrator
Sun Jun 1 14:33:43 2014, original submission:

Matlab includes a file called "penny.mat" with measurement data of a penny, see [1]. I have contacted NIST who conducted the original measurement. They sent me an Excel file with the measurement data which I converted into an Octave data file (attached). According to an email exchange I had with NIST, they do not own copyright on such measurement data and we are therefor free to redistribute the file.[2]

It would be nice if Octave could ship the data file, as matlab2tikz uses it during its test suite and does not ship the file on its own (as Matlab contains it in a default installation).

I only attach the file and not a patch, as I have no idea where the file should go in the source code tree.

[1] http://www.mathworks.nl/products/matlab/examples.html?file=/products/demos/shipping/matlab/penny.html
[2] I can forward the email exchange, if needed - but I am hesitating to do so on a public tracker, as I have not asked for permission to publish the email exchange.

Thomas Weber <tweber>

 

(Note: upload size limit is set to 16384 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

Attached Files
file #31912:  mail-exchange.mbox.lzma added by tweber (9kB - application/x-lzma)
file #31564:  penny.xls added by tweber (190kB - application/vnd.ms-excel)
file #31481:  penny.mat added by tweber (54kB - application/octet-stream)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by jwe (Posted a comment)
  • -unavailable- added by andy1978 (Posted a comment)
  • -unavailable- added by rik5 (Posted a comment)
  • -unavailable- added by tweber (Submitted the item)
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 5 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Tue Aug 19 21:33:43 2014rik5StatusNone=>Done
      Open/ClosedOpen=>Closed
    Mon Aug 18 17:08:44 2014tweberAttached File-=>Added mail-exchange.mbox.lzma, #31912
    Mon Jun 16 11:35:13 2014tweberAttached File-=>Added penny.xls, #31564
    Sun Jun 1 14:33:43 2014tweberAttached File-=>Added penny.mat, #31481

    Back to the top


    Powered by Savane 3.1-cleanup