[Insight-developers] ITK 2.8 Repository Tagged & Branched

Gaetan Lehmann gaetan.lehmann at jouy.inra.fr
Wed Jun 7 12:14:28 EDT 2006 

On Tue, 06 Jun 2006 19:38:16 +0200, Hans Johnson <hans-johnson at uiowa.edu>  
wrote:

> Julien,
>
> Categories for each paper would also be helpful.  For example, I may  
> want to query for all papers on segmentation, but not on registration.   
> People could then sign up as reviewers for all IJ documents that are on  
> segmentation and are desired to be included in Insight/Code/BasicFilters.
>
> 2 sets of categories:
>
> 1) Relevant topics
> 2) Desired audience (Insight developers, image processing end users)   
> i.e. is this submission a stand-alone tool, or a new feature for ITK?
>
> Just my 2c.
>
> Hans
>
>
> On Jun 6, 2006, at 11:01 AM, Julien Jomier wrote:
>
>> Hi Gaetan,
>>
>> Reviewing a paper can be interesting but sometimes it can be really  
>> painful (talking from my own experience here). We just have to accept  
>> that this is a necessary step.
>>
>> Couple of things we can do in order to improve the review process:
>>
>> 1) Send automatic email to the users/developers lists when a new  
>> publication is submitted to the IJ (Luis had this idea before and it's  
>> almost implemented).
>>
>> 2) Have an editor for the IJ who will select the appropriate reviewers  
>> and set a deadline when the reviews are due. (I tend to wait until the  
>> last moment to review a paper and if I don't get a reminder then I tend  
>> to forget about it...)
>>
>> 3) Send automatic email when a paper had no/few reviews for a long time  
>> to encourage users and developers to sign up for reviews.
>>
>> 4) Publish a newsletter (or something like that) with the new  
>> contributions and the names of the reviewers. This will somehow reward  
>> reviewers (as well as authors).
>>
>> If you have any other ideas I'm open to suggestions.

All the ideas above are nice.
It would also be nice to be able to discuss a publication by mail. The  
content of the email is then added to the discussion on the publication  
web page.
roundup (http://roundup.sourceforge.net/) in doing that, and it's a very  
nice feature


>>
>> For the last 6 months we've received 21 publications for ITK, which  
>> makes about one publication a week (not that bad...).
>> We didn't envision using the IJ as a tollgate for ITK contributions in  
>> the first place and I really think if we improve the IJ to fit this  
>> design it will get better.
>>

I was thinking about the result for the contribution process. How many  
publications have been integrated in the toolkit ? How many publications  
have been really discussed (discussion is a well known factor to improve  
code quality) ?
For those points, we can't say that the results of the insight journal are  
really good.
Now I was only thinking about contribution process, and the IJ is surely  
useful for other things :-)

>> Julien
>>
>> Gaetan Lehmann wrote:
>>> Hi Julien,
>>> To be clear, I thought IJ was a great idea, to improve contribution  
>>> quality, and to valorize (I'm not sure that the right word) my work -  
>>> show to everybody the work done, and have something about ITK to put  
>>> in my evaluation form.
>>> That's why I contributed articles and reviews.
>>> That's right that there is some things to change in the interface to  
>>> make IJ much better, but I don't think it will significantly change  
>>> the numbers of reviews. Reviewing a contribution take some time, and  
>>> it seems that the number of people ready to do that is quite small.  
>>> Even official ITK developers are rarely taking the time to review the  
>>> articles. How do you think to convince the normal user to do that ?
>>> With the results of the first months of the IJ, are you sure that  
>>> working more on it is a good use of your time ?
>>> Gaetan
>>> On Tue, 06 Jun 2006 16:29:57 +0200, Julien Jomier  
>>> <julien.jomier at kitware.com> wrote:
>>>> Gaetan,
>>>>
>>>> I agree with you the IJ is not reaching is full potential.
>>>> However, the main issue at this point is not the number of  
>>>> publications but the amount of reviews for new code and this is  
>>>> slowing down the process. Moreover, the submission of new revisions  
>>>> is somehow painful.
>>>>
>>>> Here are some features that we want to implement (hopefully soon) for  
>>>> the Insight Journal.
>>>>
>>>> 1) Separate CVS access for submission into the IJ so bug fixes can be  
>>>> quickly fixed without a need for a resubmission.
>>>>
>>>> 2) Improved review design for submission to ITK (and other toolkits).  
>>>> The idea is to assign two/three main reviewers and set a deadline for  
>>>> the reviews (with email reminders). This should speedup the  
>>>> integration into the toolkit. Also, at submission time, you will be  
>>>> able to specify if this is a new feature or bug fix, etc...
>>>>
>>>> I'm also collecting ideas/features on how to improve the IJ. Feel  
>>>> free to send me an email or log a feature request at www.itk.org/Bug  
>>>> (there is a project for Insight Journal).
>>>>
>>>> Thanks,
>>>>
>>>> Julien
>>>>
>>>> Gaetan Lehmann wrote:
>>>>>  Luis,
>>>>>  Do you still think the Insight Journal is the proper way to get new  
>>>>> code ?
>>>>> While some interesting contributions have been published to the  
>>>>> journal, the abscence of discussion about those contributions  
>>>>> completely kills the contribution process. The new code seems to not  
>>>>> be much integrated than before to the toolkit - even bug fixes and  
>>>>> feature/performance improvement are not. After several months of  
>>>>> existance, it seems that the journal is not able to reach the  
>>>>> minimum amount of publishers and reviewers to work smoothly.
>>>>> Perhaps the ITK community is too small for this kind of process ?
>>>>> Don't you think that going back to committing directly new classes  
>>>>> in the repository should increase the development dynamic of the  
>>>>> toolkit ?
>>>>>  Regards,
>>>>>  Gaetan
>>>>>   On Tue, 06 Jun 2006 14:21:03 +0200, Luis Ibanez  
>>>>> <luis.ibanez at kitware.com> wrote:
>>>>>
>>>>>>
>>>>>> The CVS repository has been tagged and branched for ITK 2.8.
>>>>>>
>>>>>> The repository is open for commits again.
>>>>>>
>>>>>> Please keep in mind that only bug fixes, new tests, and
>>>>>> performance improvements should be committed directly.
>>>>>>
>>>>>> Any new classes, or API changes should be posted first as
>>>>>> technical reports to the Insight Journal.
>>>>>>
>>>>>>
>>>>>> Please let us know if you find any problems,
>>>>>>
>>>>>>
>>>>>>      Thanks
>>>>>>
>>>>>>
>>>>>>
>>>>>>         Luis
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> Insight-developers mailing list
>>>>>> Insight-developers at itk.org
>>>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>>>    --Gaëtan Lehmann
>>>>> Biologie du Développement et de la Reproduction
>>>>> INRA de Jouy-en-Josas (France)
>>>>> tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
>>>>> http://voxel.jouy.inra.fr
>>>>> _______________________________________________
>>>>> Insight-developers mailing list
>>>>> Insight-developers at itk.org
>>>>> http://www.itk.org/mailman/listinfo/insight-developers
>>>>>
>>> --Gaëtan Lehmann
>>> Biologie du Développement et de la Reproduction
>>> INRA de Jouy-en-Josas (France)
>>> tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
>>> http://voxel.jouy.inra.fr
>>> _______________________________________________
>>> Insight-developers mailing list
>>> Insight-developers at itk.org
>>> http://www.itk.org/mailman/listinfo/insight-developers
>>
>> _______________________________________________
>> Insight-developers mailing list
>> Insight-developers at itk.org
>> http://www.itk.org/mailman/listinfo/insight-developers
>>



-- 
Gaëtan Lehmann
Biologie du Développement et de la Reproduction
INRA de Jouy-en-Josas (France)
tel: +33 1 34 65 29 66    fax: 01 34 65 29 09
http://voxel.jouy.inra.fr

More information about the Insight-developers mailing list