Re: [SED-ML-discuss] The 'numberOfPoints' attribute

[Chris M. <my...@ec...>]

I think as long as the libraries are updated and a backwards compatible API function call exists, then it should not break software.  So, I would say the libraries should hide this alias from the user.  I’m fine though to make the specification say “numberOfSteps” for the new version.

Chris

> On Jun 27, 2017, at 1:20 PM, Herbert Sauro <uw....@gm...> wrote:
> 
> Given that the sedml community is still very small, it would be better to get it right at this early stage, hence I prefer option
> 
> 2) change the name to set thing else, e.g. numberOfSteps
> 
> For those clinging to earlier versions they can check the version of the sedml and interpret numberofpoints in the old way. In that sense I don't think it affects backward compatibility.
> 
> Herbett
> 
> 
> On Tue, Jun 27, 2017 at 1:08 PM Kyle Medley <me...@co... <mailto:me...@co...>> wrote:
> It seems to me that breaking backwards compatibility defeats one of the major advantages of having a standard encoding, so I would prefer option 1. If people are not happy with that, I have a different suggestion:
> 
> 4) Allow 'numberOfPoints', and do not change its meaning, but also create a new attribute 'numberOfSteps' with the same semantics. 'numberOfPoints' would not be deprecated.
> 
> Cheers,
> 
> Kyle
> 
> J Kyle Medley, PhD Candidate
> Bioengineering/BPSD, Sauro Lab
> University of Washington, Seattle
> skype: jkylemedley
> On 6/27/17 12:46 PM, Lucian Smith wrote:
>> At HARMONY, the issue was brought up that the 'numberOfPoints' attribute is confusing, because the literal number of expected points in the output file is one more than that: the initial conditions, and the number of *computed* points.
>> 
>> It was proposed that we fix this for SED-ML Level 1 Version 3.  There are a few options:
>> 
>> 1) No change:  simply make it more clear in the spec that 'numberOfPoints' means 'number of computed points' aka 'number of steps'.
>> 
>> 2) Change the name of the attribute to 'numberOfSteps' or possibly 'numberOfComputedPoints'.
>> 
>> 3) Deprecate but allow 'numberOfPoints', and make a new 'numberOfSteps' attribute that people are encouraged to use instead.
>> 
>> Any opinions?  It seems like the advantage is preventing future confusion, and the cost is having a non-backwards-compatible spec, and the cost of implementation changes for the groups that currently support SED-ML.
>> 
>> -Lucian
>> 
>> 
>> ------------------------------------------------------------------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot <http://sdm.link/slashdot>
>> 
>> _______________________________________________
>> SED-ML-discuss mailing list
>> SED...@li... <mailto:SED...@li...>
>> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss <https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss>
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot <http://sdm.link/slashdot>_______________________________________________
> SED-ML-discuss mailing list
> SED...@li... <mailto:SED...@li...>
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss <https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org <http://slashdot.org/>! http://sdm.link/slashdot_______________________________________________ <http://sdm.link/slashdot_______________________________________________>
> SED-ML-discuss mailing list
> SED...@li... <mailto:SED...@li...>
> https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss <https://lists.sourceforge.net/lists/listinfo/sed-ml-discuss>