Re: st: source code of mo files in ssc or sj packages

[Nick Cox <njcoxstata@gmail.com>]

Kit Baum determines SSC policy.

I have a say in SJ policy, but what Kit said looks right to me. I
think we'd want a reason for people holding back their Mata code, but
I don't want to say that there could not be a good one.

Nick

2012/4/4 László Sándor <sandorl@gmail.com>:
> Sorry about this, probably not worth your time. By policy change I
> meant requiring SSC/SJ authors to publish the source code.
>
> On Wed, Apr 4, 2012 at 2:44 PM, Nick Cox <njcoxstata@gmail.com> wrote:
>>
>> Sorry, but I don't get what policy change you are imagining. If you
>> can't download, how is code to be delivered to you?
>>
>> 2012/4/4 László Sándor <sandorl@gmail.com>:
>> > Thank you very much, Kit.
>> >
>> > FWIW, my justification: I can only type with remote access, so I can
>> > type in the most important code I would use. Or I can script "myself
>> > typing", but I can only get text up stil.
>> >
>> > Maybe this use case is so limited that no policy will change because
>> > of this. Thanks again, I hope I'll be able to get ivqreg one day.
>> >
>> > On Wed, Apr 4, 2012 at 2:22 PM, Christopher Baum <kit.baum@bc.edu> wrote:
>> >> <>
>> >> Laszlo said
>> >>
>> >>> I need to work with quantile IV models, and I trust the work of Do Won
>> >>> Kwak on this, which seems to be imminent in Stata Journal:
>> >>>
>> >>> https://www.msu.edu/~kwakdo/ivqreg.pdf
>> >>>
>> >>>
>> >>> However, the package circulated has the readable ado file but the
>> >>> compiled mo files. Is there a repository for the source of these (at
>> >>> least after it gets onto sj)?
>> >>>
>> >>> Of course, I contacted Do Won about this, but he must be busy, he has
>> >>> not got back to me yet.
>> >>>
>> >>> I would need the original source to be able to copy it to a system and
>> >>> compile over there. I cannot upload files.
>> >>
>> >>
>> >> .mo (compiled Mata) files are not platform/OS dependent. They are only sensitive to the version of Mata that
>> >> was used to compile them.
>> >>
>> >> Although it would always be preferable for SJ or SSC authors to provide the Mata source code for their
>> >> routines, they need not do so if they wish to keep them private (just as StataCorp does for a number of commands
>> >> accessed from an .mlib, whereas others are accessible within Mata). Thus it is not an SJ submission requirement
>> >> that Mata source code be provided, although it is always welcomed. Likewise for the SSC Archive.
>> >>
>> >> If you can copy .ado files to a server, you can copy .mo or .mlib files to the server, so strictly
>> >> speaking you do not need the .mo files to use the commands that reference their contents.
>> >>
>> >> Kit
>> >> manager SSC
>> >>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/