Re: st: new outreg and error code r(3499)

[John Luke Gallup <jlgallup@pdx.edu>]

Thanks to NIck, Kit, and Richard Williams for directing me to posts about the version-specific nature of mata libraries.  I promise to do a better job of checking the archives next time.

I have located a copy of Stata 10 and recompiled -outreg-'s mata library, so the code should be compatible back to Stata 10 once Kit does his magic and posts the update.

John

On Apr 8, 2012, at 1:43 PM, Richard Williams wrote:

> At 08:16 AM 4/7/2012, Christopher Baum wrote:
>> On Apr 7, 2012, at 2:33 AM, Richard wrote:
>> 
>> >
>> > I think Bill Gould explained this (or part of this) three years ago:
>> >
>> > http://www.stata.com/statalist/archive/2009-02/msg00648.html
>> >
>> > In a nutshell, version control doesn't work when it comes to Mata
>> > compilation. I think one solution is to actually do the compiling
>> > using Stata 9.2 (if you still have a copy anyway). My memory is
>> > vaguer on this, but I believe you can also just delete the compiled
>> > libraries and let Stata recreate them, at least if you have provided
>> > the original Mata code. I've never had to do this because I usually
>> > have the latest version of Stata, but maybe somebody else can provide
>> > the exact details.
>> 
>> Notwithstanding the complexities that Bill Gould described, it would be -very useful- to have a "mata mlib query' command that could identify the
>> version of _Stata_ that was used to produce the contents of a mlib file. We have had this problem with SSC mlibs, including the mlib that Mark
>> Schaffer and I recently added to -ivreg2-: they have to be compiled on an earlier version of Stata if they are to be usable there, and given an mlib,
>> there is no way of identifying whether it is compatible with an earlier version except trying to run it there (and unlike Richard, I don't have all the prior
>> versions sitting around on my machine).
> 
> I'll seize this as an opportunity to make my periodic suggestion that people keep old versions of Stata around if at all possible. If you want programs to have backward compatibility, the only way to make sure is to run the program on an old version of Stata. Even more important is having the old help files around. If something is being run with version control, having the old help files around lets you know how commands worked under that version. When installing a new version of Stata, you don't have to uninstall the old version; and if moving to a new machine, it is pretty easy just to copy all the old program directories. I have everything from Stata 7 on installed on my machine. (This also lets you see how much Stata has improved over the years!)
> 
> 
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> HOME:   (574)289-5227
> EMAIL:  Richard.A.Williams.5@ND.Edu
> WWW:    http://www.nd.edu/~rwilliam
> 
> *
> *   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/