For fuller understanding, recall that SSC and -ssc- have different histories.
The SSC archive goes back about a decade. There are various ways by which
people can install stuff from SSC on their machines, but the "proper" way
to do it is through -net install-.
As I understand it, -adoupdate- depends on people having used -net install-
to install packages, although they might have done this unknowingly.
The standard advice since late 2001 is to use -ssc- to install from SSC,
if you can. This advice is not contradictory: -ssc- does several things,
but that most relevant here is that it calls up -net install- to install
packages.
Similarly, before 2001 there was an -archinst- program from Kit Baum
and myself which was another wrapper for -net install-. That was absorbed
in and superseded by -ssc- when -ssc- was added to Stata as an official
command.
Nick
[email protected]
Nick Cox
> Argh! Joseph has uncovered a small mess.
>
> I take the blame for not thinking carefully
> enough in 1999 about the requirements and
> consequences of -adoupdate-, added to Stata 12
> December 2005.
>
> The full explanation needs what those wonderful
> Guides Verts from Michelin call un peu d'histoire
> [a little history].
>
> As far as I can recall, through Kit Baum's good agencies
> I put a package of matrix commands on SSC as the
> -matodd- package. Then I wrote up some, but not all,
> of those commands in the STB as
>
> STB-50 dm69 . . . . . . . . . . . . . . . . . . Further
> new matrix commands
> (help matdelrc, matewm, matmad, matpow if installed)
> . . . N. J. Cox
> 7/99 pp.5--9; STB Reprints Vol 9, pp.29--34
> collection of new matrix commands providing additional matrix
> checking, management, element-wise operators, maximum absolute
> difference, and power
>
> -mstdizem- was one of the commands not written up, not
> because of any perceived
> weaknesses, but because it needed a different kind of
> write-up. (Still not
> done.)
>
> (There was a sequel,
>
> STB-56 dm79 . . . . . . . . . . . . . . . . . . Yet more
> new matrix commands
> (help matcorr, matewmf, matvsort, svmat2 if
> installed) . . N. J. Cox
> 7/00 pp.4--8; STB Reprints Vol 10, pp.17--23
> commands to produce a correlation matrix, elementwise monadic
> function of another matrix, selected subsets of matrix rows
> and columns, vec or vech of a matrix, elements sorted within
> a vector, matrix from a vector, and commands to save matrices
>
> and the article naming was a conscious reflection of Jeroen
> Weesie's pioneer work:
>
> STB-39 dm49 . . . . . . . . . . . . . . . . . . . . Some
> new matrix commands
> (help matfunc, varfunc if installed) . . . . . . . .
> . . . J. Weesie
> 9/97 pp.17--20; STB Reprints Vol 7, pp.43--48
> collection of new matrix commands; several for
> explicit matrices
> and a few for implicit matrices (i.e., variables)
>
> but I don't think they feature in Joseph's problems.)
>
> So, as some programs in the package on SSC weren't included
> in the STB
> publication, and as such were not downloadable from the Stata
> website, the
> -matodd- package remained on SSC.
>
> In the short run, Joseph can just hand-copy the particular files
> needed, using -ssc copy-. As I understand it, that circumvents
> -adoupdate-.
>
> In the long run, I should probably restructure -matodd- so that
> it includes only files not available as dm69 from STB-50.
>
> If Bill Gould is at base, he may want to give better advice.
>
> Nick
> [email protected]
>
> Joseph Coveney
>
> > Nick Cox wrote (in the "Deming adjustments" thread):
> >
> > There is also a matrix-based version
> > called -mstdizem- which is also on
> > SSC, as part of the -matodd- package.
> >
> > --------------------------------------------------------------
> > ------------------
> >
> > I noticed that I don't have -mstdizem-, so I did a -findit
> > matodd-. It
> > brought up packages on StataCorp's site (users' directory)
> > and, of course,
> > the one on SSC. I clicked on the hyperlink for SSC's
> > -matodd- to install it
> > and I got the "following files already exist and are
> > different" balk. I
> > chose one of the files (-matpow-) in the list in the Viewer
> > in order to see
> > what's up:
> >
> > . which matpow
> > c:\Program Files\Stata9\ado\STBplus\m\matpow.ado
> > *! 1.1.0 NJC 15 June 1999 STB-50 dm69
> >
> > . type http://fmwww.bc.edu/RePEc/bocode/m/matpow.ado
> > program def matpow
> > * power of square matrix
> > *! 1.0.0 NJC 20 July 1998
> > [snipped]
> >
> > When I click on option 2--"Look for an already-installed
> > package of the same
> > name (which you might then choose to uninstall)"--I get:
> >
> > --------------------------------------------------------------
> > ----------------------------------------------------
> > directory of installed user-written packages
> > --------------------------------------------------------------
> > ----------------------------------------------------
> > all packages containing "matodd.pkg":
> > --------------------------------------------------------------
> > ----------------------------------------------------
> >
> > (none)
> > --------------------------------------------------------------
> > ----------------------------------------------------
> > (click here to return to the previous screen)
> >
> >
> > Typing -adoupdate matodd- gives:
> >
> > (no packages match "matodd")
> >
> > The date given for the package on StataCorp's website is
> > January 1999, so I
> > couldn't have got -matpow- from there; regardless, I'm not so much
> > interested in where I got -matpow- from as much as: how do I
> > install -mstdizem- and other -matodd- ado-files without overwriting
> > more-recent versions of -matpow- and others?
> >
> > Does -adoupdate- take care of this automatically, or do I first
> > install -matodd- and then go back and re-adoupdate- each of
> > the ado-files in
> > the list of already-installed packages displayed in the Viewer?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
