Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Downloading all ssc modules


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Downloading all ssc modules
Date   Thu, 27 Apr 2006 23:31:42 +0100

I'd say rather it wasn't designed in. 

If you want to interpret that as designed out, 
so be it. 

Less gnomically: -ssc- is an official Stata 
command. It grew out of a package -archutil- 
by Kit Baum and myself, or versa vice. StataCorp bear full
responsibility for the current command. I 
take responsibility for the following comments. 
We means Kit and myself. He can correct me if
needed. 

I'd say that we did imagine that some people
might think that they wanted everything -- or 
at least would save themselves installation time 
by just grabbing everything -- but we would want to 
persuade them that was a bad idea. 

Note especially: the SSC archive is not a kind of 
superpackage. It contains a lot of garbage as far 
as you, the reader, are concerned. Stuff put there
is put there by the authors. If anyone maintains
it, it is the authors. There is no validation process. 
Some of the stuff is just gathering dust. This 
is all inevitable. It is quite consistent with 
the SSC containing a lot of very useful stuff. 

Most but perhaps not all of the reasons for saying
this should be clear. 

1. There's stuff that doesn't interest you, unless
you are interested in everything. 

2. ... stuff that is out-of-date. (So, why doesn't 
someone weed it? Largely because it might be useful
to people on old Statas. That's one of the functions
of an archive.) 

3. ... stuff that might clash with something else
you have already, say if it has subsequently appeared via 
the Stata Journal and you installed it. 
(So, why doesn't someone weed it? 
Partly for the reason above. Partly because no one 
has the time and energy to keep this kind of housekeeping 
going. That's not to say that there is no weeding at 
all. A few years ago, I got Kit to dump a few dozen of 
my packages that seemed completely useless given other
stuff. With just one exception, no one noticed.) 

4. ... stuff that is poor by current standards. It 
is a bit buggy, a lot buggy, trivial, superfluous, 
poorly documented, or not documented at all. We know 
this in principle. It must be true. No archive 
with hundreds of packages could be otherwise. 
But no individual knows all about which of the 
packages are good and which aren't. Also, there has 
been a learning process over the lifetime of SSC. 
I doubt, for example, that any package would now be accepted 
that didn't come with help file(s), but in the early days
some slipped in. (So, ... Partly because Kit does not want to 
be an arbiter of correctness or quality. That does not 
go with an archive function.)

So, what can you do instead? 

If you go 

log using all_on_ssc.log 
foreach l in `c(alpha)' _ { 
	ssc desc `l' 
} 
log close

You can get a complete list of what is on SSC. 

If you then look at that in the Stata viewer or your 
favo[u]rite text editor, you can make a list of what
looks interesting and you think you are missing. 

Suppose you have that list:

You can get more details of those packages by 

foreach p in goodstuff estout otherstuff { 
	ssc desc `p' 
} 

Similarly you can loop over a list of packages
to install them. 

That's not what Paul asked for. However, he 
is a Stata programmer himself, and could use the material
gathered in this way to do what he asked. 

Nick 
n.j.cox@durham.ac.uk 

Austin Nichols
 
> That functionality is not available.  By design.
 
Paul Millar 

> > Does anyone know hot to download all the ssc modules 
> (without typing each one in)?

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index