Statalist


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

Re: Re: st: Does Blasnik's Law apply to -use-?


From   n j cox <[email protected]>
To   [email protected]
Subject   Re: Re: st: Does Blasnik's Law apply to -use-?
Date   Tue, 18 Sep 2007 18:37:17 +0100

The short answer to the last question is, I suspect,
very definitely No and No. The main reason for that
is that the list is very long and the full explanation of
that list probably doesn't exist anywhere, even at
StataCorp.

Let me try a slightly more interesting longer answer. First
we need some terminology. There is a spectrum from
fully documented commands -- documentation usually
available both online and in the manuals -- through
"undocumented" commands -- which oxymoronically
can be discovered through -help undocumented- --
to everything else, which I will call "not documented",
although in turn scraps of information may exist
on them in various places. I can't give you even
an order of magnitude estimate of the ratio of
"undocumented" and "not documented", as I don't know
what I don't know and I can't remember what I've
forgotten.

Already some readers are finding this less than crystal
clear. Recall that Stata has been the subject of
development for some 22 years, in the course of which
there have been many changes. Anyone remember Stata's
idiosyncratic menus? the -gph- language? cross-product
datasets?

Why is stuff not documented? There are answers on
various levels. These categories aren't complete
or exclusive.

1. "Old" stuff, which still exists, but is thought
to be outdated, obsolete, superseded or even dangerous
in some sense.

2. Commands called by other commands, especially
utility and workhorse commands. If you stumble across
these in looking at code they might be useful to
you. In essence, however, life is far too short for StataCorp
to document all these too. Very often, there is no mystery:
different commands just exist as ways of subdividing complicated
tasks. By and large, StataCorp are really keen on exposing
tools that users would really want to use, as that takes
the heat off them, gives good publicity to Stata when users
write new stuff, and so everyone benefits.

3. "New" stuff, on its way in, or "not yet" documented.
Most of the class stuff falls under this heading.
Stata documentation is like the books and papers
every researcher would like to write: they won't all
appear even when there are the best of intentions.

4. Private hooks and handles. Every programmer of any
experience finds themselves adding extras that they
want but which may not be of widespread interest, or
which are the programmer's equivalent of magic tricks
that they can show off when the opportunity arises.

On the whole I strongly recommend asking good specific
questions to get good specific answers. I don't think
that Googling will get you very far on most of what
you want to know.

Nick
[email protected]

Sergiy Radyakin

I find it more exciting to learn about the commands which are not only
not documented, they are not even mentioned anywhere, not even in the
internet (google currently returns 0 links). Does anyone has an idea
of how the "_xt..." commands work? I mean these:

_xtarm
_xtmka
_xtmkz
_xtzw
_xtwhw
_xta2

Does anyone has a complete list of _all Stata commands and is willing
to present it to the community?


*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index