Jens M. Lauritsen
>
> Please note that with the "labjl" command
>
> STB-59 dm84 . . . . . . . . . labjl: Adding numerical
> codes to value labels
> (help labjl if installed) . . . . . . . . . . . .
> . . J. M. Lauritsen
> 1/01 STB Reprints Vol 10, pages 35--37
> adds or removes numerical codes to the value labels of a
> categorical variable
>
> The same feature as announced in Stata8 can be used in
> Stata 6 and 7:
>
> . tab kmgrp
>
> Residence |
> of runner | Freq. Percent Cum.
> ------------+-----------------------------------
> 0- 25 km | 675 18.65 18.65
> 26-62 km | 911 25.17 43.81
> 63 -120 | 954 26.35 70.17
> 120+ km | 1080 29.83 100.00
> ------------+-----------------------------------
> Total | 3620 100.00
>
> . labjl kmgrp
> Numerical codes added to Label: kmlbl
>
> * note that labels will be properly aligned:
> . tab kmgrp
> Residence of |
> runner | Freq. Percent Cum.
> -------------+-----------------------------------
> 1 0- 25 km | 675 18.65 18.65
> 2 26-62 km | 911 25.17 43.81
> 3 63 -120 | 954 26.35 70.17
> 4 120+ km | 1080 29.83 100.00
> -------------+-----------------------------------
> Total | 3620 100.00
>
> . labjl kmgrp, remove
> Numerical codes removed from Label: kmlbl
>
> Within a few days labjl will be removed from ssc-ideas and
> my own site to
> diminish confusion.
>
I understand this to be in response to the new -numlabel-
command in Stata 8.
There is a general issue here. Official Stata upgrades
can make previous user-written programs largely or even entirely
redundant.
However, it can also be true that the user-written
program has some functionality not provided with
the new official command.
In addition, some people may have do files
or ado files which call upon such programs. If
other people want to use those do files or
ado files, then it is necessary to get
copies of those programs.
Moreover, many users may not be upgrading to Stata 8
immediately. I imagine that all, or almost all, Statalist
members will want to upgrade, but not all will have
the control over their system, or the finances, to
do so immediately. Also, there are probably several
users of versions earlier than 7 still out there. My wild
guess is ~10% of Statalist members, for example.
As Jens' -labjl- has a completely different name,
I can't see that the introduction of -numlabel-
poses any serious risk of confusion.
As another author of user-written software, I intend
to let my programs written for Stata 7 and earlier
versions remain in archives, principally SSC, even when
they are superseded by what is in Stata 8. This does no harm;
it requires minimal resources; and it is potentially still
of some use to people using earlier versions. That
is, or should be, one of the functions of a software
archive.
I urge Jens to reconsider his decision.
Nick
[email protected]
*
* 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/
