Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Making saveold a permanent option

From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   RE: st: Making saveold a permanent option
Date   Tue, 2 Mar 2004 23:52:38 -0000

We're taking up debating roles here. I naturally agree 
about the attractions of maximum compatibility, and the 
history of StataCorp has been very good in this respect. 
I can't comment on MS Word, which I try very hard not to use. 
However, MS seems to have a problem in that many users cannot even 
see the advantage in upgrading, while it is hard to imagine
a Stata user who would not want to upgrade if cost were no 
issue (unless they're about to abandon statistical computing
for a real life). 

A few extra points: 

1. The differences between Stata 7 and Stata 8 are more
than just technicalities that StataCorp knows how to handle. 
It's misleading to downplay this given the differences in 
content between 7 and 8 data files, e.g. in representation 
of missing values, etc. 

2. The next equally understandable requests are that someone 
wants to be able to read Stata/SE files on their Stata, 
or Stata 8 files in Stata 6, etc., etc. (And I can sympathise
with those whose domestic budgets don't stretch to getting fresh  
personal licenses for home.) But where is the line drawn? 
More generally, the marketing case seems dubious here. Stata gives 
freebies when it suits -- not least updates between releases -- 
but we're a long way into Stata 8 now. Why should
StataCorp divert attention on behalf of people who won't upgrade? 
The cost here is diversion of developer resources from worthwhile 
projects that favour loyal customers, and people who won't pay 
have no leverage on StataCorp -- especially if the freebie makes 
it even less likely that they will upgrade! 

3. Sharing data files is often best done using some ASCII 
file, as may be essential if the other people are not Stata users. 
That can make the issue of versions irrelevant. 

[email protected] 

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Richard
> Williams
> Sent: 02 March 2004 23:16
> To: [email protected]; [email protected]
> Subject: RE: st: Making saveold a permanent option
> At 10:30 PM 3/2/2004 +0000, Nick Cox wrote:
> >Essentially, 7.0 is history. The only obligation
> >it exerts over StataCorp is providing tech-support
> >to registered users.
> >
> >I've no quarrel with felt needs, but I want StataCorp
> >to look ahead, not backwards.
> >
> >Nick
> >[email protected]
>  From a marketing standpoint though, I think it is a good idea when a 
> company makes it as easy as possible for users of different 
> versions to 
> exchange files.  It used to be incredibly annoying when 
> Microsoft would 
> make changes in Word's .doc format to add features that 90% of the 
> population did not want anyway, making it a big hassle to exchange 
> files.  It was smart of MS to minimize incompatibilities and to offer 
> freebie routines for conversion.  I'm sure Stata would prefer that 
> everybody just immediately upgrade to V. 8, but an 
> alternative possibility 
> is that people will stick with 7 if it is too much hassle to upgrade 
> everyone and if it is a hassle to exchange files between versions.
> Stata obviously already has the code for switching back and 
> forth between 7 
> and 8, so it presumably wouldn't be too hard to come up with 
> a separate 
> utility for this purpose.
> In the present case, Karen can admittedly solve her problem 
> just by being a 
> little more careful!  But if you are in a situation where you 
> exchange 
> files a lot (with students, colleagues; or perhaps if you 
> place things on 
> the web) it would be nice not to have to worry about 
> compatibility issues 
> or always being sure to use the lowest common denominator when saving 
> things.  I occasionally work on a machine that has Stata 7, 
> so it is a 
> nuisance when I can't open a file and have to convert it with 
> Stat/Transfer 
> or switch to a Stata 8 machine.

*   For searches and help try:

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