R loop

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

R loop

ivo welch-2
I am not an R expert, but I think the R idea of this task is to work
with a data frame (where you make a and b are columns), and then to
iterate over the rows.

data <- read.table( textConnection(
"a b
1 name1
2 name2
3 name3"), header=TRUE) ;  # just learned this one

for (i in 1:nrow(data)) print( data[i,] );
or just  data[1:3,]


apropos, R does not support native hashes (like perl), unfortunately,
but it has something similar:

data$a=="b"   # gives F T F
data[ data$a == "name2" ]  # does not work and gives a weird result,  yuck
subset(data, data$a=="name2")   # this does work.


data frame sorting by column (using order) could also be simpler to
understand for novices---IMHO, there should be a wrapper function that
is in standard R like
  data.sorted = sort( data, by=data$b );

regards,

/iaw

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

Sean Davis



On 2/6/06 3:06 PM, "ivo welch" <[hidden email]> wrote:

> I am not an R expert, but I think the R idea of this task is to work
> with a data frame (where you make a and b are columns), and then to
> iterate over the rows.
>
> data <- read.table( textConnection(
> "a b
> 1 name1
> 2 name2
> 3 name3"), header=TRUE) ;  # just learned this one
>
> for (i in 1:nrow(data)) print( data[i,] );
> or just  data[1:3,]
>
>
> apropos, R does not support native hashes (like perl), unfortunately,
> but it has something similar:
>
> data$a=="b"   # gives F T F
> data[ data$a == "name2" ]  # does not work and gives a weird result,  yuck

 data[data$a=="name2",]

Notice the difference from what you did.

> data frame sorting by column (using order) could also be simpler to
> understand for novices---IMHO, there should be a wrapper function that
> is in standard R like
>   data.sorted = sort( data, by=data$b );

Look at ?order.  

Sean

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

ivo welch-2
> > data[ data$a == "name2" ]  # does not work and gives a weird result,  yuck
>
>  data[data$a=="name2",]
>

sorry about this.  I believe a few versions back, one could not subset
data frames, so I did not even check what I wrote.  Works now.

> Look at ?order.

I know.  This is why I suggested only that we need a wrapper that
plugs an order() functionality into the sort() function (which
fortunately points in its docs to order)---but novices would find the
sort syntax easier.

regards,

/iaw

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

PIKAL Petr
Hi

On 6 Feb 2006 at 15:57, ivo welch wrote:

Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
From:           ivo welch <[hidden email]>
To:             Sean Davis <[hidden email]>
Copies to:       [hidden email], r-help <[hidden email]>
Subject:         Re: [R] R loop

> > > data[ data$a == "name2" ]  # does not work and gives a weird
> > > result,  yuck
> >
> >  data[data$a=="name2",]
> >
>
> sorry about this.  I believe a few versions back, one could not subset
> data frames, so I did not even check what I wrote.  Works now.

It depends on what you consider few versions back. I started with R
vesion 1.2.0 about 10 years ago and I believe that data frame
subsetting was done in **very** similar manner as it is performed
now.

Cheers
Petr

>
> > Look at ?order.
>
> I know.  This is why I suggested only that we need a wrapper that
> plugs an order() functionality into the sort() function (which
> fortunately points in its docs to order)---but novices would find the
> sort syntax easier.
>
> regards,
>
> /iaw
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html

Petr Pikal
[hidden email]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

Uwe Ligges
Petr Pikal wrote:

> Hi
>
> On 6 Feb 2006 at 15:57, ivo welch wrote:
>
> Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
> From:           ivo welch <[hidden email]>
> To:             Sean Davis <[hidden email]>
> Copies to:       [hidden email], r-help <[hidden email]>
> Subject:         Re: [R] R loop
>
>
>>>>data[ data$a == "name2" ]  # does not work and gives a weird
>>>>result,  yuck
>>>
>>> data[data$a=="name2",]
>>>
>>
>>sorry about this.  I believe a few versions back, one could not subset
>>data frames, so I did not even check what I wrote.  Works now.
>
>
> It depends on what you consider few versions back. I started with R
> vesion 1.2.0 about 10 years

I bet 200$ (or EUR) you have not used R 10 years ago. ;-)

People certainly remember the 1.0.0 release at the remarkable day
29-FEB-2000.
1.2.0 was released in December 2000, about 5 years ago.
I started with 0.62.x in 1998.
The oldest version I found on CRAN is a pre-alpha R.sea.hqx for the Mac
               dated 07-Nov-1996.


Uwe Ligges





> ago and I believe that data frame
> subsetting was done in **very** similar manner as it is performed
> now.
>
> Cheers
> Petr
>
>
>>>Look at ?order.
>>
>>I know.  This is why I suggested only that we need a wrapper that
>>plugs an order() functionality into the sort() function (which
>>fortunately points in its docs to order)---but novices would find the
>>sort syntax easier.
>>
>>regards,
>>
>>/iaw
>>
>>______________________________________________
>>[hidden email] mailing list
>>https://stat.ethz.ch/mailman/listinfo/r-help
>>PLEASE do read the posting guide!
>>http://www.R-project.org/posting-guide.html
>
>
> Petr Pikal
> [hidden email]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

PIKAL Petr
Hi

you are correct as usually. It is about 10 versions 1.2.0 - 2.2.1 and
as there are 2 versions per year it is about 4-5 years.

But nevertheless it seems to me as if I used R forever (and hopefully
it will continue)

Anyway subsetting data frames **was** similar in the old versions as
it is now, wasn't it?

Cheers
Petr

On 7 Feb 2006 at 8:56, Uwe Ligges wrote:

Date sent:       Tue, 07 Feb 2006 08:56:10 +0100
From:           Uwe Ligges <[hidden email]>
Organization:   Fachbereich Statistik, Universitaet Dortmund
To:             Petr Pikal <[hidden email]>
Copies to:       [hidden email], ivo welch <[hidden email]>
Subject:         Re: [R] R loop

> Petr Pikal wrote:
>
> > Hi
> >
> > On 6 Feb 2006 at 15:57, ivo welch wrote:
> >
> > Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
> > From:           ivo welch <[hidden email]>
> > To:             Sean Davis <[hidden email]>
> > Copies to:       [hidden email], r-help
> > <[hidden email]> Subject:         Re: [R] R loop
> >
> >
> >>>>data[ data$a == "name2" ]  # does not work and gives a weird
> >>>>result,  yuck
> >>>
> >>> data[data$a=="name2",]
> >>>
> >>
> >>sorry about this.  I believe a few versions back, one could not
> >>subset data frames, so I did not even check what I wrote.  Works
> >>now.
> >
> >
> > It depends on what you consider few versions back. I started with R
> > vesion 1.2.0 about 10 years
>
> I bet 200$ (or EUR) you have not used R 10 years ago. ;-)
>
> People certainly remember the 1.0.0 release at the remarkable day
> 29-FEB-2000. 1.2.0 was released in December 2000, about 5 years ago. I
> started with 0.62.x in 1998. The oldest version I found on CRAN is a
> pre-alpha R.sea.hqx for the Mac
>                dated 07-Nov-1996.
>
>
> Uwe Ligges
>
>
>
>
>
> > ago and I believe that data frame
> > subsetting was done in **very** similar manner as it is performed
> > now.
> >
> > Cheers
> > Petr
> >
> >
> >>>Look at ?order.
> >>
> >>I know.  This is why I suggested only that we need a wrapper that
> >>plugs an order() functionality into the sort() function (which
> >>fortunately points in its docs to order)---but novices would find
> >>the sort syntax easier.
> >>
> >>regards,
> >>
> >>/iaw
> >>
> >>______________________________________________
> >>[hidden email] mailing list
> >>https://stat.ethz.ch/mailman/listinfo/r-help
> >>PLEASE do read the posting guide!
> >>http://www.R-project.org/posting-guide.html
> >
> >
> > Petr Pikal
> > [hidden email]
> >
> > ______________________________________________
> > [hidden email] mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide!
> > http://www.R-project.org/posting-guide.html
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html

Petr Pikal
[hidden email]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

Uwe Ligges
Petr Pikal wrote:

> Hi
>
> you are correct as usually. It is about 10 versions 1.2.0 - 2.2.1 and
> as there are 2 versions per year it is about 4-5 years.
>
> But nevertheless it seems to me as if I used R forever (and hopefully
> it will continue)
>
> Anyway subsetting data frames **was** similar in the old versions as
> it is now, wasn't it?

Yes, AFAIR, and just tested in 1.2.2.

Uwe Ligges

> Cheers
> Petr
>
> On 7 Feb 2006 at 8:56, Uwe Ligges wrote:
>
> Date sent:       Tue, 07 Feb 2006 08:56:10 +0100
> From:           Uwe Ligges <[hidden email]>
> Organization:   Fachbereich Statistik, Universitaet Dortmund
> To:             Petr Pikal <[hidden email]>
> Copies to:       [hidden email], ivo welch <[hidden email]>
> Subject:         Re: [R] R loop
>
>
>>Petr Pikal wrote:
>>
>>
>>>Hi
>>>
>>>On 6 Feb 2006 at 15:57, ivo welch wrote:
>>>
>>>Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
>>>From:           ivo welch <[hidden email]>
>>>To:             Sean Davis <[hidden email]>
>>>Copies to:       [hidden email], r-help
>>><[hidden email]> Subject:         Re: [R] R loop
>>>
>>>
>>>
>>>>>>data[ data$a == "name2" ]  # does not work and gives a weird
>>>>>>result,  yuck
>>>>>
>>>>>data[data$a=="name2",]
>>>>>
>>>>
>>>>sorry about this.  I believe a few versions back, one could not
>>>>subset data frames, so I did not even check what I wrote.  Works
>>>>now.
>>>
>>>
>>>It depends on what you consider few versions back. I started with R
>>>vesion 1.2.0 about 10 years
>>
>>I bet 200$ (or EUR) you have not used R 10 years ago. ;-)
>>
>>People certainly remember the 1.0.0 release at the remarkable day
>>29-FEB-2000. 1.2.0 was released in December 2000, about 5 years ago. I
>>started with 0.62.x in 1998. The oldest version I found on CRAN is a
>>pre-alpha R.sea.hqx for the Mac
>>               dated 07-Nov-1996.
>>
>>
>>Uwe Ligges
>>
>>
>>
>>
>>
>>
>>>ago and I believe that data frame
>>>subsetting was done in **very** similar manner as it is performed
>>>now.
>>>
>>>Cheers
>>>Petr
>>>
>>>
>>>
>>>>>Look at ?order.
>>>>
>>>>I know.  This is why I suggested only that we need a wrapper that
>>>>plugs an order() functionality into the sort() function (which
>>>>fortunately points in its docs to order)---but novices would find
>>>>the sort syntax easier.
>>>>
>>>>regards,
>>>>
>>>>/iaw
>>>>
>>>>______________________________________________
>>>>[hidden email] mailing list
>>>>https://stat.ethz.ch/mailman/listinfo/r-help
>>>>PLEASE do read the posting guide!
>>>>http://www.R-project.org/posting-guide.html
>>>
>>>
>>>Petr Pikal
>>>[hidden email]
>>>
>>>______________________________________________
>>>[hidden email] mailing list
>>>https://stat.ethz.ch/mailman/listinfo/r-help
>>>PLEASE do read the posting guide!
>>>http://www.R-project.org/posting-guide.html
>>
>>______________________________________________
>>[hidden email] mailing list
>>https://stat.ethz.ch/mailman/listinfo/r-help
>>PLEASE do read the posting guide!
>>http://www.R-project.org/posting-guide.html
>
>
> Petr Pikal
> [hidden email]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: R loop

Peter Dalgaard
Uwe Ligges <[hidden email]> writes:

> Petr Pikal wrote:
>
> > Hi
> >
> > you are correct as usually. It is about 10 versions 1.2.0 - 2.2.1 and
> > as there are 2 versions per year it is about 4-5 years.

[That calculus only goes back to 2001, though, previous to that the
release rate was higher (and the numbering stranger). And it is of
course a coincidence that 2.0.0 followed 1.9.x.]

> > But nevertheless it seems to me as if I used R forever (and hopefully
> > it will continue)
> >
> > Anyway subsetting data frames **was** similar in the old versions as
> > it is now, wasn't it?
>
> Yes, AFAIR, and just tested in 1.2.2.

AFAIR, the current data frame code was adopted in late 1997 from the
code that John Chambers had donated to Statlib...

--
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark          Ph:  (+45) 35327918
~~~~~~~~~~ - ([hidden email])                  FAX: (+45) 35327907

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: prehistoric versions of R --> 1995!

Martin Maechler
In reply to this post by Uwe Ligges
>>>>> "UweL" == Uwe Ligges <[hidden email]>
>>>>>     on Tue, 07 Feb 2006 08:56:10 +0100 writes:

    UweL> Petr Pikal wrote:
    >> Hi
    >>
    >> On 6 Feb 2006 at 15:57, ivo welch wrote:
    >>
    >> Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
    >> From:           ivo welch <[hidden email]>
    >> To:             Sean Davis <[hidden email]>
    >> Copies to:       [hidden email], r-help <[hidden email]>
    >> Subject:         Re: [R] R loop
    >>
    >>
    >>>>> data[ data$a == "name2" ]  # does not work and gives a weird
    >>>>> result,  yuck
    >>>>
    >>>> data[data$a=="name2",]
    >>>>
    >>>
    >>> sorry about this.  I believe a few versions back, one could not subset
    >>> data frames, so I did not even check what I wrote.  Works now.
    >>
    >>
    >> It depends on what you consider few versions back. I started with R
    >> vesion 1.2.0 about 10 years

    UweL> I bet 200$ (or EUR) you have not used R 10 years ago. ;-)

    UweL> People certainly remember the 1.0.0 release at the remarkable day
    UweL> 29-FEB-2000.
    UweL> 1.2.0 was released in December 2000, about 5 years ago.
    UweL> I started with 0.62.x in 1998.
    UweL> The oldest version I found on CRAN is a pre-alpha R.sea.hqx for the Mac
    UweL> dated 07-Nov-1996.

Eehm; that has a wrong date (or then it would not be pre-alpha):
I've always entertained the prehistoric directory of R sources
at ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/
and its oldest file is
         Name           Size    Date        
         R-0.1alpha.tar.gz 861464  Feb 12 1996

(which will be 10 years coming Sunday -- what a jubilee!!)
So, even that is not pre-alpha; and yes, that was a bit before CRAN existed.

Now to the pre-alpha history.  I've digged some more and found

1) that the mac file mentioned above would have
   correct date 'Nov 6 1995' (at least that's the date I saved
   when I looked at the Auckland FTP server through Emacs ange-ftp).

2) The oldest stuff that I have is all from 1995;
   The source I (think I) had first used is dated June 20 1995;
   notably the  R-unix-src.tar.gz with accompanying README and
   INSTALL files
   (There was also ./win subdirectory which I did not use, with
    files all from July 15, 1995; AFAIK done by Robert Gentleman)

 I've now put a bit of these oldest files into
  ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/pre-alpha/

3) I've kept an e-mail that Ross had sent me on July 28 with
   two small patches to the (June 20) sources.
   
   Yes, chances are pretty high that I was the first one outside
   of the Auckland(NZ)-community to actively use R.
 
Note it might be interesting to find even older sources, but that
would most probably have to be by Robert and Ross (or a
sysadmin at Auckland).


Martin Maechler, ETH Zurich

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: prehistoric versions of R --> 1995!

François Pinard
[Martin Maechler]

>2) The oldest stuff that I have is all from 1995;

Mailing lists seem to go back into 1995 too.  I found a few messages
from around 1994 on topics to be later found within R, but I'm not sure
where I got these old messages from.  I did find a message really
related to R-pre-alpha, which itself quotes a message written in 1994.

--
François Pinard   http://pinard.progiciels-bpi.ca

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: prehistoric versions of R --> 1995!

Paul Gilbert
In reply to this post by Martin Maechler


Martin Maechler wrote:

>>>>>>"UweL" == Uwe Ligges <[hidden email]>
>>>>>>    on Tue, 07 Feb 2006 08:56:10 +0100 writes:
>>>>>>            
>>>>>>
>
>    UweL> Petr Pikal wrote:
>    >> Hi
>    >>
>    >> On 6 Feb 2006 at 15:57, ivo welch wrote:
>    >>
>    >> Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
>    >> From:           ivo welch <[hidden email]>
>    >> To:             Sean Davis <[hidden email]>
>    >> Copies to:       [hidden email], r-help <[hidden email]>
>    >> Subject:         Re: [R] R loop
>    >>
>    >>
>    >>>>> data[ data$a == "name2" ]  # does not work and gives a weird
>    >>>>> result,  yuck
>    >>>>
>    >>>> data[data$a=="name2",]
>    >>>>
>    >>>
>    >>> sorry about this.  I believe a few versions back, one could not subset
>    >>> data frames, so I did not even check what I wrote.  Works now.
>    >>
>    >>
>    >> It depends on what you consider few versions back. I started with R
>    >> vesion 1.2.0 about 10 years
>
>    UweL> I bet 200$ (or EUR) you have not used R 10 years ago. ;-)
>  
>
Is that an open bet or just for Petr?

>    UweL> People certainly remember the 1.0.0 release at the remarkable day
>    UweL> 29-FEB-2000.
>    UweL> 1.2.0 was released in December 2000, about 5 years ago.
>    UweL> I started with 0.62.x in 1998.
>    UweL> The oldest version I found on CRAN is a pre-alpha R.sea.hqx for the Mac
>    UweL> dated 07-Nov-1996.
>
>Eehm; that has a wrong date (or then it would not be pre-alpha):
>I've always entertained the prehistoric directory of R sources
>at ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/
>and its oldest file is
>         Name           Size    Date        
>         R-0.1alpha.tar.gz 861464  Feb 12 1996
>  
>
I'm pretty sure I was using R 0.16 in the fall of 1996 on Linux, which
was when I got dse largely converted from S. I had a slightly  earlier
version in the spring of 1996 on Solaris. (I think the numbering would
have been 0.11 -0.15 in the spring of 1996.)  I don't think the
numbering was the same on other platforms.

>(which will be 10 years coming Sunday -- what a jubilee!!)
>So, even that is not pre-alpha; and yes, that was a bit before CRAN existed.
>
>Now to the pre-alpha history.  I've digged some more and found
>
>1) that the mac file mentioned above would have
>   correct date 'Nov 6 1995' (at least that's the date I saved
>   when I looked at the Auckland FTP server through Emacs ange-ftp).
>
>2) The oldest stuff that I have is all from 1995;
>   The source I (think I) had first used is dated June 20 1995;
>   notably the  R-unix-src.tar.gz with accompanying README and
>   INSTALL files
>   (There was also ./win subdirectory which I did not use, with
>    files all from July 15, 1995; AFAIK done by Robert Gentleman)
>
> I've now put a bit of these oldest files into
>  ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/pre-alpha/
>
>3) I've kept an e-mail that Ross had sent me on July 28 with
>   two small patches to the (June 20) sources.
>  
>   Yes, chances are pretty high that I was the first one outside
>   of the Auckland(NZ)-community to actively use R.
>  
>
I wouldn't be really sure of that (but it's a fair bet).

>
>Note it might be interesting to find even older sources, but that
>would most probably have to be by Robert and Ross (or a
>sysadmin at Auckland).
>  
>
I used to have some floppy disks somewhere...

Paul Gilbert

>
>Martin Maechler, ETH Zurich
>
>______________________________________________
>[hidden email] mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>  
>
====================================================================================

La version française suit le texte anglais.

------------------------------------------------------------------------------------

This email message from the Bank of Canada is given in good faith, and shall not be
binding or construed as constituting any obligation on the part of the Bank.

This email may contain privileged and/or confidential inform...{{dropped}}

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: prehistoric versions of R --> 1995!

Uwe Ligges
Paul Gilbert wrote:

>
> Martin Maechler wrote:
>
>
>>>>>>>"UweL" == Uwe Ligges <[hidden email]>
>>>>>>>   on Tue, 07 Feb 2006 08:56:10 +0100 writes:
>>>>>>>          
>>>>>>>
>>
>>   UweL> Petr Pikal wrote:
>>   >> Hi
>>   >>
>>   >> On 6 Feb 2006 at 15:57, ivo welch wrote:
>>   >>
>>   >> Date sent:       Mon, 6 Feb 2006 15:57:30 -0500
>>   >> From:           ivo welch <[hidden email]>
>>   >> To:             Sean Davis <[hidden email]>
>>   >> Copies to:       [hidden email], r-help <[hidden email]>
>>   >> Subject:         Re: [R] R loop
>>   >>
>>   >>
>>   >>>>> data[ data$a == "name2" ]  # does not work and gives a weird
>>   >>>>> result,  yuck
>>   >>>>
>>   >>>> data[data$a=="name2",]
>>   >>>>
>>   >>>
>>   >>> sorry about this.  I believe a few versions back, one could not subset
>>   >>> data frames, so I did not even check what I wrote.  Works now.
>>   >>
>>   >>
>>   >> It depends on what you consider few versions back. I started with R
>>   >> vesion 1.2.0 about 10 years
>>
>>   UweL> I bet 200$ (or EUR) you have not used R 10 years ago. ;-)
>>
>>
>
> Is that an open bet or just for Petr?

Was intended just for Petr, I know Martin came in very early and would
not have made a bet for him. I try not to waste money. ;-)

Uwe



>
>>   UweL> People certainly remember the 1.0.0 release at the remarkable day
>>   UweL> 29-FEB-2000.
>>   UweL> 1.2.0 was released in December 2000, about 5 years ago.
>>   UweL> I started with 0.62.x in 1998.
>>   UweL> The oldest version I found on CRAN is a pre-alpha R.sea.hqx for the Mac
>>   UweL> dated 07-Nov-1996.
>>
>>Eehm; that has a wrong date (or then it would not be pre-alpha):
>>I've always entertained the prehistoric directory of R sources
>>at ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/
>>and its oldest file is
>>        Name           Size    Date        
>>        R-0.1alpha.tar.gz 861464  Feb 12 1996
>>
>>
>
> I'm pretty sure I was using R 0.16 in the fall of 1996 on Linux, which
> was when I got dse largely converted from S. I had a slightly  earlier
> version in the spring of 1996 on Solaris. (I think the numbering would
> have been 0.11 -0.15 in the spring of 1996.)  I don't think the
> numbering was the same on other platforms.
>
>
>>(which will be 10 years coming Sunday -- what a jubilee!!)
>>So, even that is not pre-alpha; and yes, that was a bit before CRAN existed.
>>
>>Now to the pre-alpha history.  I've digged some more and found
>>
>>1) that the mac file mentioned above would have
>>  correct date 'Nov 6 1995' (at least that's the date I saved
>>  when I looked at the Auckland FTP server through Emacs ange-ftp).
>>
>>2) The oldest stuff that I have is all from 1995;
>>  The source I (think I) had first used is dated June 20 1995;
>>  notably the  R-unix-src.tar.gz with accompanying README and
>>  INSTALL files
>>  (There was also ./win subdirectory which I did not use, with
>>   files all from July 15, 1995; AFAIK done by Robert Gentleman)
>>
>>I've now put a bit of these oldest files into
>> ftp://stat.ethz.ch/Software/R/alpha/PreHistoric/pre-alpha/
>>
>>3) I've kept an e-mail that Ross had sent me on July 28 with
>>  two small patches to the (June 20) sources.
>>  
>>  Yes, chances are pretty high that I was the first one outside
>>  of the Auckland(NZ)-community to actively use R.
>>
>>
>
> I wouldn't be really sure of that (but it's a fair bet).
>
>
>>Note it might be interesting to find even older sources, but that
>>would most probably have to be by Robert and Ross (or a
>>sysadmin at Auckland).
>>
>>
>
> I used to have some floppy disks somewhere...
>
> Paul Gilbert
>
>
>>Martin Maechler, ETH Zurich
>>
>>______________________________________________
>>[hidden email] mailing list
>>https://stat.ethz.ch/mailman/listinfo/r-help
>>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>>
>>
>
> ====================================================================================
>
> La version française suit le texte anglais.
>
> ------------------------------------------------------------------------------------
>
> This email message from the Bank of Canada is given in good faith, and shall not be
> binding or construed as constituting any obligation on the part of the Bank.
>
> This email may contain privileged and/or confidential inform...{{dropped}}
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: prehistoric versions of R --> 1995!

Martin Maechler
In reply to this post by François Pinard
>>>>> "FrPi" == François Pinard <[hidden email]>
>>>>>     on Tue, 7 Feb 2006 08:43:28 -0500 writes:

    FrPi> [Martin Maechler]
    >> 2) The oldest stuff that I have is all from 1995;

    FrPi> Mailing lists seem to go back into 1995 too.  I found a few messages
    FrPi> from around 1994 on topics to be later found within R, but I'm not sure
    FrPi> where I got these old messages from.  I did find a message really
    FrPi> related to R-pre-alpha, which itself quotes a message written in 1994.

Hmm, I am interested to find these.

I am pretty sure the first "R list" was the one Ross and Robert
initiated with the following e-mail (the date of which also
points to that 10th anniversary date coming Sunday!):

  >> From: R Home <[hidden email]>
  >> To: [hidden email]
  >> Subject: R Code Alpha Test
  >> Date: Mon, 12 Feb 1996 16:15:53 +1300
  >>
  >> At some point in the past you have provided comments on R which
  >> indicated that you would make a useful alpha-tester for the version
  >> we plan to release to statlib.
  >>
  >> We have put your name on a mailing list
  >>
  >>    [hidden email]
  >>
  >> which we will use for discussion of problems and for distributing
  >> bug-fixes and patches.  We intend that that the list will only
  >> function for a short testing period.  Assuming that the testing
  >> isn't too much of a disaster, we will set up a real discussion
  >> mailing list.
  >>
  >> If you don't wish to take part in the testing cycle and would like to
  >> be taken off the list please send some mail to "[hidden email]".
  >> If you have suggestions for other who could take part in testing let
  >> us know too.
  >>
  >> There are a number of types of input we are interested in.
  >>
  >>    1) PORTABILITY
  >>    We are interested in seeing R ported to as many Unix
  >>    platforms as possible.  We are interested in hearing about
  >>    portability and any fixes neccessary to get R running on
  >>    platforms other than those we have available  (SunOS,
  >>    Solaris, FreeBSD, Linux, HP, SGI Irix).
  >>
  >>    2) DIFFERENCES
  >>    We would like feedback on differences between R and S.  We
  >>    have made the decision to make R as compatible with S as
  >>    possible, but have not had the chance to systematically
  >>    look for differences.
  >>
  >>    3) BUGS
  >>    We want to know about the bugs which are present in R.
  >>
  >>    4) EXTENSIONS
  >>    What should we do next?
  >>
  >> The source code is now available by anonymous ftp from
  >>
  >>    stat3.stat.auckland.ac.nz
  >>
  >> (note that this is not the usual place).
  >>
  >>    Ross Ihaka + Robert Gentleman
  >>

BTW, note the many (Unixy) platforms mentioned on which R was
running even then.

Definitely older than the above are private e-mails between
Ross, Robert and various people, but these were private.

An interesting one would be the first announcement of R on the
S-news mailing list which I think *was* in 1994 (as you indicate
above). Unfortunately the archives of S-news currently only go
back to 1998.

Also, further, to what I said previously on this thread,
I have a nice paperback book "Data Analysis - An Introduction
base on R" written by Alan Lee, Dept. Statistics, Auckland
labeled as "Course Notes of University of Auckland Papers 528.218/288"
(a kind of lecture notes) which I received as gift from Ross and
Robert at the Heidelberg workshop, March 25, 1995.
The book has a Copyright © 1994.  
That seems high evidence for R to have been in use at U. of
Auckland in 1994.

Now if we heard even more "history of R"  (with dates!)
from Ross / Robert / Alan Lee ?

Regards,
Martin Maechler, ETH Zurich

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html