[9fans] The freq command

[9fans] The freq command

[underspecified]

[-- Attachment #1: Type: text/plain, Size: 948 bytes --]

Greetings,

I have been using Plan 9 in various forms for a short while, and I am very
impressed with its minimalist design -- particularly its tendency not to
implement any function that can be done by combining other functions.

At first I missed rgrep et al, but I found that alternatives like 'grep $foo
`{du $bar}' worked just as well. So I was surprised to find the program
freq: it prints histograms of the character distributions in files. I am
interested in doing things like identifying the language of a file, and
thought it might be useful in tasks like that, but I wanted to ask what
people in the p9 community use it for.

Is freq used as the basis for other statistical tasks? Is it used by any
components of the OS? My apologies if this has already been discussed. I am
not interested in learning how it works (I have the man page and source),
rather I want to know the why behind its creation.

--underspecified

[-- Attachment #2: Type: text/html, Size: 1054 bytes --]

Re: [9fans] The freq command

[Rob Pike]

freq is a ken original, an old program that sat in his personal bin
for a long time.
i don't remember what caused it to become public.

the -r option was a late addition.

it's an odd tool in the old unix tradition.  its output is rarely the
final answer
but somehow encodes what you're looking for: this file is ascii; this file is
latin-1; there are no tabs in this file.

i don't think most people even know about it. it's not a central
application; just
a reminder of the halcyon days of fun little tools.

-rob

Re: [9fans] The freq command

[underspecified]

[-- Attachment #1: Type: text/plain, Size: 824 bytes --]

Greetings,

Thank you for the informative reply.

Freq is a nice little program. More people should know about it.

Hoping for a return of the halcyon days of fun little tools,

--underspecified

On 10/8/07, Rob Pike <robpike@gmail.com> wrote:
>
> freq is a ken original, an old program that sat in his personal bin
> for a long time.
> i don't remember what caused it to become public.
>
> the -r option was a late addition.
>
> it's an odd tool in the old unix tradition.  its output is rarely the
> final answer
> but somehow encodes what you're looking for: this file is ascii; this file
> is
> latin-1; there are no tabs in this file.
>
> i don't think most people even know about it. it's not a central
> application; just
> a reminder of the halcyon days of fun little tools.
>
> -rob
>

[-- Attachment #2: Type: text/html, Size: 1145 bytes --]