Live data from Hacker News

The 500-mile email (2002)

web.mit.edu

121–130 of 145 posts

Re: The 500-mile email (2002)

#121
post #99

Earlier quoted context omitted.

And this is why we can't have nice stories. I felt for the author as I got deeper into the faq, and recognized this pattern of cynicism, then decided the author was so generous and thorough, not out of obligation (make the emails stop!), but because that is the type of detailed person he is -- and good at dinner parties too!

Writing stories for a technical audience is tricky. I've been doing it for going on 10 years now, and I'm still not very good at it. A critical rule, however, is to omit detail, (a reader is unlikely to question an explanation they make up themselves) and most importantly, to omit details you know to be wrong . (It is impossible to nitpick a statement that is never said) An odd feature of our campus network at the ti…

Reminds me of the old saying, "it's better to stay silent and be thought a fool, than to open your mouth and confirm it."

It's also a reason why short business emails are better than longer ones. You can always go more in depth. It takes skilled restraint to touch on only the most relevant details without losing the larger point.

Re: The 500-mile email (2002)

#122
post #99
post #94

Earlier quoted context omitted.

The FAQ answers this question. Basically; it was a long time ago, and the point of the story isn't in the detail. :) http://www.ibiblio.org/harris/500milemail-faq.html

And this is why we can't have nice stories. I felt for the author as I got deeper into the faq, and recognized this pattern of cynicism, then decided the author was so generous and thorough, not out of obligation (make the emails stop!), but because that is the type of detailed person he is -- and good at dinner parties too!

Yes, I found this to be one of the most refreshing technical anecdotes I've ever read. The tone and style actually put a smile on my face as I read. I enjoyed how the author guided us through the process of discovery, one which we all know so well, driven by an insatiable curiosity to go continually deeper down the rabbit hole until we find the bottom.

Re: The 500-mile email (2002)

#123
post #72

Earlier quoted context omitted.

What do you use it for?

One use is converting columns of numbers into math strings for bc. Example (contrived): $ seq 10 20 | paste -s -d + 10+11+12+13+14+15+16+17+18+19+20 $ seq 10 20 | paste -s -d + | bc 165 Or converting columns of strings into regex 'or' clauses for searching (contrived example again): $ cut -f 3 -d , something.csv | paste -s -d "|" a|b|c|d|e|f $ egrep "$(cut -f 3 -d , something.csv | paste -s -d "|")" another_file ...…

I've used paste all my life, but I never knew you could do

  paste - -
to convert stdin into 2 columns (or "paste - - - -" to get 4 columns!). TIL ...

Re: The 500-mile email (2002)

#124
post #3

This one comes up ever 3-4 years or so in sysadmin communities, and I read it every single time. because it's worth it. It's one of those things that I highly doubt would have occurred to me to have even checked, or given even a moments thought to, under normal circumstances.

I was looking for another famous sysadmin story, where the guy who also happens to be a top Linux developer (so maybe Alan Cox?) rescues a deeply broken Linux system where even glibc is no longer accessible by manipulating inodes in a running process. Or something. Over the years, my Google-fu has failed me. Any clue? :)

That's a classic.

Best I can claim is zmodem transfers of uunecoded packages over a PLIP link as I tried to get ethernet support up on an old but fairly reliable box.

Re: The 500-mile email (2002)

#125

Earlier quoted context omitted.

ls /usr/bin

Unless `units` doesn't happen to be installed by default, which is the case at least for Arch Linux. Though it doesn't contain `units` either, here's a Wikipedia list of the standardized (IEEE 1003.1-2008) unix commands. http://en.wikipedia.org/wiki/List_of_Unix_commands

units(1) was in V7, and systems that aren't a superset of V7 are objectively wrong.

Re: The 500-mile email (2002)

#126
post #3

This one comes up ever 3-4 years or so in sysadmin communities, and I read it every single time. because it's worth it. It's one of those things that I highly doubt would have occurred to me to have even checked, or given even a moments thought to, under normal circumstances.

This and the story of Mel never get old.

http://www.catb.org/jargon/html/story-of-mel.html

Re: The 500-mile email (2002)

#127
post #60

Another email incident at Microsoft worth reading [1]. [1] http://blogs.technet.com/b/exchange/archive/2004/04/08/10962...

I also found that to be evidence of pretty horrific architecture in Exchange. Two actual recipient lists with a secret internal one? Bloating headers to 13K? At the very least, it seems to me like they chose to put the distribution logic at the wrong layer...

> Two actual recipient lists with a secret internal one?

How else do you propose handling BCC and mailing lists?

Re: The 500-mile email (2002)

#128
post #72

Earlier quoted context omitted.

One use is converting columns of numbers into math strings for bc. Example (contrived): $ seq 10 20 | paste -s -d + 10+11+12+13+14+15+16+17+18+19+20 $ seq 10 20 | paste -s -d + | bc 165 Or converting columns of strings into regex 'or' clauses for searching (contrived example again): $ cut -f 3 -d , something.csv | paste -s -d "|" a|b|c|d|e|f $ egrep "$(cut -f 3 -d , something.csv | paste -s -d "|")" another_file ...…

seq has the -s flag which voids the need of the paste for that command: $ seq -s + 10 20 10+11+12+13+14+15+16+17+18+19+20 But I agree that the paste is very useful. # a few random samples for an IN SQL statement $ shuf -i 1-500000 -n 5 | paste -s -d , 371492,250061,266669,455846,295852 # we can even get PI $ ( seq -s + -f '4/%g' 1 4 100000 && seq -s - -f '4/%g' 3 4 100000 ) | paste -s -d - | bc -l 3.14157265358979523…

There is a faster way to get pi in bc:

  echo '4*a(1)' | bc -l
(That is, 4×arctan(1)=4×π/4=π.) But your way is truly an awesome use of Unix!

Re: The 500-mile email (2002)

#129
post #99

Earlier quoted context omitted.

And this is why we can't have nice stories. I felt for the author as I got deeper into the faq, and recognized this pattern of cynicism, then decided the author was so generous and thorough, not out of obligation (make the emails stop!), but because that is the type of detailed person he is -- and good at dinner parties too!

Writing stories for a technical audience is tricky. I've been doing it for going on 10 years now, and I'm still not very good at it. A critical rule, however, is to omit detail, (a reader is unlikely to question an explanation they make up themselves) and most importantly, to omit details you know to be wrong . (It is impossible to nitpick a statement that is never said) An odd feature of our campus network at the ti…

Hi, ceequof, the original author here. I agree with you completely in concept; it was a stupid thing to include and I should have cut it.

But as I wrote in the FAQ, I fired off that email in under an hour in reply to a fast-moving thread on an email list where people knew me by reputation and wouldn't question my skills; the totality of my "research" was trying to reproduce the original numbers from memory ("500 miles" stuck in my head, but the distances to the places I remembered pinging did not); and I didn't ask anyone else to edit it for me.

All of that would have been ridiculously unprofessional of me as a writer for something intended for as wide an audience as it went to. But I had no idea it would be forwarded so much and so often (nor so many years later, now decades after the original event!).

And that's why I really prefer it when people link to the canonical version I maintain at http://www.ibiblio.org/harris/500milemail.html with a link to the FAQ.

Re: The 500-mile email (2002)

#130
post #99
post #94

Earlier quoted context omitted.

The FAQ answers this question. Basically; it was a long time ago, and the point of the story isn't in the detail. :) http://www.ibiblio.org/harris/500milemail-faq.html

And this is why we can't have nice stories. I felt for the author as I got deeper into the faq, and recognized this pattern of cynicism, then decided the author was so generous and thorough, not out of obligation (make the emails stop!), but because that is the type of detailed person he is -- and good at dinner parties too!

Thanks.
Post reply on HN