From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Computing (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.

Outdated Tape info[edit]

It says LTO-5 is the largest tape format, but this has been outdated for almost two years now. Perhaps add a new example with the 185TB tapes from Sony?


Would it be a good idea to have a table of the bandwidth of certain Sneakernets ? If this is a crazy idea tell me so .. but just how fast *is* a "...station wagon full of tapes hurtling down the highway."? --2mcm 22:54, 4 Apr 2005 (UTC)

Assuming that the average packing case of DDS tapes can carry 20 tapes, and a yank tank station wagon can carry about 10 packing cases, you could have 20x10x36GB of data transferred. 7,200GB. Now, assuming your house and your ISP are about 5 miles away, taking about 15 minutes of journey time, you could move 7,200GB in 15 minutes, or 8 megabits a second. However, a ping return would take 30 minutes.
And thats not counting the time to write the tapes, load the car, unload the car, read the tapes back in, etc...
Kiand 19:32, 1 May 2005 (UTC)
would that be 8 Giga bytes a second ? kcalc says 7200GB/15/60 = 8GB/s --2mcm 09:27, 2 May 2005 (UTC)


Should we make a see also or take up Courier?

Perhaps a link in the prose might be better than a "see also". --Damian Yerrick (serious | business) 14:52, 8 November 2007 (UTC)

Are we serious?[edit]

I was expecting something more along the lines of "physically carrying data between computers, sometimes also humorously referred to as sneakernet", and then I get this dead serious article instead.

It is a good article though, nice work. But my question: Do people actually seriously use the term sneakernet to refer to carrying data around?

I would have thought that at least references to hacker culture and to the term's origin as a joke would be the first thing you see.

Anyway, I'll put in a reference to Hacker culture, fix the reference to Jargon File, and such. Hmm, should I link to the Jargon File entry or just include it? It's short. I'll just link to it for now.

-- magetoo 10:09, 5 December 2005 (UTC)

Three suggestions for the page, though I'm not really a wikipediazen, so I don't (yet) feel comfortable editing the page directly.

First, yes, people do "seriously" use the term sneakernet; it's succint and clearly expresses the idea of hand-carrying data via physical media. Of course we all know the joke and the origin of the term.

Second, moving data by sneakernet (I can't quite bring myself to use "sneakernetting") is quite common in large corporation environments, which tend to have their networks extremely locked down at the average user level, and tend to be quite uptight and likely to react out of proportion to attempts to get around such control. Fortunately, USB thumb drives are quite common, these days. Sooner or later I'm going to cut one open and install it in a little plastic sneaker...

Third, the "station wagon full of mag tapes" quote is much older than 1996. Tannenbaum's book has an original printing in 1985, which is also the year I first read the quote - as output from the Unix fortune program, in the form "Never underestimate the bandwidth of a station wagon full of mag tapes". Not sure which came first, the book or the fortune, or whether it's truly Tannenbaum's quote or somebody else (though Tannenbaum certainly is the right sort of person in the right sort of time and place to be the originator).

Steven J. Owens

To echo Mr. Owens, yes people do use the term sneakernet including official government documents and procedures which requrie the use of removable media data transfer. No longer a 'hacker' term, it has been approved for use by US Gov.


-- I still use the term sneakernet, though I've always written it as 'sneaker net'. This has got to be one of the greatest wiki entries I've seen in a long while. The seriousness of the entry (as it should be) is so deadpan, I find it totally hilarious.

PunkFloyd 17:55, 11 October 2007 (UTC) --

Even if people DO use the term sneakernet in professionally environments, the many references to people "sneakernetting" data to thier friends are out of place. A common user shareing a file over a usb key isn't consiously deciding to sneakernet data like the article implies. It would be better written if the article cited places where protocol called for a sneaker net (such as these gov. procedures) and times when an avereage joe would chose to "sneakernet" such as the high/low bandwidth work/home duality. But saying that people have been using tem for ages (even when most people didn't have access to internet connections....) seems like your putting actions into peoples sneakers. They weren't/aren't sneakernetting, they're jsut giving thier buddy a picture.

Also perhaps we should mention that the Wikipedia CDs are an example of a sneakernet —Preceding unsigned comment added by (talk) 17:37, 13 November 2007 (UTC)

TCP (Transmission by carrier pigeon)[edit]

Sneakernet in Minority Report[edit]

A bit esoteric, but this mock memo inspired by Minority Report and response offer an insightful hypothetical justification of a sneakernet:

Memo from Lily Wong of Wong and Associates Industrial Design Consulting Group

In Response to Your Memo by Rich Thomas

Tatwell 23:42, 18 August 2006 (UTC)


I would like some precise information where the article needs copy-editing. I would like to refrain from copying it right now because I want to take part in its assessment process. --Click me! 15:32, 8 August 2007 (UTC)

Here's a sourced quote, in case anyone cares to use it:

sneakernet n. Term used (generally with ironic intent) for transfer of electronic information by physically carrying tape, disks, or some other media from one machine to another. "Never underestimate the bandwidth of a station wagon filled with magtape, or a 747 filled with CD-ROMs." Also called 'Tennis-Net', 'Armpit-Net', 'Floppy-Net', or 'Shoenet'.

[1] E. S. Raymond, Ed., The New Hacker's Dictionary, 3rd ed. , Cambridge, MA: The MIT Press, 1996.

I'm using IEEE citation style, since that's what I'm used to. Evanturner (talk) 20:59, 9 February 2008 (UTC)

references needed in Copyright concerns section[edit]

A {Refimprove|section|date=August 2008} was added to the =Copyright concerns= section because there are no references at this time, and this section is somewhat likely to be deemed as having a non-neutral point of view without references. - Bevo (talk) 22:32, 24 August 2008 (UTC)

While I did not find a reference for the "Don't copy that floppy!" campaign, I did cite instances where some kind of DRM is used in recordable media. (talk) 18:32, 14 September 2009 (UTC)

Okay, Now I see that the "Don't copy that Floppy!" campaign is listed in the "External Links" section ( (talk) 18:43, 14 September 2009 (UTC)

Security enhancement?[edit]

In the Theory section is seems to mention that using a sneakernet for transferring encrypted files would improve security. Isn't the neccesity of transmitting a key obsolete due to public key encryption? ce1984 (talk) 08:18, 31 August 2008 (UTC)

You still need sneakernet to transmit some public keys so that you can authenticate their owners. These can be the public keys of certificate authorities, or the public keys of key signing party participants. --Damian Yerrick (talk | stalk) 11:10, 31 August 2008 (UTC)

Feeble Humour[edit]

I came across this page from the disk enclosure page, and just wanted to congratulate you nerds on your feeble humour. What a pathetic page, you people should be ashamed of yourselves. —Preceding unsigned comment added by (talk) 04:55, 4 January 2009 (UTC)

Good article[edit]

This is a well written article dispite having the unfortunate and uncontrolable urge to collapse into hystrical giggles due to its inexorable seriousness. Overall it receives my rating of an 8 out of 10. -- Erin —Preceding unsigned comment added by (talk) 02:40, 6 September 2008 (UTC)

Well, gosh, we were all breathlessly waiting to hear how you'd rate it. —Preceding unsigned comment added by (talk) 00:35, 18 February 2009 (UTC)

The good and the bad[edit]

All that being said, one might make an argument to ignore the rules, but I think the better course of action would be to try and source most of the claims made. Don't let me sound all negative though... it's very well done!  :-) //Blaxthos ( t / c ) 12:38, 17 April 2009 (UTC)

Amazon follows a similar model ?[edit]

From this article, it looks like Amazon is trying a similar approach to load massive amounts of data from customers. "Amazon cloud uses FedEx instead of the Internet to ship data" . (talk) 13:18, 11 June 2010 (UTC)Ranga [1]

Statement which doesn't make sense[edit]

There is also the limitation of read/write speeds on a computer. There are three ways disk speed can be increased. The speed of the drive and/or the media may be increased, multiple disks may be used (one disk may be read as another is written to), or simultaneous use of multiple disks. This is generally not an issue for a sneakernet, as it often makes use of removable media.[3][not in citation given]

The last sentence which I see was only added recently [2] makes little sense which probably explains why it isn't mentioned in the citation. At best, the citation supports the claim 'sneakernet' makes use of removable media although even that is questionable since it's really only discussing certain types of sneakernets (having said that, I don't think this really matters since there's no much disagreement sneakernets generally use removable media). However the bolder claim isn't supported and as I said makes little sense. Removable media still have read/write speed limits. Multiple media may help speed things up, it may not, it depends on where the limitations come from. How easy it is to use multiple media depends on the type of media (memory cards or tape drivers for example would require multiple readers at the destination). Whether the read/write speeds will be significant will also vary. For example if you're transferring a full 128GB USB key or memory card and it takes 30 mins to move it, the read/write times could easily be significant. Similarly if you're transferring a full 2TB HD and it takes a 4 hours to move it between locations the read/write times could easily add 2 hours. On the other hand if you're transferring a 2TB HD and it takes 80 hours to move from one location to the other, the read/write times aren't likely to that significantly. Nil Einne (talk) 11:19, 10 April 2011 (UTC)

'sneakernet calculation'[edit]

"Boeing 747-400 carrying capacity in cargo configuration, in cubic meters:159 Volume of 3.5" Drive ( 102mm x 25mm ) in cubic millimeters: 0.00255 Number of 3.5 Drives per 747: 62,353 Data capacity of 3.5 Drive ( TB ): 3 Data capacity of 3.5 Drive ( GB): 3,000 Data capacity of 747 in Terabytes:187,059 In Gigabytes: 187,058,824 bits per byte: 8 In Gigabits: 1496470588 Distance between LAX and JFK in miles: 2470 Speed of a Boeing 747-400, in knots: 507 Time to fly between LAX and JFK at 507 knots, in seconds: 16080 Bandwidth of the 747 in Gbits/Second: 93064 So our 747-enhanced Sneakernet Bandwidth: 93 Terabits per second"

This calculation leaves out the dimension of write speed X amount of data.

I'm not aware of the write speed of a BD drive.

And also needs sneakernet citation :-P

Should this updated factoid be added to article (perhaps replacing the BD factoid)? Also, this person got a different number in similar calculation: Bratling (talk) 14:10, 9 June 2011 (UTC)

Only gives 2 dimensions for volume of hard drive. Should be 102mm x 25mm x 145mm or 0.00036975 cubic meters. — Preceding unsigned comment added by Wut44 (talkcontribs) 14:47, 9 June 2011 (UTC)

A MicroSD addendum[edit]

I'd just like to point out that MicroSD cards are ~166 mm³, which, with 32 GB MicroSD cards filling 1.66 x 10^-7 m³ and requiring no protection, leaves us with, I think, a data density of 192.8 PB/m³.

If our 747 has 159 m³ of usable space in it, that's 30.66 EB. Iain Dawson (talk) 23:48, 6 January 2012 (UTC)

Residential internet access[edit]

I use sneakernet (thanks to Macomb Community College's public computer lab with non-diskless workstations) as a cheap substitute for residential Internet access.

Fedex Bandwidth[edit]

since i am too lazy to write anything meaningful on the subject, i am just going to drop a link to randal's what if? article, where he does some math on the subject of SneakerNet -- also too lazy to sign in, 01:12, 16 July 2013 (UTC) — Preceding unsigned comment added by (talk)

Comment on the non-fiction section[edit]

The following text was added to the section Sneakernet#Non-fiction by in 2013, but it's written from a personal perspective and as a commentary on the previous text, so it rather belongs on the talk page. Some of it could maybe be moved back after a rewrite. //Essin (talk) 17:09, 6 October 2015 (UTC)

The 1970s "old joke" referred to above was told to this poster in the NASA / Jet Propulsion Lab cafeteria in about 1975-1976. He worked in the Digital Maintenance group in the JPL Space Flight Operations Center from 1974 to 1978. The story / joke was a classic regularly used at JPL to explain ping time, and differentiate bandwidth from latency (and, by the way, the need to document where your cables ran, and that you needed to distribute your data circuits across multiple cables in different trenches - or somehow via multiple paths).

The NASA Deep Space Network tracking station at Goldstone is just outside of Fort Irwin, just east of Barstow, California. When you leave the highway you have to go through Fort Irwin to get to any of the Goldstone facilities. Depending on the highway route taken, and which Deep Space Network dish at Goldstone you are driving to (or starting from) it was about 160-185 miles (255-298 km) from JPL. At freeway speeds (65 mph, about 100 km/h) it was a minimum of three-and-a-half hours, usually four, and frequently more, depending on the traffic. If you ignored the speed limit while out in the desert (risky) you could get closer to three and-a-half hours. This distance and speed also explained how the "ping time" was 7 to 8 hours. Several of the freeways now in existence were not there then.

At the time (early 1970s), the data links from JPL to Goldstone ranged from as low as 1200 and 2400 bps (several of each) to 9600 bps (one or two). The 9-track magnetic tapes of the day recorded at a maximum density of 6250 bits per inch (but some older drives were limited to 800 or 1600 bits per inch). The tape reels were made in different sizes, the largest held about 2400 feet of tape, but due to the data being written in records, with gaps between the records, the maximum data capacity of a 2400 foot reel, blocked at 32,767 bytes per record and recorded at 6250 BPI was 170 megabytes per reel.[1]

As the story that your contributor heard went, one day a plumbing contractor's backhoe dug up and broke the underground cable that carried ALL of the JPL-to-Goldstone data and voice lines through Fort Irwin, and it would take at least a day, maybe longer, to repair. So someone was designated to drive two boxes of 12 reels each of magnetic tape down to JPL, and quickly. The first available vehicle was a white NASA station wagon. Hence the punch line: "Never underestimate the bandwidth of a station wagon full of magnetic tapes hurtling down the highway".

Rounding off the numbers, twenty-four reels of tape at 170 megabytes each is 4080 megabytes. Three and a half hours is 210 minutes. 4080 megabytes divided by 210 works out to about 19.4 megabytes per minute, or roughly 323 kilobytes per second (2584 kilobits per second) - over 1000 times faster than a 2400 bps data circuit of the time. Note that the incident above involved only 24 reels - which didn't come anywhere near filling the station wagon, in fact the two boxes of tapes didn't even fill the front passenger seat. (As an aside, a station wagon is known as an estate car or estate in other parts of the world).

Incidentally, that conversation was the first time your contributor ever heard the term backhoe fade used to describe accidental massive damage to an underground cable (compare it to the term rain fade used to describe a fade-out of a point-to-point microwave radio path due to the absorptive effect of water in the air).[2]

  1. ^ 9 track tape
  2. ^ Conversation in the Jet Propulsion Laboratory cafeteria around mid-1976 related by Michael R. Morris.

External links modified[edit]

Hello fellow Wikipedians,

I have just added archive links to 3 external links on Sneakernet. Please take a moment to review my edit. If necessary, add {{cbignore}} after the link to keep me from modifying it. Alternatively, you can add {{nobots|deny=InternetArchiveBot}} to keep me off the page altogether. I made the following changes:

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

YesY An editor has reviewed this edit and fixed any errors that were found.

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

If you are unable to use these tools, you may set |needhelp=<your help request> on this template to request help from an experienced user. Please include details about your problem, to help other editors.

Cheers.—cyberbot IITalk to my owner:Online 12:31, 9 February 2016 (UTC)

Origin of the word[edit]

I strongly doubt the reference to "sneakers" in the opening paragraph. The word obviously comes from the verb "to sneak" freely joined with the "internet/ethernet/cybernet" class of words. Unfortunately I don't have any reference to proof my point. — Preceding unsigned comment added by (talk) 11:09, 9 September 2016 (UTC)

It's sourced now. --Zac67 (talk) 10:08, 10 September 2016 (UTC)