Does "data" have mass?
August 11, 2005 7:00 PM   Subscribe

Does "data" have mass? I'm defining "data" here in the computing sense, transfered electronically. Do electrons serve as a means to alter the storage in a USB drive, or are they physically added to it?

Does "data" have mass?

As always, this question is coming from someone outside the field of knowledge. My brother proposed this question from an engineering standpoint, as in energy/weight output/input. A couple weeks later, I remembered it and I'm still curious.

I'm defining "data" here in the computing sense, transfered electronically.

In our hypothetical experiment we'd weigh a "new" or reformatted USB thumb drive and weigh it on some super crazy atom-sensitive scale. Then we'd put a bunch of files on it and weigh it again. Energy is necessary, so there is of course a passage of electrons, which have a mass. Do the electrons serve as a means to alter the storage in the USB drive, or are they physically added to it?
posted by deep_sea_diving_suit to Technology (29 answers total)
Wasn't this a Dilbert cartoon? The PHB said his laptop was too heavy, so someone suggested that he delete some files so it would weigh less...

Anyway, it depends entirely on the storage medium. Generally they don't work by "storing" electrons; instead, the electronic charge is used to flip the polarity of some magnetic medium. The energy is lost as heat.

(I honestly don't know the details of how flash memory works, so if anyone can fill in here...)

In any case, the amount of mass you're talking about is incredibly tiny, compared to the amount of energy that's released as heat by all the support electronics, and the computer in general.
posted by xil at 7:16 PM on August 11, 2005

Generally what xil says is true. In Flash memory in particular, electrons actually reside in cells to alter the charge and thus are "stored" This is covered at a basic level in the wikipedia entry on flash memory

However, in this case its a fluke of the implementation. Its not a good idea to think of data as something with mass, or a thing in general. Its more of an abstraction that arises from the configuration or relationships of materials rather than the materials themselves.
posted by vacapinta at 7:21 PM on August 11, 2005

"Data" is plural.
posted by interrobang at 7:27 PM on August 11, 2005

posted by mr_roboto at 7:37 PM on August 11, 2005

deeper explanation please:

In HS physics, we were taught that electrons have no mass. Is this absolutely true, or is it an oversimplification like so many things I learned in high school?
posted by Mayor Curley at 7:42 PM on August 11, 2005

Perhaps data are plural, but "data" is plural. Anyway, plurality of "data" is no longer a solid rule. These days, either construction is standard, and it mostly depnds on what you want to do with it. See usage notes here.

When it comes to the actual question, I think xil and vacapinta have given some great anwers.
posted by musicinmybrain at 7:48 PM on August 11, 2005

Electrons have mass, it's just much smaller than that of protons and neutrons (9.1*10-31kg for electrons, 1.7*10-27kg for protons). Generally speaking, in chemical terms macroscopic objects aren't charged, i.e. even though they may have some small charge only a small portion of molecules are actually charged and so the weight difference is a lot smaller than the weight of one electron per molecule.

Or, as the joke goes: "Electrons have mass? I didn't even know they where catholic."
posted by fvw at 7:53 PM on August 11, 2005

Mayor Curley: You were probably thinking of photons (the particles representing discrete "packets" or quanta of light), which are believed to have no mass.

Electrons do have mass: about 1.67262 x 10^-31 kilograms each.
posted by musicinmybrain at 7:53 PM on August 11, 2005

Apologies: mixed up my numbers. Protons 1.67262 x 10-27 kg and electrons 9.10938 x 10-31 kg, as fvw said.
posted by musicinmybrain at 7:55 PM on August 11, 2005

Flash memory: Yes, maybe, I think.
Disk drive: No. Magnetizes some iron particles, which just means moving them around.
CD-ROM: More data actually weighs less, I think, since you phsyically burn away some stuff with a laser.

In transit, yeah, electrons weigh something. Not a whole lot, and you aren't really adding mass to anything, just moving it around. In transit via a wireless link on the other hand, photons have no rest mass, though I think maybe they have something sort of resembling mass, some of the time. But not really.
posted by sfenders at 8:06 PM on August 11, 2005

Well, to be really technical and nitpicking, photons have mass. Anything that has energy has mass. This was what the famed E=mc^2 was all about: E=energy in joules, m=mass in kg, and c=speed of light in meter/second. It's a relationship between mass and energy. But really, it's beyond the scope of the question here by many orders of magnitude.

Sorry to have turned a computer question to a physics one.
posted by state fxn at 8:20 PM on August 11, 2005

On a CD-R, you aren't burning away data as much as you are changing the state of a layer of dye between the foil label and the plastic disk.

On a pressed CD-ROM or Audio CD, more data would mean an terribly little bit more mass as the foil would have marginally increased surface area from the itty bitty bumps and pits that store the data.
posted by blasdelf at 8:49 PM on August 11, 2005

Think of a row of pennies on the ground. Half of them are heads and the other tails. Then, you flip some. The pennies still weigh the same, right?

Eventualy you get into quantum mechanics, which make no sense to people.
posted by delmoi at 9:15 PM on August 11, 2005

I think I gotta call crap on that last one, blasdelf - more surface area doesn't equal more mass. For a poor example, consider a balloon - as you blow it up, the surface area increases, but the mass doesn't (of just the balloon, don't count the air - that's why it's a poor example, it's not really the same thing, but you get the idea).
posted by attercoppe at 9:27 PM on August 11, 2005

I would disagree, state fxn, about photons having mass. Yes, there is an equivalance between mass and energy, but that doesn't mean that all energy is mass -- simply that you can convert one into the other.

Photons travel at the speed of light (for obvious reasons). Anything which has mass by necessity does not.
posted by vernondalhart at 9:28 PM on August 11, 2005

The average of all the data you have stored at any time is almost always half way between on and off - just like the expected value when you flip a coin. The more memory you have the more coin-flips you are doing, and the more likely that the result will be average.

If the memory format is single ended, such that a '0' is stored as no charge and a '1' is stored as a charge then you would expect a change in the number of electrons proportional to half of the data size. It could be an increase or a decrease though, depending on weather the logic is inverted or not. Of course when you first turn the memory module on you don't care what state each bit is in, they might all be set one way, or they might be random, who knows...

If the memory format is differential then one symbol will be marked with an electron hole. In this context holes have negative mass, since the electron had to be taken away to assign a value to that location. A fifty fifty split of equal negative and positive masses is zero, of course.

I don't agree with blasdelf, pressed CDs are made by pouring aluminum on the plastic and then stamping the aluminum with a pattern. If any material was deposited on or removed from the aluminum layer the stamp wouldn't last very long (either it would get cruddy, or it would be eaten away). Of course a CD with very little data could be made with aluminum covering only a portion of the plastic, in practice this isn't done, it would cause a lot more trouble than the material savings is worth. There are those business card CDs though.

vernondalhart, photons have inertia, otherwise solar sails wouldn't work. As far as I know inertia implies mass. At some point it degrades to a question of definitions of course, because mass and energy are really the same thing...

I have only touched a couple of storage technologies. Other technologies have been mentioned, and there are more. There might be other ways of looking at the question too.

Finally, others have mentioned quantum mechanics... Well, information does behave a lot like energy, and quantum mechanics types will tell you that information is conserved. Who knows, maybe the information itself has mass...
*queue spooky music*
posted by Chuckles at 1:38 AM on August 12, 2005

CDs are injection moulded with the data on the mould and therefore on the polycarbonate substrate. They are then coated on top with aluminum and varnish. The aluminum layer is thinner than the depth of the pits, therefore its surface area (and volume) is greater when there are more pits present. Blasdelf is right.

(OTOH, I have no idea about the relative densities of polycarbonate, aluminum and varnish, so the amount of aluminum may not be the most significant factor determining weight)
posted by cillit bang at 3:25 AM on August 12, 2005

For those who have said that photos have mass while in motion: I believed this too until not too long ago, but it seems that according to the current definition of mass the photon is in fact massless regardless of velocity. This, counterintuitively, does not mean that it does not have momentum.
posted by musicinmybrain at 3:47 AM on August 12, 2005

Sorry about the crappy photons have inertia link. Also, I shouldn't have criticized, vernondalhart's comment, it correct - sorry about that too.

Anyway, if someone wants more on the light/mass aspect try "Does light have mass?". That page talks about light in a box with perfect mirrors. They make the point that even though the light doesn't have mass, the box is heavier, which is the important issue if we are talking about storage media :P
posted by Chuckles at 3:48 AM on August 12, 2005

musicinmybrain, please get out of my brain. Thank you!
posted by Chuckles at 3:49 AM on August 12, 2005

posted by musicinmybrain at 3:51 AM on August 12, 2005

Argh, not doing too well here, I blame live preview. (I thought musicinmybrain's link was the same as mine, both pages are at, having searched and searched for a link, to have the same link show up 1 minute before my post... well, I will quite now, I am not doing well today. Yesterday actually, whatever).
posted by Chuckles at 3:52 AM on August 12, 2005

CDs are injection moulded with the data on the mould and therefore on the polycarbonate substrate. They are then coated on top with aluminum and varnish.

So why would the volume of the aluminum layer change? If it was stamped on somehow, it wouldn't change mass, just surface area. In which case the change in mass would just depend on the relative density of varnish coating and substrate.
posted by sfenders at 6:49 AM on August 12, 2005

delmoi's answer is the best so far. Moving data around in most computer storage (cd storage, magnetic disc, tape, etc...) is very analogous to flipping coins. No net mass change at all.

Data alteration in silicon does move charge a tiny distance (or, if you prefer, electrons and electron-holes). This is better thought of as electrons sloshing about the chip however, moving from one bucket to another, rather than a net gain or loss of charge.

So no, data movement, on the scale of your usb drive does not move mass around. You change the configuration of the device, not it's mass.
posted by bonehead at 7:01 AM on August 12, 2005

Electronic storage (flash) cells are usually comprised of a complementary pair of transistors. In the binary '0' state, transistor A is 'off' and transistor 'B' is 'on'. In the binary '1' state, transistor A is 'on' and transistor 'B' is 'off'. So, even if a transistor in the 'on' state has a few more or less electrons than one in the 'off' state, the total mass of the complementary pair will always be constant.
posted by rocket88 at 8:26 AM on August 12, 2005 tells me that in CD manufacturing it's the master that's etched with a laser. Copies are then stamped out, then vacuum-coated in aluminum or some other metal.
posted by sfenders at 10:54 AM on August 12, 2005

doesn't it stand to reason that a hard drive is going to have a constant weight? and you're not adding or subtracting from that weight when you save or delete data, you're just rearranging the magnetic whatchawhosits.

kind of like in an etch-a-sketch.
posted by eatcake at 11:14 AM on August 12, 2005

So why would the volume of the aluminum layer change?

Because it's evaporated on, not pressed. More aluminum would stick to a CD with lots of pits (actually lumps) because it has to coat the sides of them.
posted by cillit bang at 1:53 PM on August 12, 2005

Ah. I stand corrected. I apologize.
posted by state fxn at 8:26 PM on August 12, 2005

« Older What are some good resources to learn about basic...   |   Cedar Plank Salmon Newer »
This thread is closed to new comments.