The Physical Hard Drive (PHD)

[TheEinstein]

Short version (not the provisional patent version)

A hard drive made up of physical parts, such as plastics or metals.

Colors can be used to encode more data than a standard hard drive. 3D printing is close to making 'nodes' smaller than current magnetic drive pits, though color 3D printing has a few more years to go.

shapes inside a color format can, mattering how they are utilized, result in more data per square inch or result in serious encryption.

By altering the max length, height, and width of a given drive (or the equivalent if working in 1D or 2D) you can dramatically increase the possible outcomes while accepting some drives will differ in size if in 3D. Subdividing can further allow for higher data yeilds.

Further you can, since we are working in the physical realm, leave some areas empty instead of filling them. This further increases possible outcomes. Possible outcomes in this regards is potential data sizes. 8 otcomes is 3 bits, 1024 outcomes is 10 bits. The math is log(possible outcomes) / log(2).

Combining all three has allowed me to demonstrate up to 2 times the capacity of conventional drives with a strong possibility of exceeding that. The systen is akin to hardcoding except it will be far cheaper than conventional methods. The proces can also be utilized on most, if not all, memory models to add extra total capacity (albiet a hard-coded amount) in large sums in exchange for small amounts reduced from the memory type in question. This capacity is per inch squared of drive space.

For instance removing 1 bit from a terabyte drive allows for 5 bytes of hardcoded data. Removing two would dramatically increase this total (my computer wont model that high). This is probably near a kilobyte after a half dozen bits are removed and the pattern of removal designates what the value of the hard-coded bits are.

The system is fully compatible to convert to binary but will become larger in size with Shapes and Scaling. As 3D printing matures this will become far cheaper than regular memory types. Ironically the best demonstration tool for the system would be a box of lego's.

PATENT PENDING - this is not the write up of the patent, it covers a bit more

I intend to monetize this rather than keep it. Ideal customers are banks and financial institutions, schools, hospitals, governments, data storage facilities, etc.

 

[TheEinstein]

Now how about that Peer review minus the troll and friends?

 

[_Rick_]

Combining all three has allowed me to demonstrate up to 2 times the capacity of conventional drives with a strong possibility of exceeding that.

Before this goes anywhere, clarification of what "demonstrate" means in this context is required.

Also, please rewrite your description, it is currently very difficult to figure out your back-references in sentences. Sentences also lack a logical structure or explanation.

From here, it looks as though you want to use a shape to encode information. Do you have a partial implementation of the process required?
Furthermore, do you think the patent is worth the application fee? I fear it may not be, as the technology required to make your approach cost effective (even reading data at any relevant density requires insanely high resolution laser rangefinding, actually adding on physical features is not feasible at but very slow speeds, for decades to come - way beyond the term of a patent). Furthermore, using color is somewhat similar to what multilayer optical disks already do. Technically, a second color is used to access the second layer, which therefore encodes additional data in color, and may be partially prior art.

As 3D printing matures this will become far cheaper than regular memory types.

You are basically betting on that, with this patent application, and everything I've learned over the past years tells me that this is highly unlikely. Even now nano-scale physical manipulation techniques are barely working in the lab. Scaling this over the course of 10 to 20 years to something that finds its way into a useful number of products to recoup your patent application fees sounds utopian.

 

[TheEinstein]

Before this goes anywhere, clarification of what "demonstrate" means in this context is required.

Also, please rewrite your description, it is currently very difficult to figure out your back-references in sentences. Sentences also lack a logical structure or explanation.

Being vague worked? This is to get intelligent questions and weed out amateurs.

From here, it looks as though you want to use a shape to encode information. Do you have a partial implementation of the process required?

Shape is one of three seperate possible functions. There are many ways you can seperately incorporate shapes for data storage. Imagine one possible method with me, where pins of various lengths are emplaced and the length of the pin determines the value. Shapes is not neccessary but is allowed in my invention. Consider lego's where peices can be 1x1x1 in size, or ne otherwise shaped. If using a graph paper, for example, a single cell can be a single source of data. If you somehow seperate, merge, or otherwise make it clear several cells are together you gain a shape. There are many ways on can do this, all covered in my patent, including providing a physical seperation, color outlining, and encoding methods. The methods vary for quality of course.

shapes can be several atoms but this is probably not fiscally ideal. Instead shapes will probably be somewhat obvious via grooving, seperation, or even elevation based differences.

back to graph paper, inside that graph paper make every and any possibe shape, including T, M, U, and Cross shapes. Any plausible shape is allowed. Now a shape can be multiple color encoded or singular. Since our graph paper has an encodable value if each cell is 0 or 1, merging just two cells into a single 0 or 1 value increases this. Since we can merge any two connecting cells this means there is a lot of variations possible. Since we can make multiple shapes this increases our pool of possible outcomes.

Furthermore, do you think the patent is worth the application fee? I fear it may not be, as the technology required to make your approach cost effective (even reading data at any relevant density requires insanely high resolution laser rangefinding, actually adding on physical features is not feasible at but very slow speeds, for decades to come - way beyond the term of a patent).

The costs are actually feasible due to the way I plan tl propose a company handle it, though I covered as wide a variety of possible methods as I could, to protect my IP.

Furthermore, using color is somewhat similar to what multilayer optical disks already do. Technically, a second color is used to access the second layer, which therefore encodes additional data in color, and may be partially prior art.

My methods differ in the extreme to optical disks. My method has no tracks normally, in the manner tracks is described. That I can apply scaling to multimedia disks is seperate from colors. Yes I can enhance optical disks.

You are basically betting on that, with this patent application, and everything I've learned over the past years tells me that this is highly unlikely. Even now nano-scale physical manipulation techniques are barely working in the lab. Scaling this over the course of 10 to 20 years to something that finds its way into a useful number of products to recoup your patent application fees sounds utopian.

There is a functional 3D printer able to print accurately at 10 microns. There is a full color printer able to print at 25 microns. Viewing color at those levels is trivial to current science. Yes there is some costs but I have planned a functional way to mitigate those costs. No I wont give all methods and manners away but reasonable questions are plausible for me to answer.

Now excuse typos, I just am now taking a breather from being 2nd person at a serious rollover accident involving 3 people, got a major asthma attack for my efforts, and now while I sit and rest here my body is suffering adrenalin withdrawls. I will look back later when I feel better.

 

[KingFatty]

The description gives me the impression that these are desirable principles to be combined. However, the problem is there is no conceivable way to actually implement things.

So to me, any breakthrough would mean inventing something that is not contained in the explanation. For example, using colors or shapes is mentioned as being capable of increasing capacity. However, there is no explanation how that could actually be used on a hard drive. The reader is left to invent these details on his/her own. Perhaps that is the issue - the current idea is still about principles to be used in an ideal best-case situation, but hasn't actually gotten the break through of how to implement that?

Analogy: there is an abstract principle of creating a light-speed warp drive, and we can harness nuclear energy to make current conventional drives faster. But can you see there is no invention here, because I haven't 'invented' a way of actually achieving light speed, or explained how to harness nuclear energy specifically?

Here, it seems like there are abstract conclusions how some basic physics principles could potentially increase storage, but no tangible way to actually accomplish that.

 

[KingFatty]

Oh I see, you are describing use of a 3D and/or color printer to encode/store data? I was mistaken by your title of a hard drive by thinking it was re-writable. But this is more like a printout that is permanent?

Ah, what about past physical encoding schemes such as phonograph records (grooves), braille (raised bumps), computer punch-cards (holes), etc.? There have been many previous ways to encode data using physical shapes. Even the little jewelry music box that plinks out sounds based on a song encoded as raised bumps on a metal spool? Did you know that optical CDs, like music CDs, are pressed in a mold too, like physically stamping the optical pits/lands into the disc?

Also, consider past data encoding schemes using color? The most basic would be a UPC bar code (black and white lines), and now there are 2-D matrix codes, such as a QR code you scan with your phone? Microsoft invested heavily in a color-based QR code system, forget the name, but it used shape patterns and color patterns.

So now we consider, is your invention combining the two above principles, based on using a 3D printer? Then the patent office might try to reject your invention based on an obviousness argument combining the two principles above. Or, is there another feature not contained anywhere in the above principles? That is good to bring out/elaborate on.

But also the practical realities. Did you calculate how much data could be actually stored by your invention, such as the data density? It seems like it would be far lower than anything currently available, so I don't see how we could justify the expense of a fancy 3D printer to create storage output that cannot store more than a Blu-ray or microSD memory card?

Why not just use those super-cheap available methods to store vast amounts of data with reliable error correction protocols etc. and very cheap readers/writers/burners? Why sacrifice the ability to re-write/erase reliably? Are you providing a solution, but there is no problem that needs to be solved?

I don't mean to be hard on you, I just want to play devil's advocate because you will face much harder questions from investors who you'll need to convince to give you funding, so getting a handle on answering these tough questions are probably a good start.

 

[TheEinstein]

Kingfatty, while other methods have lightly utilized some features people could call shapes this is a pretty radical departure.

The various aspects in turn could be quoted a little perhaps in some regards but the aim, methods, scale, and more make this a unique product.

Yes this is a hardcoded drive, permament storage system.

If somehow recycling gets to nanoscale then there might be a re-write function

I did do some calculations on the data density and my efforts stalled at about 175% of current density. The problem lies in modeling 'Scaling' and 'Shapes', and determining a cost effective bound on different aspects. I sincerely expect 200% to be passed and 300%+ is even a plausible outcome.

Thereis a target market for perma entdata storage, since the media would last concievably far longer than most other medias, be cheaper to produce, have some serious nasty encryption capabilities built in...those customers would have incentive to use my system. Banks for instance keep records for many years, and they invest a lot in this effort. The NSA is trying to store -everything- and this makes storage of less essential data a real possible.

Now the read times would be slow, but it is mitigatable and will improve as GPU, CPU, and RAM improve.


the fact I can hard-code data onto an existing storage medium also has some applications, the read speed on that would be faster, and the advantages to OEMS could be to hardcode certain functions in the hard-drives to reduce space issues elsewhere, or other potential benefits.

Spycraft of course would significantly benefit... a USB with hidden hard-coded data that a 3rd world nightmare nation would not know to look for.... or know how to read if they did find it...

I am not worried about the marketability, the ideas that other patents could have leeway (like someone saying a torch patent covers a neon light), nor in any ideas that I wont get a good value for this.

 

[TheEinstein]

Oh and devils advocate is ok, I have no problem with even questioning the math of it. I care when someone brings up a hurtful time in my past with no need, to try to say I have nothing, but a devils advocate is fine.

 

[_Rick_]

Spycraft of course would significantly benefit... a USB with hidden hard-coded data that a 3rd world nightmare nation would not know to look for.... or know how to read if they did find it...

I don't think you can sell something based on the broken "security through obscurity" paradigm.

I also am still confused about whether you want to create a writable medium, or a permanent medium. CDs, as permanent medium are not writable - which is why they are not used for storage, but instead for distribution.
This is something that needs to be cleared up.

Micron-scale is also still ruddy large, compared to nano-scale flash memory, that is re-writable hundreds of times. While you may be able to match, after many years of development, what is currently possible in the flash based memory sector, I doubt your proposed technology will be able to catch up.

Quite simply the proposed complexity of the read and write mechanism exceeds what any reasonable person is ready to accept.
Data density also has no advantage, if the read speed is slow. Even tape back ups, that used to serve that part of the market are on the decline.

Oh and:

Being vague worked? This is to get intelligent questions and weed out amateurs.

I would claim you achieve the opposite. Intelligent people know better than to bother with someone who cannot put his ideas into a clear and concise format. Professionalism and intelligence attracts its kind. If you express yourself in a way which hides this, then you will not gain the answers you seek.

Also, you say "no tracks are present". Then how is the data ordered? Now, don't wiggle out, saying it could have arbitrary order, because that will not make it a storage medium.
And I am still not convinced that technologically, you are doing something significantly different from what a multi-layer optical medium does. They are a 3D structure, using multiple colors to encode data, arranged in a specific way (tracks).
What I gleaned from your exposition so far, is this very same concept, but quasi-analogized, with the hope the larger features of greater variety can achieve the same density as smaller features of little variety. Everything I know about A/D conversion tells me, that this is not going to work, simply because it is easier to build a high frequency simple A/D-converter, than it is to build a good, but lower frequency converter.

What is your error correction strategy? What is the complexity of the read-head? I would be assuming that you'd require an optical approach - and thus be limited in the color range you can use, simply because at certain resolutions you wouldn't be able to form a beam of light that can excite the colored area to reflect, due to wavelength limitations.

Do you plan to use a spinning disk format (how do you balance the disk?), a tape format (how do physical features resist the wrapping over a tape reel?) or something else entirely?

Oh yeah, here's a final question: How do you think this will work, when we cannot even get OCR on printed text to work reliably?

 

[TheEinstein]

Sequential, or for encryption user defined.

OCR has advanced a long ways, where in sometimes it has problems with written text or for certain fonts and angles. I am not worried about the reading ability since a perfect order will be easy for software to read.


Wavelength complaint laughed at since colors in human range are far smaller than talked of here. also see 'invisibility cloak' for an alternative way to manage light precisely (not reccomending that, just a possibility).


A simple pascals triangle answers storage density in scaling regards (scaling having two components and this being one of them). it also can help with shapes

1 1
1 2 1
1 3 3 1
1 4 6 4 1
1 5 10 10 5 1
1 6 15 20 15 6 1
1 7 21 35 35 21 7 1
1 8 28 56 70 56 28 8 1

Now where if you remove a bit you still have some potential variations within. so 8 bits is 256, 7 bits is 8 arrangements of 7 bits or a value of 1024. 6 bits would be 28 variations of 6 bits or 1792 outcomes (roughly 10.8 bits). 5 bits would be 56 outcomes times 32, again 10.8 bits with rounding. and so forth. in this demonstration there is (rounded) 12.67 bits of data encodable just with scaling or minor shape efforts. I am probably going to map out a sequence of 64 bits and try to get the exact "possibilities" but it will take time.

As for the rest, read speed will be read speed, I am not going to claim expertise in spinning versus tape versus the idea I have. This is not a worry, for the physical act of reading will probably not be as time consuming as the act of "recreating" which will be the big time consumer.


Seriously what is the problem? I do not feel you are devils advocating so much as trying to make me give up hope so you can yoink the patent from me.

 

[aigomorla]

your type of storage would be excellent if we had quanitum computers.
Because your type of storage allows the different types of code a quanitum pc would utilize. +, -, neutral.

however because today's x86 structure is based on binary, as well as ARM, we only need 2 gate channels. Open and a Closed, which translates to binary 0 1 0 1 1 1 0 .

This type of storage ur suggesting is about 50 yrs too advance to utilize fully, and a translator to make binary into a data format allowable to be stored on your matrix would probably require a PC stronger then the one using to transfer the storage.

Seriously what is the problem? I do not feel you are devils advocating so much as trying to make me give up hope so you can yoink the patent from me.

i think your assuming too much on hardware.
As you got to the upper levels of your data matrix, it would translate to a massive cpu overhead to translate the binary into the upper level matrix.
To us, that's just not efficient for the increase of storage space.
Also you told us, it significantly reduces shelf life, which in the IT world is considered TABOO.
Anyone who has IT experience finds it very difficult to do something called "delete" unless we know a backup of it is somewhere.
To tell us we can lose our data after a set amount of time.. is like telling us to plan ahead for an unavoidable doomsday.

 

[TheEinstein]

Naw it is not that complicated, all data is data. Binary is data.

A processor will interpet the data as it comes in without these worries, yes as stated a long 'translate phase' but instead of paying $40++ for a terabyte of data storage your going to pay $2 (not right away, supply and demand is a skewed right now, but a heightened demand means more manufactoring supply, and this would be planned out by a company in position to push the PHD).



Besides GPU's would be better to push the code either way.


And Let me be clear, storage life might be up until our sun scorches Earth mattering on materials used. There is no problem with shelf life.

 

[KingFatty]

Just FYI the patent office Examiner will try to classify your invention, by assigning it into one or more of these classes (which each have their multiple sub classes):
http://www.uspto.gov/web/patents/classification/selectnumwithtitle.htm

Do you think it's appropriate to fit into class 365, and if so, do any of those subclasses seem to really hit the nail on the head as far as what this is?
http://www.uspto.gov/web/patents/classification/uspc365/sched365.htm

Or maybe you can find a more appropriate way to classify it. But, searching based on the classification system is another approach to help find potentially similar technology patents/application.

Then, it's extremely helpful for our understanding, to describe things in terms of how it might differ from something in the classification system we are already familiar with.

 

[TheEinstein]

My lawyer is handling classification

 

[Haserath]

I might have missed it, but how do you plan on reading the data?

 

[Revolution 11]

TheEinstein, no one here is interested in stealing your patent or simply discouraging you for the sake of being a troll. We are trying to help by asking the hard questions before other more important people do the same.

But like _Rick_ said, you have a interesting idea that is vaguely explained, with certain principles that others have thought of before (prior art concerns). When someone mentions a valid technical concern like data density, material properties, or error correction, you just hand-wave it.

Certainly a storage company can license the patent from you and handle all the really hard problems associated with storage. But then the patent itself is not worth much or the licensing must be cheap. And the implementation is key here, or the customers you mentioned like banks and hospitals will never touch this.

 

[TheEinstein]

Well part of the problem is my lawyer has advised me to not be boxed into marginalizing a claim


Patent is secure.. feather quill pen, fountain pen, ballpoint pen all are used for the same purpose, all use ink, they are all gripped in a similar manner, and they are roughly the same shape.

Yet they are different in a novel and new manner.


A punch card has holes yet the means to read is different.

An optical disk uses some colors but the aim is no where near my level, nor is the possible bounds of their data retention.

All other products differ in significant ways and therefor this is new and novel which meets the criteria set by the patent systems of the world.



Now as to the product it will not be tied down to water down the patent, it is flexible in many ways and you need to accept it is flexible.

Reading the drive can be multifaceted... if there are colors then clearly a means to identify the colors is needed. identifying a specific tool to do this is non-essential to the patent.

If there are shapes then a means to detect them is also needful. will it be imaging? sound? feel? it is a minutae item which is best left to the customer desires and/or manufacturer choice.

Which direction will you read the bits? again an amateur question.. if encryption is desired make a different way than others will. If speed is desired do it in a logical manner...

If the method is 'trinary' instead of binary well then 2 spots equals 9 and 5 is 273. pick which level you like and make the number outcomes fall inside if you are worried about overhead on a CPU. All of it is trivial and up to the next owner of the patent.

I am not going to be the optimizer, presumably the next owner will pay top dollar for people able to optimize it. All i am is the inventor, the person who made the depth and scope as wide as posible so i am the man who profits and not others.

The sad thing is i am not looking for critics but peer review, even sales people who cab get a % of the whole. seems i wont find either in here.

 

[Hellhammer]

The sad thing is i am not looking for critics but peer review, even sales people who cab get a % of the whole. seems i wont find either in here.

No offense but I don't think this is the best way to find what you are looking for. While most of the member here are knowledgeable about technology, I bet most are just enthusiasts with no professional background in the industry.

If you're serious about selling or licensing your patent, you should look into hiring a marketing firm with experience in the storage industry. Sure it's an expense but it will be hard trying to get in touch with the right people without help from an insider.

 

[corkyg]

Agree with Hellhammer. At the very least, see if you can get this moved to "Highly Technical."

 

[_Rick_]

All of it is trivial and up to the next owner of the patent.

And that there is the problem. A patent has a limited lifetime, during which it has to generate enough revenue to recoup the investment.
A patent that has no current implementation is going to be of no interest to someone, that can not expect to recoup the investment, before the patent protection runs out.

You used the word "demonstrate", yet from what I gleam in your descriptions, there is no such thing as a technology demonstrator. Merely a general concept. Sure, you can patent this concept, but if you want to sell it, you need to make a more coherent case than you currently do.

I would suggest you sit down and write a two-page white paper, that clearly shows potential applications, the challenges that need to be overcome to monetize on the patent as well as relevant projections of data densities and read/write speeds. These are things that may not be important to you, but they are important to the person you are selling to - which means you have to convince them on these points.

Peer review in the classical sense, by the way, means that you propose a technique which is then examined, to see whether it is a workable solution. Knowing this is important before filing for a patent, as an unworkable patent may either be rejected, or accepted, but you will not be able to monetize on it, as there is no product that can be made of it.

Nebulous claims, presented in what is quite honestly a rather haphazard way, do not lend themselves to the analysis a proper peer review would require.
I'm not even sure if what is described is sufficient to make up a patent, as this requires a technical solution, which I cannot currently identify.

 

[TheEinstein]

I will make a fresh post in highly technical, better written.