Talk:List of prolific inventors

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Invention (Rated List-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Invention, a collaborative effort to improve the coverage of Invention 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.
 List  This article has been rated as List-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.

Prolific inventors article[edit]

I started this article as there was no up-to-date information available anywhere I could find on this topic. The four main lists that I found were an article in Popular Science (January 1936), and article on the top ten living US patent holders in USA Today (and later revised in Portfolio magazine in 2007), and a 2000 article in Time magazine on the top 5 inventors. The USA Today and Portfolio articles did not include historical prolific inventors such as Thomas Edison. (a significant oversight!), and the Time article did not include inventors who have recently surpassed Edison in terms of the number of patents. I had intended a succinct article with just the list of the top ten prolific inventors in history, but this attracted "globalize" and "original research" tags, so I thought it necessary to show that the information was not original, but was the combination of several published lists. All of the patent numbers are verifiable using the links provided.

Another feature that could be added to this article is the distinction between Utility patents and Design patents. (Done)

The patent numbers of the active inventors may change on Monday nights, when the USPTO publishes new patents. I will endeavor to update the list on a regular basis. AlexBartlett4 (talk) 13:57, 10 October 2009 (UTC) — Preceding unsigned comment added by (talk) 05:52, 6 August 2014 (UTC)

Design patents[edit]

This article is useful information as regards utility patents, but is flawed in omitting inventions of new, original designs for an article of manufacture. Design patents are, by law (35 USC §171), inventions.

The PTO itself says it succinctly on their patents pageUSTPO definition of patents, including design and plant patents, as being for inventions What is a patent?

A patent is an intellectual property right granted by the Government of the United States of America to an inventor “to exclude others from making, using, offering for sale, or selling the invention throughout the United States or importing the invention into the United States” for a limited time in exchange for public disclosure of the invention when the patent is granted.

There are three types of patents. Utility patents may be granted to anyone who invents or discovers any new and useful process, machine, article of manufacture, or composition of matter, or any new and useful improvement thereof. Here is the process for obtaining a utility patent. Design patents may be granted to anyone who invents a new, original, and ornamental design for an article of manufacture. Plant patents may be granted to anyone who invents or discovers and asexually reproduces any distinct and new variety of plant. Burdlaw (talk) 23:04, 7 February 2012 (UTC)burdlaw

Your own reference 35 USC 171 provides an explicit distinction between patents for inventions and patents for design: "The provisions of this title relating to patents for inventions shall apply to patents for designs, except as otherwise provided. " My humble interpretation of it is that "patents for design" are different from "patents of inventions" (which is, BTW, very consistent with the rest of Wikipedia on this subject). Ipsign (talk) 06:04, 8 February 2012 (UTC)
Whether "Design patents are, by law, inventions" is not relevant. This is not an issue of some legal definition of the word "inventor" in some USPTO regulations. Using the logic: "a new design is the invention of a design, therefore designers are inventors", inventors of new words are also inventors, and should be included in List of prolific inventors. Lewis Carroll invented 28 new words in "Jabberwocky" alone, and many others in other writings. However, Carroll would be far the most prolific inventor - a madman talking complete gibberish may invent thousands of words every day, and should be listed as the most prolific inventor. While this is reductio ad absurdum, it illustrates that this is clearly not what the Wikipedia readership would expect of this article.
Calling an industrial design right a design patent is US centric terminology. In many countries they are called registered designs. Wikipedia's own description (from the article Patent) is worth quoting:
"The term patent usually refers to an exclusive right granted to anyone who invents any new, useful, and non-obvious process, machine, article of manufacture, or composition of matter, or any new and useful improvement thereof, and claims that right in a formal patent application. The additional qualification utility patent is used in the United States to distinguish it from other types of patents (e.g. design patents) but should not be confused with utility models granted by other countries. Examples of particular species of patents for inventions include biological patents, business method patents, chemical patents and software patents.
Some other types of intellectual property rights are referred to as patents in some jurisdictions: industrial design rights are called design patents in some jurisdictions (they protect the visual design of objects that are not purely utilitarian), plant breeders' rights are sometimes called plant patents, and utility models or Gebrauchsmuster are sometimes called petty patents or innovation patents. This article relates primarily to the patent for an invention, although so-called petty patents and utility models may also be granted for inventions."
Just as the Wikipedia article Patent relates primarily to patents for inventions, List of prolific inventors relates primarily to inventors of patents for inventions. Design patents, plant patents, and patents pending are included in the numbers in the "Total, INPADOC" column. AlexBartlett4 (talk) 10:22, 10 February 2012 (UTC)


I have been tasked with 'Globalizing' this article, which may be a little tricky - any advice greatfully received!

The most common inventor on a European Patent Office search on Boliven is "Die Erfindernennung Liegt Noch Nicht Vor". This didn't seem to be a very likely name, so I used Google to translate this "name" into English. The result:"The inventor is still not available", which, I expect, is a pretty good translation! This search included applications as well as granted patents, and this particular inventor did not appear on granted patents, so I am being a little unfair! However, it shows that a database search is perhaps not as easy as one would like it to be.

The problem in Japan seems more severe. There are a massive number of 'patents' is Japan, but they are mostly not for what is normally called 'inventions'. The most prolific filer of these patents in Japan (according to a Boliven ranking) is Ugawa Shohachi, with 'inventions' seemingly exclusively related to Pachinko machines manufactured by Sankyo. These inventions are things such as color coded ball trays and patterns of pins for the balls to bounce off. While it is a bit far fetched to call these 'inventions', the real problem is that this lists applications, not granted patents. As the article is about prolific inventors, it is inapprpriate to include either:

  • Designs
  • Patent applications. The patent office decides whether something is an invention or not - anyone can file anything as an application for a patent. As as result, I have only included granted patents in this list.

I have included an international section. So far, this only includes Japanese, Korean, European, and US patents. I will expand this as I find appropriate sources. As the "globalization" of this article to major international patent countries did not alter the ranking or prolific inventors, I took the liberty of removing the {{globalize}} tag. (Edcollins, if you disagree, please let me know what extra I should do)

AlexBartlett4 (talk) 03:22, 28 October 2009 (UTC)

I added INPADOC and PCT patent counts to the main table. AlexBartlett4 (talk) 02:15, 18 March 2011 (UTC)

To further internationalize, I changed the main tables to concentrate on granted utility Patent families rather than granted US utility patents. In most cases, this is the same, as the highest number of patents is in the US. There are exceptions, though, such as Béla Barényi, who has more patents in Germany than the U.S. AlexBartlett4 (talk) 00:52, 15 May 2011 (UTC)

Proposal to remove INPADOC data A proposal by Keithrwalker (talk) 16:10, 1 November 2017 (UTC)

Having maintained the data for this article for many years, I have come to realize that the effort to meet a globalization goal have not been met through INPADOC, and in fact INPADOC reduces article quality for the following reasons:

  1. The data is stale because the host, ESPACNET, started blocking automated data pulls and what they consider excessive queries, both of which are necessary to practically maintain the page. They allow registration for an ID that can be used for this purpose, but the queries require learning a specialized syntax and using a different method than I use. I'm unwilling to invest time sorting this out since I pull data and produce the wiki code using simple web queries via a Microsoft Excel spreadsheet with macros.
  2. The database has a 10,000 patent family limit, and so it is impossible to get a count for the most prolific inventors.
  3. The database has inexplicable record loss. For example, one week an inventor may have 1,000, but the next week only 900 using the same query.
  4. The database cannot be reliably queried by inventor. For example, searching for inventor "John Smith" will return not only John Smith's patent families, but also those that have any combination of inventors that include a John and a Smith in their names, such as John Doe and Jane Smith.

The globalization goal is already met by adding counts outside of the USPTO database to the USPTO column, such as is the case for Béla Barényi. So I will use that method going forward and rely on others updating the article with references to their sources (or they can update this Talk page), and of course inspect the references to ensure there are no double count errors. I use a similar approach for inventors who have patents prior to 1976, which cannot be queried from the USPTO database, such as for Shunpei Yamazaki. I will update the article to make clear the limitations of the article and the use of sources. It needs to be clear that it is impossible to get a perfect count for prolific inventors due to record-keeping, inventions that are never patented, and disagreements about valid patents.

Before I implement this change, I welcome any comments below, and of course will welcome edits to the main article should I proceed with the change.

Comments about the Proposal to remove INPADOC data

Please add any comments here.

Carleton Ellis[edit]

I have just encountered the article on Carleton Ellis which claims he was the holder of 753 patents, which would put him on the top ten list. However, the article is not clear that they are 753 US patents. If 753 is the sum of all patents filed internationally, then it is not a comparable number, as the same invention would be counted many times. As all of his patents predate 1976, a USPTO search doesn't retrieve any. Does anyone know the true number? AlexBartlett4 (talk) 12:28, 11 October 2009 (UTC) After finding a Popular Science article from 1936 on the most prolific living inventors of the time, I included Carleton Ellis.

Yoshiro Nakamatsu[edit]

I am trying to track down the patents of Yoshiro Nakamatsu, aka Dr. Nakamats. He claims to have over 3000 patents, and appears to be somewhat of a (controversial) celebrity in Japan. He also claims to have invented many things which he clearly did not invent. A search of the USPTO website gives only six patents, but there may be many more patents issued before 1976. If his claim of 3000 patents is true, it would likely be for a total of worldwide patents, and almost certainly mostly Japanese patents. Even more likely, this would refer to Japanese patent applications, as in Japan a patent can be filed without being examined. It is published after 18 months, and then is "laid open" for up to 7 years before the inventor elects whether to pay for examination. Does anyone know a source of some reliable information on Nakamatsu's patents? AlexBartlett4 (talk) 06:42, 12 October 2009 (UTC)

I get 107 Japanese patents for the name "Nakamatsu Yoshiro" but I am not sure how many of these are referring to the same Dr. Nakamats: More notably, there are hardly any citations from US patents on these. He has only been cited a total of 7 times. Not much for a 'prolific' inventor. markashworth (talk) 15:36, 29 July 2010 (UTC)

Nakamatsu's claim is false on two counts: Nakamatsu does not have 3,200 patents, and even if he did, he would still not be the most prolific inventor in history. Nakamatsu has been granted just 14[1] U.S. patents. Even when counting all international patents and patent applications, Nakamatsu does not approach his claimed patent count. A search on the International Patent Document database (INPADOC), reveals that Nakamatsu has a total of 395[2] worldwide patents or patent applications. AlexBartlett4 (talk) 12:02, 27 November 2011 (UTC)

Article Title[edit]

The original title of this article was "Prolific inventors". This was improved to "List of prolific inventors", certainly more accurate, and that is the current title. The title was then further changed to "List of prolific individual patent grantees" by Reboot with the explanation "moved List of prolific inventors to List of prolific individual patent grantees: the number of patents issued and the number of inventions do not correlate very strongly. The title is misleading at best and forcing a POV at worst". This is actually incorrect. The number of patents and the number of inventions correlate very strongly, if one uses the actual definition of inventions and patents. What does not correlate strongly is the public's perception of what an invention is, and the real definition of an invention. The general public often associates an 'invention' with a 'product', as if there were a one-to-one correlation. However, some products are the accumulation of tens of thousands of inventions - a smartphone is a good example, where there would be tens of thousands of inventions relating to the CMOS chips used, many thousands more relating to the ways of manufacturing these, thousands more relating to the GaAs chips used for the radio circuits, thousands for the processor architecture, thousands for the Flash memory, thousands for the WiFi and cellular radio circuits, tens of thousands for the LCD screen, and thousands for the software. Various other products involve no recorded inventions, as the product may have essentially been invented in prehistory, with no patentable improvements being made since then. In other cases, there may be a wide range of products, all based on one single invention. As a trivial example, a product (for example, a shoe) produced in a range different styles, colors, patterns and sizes is a range of separate products, but usually these would not rate as separate inventions. Each shoe, however, may incorporate a number of inventions. Nike, for example, has a total of 2213 U.S. shoe patents (as at 30 March 2010). Of these, 263 are utility patents, 1949 are design patents, and 1 is a reissue patent. Products that only differ from the prior art in their design are eligible for design patents. Designers are very valuable to society, but a designer is not the same as an inventor. This is why I have ranked the inventors according to utility patents - patents for inventions.

The general view is that the number of utility patents is the only available objective measure of a prolific inventor. For all of the 20th century, Thomas Edison was considered to be the most prolific inventor in history, in part because he had the most patents. While he was granted 1084 U.S. utility patents, these related to a dozen or so products. For example, he was granted over 200 patents on the incandescent light bulb, a very simple product involving a glass envelope evacuated of air, with a carbon filament in it (incidentally, Edison did not invent any of the glass bulb, evacuation, or carbon filament aspects of light bulbs - his inventions were in the details of these aspects). The public may generally consider the incandescent light bulb to be one single invention, but again, that is because the public conception of an invention is not the same as the actual definition of an invention. The Wikipedia definition is "An invention is a new composition, device, or process. An invention may be derived from a pre-existing model or idea, or it could be independently conceived in which case it may be a radical breakthrough."

Some may think that Edison's light bulb is a classic case of an "independently conceived radical breakthrough". This is not the case. The basic concept of an incandescent light bulb was made Sir Humphry Davy in 1802. Many subsequent inventors improved Davy's invention prior to the successful commercialization of electric lighting by Thomas Edison in 1880, 78 years later. There are many other inventors listed in the brief history of the light bulb given in Wikipedia's article on the incandescent light bulb. Edison's two hundred inventions on the light bulb combined with his other inventions on electricity generation, batteries, and distribution to make the first commercialized electric lighting system. It was not 'radical breakthroughs on the part of Edison. Although Edison had few "independently conceived radical breakthroughs", I think that few people would deny that Edison was a prolific inventor, based on the number of U.S. patents he was awarded. Also from Wikipedia's invention article: "An invention that is novel and not obvious to others skilled in the same field may be able to obtain the legal protection of a patent." There is no tally or record of inventions other than those at the patent offices. There are many, many inventions made that are not novel - that is, they are re-invented, as the inventor did not have knowledge of the original invention. However, a prolific 're-inventor' hardly tallies with the public perception of a 'prolific inventor'. Also, a prolific inventor of inventions that are obvious does not inspire the public to think that they are a 'prolific inventor'. So in these two aspects, the USPTO definition of a patentable invention actually correlates more strongly with the public perception of an invention than does the actual definition of invention (which does not require novelty or non-obviousness). All of this combines to mean that U.S. Uutility patents are our best available measure of inventions, and correlates reasonably well with the public's idea of prolific inventors. The article already contains a discussion on the significance or otherwise of inventions. Perhaps this section should be expanded to cover any misconceptions. Note that this is not just my opinion. Lists that I've managed to find and reference in the article talk about prolific inventors, rather than 'individual prolific patent assignees'. For example:

In January 1936, Popular Science Magazine published a list of the "most prolific living inventors to be found in America today"
On December 4, 2000, Time Magazine published a list of the "top five inventors".
On October 15, 2007 Condé Nast Portfolio Magazine published a list of "the world's most prolific inventors alive"

The proposed new title "List of prolific individual patent grantees" may be technically accurate, but is no more accurate than the previous title, and is overly pedantic and obscure. AlexBartlett4 (talk) 02:00, 26 March 2010 (UTC)

Reboot's proposed rename is nonsense at best, and is closer to anti-patent POV propaganda. He's also attempted to push his POV in articles on individual inventors. AldaronT/C 04:34, 26 March 2010 (UTC)

Biographies of the top ten inventors[edit]

Seven of the top ten most prolific inventors in history have no biographies in Wikipedia. Over the last few months, three people have added links to their names - however, they are all red links, going nowhere. Originally, I did not link these names, as there were no corresponding articles, so there seemed no point in creating dead links. However, others seem to think they should be linked - perhaps as a spur for someone to create the entries. I think that it would be useful for Wikipedia to include such biographies, but I don't think that I would be the right person to write them. Does anyone want to take up the challenge? AlexBartlett4 (talk) 13:29, 29 June 2010 (UTC)

List limited to US patents - other inventors with >1000 patents are not listed[edit]

This list is misleading as it does not rank the most prolific inventors, but those with the most U.S. utility patents. Two very different things! For example, Béla Barényi is not even listed although he had over 2,500 patents (mostly in the auto industry): Quiname (talk) 20:30, 16 April 2011 (UTC)

Thank you for bringing Barényi to my attention. I have used granted U.S. utility patents as a proxy for worldwide granted utility patent families. There are several reasons for this:
  • It is inappropriate to use the total worldwide number of patents, as patents are national only - there is no world patent. To obtain patent coverage in more than one country, a patent for the same invention must be filed in each country of interest. PCT (Patent Cooperation Treaty) patents are not worldwide patents - the patent must still be granted by each national office.
  • Design patents should not be included, as they are patents for designs, not inventions. Likewise, reissue patents and plant patents should not be included.
  • Only granted patents are included, not patent applications (patents pending). Until a patent is granted by the appropriate patent office it is not known that the patent is actually novel, or patentable.
However, you have a very good point - being ranked on U.S. patents gives an undue U.S. perspective to the list. What is the answer to this? I think, to rank on Patent families. A patent family is a set of patents taken in various countries to protect a single invention. So, the question becomes, how to determine the number of patent families? As the number of patent families must be equal or greater than the maximum number of patents in any one country, then the minimum number of patent families will be the largest patent count in single country. This will often be the number of US patents, but it will not always be. Béla Barényi is a case in point, as he has more German patents than U.S. patents. There are also some Chinese inventors who are rapidly advancing in patent count, and who typically have more Chinese patents than US patents. I shall adapt the list to be ranked by granted utility patent families rather than granted U.S. utility patents. AlexBartlett4 (talk) 05:55, 17 April 2011 (UTC)

Béla Barényi > 2500 patents[edit]

When Béla Barényi retired on 31. Dezember 1972, he already had more than 2000 patents, twice as many as Thomas Edison (in 2009 Barényi had over 2500). So the article's claim is not true that "For all of the 20th century, Thomas Edison was the most prolific inventor in history." I do not know when exactly Béla Barényi surpassed Edison, but it seems likely that Barényi was the most prolific inventor for most of the 20th century. Quiname (talk) 20:39, 16 April 2011 (UTC)

The patent count for Barényi is for worldwide patents. This is not equal to the number of inventions, as each separate invention must be filed as a separate patent in each country for which patent protection is sought. As Barényi was European, and worked for Daimler-Benz, many more patents for his his inventions were filed in various European countries than is typical for most inventors. From INPADOC[1] the number of patents that Barényi has in various countries is as follows:
  • Germany: 595
  • France: 334
  • USA: 178
  • Austria: 80
  • Switzerland: 39
  • United Kingdom: 10
  • Other: 2
INPADOC makes an attempt to list each patent family once only. However, small variations in a patent name or filing sequence can cause patent families to be listed more than once, so INPADOC's count of 1,238 likely corresponds to the 2500 total quoted for individual patents. In any case, the 595 granted patents in Germany certainly must be separate inventions. With at least 595 patent families, Barényi should certainly be on the list. I will add him at the appropriate ranking.
The statement regarding Edison being the most prolific inventor for all of the 20th century still stands: Edison had 1084 separate inventions (and 2332 worldwide patents) versus 595 separate inventions for Barényi. AlexBartlett4 (talk) 05:55, 17 April 2011 (UTC)

Inventors to watch[edit]

There are a number of inventors who are either close to 200 US utility patents, or are rapidly rising in patent count, and could therefore exceed 200 patents soon, or need to be included if the cutoff level for inclusion is altered. These inventors are tracked in the list below. USPTO references should be verified prior to inclusion, especially to ensure there are not multiple inventors with the same name included in the reference.

Issued US Utility Patents

Inventor Last update Last patent
Benzion Landa 198 [3] Mar. 13, 2018
Robert E. Fischell 196 [4] May 15, 2018
Michael Fripp 191 [5] May 15, 2018
Dilip Chatterjee 176 [6] Jan. 9, 2018
Shinichiro Koto 161 [7] Feb. 6, 2018
Rabindranath Dutta 155 [8] Nov. 22, 2016

This table was last updated on April 12, 2018. Note that this list is quite incomplete. I will add to it as I find more information. The high rate of invention has been determined from the large number of pending patents, more than the number of granted patents. Many of the names are derived from a list published on the web by Patentdocs, titled "top 100 inventors", which does not list granted patents, but lists patent applications filed over the last few years. Others were derived from the last list of prolific inventors published by the USPTO, back in 1997.[9]. Others have been derived from annual lists published at Alex Bartlett4a (talk) 05:41, 1 January 2013 (UTC)

Misleading Statement about Edison[edit]

The opening statement is flawed in many ways: "For all of the 20th century, Thomas Edison was the most prolific inventor in history, with 1084 U.S. utility patents."

1. To claim that someone is the "most prolific inventor in history" based on his number of U.S. utility patents is ridiculous, as this automatically excludes all those who invented things before the age of U.S. utility patents, or those who used foreign patent systems.

I had already changed it from ranking by U.S. to ranking by (international) patent families. However, for the top ten list, these two numbers are the same, as they all have more utility patents in the US than any other country. Some inventors, such as Béla Barényi have more patent families than they have patents in the U.S. The list now reflects that. AlexBartlett4 (talk) 01:58, 23 June 2011 (UTC)

2. It is well-known that most of Edison's patents were not invented by Edison himself but by members of his large lab; Edison just was the one who put his name on the patents. One could rightly claim that for a long time Edison was the man with the most U.S. utility patents, but this does not automatically make him "the most prolific inventor".

3. A comparatively minor syntax issue: in the early century he had fewer than 1084 U.S. utility patents, while the sentence says otherwise.

To fix this, one could write: "Throughout the 20th century, Thomas Edison held more U.S. utility patents than anybody else; his final count was 1084."

Quiname (talk) 18:35, 22 June 2011 (UTC)

Just for reference, Edison had 736 U.S. patents at the beginning of the 20th century. However, I think it would be pedantic to include this fact - your wording is accurate and sufficient. AlexBartlett4 (talk) 01:58, 23 June 2011 (UTC)

Article is full of Original Research[edit]

The article says: "While the U.S. Patent and Trademark Office (USPTO) is the primary source for U.S. patent information, only patents issued since 1976 can be electronically searched by the inventor's name at the USPTO website.[52] For some of the listed inventors, such as Thomas Edison, all of their patents predate 1976, so other sources must be used. For some inventors, such as Shunpei Yamazaki, some patents predate 1976. The earlier patents must be added to the results of a USPTO search to obtain the complete number."

It seems clear that all of this involved Original Research which is prohibited by Wikipedia - probably the entire article must be deleted, or at least substantially rewritten. Quiname (talk) 18:54, 22 June 2011 (UTC)

I concur (it is WP:SYN which is kind of WP:OR) Ipsign (talk) 13:56, 22 August 2011 (UTC)
Please see reply below regarding WP:SYN AlexBartlett4 (talk) 20:35, 22 August 2011 (UTC)
The article is not Original Research. The list is the concatenation of the five published lists that are included in the article. The patent numbers are then updated on a (semi-regular) basis by reference to other published data - that is, the patents published by the appropriate official government patent offices, such as the USPTO. As these patent offices effectively define what a patent is, there can be no more reliable reference. The note quoted above is just a precaution to those who might rely on an electronic search of the USPTO records, and an explanation of why the references add extra patents to those found in searches of a few of the listed inventors whose inventing careers span 1976. While the USPTO records are essentially complete, they are not electronically searchable searchable by name before 1976. This is why you must add 56 patents to the USPTO search results for Shunpei Yamazaki. These 56 are in the USPTO records, but simply don't show up in an electronic search. Likewise, the patents of Thomas Edison don't show up in an electronic search, but there are plenty of other references to these, as entire volumes have been published on them. I used the convenient Wikipedia List of Edison patents which lists them all, though there are is no shortage of other references to the number of Edison patents. Many of these sources quote 1093 as the number of patents. However, 9 of these 1093 patents are design patents, so are not patents for inventions. AlexBartlett4 (talk) 01:48, 23 June 2011 (UTC)
90% of the current article clearly qualifies as WP:SYN which is a kind of WP:OR, and is prohibited. Creating a ranked list on your own is obvious WP:SYN. The only way I can see to avoid AfD for this article is to remove everything, except for already published lists (and tiny bit of explanation, though even with this you should be very cautious). BTW, personally I admire your efforts, and if you could publish it somewhere, wouldn't object to include reference to your research to Wikipedia, but Wikipedia is not the right place of doing the research itself :-(. Ipsign (talk) 13:56, 22 August 2011 (UTC)
My reading of WP:SYN is that it prohibits the combination of material from multiple sources to reach or imply a conclusion not implicitly stated in the sources. It does not say that editors should not create simple ranked lists of well-referenced data. The creation of ranked lists has been a staple of encyclopedias for centuries - Britannica certainly had them, and the World Book encyclopedia was replete with them. Wikipedia is also. There are many ranked lists which are collated, ranked, and updated within Wikipedia, based on individually sourced data.
Type in "list of largest...", "list of longest..." or similar into the Wikipedia search field, and many will pop up. The first "list of largest" is List of largest power stations in the world, a well researched and up-to-date article which contains 18 ranked lists which are not published elsewhere. Virtually all of the 111 references are to individual articles about specific power stations. An exception is reference 19: "Top 100 largest power stations (retrieved 2010-08-31)" published by Comparing the top 10 of this published industcards list with the top 10 in List of largest power stations in the world, there are significant differences. The rankings of numbers 4 and 5 are reversed, and No 7 (Sayana-Shushenskaya) and No 10 (Futtsu) of the industcards list do not appear in the Wikipedia list. This is because a power plant must be at least 6,000 MW to be included in Wikipedia's top 10 list - Futtsu is 4,534 MW, and Sayana-Shushenskaya, a Russian hydroelectric plant which had ten 640 MW generators, suffered a major accident in 2009 destroying much of its capacity, and is only partly rebuilt, so is currently less than 6,000 MW. Wikipedia's no 9 - Ulchin in Korea, appears as No 14 in the industcards list. Some of the references are to national lists of power stations, which have been collated and ranked in the Wikipedia article. It would appear likely that the Wikipedia list is the most accurate and up-to-date published list available.
To check whether List of largest power stations in the world was just an aberration, I looked further down the set of lists titled: "list of largest...":
List of largest monoliths in the world is a similar collation of data
List of largest California cities by population is based on 2010 US census data, and involved no collation
List of largest buildings in the world is a collation
List of largest optical reflecting telescopes is a collation
List of largest empires is a collation
List of largest enclosed shopping malls in Canada is a collation
This cursory look was certainly not exhaustive, but indicated that it was likely that most ranked lists in Wikipedia are collations of other lists and individual data points.
I think these lists are very useful, and are a significant part of Wikipedia. I certainly use them all the time, and have no objection to the fact that Wikipedia's list is typically more accurate and up-to-date than any other source. I strongly believe that Wikipedia should be allowed to have the most accurate lists that can be created, and not be restricted to slavishly republishing existing lists, even when those lists contain known errors and omissions, or are simply out of date.
Nonetheless, I shall do my best to improve the article with your comments in mind, and shall be vigilant about excluding any WP:SYN or other WP:OR. AlexBartlett4 (talk) 20:35, 22 August 2011 (UTC)
Thanks, it is much better now; I am not sure whether remaining lists are WP:SYN or not, so I won't argue (as long as you don't introduce ranking into them - which would be promoting a point which wasn't made by original sources, making them WP:SYN). I will also perform further minor clean-up. Ipsign (talk) 06:08, 24 August 2011 (UTC)

I am afraid the first table is pure original research. It cannot stay in the article. Querying specific databases based on specific spellings (of inventor's names) directly results from your personal choices and your original research. --Edcolins (talk) 22:06, 7 October 2011 (UTC)

When I first read WP:NOR I assumed that it's purpose was to prevent Wikipedia being used to publish crackpot theories, scuttlebutt, or politically motivated rants. Here is the first paragraph of Wikipedia's prohibition on original research:
Wikipedia articles must not contain original research. The term "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas—for which no reliable, published source exists.[10] This includes any analysis or synthesis of published material that serves to advance a position not advanced by the sources. To demonstrate that you are not adding OR, you must be able to cite reliable, published sources that are directly related to the topic of the article, and directly support the material as presented.
The article: Wikipedia:What Wikipedia is not contains the following: "Per our policy on original research, please do not use Wikipedia for any of the following:
  1. Primary (original) research, such as proposing theories and solutions, original ideas, defining terms, coining new words, etc. ...(continues)"
List of prolific inventors is just a list. I don't see how it proposes a theory or solution, has any original ideas, defines any terms, or coins any new words.
I did put quite a bit of effort in to "fact check" every entry in the table. All of the names initially came from published lists (which were all included in the article, though some have since been deleted). As the patent numbers are slowly and constantly changing, I periodically update the patent numbers using searches at the two most reliable publicly available sources that I am aware of - the USPTO database (surely the source of sources for US patents) and Inpadoc the international patent document database established by the World Intellectual Property Organization (WIPO) and maintained by the European Patent Office (EPO). Surely, as the official government patent databases, these count as reliable, published sources.
Given that the sources are probably acceptable, there is the issue of "Querying specific databases based on specific spellings (of inventor's names)". In the database queries, I have consistently used used the form: ("Lastname, Firstname" or "Firstname, Lastname"). Although the USPTO uses "Lastname, Firstname" there is a small percentage of errors (substantially below 1%) where the names have been accidentally reversed at the USPTO, and the patents are therefore filed incorrectly. I include this reversal in the search. I have checked for spelling variations, and in general, they represent a small error at the USPTO. For example, "Silverbrook" appears in around twenty variations - Sliverbrook, Silverbook, Siverbrook, etc. However, as there are no more than one or two patents under each variation, I thought that these misspellings did not materially alter the relative patent counts. The exception to this is for "Shunpei Yamazaki", where I also include the alternative transliteration of "Shumpei Yamazaki". Yamazaki has 20 patents at the USPTO listed under "Shumpei".
I have done quite extensive checking to ensure that the inventions listed under a particular name are actually by the one inventor, and not by two or more inventors of the same name. In September 2011 I cleaned up the list, as there were 6 entries that were, in fact, by multiple people of the same name. These are listed in the section below: "Common names". I deleted these, as the most prolific of any of the people with the same name did not reach the cut-off for the "top 40" list. I am certain that all of the people that remain in the list now are individuals. Now, this may have been "original research" - but I actually think of it as being "fact checking".
Alternatively, if it really is "original research", maybe there are kinds of original research that Wikipedia really should allow. I note that Wikipedia has a policy Wikipedia:Ignore all rules. There is a nice quote on the page Wikipedia:What "Ignore all rules" means:

By all means break the rules, and break them beautifully, deliberately and well. That is one of the ends for which they exist.

Inventions and patents are an area that is not well reported in the media. Normally, Time Magazine would be considered a good source to quote in Wikipedia. The article published on December 4, 2000, by Time - "Man-Made Marvels" - was a list of what Time claimed to be the "TOP FIVE INVENTORS":
Rank Inventor U.S. Patents
1 Thomas Edison 1,093
2 Melvin De Groote 925
3 Francis H. Richards 894
4 Elihu Thomson 696
5 Jerome Lemelson 554
As usual, this list used patent counts as the metric for determining the top five inventors. However, the list omitted five inventors with more patents than the last inventor ion the list (Jerome Lemelson). When the Time article was published, George Albert Lyon had 993 U.S. patents, John F. O'Connor with 949 U.S. patents, Carleton Ellis had 753 U.S. patents, Shunpei Yamazaki had 745 U.S. patents, and Béla Barényi had 595 German patents. These errors are readily verifiable.
If Wikipedia is constrained to do no more than republish error filled and out-of-date articles that have appeared in other media, I think that would be a terrible shame. AlexBartlett4 (talk) 08:08, 22 October 2011 (UTC)

Two sections[edit]

The list is in two sections: the top ten, and the subsequent list which is effectively the top 50 (Inventors with more than 400 utility patent families worldwide). The lists are formed on the same basis, so the top ten list is really a subset of the top 50 list. The difference is the level of detail in each list - the first list has more detail (total patents of all types, dates, and areas of invention. I am trying to collect this information for the others in the list, but it can be hard to find (I recently added countries). When the second list has all of the details of the first list, then I propose that the lists be merged into a single list. In the meantime, if anyone has any such information, can you please add it below. AlexBartlett4 (talk) 02:54, 26 September 2011 (UTC)

The two sections have now been merged into a single section, containing a "top 40" list. There were 46 in the previous second list, but some of these were deleted as they were actually a number of inventors with the same name, none of whom quite reached the 'Top 40" list (see below) AlexBartlett4 (talk) 08:11, 22 October 2011 (UTC)
It is now no-longer a top 40 list, but a list of inventors with more than 400 patent families. Note that the cut-off of 400 is entirely arbitrary - I will try to extend it as I find more information. AlexBartlett4 (talk) 12:13, 27 November 2011 (UTC)
The threshold for inclusion is now reduced to 300 patent families. AlexBartlett4 (talk) 04:04, 18 March 2012 (UTC)
The threshold for inclusion will be reduced to 200 patent families once the INPADOC source feed is automated like the USPTO source feed. This will not add many inventors to the list. Keithrwalker (talk) 23:07, 15 March 2015 (UTC)
We might move the annual Top 10 lists to a different page or remove them because recent years are no longer published (at least I can't find them) and the Top 10 lists have not been entirely accurate. However, I would appreciate comment, especially from AlexBartlett or anyone who has used the article, as to whether this might reduce article value. For example, is there value in learning that a certain inventor had (or "might have had") the most patents issued in 2011? Keithrwalker (talk) 23:07, 15 March 2015 (UTC)

Common names[edit]

I have been checking through the list to ensure that all of the patents attributed to a person are to the same person, that is, they are not to more than one person with the same name. Generally, the list has not had this problem, with the exception of some of the Japanese inventors. Certain personal names, such as Hiroshi and Takashi, are very common in Japan. Likewise, certain family names such as Tanaka, Suzuki, Inoue, and Sato are also very common.

  • At least 20 different inventors with the name Takashi Suzuki were responsible for the 636 patents listed - the highest number to any one inventor with this name was 58.
  • At least 25 different inventors with the name Hiroshi Suzuki were responsible for the 572 patents listed - the highest number to any one inventor with this name was 73.
  • At least 20 different inventors with the name Hiroshi Tanaka were responsible for the 591 patents listed - the highest number to any one inventor with this name was 52.
  • At least 25 different inventors with the name Hiroshi Watanabe were responsible for the 570 patents listed - the highest number to any one inventor with this name was 96.
  • At least 20 different inventors with the name Hiroshi Sato were responsible for the 515 patents listed - the highest number to any one inventor with this name was 144.
  • At least 20 different inventors with the name Hiroshi Inoue were responsible for the 491 patents listed - the highest number to any one inventor with this name was 108.

The other Japanese inventors in the list are individuals. AlexBartlett4 (talk) 03:19, 30 September 2011 (UTC)

Thomas J. Kennedy III[edit]

Section added by Keithrwalker (talk) 19:08, 11 February 2016 (UTC)

On February 2, 2016, Tjkennedy3 added inventor Thomas J. Kennedy to the article with an atypical USPTO query. Sometimes atypical complexity is required to ensure multiple inventors are not included in a single query. The query included a partial inventor name of kennedy$ (the $ is a wildcard that will pick up all generational suffixes, first names, middle names, and middle initials). The query also limited results to two cities, wilbraham and chicopee, so as not to return every Kennedy. Given the common name and how the USPTO search works, I investigated.

The USPTO query of an inventor name and city queries the entire inventor list field, and so a query of an inventor name and city will include results where the inventor's name is matched, but where the queried city matches a co-inventor of the patent and not the matched inventor. So if one queries for an inventor Kennedy from Wilbraham, one will inadvertently retrieve a patent that has an inventor Kennedy from Boston with a co-inventor Dave from Wilbraham. In fact, the query did just that, pulling in patent 5,915,453, which has an inventor named Paul Kennedy and a different inventor from Chicopee. By correcting the query, Thomas J. Kennedy III has 199 patents, which is at present below the 200 threshold. Therefore, Thomas J. Kennedy will be moved to the watch list until he is issued one more patent, which is likely to be soon.

I refined the query by making the inventor name more precise: (kennedy-thomas$ or kennedy-III-thomas$ or kennedy-II-thomas$). There is a single patent that incorrectly gives him a generational suffix of II. The inventor's city, assigned company, co-inventors, and technology all demonstrate this was a filing mistake and not a different inventor. Here is the correct URL:$+or+kennedy-III-thomas$+or+kennedy-II-thomas$%29+and+ic%2F%28wilbraham+or+chicopee%29

The refined query is still imperfect and will need to be watched. For example, Thomas J. Kennedy III might move to a different city or inadvertently be listed with an incorrect city name (which happens due to typos, use of a corporate headquarter city, or other filing mistake). Further, we have the potential, albeit reduced now, that there will be a Thomas Kennedy who is not this Thomas J. Kennedy III who is included purely because one of the inventors listed is from one of the queried cities. It is generally easy to spot such anomalies by investigating the subject matter, assigned company, and co-inventors.


The headings for the first table are not defined. Updating Stanford Ovshinsky's entry (because of his death) led me to question the meaning of the "Years" column. Is this their lifetime, the first and last patent (application, grant, or ...?), or something else? Any of them, other than their lifetime, are pretty hard to verify/cite. —[AlanM1(talk)]— 07:37, 19 October 2012 (UTC)

  • The definitions for the headings of the first table appear at the bottom of the table. If they were at the top, then the whole article would appear to be definitions of table headings to anyone with a relatively low-resolution screen. The heading "years" is defined as: "Years: These are the birth and death years of the inventor, where known." I was very sad to hear of Stanford Ovshinsky's death (from your update). He made some fantastic contributions. Alex Bartlett4a (talk) 05:44, 13 November 2012 (UTC)
  • I propose changing the Lived column, which is almost always unknown, to the first and most recent years of patent issuance (labled Active), which is easy to determine except for those who pre-date digital records. Besides allowing a more complete representation, the years a person was active is more relevant for researching prolific inventor characteristics than the years they lived. This might give rise to interesting insight by readers, such as identifying averages and deviations in issuances per inventor per year. More importantly, it will give readers a better clue as to whether the inventor is likely to have many more issue. The column will be labled "Active" and be described as: The first and most recent year in which the inventor received a patent issuance. In cases where an inventor was active prior to digital records, the years lived may be substituted, in which case they will be signified with "(b)" for born and "(d)" for died: for example, 1882(b)-1944(d).Keithrwalker (talk) 22:30, 16 April 2015 (UTC)