Talk:Robot/Archive 4

From Wikipedia, the free encyclopedia

Contents

[edit] Good article candidacy

I just marked this article "failed" as a good article candidate. There are no footnotes (not counting the single one that doesn't function) and very few references to printed works. The list of references seems to be quite random (Tesla, Man Out of Time??), rather than being a representative collection of the literature on the subject. The inline links to external URLs should be replaced with footnotes, and more of the citations should be to "hard" sources; i.e. books or peer-reviewed articles. The "History" section should come before the "Contemporary uses" section, and the History section should include a picture of something historic, rather than a picture of a contemporary toy. The information in the "Robots and Human-Machine interfaces" section is too short and should be incorporated into some other section. And to name the most obvious flaw last, there's a dead picture link. This isn't an article that any editor or editors have devoted concerted attention to. KarlBunker 16:09, 17 December 2006 (UTC)

[edit] Timeline

I just added a simple timeline to the page. Not many items in there yet, and I'm not sure about the structure. If anyone's got a better idea, feel free to replace it.

There are a lot of other items, mentioned in the pages below, that we can add to the timeline. But we need to get references for each one before they are added.

Rocketmagnet 17:33, 5 May 2007 (UTC)

Joseph Engleberger is not an inventor and should not be listed in the timeline as such. George Devol is the only patent holder and Engleberger started the first robot company, Unimation, with him. Bangthedash101 22:50, 18 May 2007 (UTC)
You probably know more about it than me. However, from looking on the web, Joseph Engleberger does seem to be more than a businessman. He seems to be an engineer. In fact, Devol won the Engleberger Robotics Award in 1982 [1]. Can anyone shed and more light on this? Rocketmagnet 00:13, 18 May 2007 (UTC)

Joe Engleberger holds an engineering degree from Columbia and is very intelligent, but did not invent any part of the Unimate and in fact did not even know Devol at the time Devol was applying for the first robotics patents. The only inventor is George C. Devol who held a variety of patents before applying for the ones forming the basis for the Unimate. If you need more confirmation or information on robots, read Isaac Asimov's book "Robots, Machines In Man's Image" Harmony House 1985.

Also, the inclusion of Tesla on the timeline is erroneous, as a radio controlled boat doesn't meet any of the terms set forth as the definition for robotics (e.g. neither autonomous nor programmable). As such, I have deleted all mention of Tesla from this page. If someone wants to make a page about the origins of radio controls, then they should.

The problem is this: there are many devices which are not programmable (from the user's point of view) or autonomous, but are still referred to as robots. For example iRobot's Packbot, or the Foster-Miller Talon. They are basically remotely controlled vehicles with no smarts of their own. If Tesla's robot boat is to be removed, then those other robots should be removed too. Rocketmagnet 12:54, 7 August 2007 (UTC)
As the article tries to explain, there is no hard and fast definition of a robot. Rather, there are machines which more or fewer people consider to be robots. I agree that a robot boat is way on the lower end of the spectrum of what I consider to be a robot, but I don't agree that it's solidly non-robot. I really don't want to start an edit war on this subject, and would like to hear from other people to know their opinions. Rocketmagnet 12:54, 7 August 2007 (UTC)
Bangthedash101, I am keen to resolve this issue of Tesla's boat, since that directly affects the article. I believe that the article should take an inclusive attitude towards robots, accepting that there is no hard definition. I respectfully ask you to replace the paragraph about Tesla's boat. Rocketmagnet 18:53, 7 August 2007 (UTC)

Under any definition, Tesla's boat is not a robot because it is neither programmable nor autonomous nor can it move in three axes. I understand the fascination with Tesla and am a great admirer, as is Mr. Devol, but what Tesla built was a radio controlled boat and in no way can be properly considered a robot, as it fails every criteria. (Tesla's contributions to the production of electricity, particularly alternating current, are fundamental and cannot be overstated) I appreciate your relucatance to limit the definition of a robot, but there must be some minimum criteria. In my opinion, the ISO standard sets forth that minimum criteria when they say that a robot is "An automatically controlled, reprogrammable, multipurpose, manipulator programmable in three or more axes" Tesla's boat was not automatically controlled, was neither reprogammable nor programmable in the first instance, did not manipulate anything, and only moved in two axes. With all due respect, I must stand by my original position.

When you say "under any definition", you are only referring to the ISO definition. The problem with the ISO definition is that it excludes a great many machines which are widely considered robots, which is why the Wikipedia robots article does not use that as its definition. The use of the ISO definition has already been discussed here (please see the archives). The ISO definition really only applies to industrial robots, and not to the wider world of robots and so is totally inappropriate. So please do not enforce that definition on the robots of this article. I suspect that one reason you are keen to remove Tesla's robot boat is to secure Devol's position as inventor of first "real" robot. If this debate continues much longer, I will have to ask for mediation from other editors. Rocketmagnet 23:34, 7 August 2007 (UTC)

I have read the archives and they don't add anything to this discussion. The point of an encyclopedia is to provide objective facts to the knowledge-seeking public. Not to appease everyone with an opinion. My point is that the definition of "robot" needs a minimum starting point and the ISO defition is exactly that. It is not some science-fiction fanatic's Bladerunner idea of what makes a robot, nor is it some historical relativist's idea of any machine that is controlled by a human at a distance. It is an internationally accepted minimum standard. I understand your sense of "ownership" with this entry, but it is misplaced. While you deserve credit for starting a very important entry and have done a terrific job so far, you do not "own" the definition.

To correct your last posting, Telsa did not invent a "robot boat", he invented a radio-controlled boat. By your definition, the radio itself is a robot because it actuates a mecahnical speaker at a distance. That is nonsense. Why are you so "keen" to have Tesla in the timeline?

To give credit that is not due is just as bad as denying credit that is due.

All references to machines that do not meet the ISO minimum standard should be deleted.

While I am on the subject, the inclusion of the first "deaths by robot" are pandering and inappropriate in the context of the historical timeline. When a robot "intentionally" kills a human, then it deserves an entry. Until that time, the "deaths by robot" should at least be separated from the historical timeline.

Please request mediation from other editors post haste as you have obviously dug in your heels. Bangthedash101 04:51, 8 August 2007 (UTC)

Can't you see that the ISO definition is designed to apply to industrial robots only? It does not apply to the vast world of robots out there. If this article were to enforce the ISO definition, we would have to get rid of all robots with only one or two axes, e.g. the Ball bot, some swimming robots, some nanobots, many swarm robots, many balancing robots, 1DOF hopping robots and others. We would have to get rid of all robots which were not manipulators, e.g. robot cars, robot snakes, flying robots, walking robots, kismet, and very very many others. Also, all robots which are not multipurpose, e.g. the great majority of research robots which are made for one purpose only. In fact, we would have to get rid of all robots which are not for industrial automation applications, as the ISO definition states: "...for use in industrial automation applications.". We would be left with only industrial robot arms (which some people argue are not robots either). So, please, think about what you are saying. Rocketmagnet 10:32, 8 August 2007 (UTC)
Personally, I think that the boat is right at the very lowest end of what could be considered a robot. The only reason I didn't delete it myself was because then all telerobots would have to be removed too. A case would have to be made that other telerobots were robots, and this one wasn't. Rocketmagnet 13:22, 8 August 2007 (UTC)

Correct me if I'm wrong, but Tesla's radio-controlled boat fails to meet any part of any definition of robot, save for the fact that is it artificially created. As such, his inclusion in the historical timeline is ridiculous. At the very very least, a robot must have autonomous motion, and be either teachable/programmable or responsive to its environment. Can't you at least agree on that?Bangthedash101 16:25, 8 August 2007 (UTC)

I almost agree with you. Whether it's a robot is very debatable IMHO. I would actually agree with you were it not for the fact that there are many teleoperated robots which are not at all autonomous. For example, the da Vinci surgical robot, the Packbot and the Foster-Miller_TALON. None of these are autonomous, but are widely referred to as robots. Should all of these be removed from the article? Rocketmagnet 17:42, 8 August 2007 (UTC)
Furthermore, Tesla's boat does meet the criteria for the Wikipedia definition of robot: it conveys a sense of agency. Granted, by todays standards it's a weak sense of agency, it barely makes it. But by the standards of the day, I reckon would have been really quite spooky. And, AFAIK Tesla imagined this as the first of many machines which could do things for themselves. Interestingly, there are people who do not consider the PUMA arm to be a true robot, because in their particular definition, a robot must sense and respond to the environment, rather than carrying out a sequence of commands. IMHO, the sense of agency conveyed by a machine is critical to people deciding to call it a robot. As Engleberger supposedly said: "I can't define a robot, but I know one when I see one." Rocketmagnet 17:42, 8 August 2007 (UTC)

How does watching someone move controls for a radio controlled boat "convey a sense of agency"? The fact that he probably disguised his use of the controls does not change the fact that a radio controlled boat, when seen in total, cannot convey a sense of agency.

I have no idea what you mean by IMHO and AFAIK.

IMHO means "In my humble opinion". AFAIK means "As far as I know". Rocketmagnet 23:42, 8 August 2007 (UTC)

If you want to include "teleoperation" in this article, and Tesla's obvious fundamental contribution to that field, then it should be in a separate section with the disclaimer that the term "telerobot" is a misnomer.

Telerobot is not a misnomer according to a great many people.

You claim that "Tesla imagined this as the first of many machines which could do things for themselves." Upon what authority is that statement based? It seems very suspect given the fact that the radio controlled boat could not do anything by itself. The Cult of Tesla wishes to overcompensate for his lack of recognition in his lifetime by expanding his sphere of influence well beyond what is actually was. I laughed out loud when I read an article about his boat last night which claimed:

"When a New York Times writer suggested that Tesla could make the boat submerge and carry dynamite as a weapon of war, the inventor himself exploded. Tesla quickly corrected the reporter: "You do not see there a wireless torpedo, you see there the first of a race of robots, mechanical men which will do the laborious work of the human race."

http://www.pbs.org/tesla/ll/ll_robots.html

The word "robot" was not even coined until at least 22 years after this supposed conversation, and certainly was not well known until much later. It makes a more interesting story but is obviously BS.

Whether a robot "must sense and respond to the environment, rather than carrying out a sequence of commands" really misses the point. There needs to be a minimum standard before a particular machine can even be considered, and Tesla's boat (or any manually operated machine) cannot even meet this threshold burden.

It is worth noting that Devol also obtained patents on visual and tactile sensors for robots. Where these fall in the timeline, I am uncertain but will research when I get some time.

The bottom line is that any automatically controlled, reprogrammable, multipurpose, manipulator programmable in three or more axes is internationally recognized as a robot, regardless of how the Wikipedia contributors feel about it.Bangthedash101 20:55, 8 August 2007 (UTC)

I cannot claim to be a member of such a cult. I know next to nothing about Tesla and have no feelings about him or his place in history. What I am defending here is the large number of machines which people generally consider to be robots, but do not fall under the ISO definition.
  • Is ballbot a robot?
  • Is ASIMO a robot?
  • Is the Epson micro helecopter a robot? [2]
  • Is this mechanical snake a robot? [3]
  • Is this skating machine a robot? [4]
  • Is Kismet a robot?
  • Are these swarming machines robots? [5]
  • Is a 2 axis programmable manipulator a robot?
  • Are molecubes robots?
  • Is Dexter a robot? [6]
  • Is the daVinci surgical 'robot' a robot?
  • Is the Roomba a robot?
  • Is a cruise missile a robot?
Think carefully before you answer. None of these machines are robots according to the ISO definition. But they are all refered to as robots by their inventors, and other people in the robotics community (of which I am an active member). Ballbot and the swarmers are only 2 DOF. ASIMO is not for industrial automation applications, and is a pretty rubbish manipulator. At one point it was even teleoperated, it still might be for some demos. A cruise missile, the snake, the skating machine, Kismet and Roomba are not a manipulators. daVinci is teleoperated. etc. Again, think carefully, because it's your opinion against most of the robot research industry. Why not try telling someone that ASIMO isn't a robot? See if they take you seriously. Rocketmagnet 23:42, 8 August 2007 (UTC)
The ISO definition is not a 'minimum' standard. It does not sit at some particular place on the spectrum of robots and divide machines into robots and non-robots. What you must ask is why did the ISO choose that definition of robot? It was chosen to keep track of the number of 'robots' are being used in industrial settings [7]. They were not interested in other kinds of robots used for entertainment, domestic chores, research etc.. The ISO definition is only supposed to apply to industrial robots, and it is totally inappropriate for other types of robot. Rocketmagnet 23:42, 8 August 2007 (UTC)
Furthermore, Wikipedia articles must take a neutral point of view, which I have taken great pains to do, explaining that there is really no correct definition of robot, and everyone has their own definition. What we can say is that there are machines which everyone calls robots, machines which nobody calls robots, and lots inbetween. I went to the trouble to state three definitions of robots, one from the industrial sector, one from a famous roboticist, and one from a dictionary. If the article takes a hard line on the ISO definition, it totally fails in its duty to be neutral in its point of view. Rocketmagnet 23:47, 8 August 2007 (UTC)
I have requested mediation on this dispute. The relevant page is here [8]. On that page I have tried to present both arguments very briefly. Please go to that page and edit the summary of the dispute so that you agree with your side of it. Rocketmagnet 00:10, 9 August 2007 (UTC)
The thing is, Bang, I agree with you about the boat. I wouldn't call it a robot either. But the Wikipedia is not a place to put my personal opinion, but rather a place to record a neutral point of view. The editors of this article have to accept that, whatever their definition of 'robot', they are probably the only one who uses that definition. The fact that teleoperated machines are widely referred to as robots, by both experts and laymen is something I, we, have to accept. So, if teleoperated bomb disposal mobile manipulator gets called a robot, we have to accept that a teleoperated floating mobile machine without a manipulator gets called one too. Sorry.Rocketmagnet 20:21, 9 August 2007 (UTC)

Why don't you start with providing some authority for your contention that "teleoperated machines are widely referred to as robots, by both experts and laymen" rather than purporting to speak on their behalf. I can't imagine any robotics expert referring to a radio-controlled vehicle as a robot. In fact, I can't imagine calling anyone a "robotics expert" who misses such an obvious distinction!

As far as what "laypeople" consider robots, laypeople are the ones who are likely coming to the robot page seeking information from the experts and it does them a great disservice to supply erroneous information.

I am sorry I sound agitated, but this is getting ridiculous. Someone has to put their foot down, and I am not about to allow the Cult of Tesla usurp credit for the first "truly modern robot" based on a radio-controlled manually-operated non-programmable non-teachable toy boat.

Again, I am a huge admirer of Tesla, and in fact Mr. Devol and I drove to Colorado Springs (near where I grew up) a couple years ago looking for the Tesla museum only to find it had been closed due to lack on interest and the displays were in storage at the local college.

Let's put this thing to rest. You say you agree with me, but obviously are more worried about offending the sensibilities of laypeople. Editors make tough decisions all the time and one needs to be made here. Hopefully the mediators will agree with both of us and remove Tesla and "telerobots" from the page, save for a link and perhaps a disclaimer. ~ Bangthedash101 4.228.21.35 22:57, 9 August 2007 (UTC)

Bangthedash101, are you still here? The mediation is happening at the bottom of the page, under the heading The Issue. Rocketmagnet 14:39, 12 August 2007 (UTC)

[edit] Father of Robotics

It seems that both Devol and Engleberger are the "Father of Robotics". Did robotics have two dads? They must have been liberals. Rocketmagnet 00:13, 18 May 2007 (UTC)

George Devol is anything but a liberal and would be profoundly offended by the suggestion. To answer your question, Engleberger promoted himself as the "father of robotics" but never held any patents for robotics. The patents on which the first Unimate was based on were held by Devol who applied for them years before meeting Engleberger.

There is only one "father of robotics" and that is George C. Devol. How do I know? He is my grandfather and is alive and well at age 95. He and Joe Engleberger are still great friends.

I am slightly concerned that you may not be objective here, as you have a strong emotional connection to George Devol. In my opinion, there are several people who contributed significantly to robotics (as to almost any field) and it's tricky to pin down one as being the only one. Any chance of hearing from George and Joe? Rocketmagnet 12:54, 7 August 2007 (UTC)

If you have any specific questions for Mr. Devol, I would be happy to ask him, as I talk to him about every week. I am not sure how often he and Joe get together, but I think it is probably a few times a year. If you do not feel I am objective, please refer to the Isaac Asimov book I referenced above. As stated in the History section, no other patents were cited against Mr. Devol's original robotics patent. If you have any experience with patents, you know that is extremely rare. As such, in this unique case, it is appropriate to give credit for the first robotics patents to Mr. Devol.

Joe Engleberger is certainly a visionary and persuasive promoter who convinced others to invest in robotics when Mr. Devol had hit the wall. As stated in the March 21, 1982 New York Times article, "He Brought The Robot To Life" by Barnaby Feder: http://query.nytimes.com/gst/fullpage.html?res=9502E0DC1639F932A15750C0A964948260

"Actually, it was George C. Devol, not Mr. Engelberger, who developed and patented the basic technology on which the industry is founded. But since meeting Mr. Devol in 1956, Mr. Engelberger has preached the gospel that smart machines were the key to getting people out of dangerous or tedious production jobs and a key to improving productivity."

Further, as Mr. Engleberger states: "GEORGE DEVOL was unable to restrain himself from spilling the whole dream out, which scared most businessmen off, said Mr. Engelberger during an interview last week at Unimation's headquarters. I kept myself from talking about some of the things that have happened, which he envisioned."

In other words, Mr. Devol saw the big picture long before anyone else, but as anyone who has talked to him knows, he is very blunt and not particularly patient or forgiving, and if you don't agree with him, you had better know what you are talking about and be able to keep up. This scared off potential investors.

Joe Engleberger deserves tremendous credit as a man who advanced the "cause" of robotics, and preached the "gospel" of robotics while most people considered it science fiction.

Mr. Devol deserves sole credit as the inventor, and the patents provide evidence of this.

I invite anyone who wants to ask questions for Mr. Devol in this forum. I can get them answered in short order, as I will check back periodically. I can try to get answer to questions for Mr. Engleberger as well, but cannot promise anything.

Well, what would be brilliant is to have a picture of George and/or Joe for the article. Rocketmagnet 18:53, 7 August 2007 (UTC)

[edit] Mediator

I want to establish myself first, so here goes:

Hi, my user name is Sr13 (will likely be Singularity in the next couple of days because I am filling in a usurpation request, so don't get those confused) and I will be the MedCab mediator for this dispute. The discussion will take place on this page, below this message.

I only ask of one thing: During discussion, please do not be uncivil, as it may slow down the discussion and prevent an agreement between the two parties. I will mediate this dispute as best I can. Sr13 is almost Singularity 21:27, 9 August 2007 (UTC)

Hi Sr13. Thanks for taking the time to mediate this dispute, which, as you can see, rages on up there. So, how does this work? Do we take it in turns, or do we all join in? Rocketmagnet 23:49, 9 August 2007 (UTC)
Anyone can join in the discussion. Unlike WP:MEDCOM and WP:ARBCOM, this is purely made to be a informal way of resolving dispute. Singularity 20:50, 10 August 2007 (UTC)
I was just wondering how you mediate. Do the disputing parties just keep on arguing, and then you jump in at some point and say "easy guys"? How will it be resolved finally? Rocketmagnet 21:15, 10 August 2007 (UTC)
In the end, hopefully there will be some sort of compromise between the two parties. Singularity 21:53, 10 August 2007 (UTC)

[edit] The issue

I think both users are informed now (both parties edited the MedCab page), so let's begin.

Since I don't want to repeat what has been discussed already, I'll summarize what has happened so we can work towards a compromise.

  • The issue itself is the definition of a "Robot".
    • Bangthedash101 says: "I do not feel the ISO standard is the minimum, but as I stated in my penultimate posting, at the very least, a robot must have autonomous motion, and be either teachable/programmable or responsive to its environment."
    • Rocketmagnet says: As the article tries to explain, there is no hard and fast definition of a robot. Rather, there are machines which more or fewer people consider to be robots...The problem is this: there are many devices which are not programmable (from the user's point of view) or autonomous, but are still referred to as robots. Singularity 21:24, 10 August 2007 (UTC)
  • There are several possible solutions.
    • 1: We use the ISO definition only (disagreed with by both parties)
    • 2: We use the ISO definition, but as a minimum requirement (Bangthedash101's stance)
    • 3: We use the current definition in the article
    • 4: We use no definition, and include all machines which are referred to as robots by their creators or reporters (Rocketmagnet's stance).

Objections? Comments? Singularity 21:53, 10 August 2007 (UTC)

1: The ISO definition is not intended to be a definition of a robot in the general case. It is simply used when counting the number of industrial robots in each country. Such official definitions (eg the JIRA and RIA) are intended for legal purposes. e.g. when deciding which safety laws apply to particular machines. So we should not apply the ISO definition to all robots in the article.

2: BTD did say "All references to machines that do not meet the ISO minimum standard should be deleted." but also said "I do not feel the ISO standard is the minimum..." So I'm not exactly sure what his stance is. I'm not sure what the difference is between 1: and 2:.

3&4: The "definition" in the article is supposed to be less a definition, and more a description of the types of machines which get called robots. Also, the article should not just include the robots, but also state that some machines are called robots by nearly everybody, and others (like telerobots) are often debated. The article already does this.


Now to tackle BTD's request: "Why don't you start with providing some authority for your contention that "teleoperated machines are widely referred to as robots, by both experts and laymen" rather than purporting to speak on their behalf. I can't imagine any robotics expert referring to a radio-controlled vehicle as a robot. In fact, I can't imagine calling anyone a 'robotics expert' who misses such an obvious distinction!"

I refer you to three examples of teleoperated machines which are widely referred to as robots: iRobot's Packbot, the Foster-Miller TALON, and the Da Vinci Surgical System. The fact that these are widely referred to as robots means that, in the interests of NPOV, the article should also call them robots. It is not for the editors to impose their views on the article. Rocketmagnet 23:24, 10 August 2007 (UTC)

Let's wait for BTD to join in. Singularity 23:51, 10 August 2007 (UTC)

May I add a comment here? I've been in industrial robotics for 25 years and I contribute to a few robot articles. I do believe the ISO definition only applies to industrial robots. The definition includes the word 'manipulator' and a robotic vehicle might not even have one. On the second point it does annoy many to see radio controlled vehicles described as robots. I do believe there must be a high degree of autonomy. It is a fact that many things in life are misnamed by the public in general and in some cases the wrong name is accepted and in some it is not. For example laptop computers were replaced by 'notebook' computers but people keep right on calling them laptops. Who is right? In this case I think we should include autonomous in the definition of robot. Just my 2 cents! Robotics1 19:36, 12 August 2007 (UTC)

Thanks for your input; we're still waiting for BTD to join so we can continue to discuss. Singularity 22:33, 12 August 2007 (UTC)
I said "by both experts and laymen". People seem to have forgotten that I said experts. Robotics1, please refer to the iRobot Packbot, the Foster-Miller TALON, and the daVinci surgical system, which are all teleoperated and all referred to by their manufacturers as robots. ASIMO used to be teleoperated. Would anyone declare ASIMO as a non-robot. Rocketmagnet 12:17, 13 August 2007 (UTC)
Bangthedash101 where are you ? Rocketmagnet 12:17, 13 August 2007 (UTC)

Well yes, I would. It *looks* like a robot but it isn't one. (sorry *wasn't* one when radio controlled.) I think some manufacturers use the term to enhance their sales quite frankly. So it conforms to the laymen's definition of robot rather than the purist one. Maybe that's OK. That's the nature of language. Now I ask you, is a radio controlled toy robot (or Asimo) subject to the technology of robotics and should it conform to the 3 laws of robotics? There is another word for 'robots' that are only radio controlled - automaton. The Oxford dictionary lists 1 a moving mechanical device resembling a human being. Just like Asimo. But then 2 a machine which operates according to coded instructions. Just like a robot. For more bad news there is a Wikipedia page on automaton. That describes an automaton the way I would describe it but then includes robot. Personally I think we're in the s**t. Robotics1 16:10, 13 August 2007 (UTC) Robotics1 16:10, 13 August 2007 (UTC)

I don't think we're in the s**t. The problem, as the article tries to explain is that there really is no definition of robot, and everybody's personal definition is a little different. Because of this, the article should not try to impose one particular definition, but take a neutral point of view, accepting and explaining that there are shades of grey. This is what it tries to do now. Which was the first "real" robot? Dunno, depends what you mean by "robot". This is the same way it is in the battle for first real computer.
Actually, looking at the various machines in the timeline, I would think that Al-Jazari's automata satisfy the conditions for first robot better than Tesla's or Devol's. They are automatic, and possibly programmable (maybe by changing cams inside). Another candidate would be Heron of Alexandria's robot which could be programmed to travel across the floor in different and predictable patterns by placing pegs and winding string [9]. Rocketmagnet 17:08, 13 August 2007 (UTC)
Perhaps what I'll do is to add a paragraph entitled: "Candidates for first real robot", and discuss several machines, each of which introduced a new feature, bringing them further up the spectrum. I'll refer to Devol's robot as "first industrial robot", rather than first real robot.

Actually I wish I hadn't written that about Automata. A radio controlled "robot" is not an automaton. Error 101. I like the idea of first real robot and Devol's first industrial robot. That would be a wonderful addition. Of course industrial robots are essentially an example of manufacturer's romance because they are not robots either as Asimov and Capek envisioned them. They are just the arms of a robot, hence robot arm rather than robot. Then if we are to include RC look-alikes as robots then we need to work on the definition. That is the s* that I was thinking of. Robots as I understand it are programmable, including programmable with the 3 laws. Automata are not programmable as such but when you look at the cams etc. that control many of them those are kind of mechanical programs. Sheesh the more you look at this the worse it gets. Robotics1 19:11, 13 August 2007 (UTC)

No worries, I'll pretend you didn't say it. As for the 3 laws, I really don't think they apply to any robots yet created. Don't forget, they were just a literary tool, used to help generate stories about robots. The 3 laws don't apply to RC cars, industrial robots or ASIMO. The only laws that apply are the laws of physics, and the health and safety laws. If BTD doesn't come back by wednesday, I might just go ahead and write that section about "Candidates for first real robot". Rocketmagnet 19:54, 13 August 2007 (UTC)

Hi, R, the trouble is the literary tool, the romance, is what has led to the reality. If it were not for Asimov and Capek the industrial robot would be called a programmable manipulator, which is what it is, no more. There seems to be several categories of which industrial robots are one, automata are another, RC 'bots' such as those in robot wars would be another, walking robots another. Would addressing the various categories or lay perceptions of robots be useful somewhere? As for the 3 laws they are OSHA / COSH under another name. Such concepts inspire engineers to look forward to possible realities. Robotics1 20:07, 13 August 2007 (UTC)
Rocketmagnet has a point in an earlier post: It's difficult to determine the first robot like determining the first computer. Quoting the computer article:

It is difficult to define any one device as the earliest computer. The very definition of a computer has changed and it is therefore impossible to identify the first computer. Many devices once called "computers" would no longer qualify as such by today's standards.

I'm not really familiar with robot history and whatnot (I'll leave that to you guys), but I think this statement can be somehow applied for robots. We can definitely see that there are many definitions for what a robot can be. Restriction by a rule set in stone here may prevent views from being expressed in the article. The definitions for robot are much different than the definitions of a robot 30 or 40 years ago. Singularity 21:48, 13 August 2007 (UTC)
That's a good point Singularity. People's ideas of what counts as a robot would have changed through time. The same is true with AI. Once an AI problem has been solved, it stops becomming Intelligence, and simply becomes an algorithm, and so people chase a moving goal. As robots become more sophisticated, I reckon that people must demand more and more autonomy from a machine before calling it a robot. I've even heard people say that a machine must have emotions before it's a robot! As I mentioned in the Timeline section (i think), I'm sure that when Tesla first showed his boat, people must have been quite amazed by it, commenting that it seems to move around all by itself. (BTD rightly pointed out that this is total conjecture. But I still reckon it's true.) Now that we all know how RC works, it's not in the slightest bit amazing, and practically falls off the end of the robot spectrum. (They still scare the c**p out of my cat though, who probably believes it's some kind of animal.) Rocketmagnet 22:03, 13 August 2007 (UTC)
I'm not sure about the wisdom of having sharply defined categories because there are an infinite number of gradual steps from one 'category' of robot to another. There would be so many robots which fall on the edges that it would lead to more arguments. I tried to give an idea of the 'robotyness' of various machines in the "Defining characteristics" section, ranking several machines by how many people (estimated) called them robots. One possibility would be to draw this explicitly as a spectrum of robots, trying to rank them by robotyness. Or, you could have a 2D spectrum, with one axis showing how roboty a machine looks or functions (physical agency), and the other axis showing how how autonomously it thinks (mental agency). One corner would be definitely not robot (like a lamp), the other corner would be totally robot, (like the Terminator). Rather than plotting machines as points on the 2D spectrum, they could be shown as fuzzy areas to show that they are not sharply defined. One foreseeable problem with this is that it's not referenceable, and could count as original research (or just made up). Rocketmagnet 22:03, 13 August 2007 (UTC)

Rocketmagnet gave me a link to the mediation page and I have been checking there for a week waiting for the mediation to start. Now it looks as though the issue has pretty much been decided without me.

If you can't agree that robots need to function autonomously, then this is hopeless. If you erase that BRIGHT LINE then Wikipedia can let its readers swim in a stew of retarded relativity for eternity. Now if anyone with a mind that can actually SEE black and white as opposed to "shades of grey in every direction" is still with me, then there is another line just past autonomy. This line is there so that a gyroscope is not considered a robot. That line is programmability. Fine, you say, a kitting loom is programmable and functions autonomously, so now it is a robot. Well, I agree that is a problem, so there needs to be yet another line. A robot needs to be able to move independently or manipulate something, and furthermore, in three dimensions (otherwise the kitting loom still qualifies). Now we have arrived. You call it an industrial robot. I call it the first truly modern robot because all subsequent modern robots stem from George C. Devol's original patent (against which not a single other patent was cited).

The Cult of Tesla wants to usurp credit from Mr. Devol based on a radio-controlled manually-operated non-programmable non-teachable toy boat that only moved in two dimensions and didn't manipulate anything. The historical relativists want to include every tool since the dawn of time. The science fiction fanatics will not rest until they can take an android on a date.

So fine, you say, see, Bangthedash101 has proven my point that nobody can agree on a definition.

Well, no. I did not prove that point. I have proven that there are a lot of WRONG opinions out there. And yes, laypeople can be WRONG and very often are WRONG and come to resources like Wikipedia to find out the RIGHT answer.

Oh, and don't get me started on advertisers. You are telling me because Irobot and packbot and TALON all call themselves "robots", that the robotics experts should follow THEIR lead?!?!?!

And believe it or not, even "experts" can be wrong. [Further, just because someone calls himself an "expert" does not mean he is.]

My point in entering this discussion and editing the robot page in the first place is that I was disturbed that George C. Devol's credit for inventing the first truly modern robot was being usurped by the Cult of Tesla, claiming that a radio-controlled toy boat spawned an entire industry.

Tesla's boat does not deserve to be called the "first truly modern robot". The Unimate deserves that distincion and Unimation deserves the distinction of being the world's first robot company.

The toy boat and the Unimate are not even comparable, and, as such, do not deserve EQUAL STANDING.

THAT is my point.

I couldn't care less what you consider all the things that predate the Unimate. Yes, many were robots but they were not "modern" robots in the sense that the Unimate was.

The Unimate was a huge innovation because it made the leap from "mechanical" robots to a digitally controlled, teachable robot that could manipulate things in three dimensions. Mr. Devol advised me that it was accurate to within 1/10,000 of an inch!

[as I stated above, if anyone has questions for Mr. Devol, I would be happy to get them answered. In spite of his 95 years, he is sharp as a tack and has vast knowledge of the "halcyon days" of the electronics industry (not just robots)] Feel free to email me at godd@kaom.com. I will post the questions and responses under the "Father of Robotics" section of this page.

Rocketmagnet stated above: "The thing is, Bang, I agree with you about the boat. I wouldn't call it a robot either."

In other words, Rocketmagnet agrees that Tesla's robot does not deserve the credit it was getting. Robotics1 also seems to agree. As far as I can tell, everyone who has chimed in all has the same opinion, but the debate rages on.

As far as the ISO standard, as I stated on the mediation page (where I thought the debate would take place), "I do not feel the ISO standard is the minimum, but as I stated in my penultimate posting, at the very least, a robot must have autonomous motion, and be either teachable/programmable or responsive to its environment. That standard is far broader than the ISO standard on which Rocketmagnet presumes this dispute is based and covers virtually every type of robot listed on the page, save for so-called "telerobots" which are merely radio controlled machines. ~ Bangthedash101 4.228.27.82 03:09, 9 August 2007 (UTC)"

Tesla's "contribution" to robotics (if any) needs to be put in its proper perspective and George C. Devol should be given credit for the first truly modern robot, as proven by his singular patents and his company, the worlds' first robot company, Unimation - neither of which can be disputed.

Say what you will about the mechanial robots that predate the Unimate. Say that the Unimate is "just a robotic arm" and only bi-pedal androids that smile back at you are really robots. But put the Unimate in its proper perspective as the seminal machine that spawned the modern robotics industry. Bangthedash101 05:00, 14 August 2007 (UTC)

I think it is already pretty well established that the Unimate was the first industrial robot. Singularity 09:16, 14 August 2007 (UTC)
Bang, nobody here is saying that your grandfather's contribution to robotics was anything less than extraordinary. More than Tesla's, by a very long way indeed. You are rightly very proud of Mr Devol, but I do think that the family collection is giving you an emotional involvement here. I don't know what this "cult of Tesla" is, but I am not one of them, and am not interested in Tesla in the slightest. Who invented the radio controlled boat is totally irrelevant; it could have been Jack the Ripper for all I care. So, please stop talking about this cult.
When I hear someone saying that they alone know the true definition of robot, and all the other people are wrong, even the manufacturers and experts, I think it sounds more like religious fundamentalism than objective reason. How did you come across this knowledge? Was it revealed to you in a divine vision, or do you have a reference for it that anyone can refer to? If I were to say something like that, you would jump down my throat with: "Upon what authority is that statement based?".
Now, the authority I would like to call is the Japanese Industrial Robot Association (JIRA). They would disagree with you. JIRA recognizes six types of robot, including teleoperated robots (class 1). Here is a link to a PDF which contains a list of the classes. I believe that the Robotics Institute of America (RIA) have different definitions, which do not include teleoperated robots. But to say that nobody calls telerobots robots, or that you are right and they are wrong is just weird. Rocketmagnet 12:46, 14 August 2007 (UTC)

I guess you all have missed my point again.

I have never been anything less than candid about my reasons for wanting Tesla put in his proper perspective. Mr. Devol has no idea that I am involved in this dispute and frankly would be mortified to find out. He would also be beside himself to learn that the Japanese definition of "robot" has pushed aside his contribution in favor of Tesla's. Particularly in light of the fact that the Japanese copied the Unimate to create their robotics industry.

Yes, the "family connection" creates an emotional involvement when you have grown up reading that Joe Engleberger is the "father of robotics" when you know damn well that your grandfather invented the first modern robot and Engleberger has simply promoted himself as the "father" because he was the spokesman (albeit a very effective one and a very nice man as well).

Then some smart-ass named Rocketmagnet, calls your grandfather a "liberal" which would make him furious beyond belief.

The authority on which I base my opinion is that of Mr. Devol who insists that a robot must be programmable (noting that all American manufacturing definitions of robot are in agreement). Obviously this authority is unavailing in light of the Japanese manufacturers wishes to have radio-controlled toys sold as robots, and Rocketmagnet's fervent desire to let everyone in the door who calls his product a robot.

There are no definitions in this world. The sky is grey and the sun is grey.

Bangthedash101 is "weird" because he can see black and white. Rocketmagnet is "normal" because he can see subtle distinctions between machines and recognizes a vast continuum of machines stretching from the first stone tools (which are just as robotic as Tesla's boat) all the way into the future when robots will weep softly at the sight of the grey sun dipping below the horizon.

You win Rocky, you have rewritten history. Now let me hear you scream for Adrian:

http://www.teslasociety.com/ http://www.tesla.org/

Bangthedash101 15:32, 14 August 2007 (UTC)

Hi, Bangthedash101. May I first say that you do yourself no good by expressing such emotion. If I may pour oil on troubled waters: first I read Rocketmagnet's comment about liberals as being entirely lighthearted and not intended to cause offense at all. That you are Devol's grandson is great; pleased to meet you. I for one find it sad that Engleberger seems to have taken all the credit, just as no-one has ever heard of Paul Allen, Bill Gates' partner.
Like it or not people are calling things robots which are not robots and because of that they become robots. I don't think there is much we can do about that but an answer may be to raise 6 categories like the Japanese robot classes. Then we can put tele-operated robots in a class of their own and not confuse them with what we purists would call 'real' robots. Incidentally tele-operated robots in the nuclear industry are usually wire controlled not radio. And alas they call them robots.
Rocketmaget may be right that there is no definition of robot. You are right that there should be. Maybe now is the time to write one.
I didn't know there was another debate on mediation. I think we can come to a consensus here.

Robotics1 16:08, 14 August 2007 (UTC)

Yes, Robotics1. Actually it was a reference to the 1980s sitcom My Two Dads, and how, if robotics had two dads, it would probably have to have been quite a liberal household. It probably got teased at school about it.
"Maybe now is the time to write one." - It is not the job of the Wikipedia to make up definitions. That would count as original research. The only thing we can do is to try to record the state of the world as it is, with all its faults.
Bang, the RIA have no more or less authority to define robot than the JIRA. If there is one person in the world who could claim this right, I guess it would be Carel Kapek. What would he say? I don't know. Maybe his idea only stretched as far as artificial people, ruling out most robots today.
What I might try to do is to re-write the history section, removing all mention of first "real" or "modern" this or that, and simply talk about the significant developments which led up to robots as we know them today. And I'll try to do it in the farest possible way. I'll post it on a scratch page to people can comment on it before it goes live.
Bang, one more thing. I originally put up a picture up Tesla, because his was the only picture I could find on the Wikipedia. I would much rather have put a picture of Devol, as his contribution was far more significant. If you have any pictures of him, especially standing next to his robot, that would be awesome to have in the article. Rocketmagnet 16:54, 14 August 2007 (UTC)
I agree it's not our job to write definitions, but we are allowed to break down what is fact into categories that more clearly describe the facts. Now you said that JIRA had 6 classes. Can we not use those? Could we/you not say: robots fall into 6 distinct classes (subheading 'classes of robots') 1. teleoperated robots, 2. industrial robot arms 3. etc. (no order preferred). I would have thought that would clarify things for the reader without breaking new ground. What do you think? I think what is troubling Bang is that readers will want to know 'what is a robot'. Are we not obliged to tell them? Robotics1 17:24, 14 August 2007 (UTC)

According to the Oxford English Dictionary, "robot" is defined as "One of the mechanical men and women in Capek's [1923] play, hence, a machine (sometimes resembling a human being in appearance) designed to function in place of a living agent, esp. one which carries out a variety of tasks automatically or with a minimum of external impulse."

Frankly, Rocky, you need to read about the history of robotics before you start writing the history of robotics.

Start with Asimov's book I mentioned above. Then please post a link to this "scratch page" you are developing.

This "off-the-cuff" style of creating Wikipedia entries creates big problems because the authors get married to their original postings to the exclusion of others; or in this case, demand equal standing for their original postings no matter how tangential and can enforce those demands because they somehow "own" the page since they were the first to create the entry, and can therefore initiate a "cabal" when opposing viewpoints threaten their worldview.

Your desire to erase any distinction between what is "modern" and what is not is nonsense. Even at the "computer" Wikipedia entry, something even more amorphous than "robot", the authors state "...none of those devices fit the modern definition of a computer because they could not be programmed..."

The history of modern, and I mean Modern with a capital "M", robotics does have a starting point, and that is the Unimate. Is this fact in dispute? If not, then when you are drafting the History of the development of the robot, you need to identify it as such.

Bangthedash101 17:41, 14 August 2007 (UTC)

With respect to the overall definition, Robotics1 has the right idea, but I think using the Japanese definition as a starting point would unfairly give radio-controlled machines "equal standing" when the American definitions don't. As such, when the Japanese definition is set forth, it should be emphasized that their inclusion of "telerobots" is in direct conflict with the accepted American definitions that exclude such devices, and explain why, i.e. that such machines are neither autonomous nor programmable. Bangthedash101 17:47, 14 August 2007 (UTC)

Chaps. It was done a couple of days ago. Check out the robot article now, and you will see a section about the JIRA and RIA definitions of robot. Rocketmagnet 18:19, 14 August 2007 (UTC)
Bang, before I rewrote the first sections of this article, I extensively researched people's definitions of robot. I looked at dictionary definitions (none of which agree exactly, though I think the Oxford one is my personal favorite), many books on robotics at our company, people's discussions on forums about the definition, quotes from well known roboticists, and articles discussing the possibility of finding a definition. Please don't presume that I did no research before I came into this. I admit that the history of robotics (or any kind of history) is not my forte, which is why I barely touched the history section, except to tidy it up.
I am actually surprised that you mentioned that particular definition, as I interpreted it to include Tesla's boat, as it says "minimum" of external impulse, which could include radio based instructions. Rocketmagnet 18:19, 14 August 2007 (UTC)

You certainly are open-minded, Rocky. Did you also see that part that says "one which carries out a variety of tasks", and that other bit which says "designed to function in place of a living agent". I guess not, because Tesla's boat would not fit into the remainder of the definition, now would it?

Also, "a minimum of external impulse" seems to logically exclude any machine that is constantly under the control of a human, doesn't it?

Just so we are clear Rocky, does your company manufacture "telerobots" and call them robots? I laid my cards on the table long ago but you have yet to do so. What is the name of your company? Whose interests are you really advocating? - Bangthedash101 Bangthedash101 18:34, 14 August 2007 (UTC)

Bang stated in an earlier post: "The history of modern, and I mean Modern with a capital "M", robotics does have a starting point, and that is the Unimate. Is this fact in dispute? If not, then when you are drafting the History of the development of the robot, you need to identify it as such." We need to slowly come to a consensus here, so can we all agree that the Unimate will fit the current definition of a modern robot (this may not exclude the fact that it was the first robot)? Singularity 18:52, 14 August 2007 (UTC)


I am not claiming that the Unimate was the first "robot" by any means. I am saying that it was the first Modern robot and spawned an entire industry and therefore should be recognized accordingly, and Tesla should be, at most, a sidenote in the history of robotics.

Rocky, is this your company: http://www.shadowrobot.com/hand/overview.shtml

I find it interesting that your only product, The Shadow Hand, is described as "the closest robot Hand to the human Hand available. It provides 24 movements, allowing a direct mapping from a human to the robot." In other words, this "direct mapping" means that a human is constantly under control of the the "Shadow Hand", just like Tesla's boat, am I right???

Bangthedash101 19:03, 14 August 2007 (UTC)

Let's not try to find out who's advocating what; that would only create extra unnecessary discussion. Singularity 19:08, 14 August 2007 (UTC)

Hey guess what else gang, our good friend Rocky, the man who insists upon a "neutral point of view" also pasted pictures of his company products all over the robot page and even provided links!

You sack of s**t. Take all that self-serving crap off and resign. YOU are the problem.

See: Actuation A robot leg, powered by Air Muscles. Built by The Shadow Robot Company Ltd. (link under photo)

See: Manipulation Shadow Hand, an advanced robot hand system, holding a lightbulb. Touch sensors in the fingertips allow it to apply gentle pressure. (link to a page advertising his product, Shadow Hand under photo. http://en.wikipedia.org/wiki/Shadow_Hand even described by Wikipedia as "This article or section is written like an advertisement. Please help rewrite this article from a neutral point of view per Wikipedia policy. Mark blatant advertising for speedy deletion... ") Another link to his company website in the discussion part as well as two more links under the reference section.

The truth has been told. Now crawl back under your rock.

Bangthedash101 19:15, 14 August 2007 (UTC)


Easy tiger. Firstly, you are wrong, the Shadow Hand is a component of a robot. It's up to the customer whether they use it as part of a fully autonomous robot (some of them are, e.g. Bielefeld University and CMU) or you choose to have it slaved to a data glove. Secondly, you are wrong, "direct mapping" means that for each joint on the human hand, there is a similar joint on the Shadow Hand.
The reason I used two of our own pictures to illustrate the article is simply because they are the easiest for me to come by. I can upload them myself and don't have to ask for copyright. As it happens, I have written numerous e-mails to other robot companies asking for permission to use pictures of their robots on the Wikipedia article. Did you notice that I have also pasted 21 pictures of other people's robots all over the article, and provided links? Am I working for them as well? No, of course not. Is the robot article better for having a good quality image of a robot hand? Yes it is. I used ours because it's a picture I have. If I were to use someone else's, then it would take me ages to get their permission, if I ever manage to get it at all. If you have any better alternative pictures then please, please upload them and paste them all over the article. I used the image of our leg only because it was the best image of air muscles I could find. I would much rather it was just a picture of a muscle. If you have a better picture of an air muscle, please upload that and replace the picture of the leg.
About the Wikipedia page on the Shadow hand, we are actively trying to get some of our customers to edit the article themselves with their own opinions about it. When it was written, I tried to make sure that it contains only facts and no advertising waffle. However, as it stands, it does contain only information from my company, which is a shame, and I hope that one the hand will be successful enough that other people edit the article too, without me having to ask them. Rocketmagnet 19:49, 14 August 2007 (UTC)
Singularity, I don't think anyone here is arguing that the Unimate doesn't fit the definition of robot. Also, can we get some more mediation in here, it's overheating.


It's only overheating because you have been burned. Sure you can claim the photos were the only ones available, but the links. At least five separate links, four of which go directly to your website (calling it "the most advanced robot hand in the world" product of its kind! How neutral is that?!), and the other to a Wikipedia page for Shadow Hand you posted that constitutes a "blatant advertisement" of your product that is earmarked by Wikipedia for "speedy deletion".

You have lost all credibility my friend. You wouldn't know a neutral point of view if it bit you on the ass. You have no business controlling the contents of the robot page or even discussing what constitutes a neutral point of view. Please go away now. Bangthedash101 20:08, 14 August 2007 (UTC)

The fact that you have resorted to an ad hominem argument suggests to me that you have run out of decent arguments in favour of your original claim about the definition of robot.
Take a breath and think for a moment. If my motivation was simply to advertise the hand, what do you think I would have done? I would have put up pictures and links, and some gushing comments about it, and left it at that. In fact what I did was to spend tens or hundreds of hours researching and writing and finding references and e-mailing companies, proofreading and asking for advice and opinions, on a whole bunch of different robots and technologies in order to create a good quality robot article. The section containing the shadow hand was actually one of the last things I put in.
I agree, though, I may have put in a couple too many links. I'll take some out now. Rocketmagnet 20:29, 14 August 2007 (UTC) - Done Rocketmagnet 20:45, 14 August 2007 (UTC)
Guys, we must stay on topic here. This mediation was intended to resolve the issues in the article; not to debate on emotional issues and who Rocket is working for. Bang, please try to assume good faith on Rocket's work here and don't disrupt the discussion any further. Singularity 20:39, 14 August 2007 (UTC)
Also, if I was just here to advertise, why would I have spent many hours researching, drawing diagrams and contributing to the Milling cutter article? Do we sell milling cutters? No. It's because I love the Wikipedia, and believe it is an extremely valuable resource. I have also tried to encourage my friends to contribute on their own specialist subjects. I will be visiting one of the world's foremost beetle experts this month, and will try to get him to look over the beetle article. Rocketmagnet 20:45, 14 August 2007 (UTC)

[edit] What to do now

Since the discussion for the last several days has been quite confusing (and opinions may have changed), I want straight opinions on what should be changed in the article so we can come to a compromise. Please look at the article first before making any comments. No allegations, claims...etc. Singularity 20:59, 14 August 2007 (UTC)


Rocky feels I have made an ad hominem attack, but is the one who claimed, in all seriousness "Wikipedia articles must take a neutral point of view, which I have taken great pains to do".

Rocky has not done that, as proven by the evidence set forth hereinabove.

Rocky claimed my (readily admitted) family connection clouded my judgment and he didn't get called out on that, but now Singularity wishes to chastize me for calling out Rocky when he was hiding his true motivations from all of us and posting "blatant advertisements" (not my words, but the words of Wikipedia). That hardly seems neutral.

What is this debate really about?

It began and should end based not on the correct definition of a robot, but whether Rocky and the Cult of Tesla can bestow credit to Tesla's toy boat for the first "truly modern robot" thereby usurping the Unimate's place in history.

The whole issue is how the History section should read. After Rocky got done with it, Tesla was placed at the top of the mountain with a giant picture and credit for the first "truly modern robot". George Devol was given a subordinate position, followed immediately by pandering comments about "deaths by robot"

I submit that there is a correct definition of the first "Modern robot" just like there is a correct definition of the first "modern computer". http://en.wikipedia.org/wiki/Computer

Shadowhand doesn't like it because it means his robot hand is not a robot, but just a part of a robot. That's why he artfully espouses a "neutral point of view" that, of course, includes his products. (Note that even the name of his company, shadowROBOT is a misnomer under all American definitions).

Singularity, or whomever has final say, needs to decide whether there exists a correct definition of a "Modern robot". I think you have all the facts and opinions. If not, please advise. Bangthedash101 21:15, 14 August 2007 (UTC)

What do you want to change about the article in its current state? The article does not mention a sentence about Tesla's boat, but clearly mentions that the Unimate was the first modern robot. Singularity 21:30, 14 August 2007 (UTC)
Singularity, the article doesn't mention Tesla because Bangthedash101 deleted the section about Tesla, which started all of this.
Bangthedash, I did not write the history section of the robots article. All I did was to re-format it, and add the pictures. Please, please stop connecting me to Tesla, or a cult of Tesla. The reason I chose the pictures I did was because those were the only ones I could find on the Wikipedia. If I could have found a picture of Devol, I would much rather have used that. If you have any pictures of him, I would be delighted to see them on the article in place of Tesla. If you connect me to the cult of Tesla again, I'll begin to suspect you're not actually reading what I write here. Rocketmagnet 23:05, 14 August 2007 (UTC)
Also, I am not in the slightest but concerned whether people call the Shadow hand a robot, or component of a robot. Because a component of a robot is what it is. Any fool can see that. If one of our customers decides to use it on a machine which is arguably not a robot, good for them. We are called Shadow Robot because we make a) components for robots, and b) machines that even you would call robots. If you keep trying that one, I'll begin to suspect you're not actually reading what I write here. Rocketmagnet 23:05, 14 August 2007 (UTC)
Singularity, what I would like to see in the history section (which I have actually been trying to get other people to write, rather than myself, because I don't feel I am qualified), is a careful use of words which do not cause massive time wasting arguments like this one. Something like "this machine is the earliest known example of such and such". Or "this machine was the first to introduce programmability". Perhaps that kind of thing is less debatable. Rocketmagnet 23:05, 14 August 2007 (UTC)
I would also like to see a picture of Devol. BTD are you reading this? Rocketmagnet 23:05, 14 August 2007 (UTC)
I would like to see Tesla and his boat, with some non-angering fair text. Rocketmagnet 23:05, 14 August 2007 (UTC)

[edit] The conclusion

The two parties in this dispute disagree on a number of points, but I think there is a conclusion that will work out for both parties:

Tesla will be included in the history section of the article, but in the early developments section (I have already done so). In addition, it will not be dubbed the first modern robot, but simply what some may consider one of the key developments to what is now called the robot.

Because the usage of the word robot is varied and difficult to distinctly define and has changed over time, one cannot accurately define the word robot with a single definition. Any attempts to define the modern robot should be adequately cited.

Is this satisfactory enough for both parties? Singularity 02:35, 15 August 2007 (UTC)

I'm happy with that. Rocketmagnet 09:46, 15 August 2007 (UTC)

That proposal is acceptable BUT, as we discussed above, the quote about Tesla's statement of intent included in the PBS article and Cheney's book is demonstrably false. As such, the reference thereto needs to be removed, which I have done. Bangthedash101 14:08, 15 August 2007 (UTC)

Again may I butt in! The article looks pretty good to me now. The dispute has been moderated by singularity. BUT the paragraph on Tesla really is really bad and there I agree with Bang. OK, Rocketmagnet is not responsible for it but can we not get rid of it? It is a very weak connection indeed and lets the article down IMO. "A device some considered to be the first robot..." Who? I have never heard of it in 25 years of robotics. It reads to me like an RC toy ahead of its time, but a robot? Also I would be in favor of removing the picture of Tesla (famous for magnetism but not robots) and putting a picture of Devol further down. Can Bang provide a non copyright photo perhaps? Such a photo would be great on industrial robot also. I would love to see Devol more recognized. Robotics1 17:34, 15 August 2007 (UTC)

I have added a photo of Mr. Devol. I have an even better one that I can't find right now from the 1950's and will upload that one instead when I find it. I am not sure whether there are any still photos of him and the Unimate floating around. I don't have one like that but I will ask him. I know there are Unimation promotional films etc. that he has and there was a History Channel documentary a number of years back, both of which contain interviews.

I agree with Robotics1 that the Tesla connection is weak, but I am not going to be the one to remove it after all that has transpired. Rocky really wants a photo of Tesla on the page. I would much prefer a scaled-down photo of Tesla (if any) rather than a copy of the giant one that is in the headline of his own Wikipedia entry. Can users scale down photos that someone else has uploaded without violating any copyright? Bangthedash101 19:22, 15 August 2007 (UTC)

Mate, you cannot possibly have read a single thing I have written here. How many times have I asked you for a photo of Devol, and said I would much rather have him tha Tesla? I also explained that the only reason I put up a pic of Tesla was because it was the only one I could find. Why do you so love writing things you know to be false? Rocketmagnet 20:28, 15 August 2007 (UTC)
Also, I think the whole of the history section is weak. It lets the down the article. If anyone here knows anything about robot history, it would be so great if people could add some real content. Rocketmagnet 20:40, 15 August 2007 (UTC)
Bang, the image of Tesla is not giant. It is exactly the same width as all of the other pictures on the page. The width is chosen by the Wikipedia. It is possible to force a width other than the default one, using |120px| for example, to set the width to 120 pixels. However, apparently, we're not supposed to do this now. All of the px fields were removed from the images of the robot page a while ago by someone. This allows users to set their own default width for viewing the Wikipedia on things like mobile phones, which need smaller images. (Please stop throwing unfounded emotional accusations around. It does you no credit) Rocketmagnet 20:40, 15 August 2007 (UTC)
I should say 'thanks' for uploading a much needed photo of Devol. This could well be the first photo of him on the internet. I searched a lot on Google images for one, but only found millions of pictures of Engleberger. If you can find a larger image, especially one with a Unimate in it too, I would be more than happy to see the picture of Tesla removed. Rocketmagnet 20:48, 15 August 2007 (UTC)

Me again. I guess this is concluded but I will leave these thoughts from my perspective: 1. The Tesla paragraph is still too big compared to its true significance and the photo not warranted. 2. I think the link to the Shadow hand is commercially biased. I also have a company, st robotics, been making robot arms for 20 years and I resisted the temptation to insert a link. It's hard, I know. Rocketmagnet I do respect your good faith but I would ask you to be more self critical. And Bang's language took my breath away. I'm going to put the pic in industrial robots which is really my forte and wish this one all the best. Robotics1 22:40, 15 August 2007 (UTC)

I will watch this page for a couple of days if anything comes up. The inclusion of Tesla, I think, is agreed on, but what the paragraph contains is up to you guys. Singularity 01:29, 16 August 2007 (UTC)


OK, I've revised the Tesla entry consistent with what I think we decided here. I apoligize for my egregious breach of etiquette and assure you it will not happen again.

Regarding the Devol photo, yes I am sure it is the only one on the net. He didn't want a photo on Wikipedia but my mother talked him into it. I will be talking to him in the next day or so and plan to get more details about the Unimate (if not for this page then for the industrial robots page or his own page).

If anyone has suggestions, please post them here. I will be asking how the Unimate was taught (I know it was with a control pad but don't know if the arm was moved manually and the position recorded or if it was moved electronically at all times). How many steps could be recorded. How many angles/vectors were recorded at each step. How the speed of motion was controlled. How long (time-wise) the cycles could last. How much weight it could handle. Where the first one is (I think it is now in storage at the Smithsonian but was at the Henry Ford Museum on display for a long time).

Also about tactile and visual sensors he patented for use in robots, whether any others existed prior to his and the nature of his designs.

He also developed automatic controls (I think in the 30's) based on Theremin technology. This was primarily used in the context of laundry presses that would open and close based on the proximity of the user. When the war came around he was told to stop working on any proximity controls and not to talk about it. He still won't tell me any more than that, as he promised his secrecy and was very involved in the war effort, primarily with radar counter-measures -- which led to the first commercial use of the microwave oven, the Speedy Weeny. That is his secretary in the photo on his page, although he doesn't hold any microwave-related patents.

He also designed and installed what might be the first bar-coding machinery (again probably in the 30's as United Cinephone made a huge variety of photo-cell based products) for routing packages at a shipping company, whose name escapes me, using black paint, lightbulbs, and photocells. I know this was a one-off project that didn't end up in production or get patented, but is fairly interesting in the context of robotic visual sensors if not historically verifiable.

He also designed and built amplifiers and phonograph needles in the 30's. I have asked him about this but he doesn't think it relates to robotics. I think it does because it is another method of recording and controlling signals and shows the evolution of his ideas from audio products to photocells to automatic machine controls to robots. Anyway, enough of my ranting, you get the idea.

I know my mother has a copy of the Asimov robot book and will see if I can get a copy which should serve to bolster the History section. I do know that it contains one Engleberger fable about meeting Devol at a cocktail party and waking up with a hangover thinking that the Unimate "still looked like a good idea in the cold grey dawn". Though I admit GCD could put 'em away with the best of them. People run out of the room when he mixes the drinks...Bangthedash101 05:14, 16 August 2007 (UTC)

I am very glad that this is finally coming to a conclusion, and the language is calming down. Bang, I'm also glad your mother managed to convince Devol to show his face :) You seem to know a great deal of interesting information about him and the unimate, and it would be great to see the robot, Unimate and George Devol articles benefit from this. Rocketmagnet 07:20, 16 August 2007 (UTC)

[edit] The motivation for in-depth pages about specific robots

Now I would like to address the dispute about the Shadow Hand page. As a frequent user of the Wikipedia, I like to see in-depth pages about as many robots (and robotic components) as possible, eg Kismet, Packbot, Unimate, Roomba, ASIMO, Roboraptor, Robosapien, Foster-Miller TALON, Ballbot, Robonaut and many others. I was actually quite disappointed, when I was writing sections in the robot article, to find there were no Wikipedia pages about so many others, like the Schunk Hand, Barrett Hand, PUMA Robot, VersaTrax, MIT Hopping Robots, Dexter (Robot), and many others.

So, I think it is quite fair for me to add a page about a notable robot hand. Not about our company, not about other products we make, not about me, but about a notable item, alongside many pages about others. It currently contains only information from our company, which is a shame. However, I made sure the information was limited to only the factual specification about the hand.

If the contents of the Shadow Hand page should be deleted, then so should the contents of many other pages about specific robots and robot components.

Robotics1, If your company produces a notable item, then you should start a Wikipedia page about it. Rocketmagnet 08:06, 16 August 2007 (UTC)

No, that would be WP:COI. If an item is notable enough, it will be written be someone else. Singularity 08:27, 16 August 2007 (UTC)

Singularity is right. As for other grippers I would like to extend the existing article on industrial robots with various types of end effectors with names. I personally feel that if there is to be a link to, for example schunk, then it should be in the external links list and not embedded in the text. Am I right?

As for how the Unimate was programmed; we actually owned one at one time so I played with it. It was programmed in VAL (covered in Wikipedia) but the early ones may not have been. A pic of George with a robot would be fantastic for the industrial robot page. I'll use the existing one for now. Robotics1 09:35, 16 August 2007 (UTC)