Talk:Universal Robotics

Editor on this article, 69.247.131.76 has a reserve lookup of

Comcast Cable Communications, Inc. NASHVILLE-21 (NET-69-247-128-0-1)
69.247.128.0 - 69.247.159.255

Which would seem to indicate that they come from the same city as this companies HQ, given the nature of the edits by that user, I beleieve there is a high probability that there is a conflict of interest here, tagging the article as such 82.32.73.70 (talk) 11:25, 30 January 2009 (UTC)[reply]

82.32.73.70. Please clarify why you believe this is phrased as an advertisement. It is stating facts about a state-of-the-art control system architecture and the company. Explain yourself and stop simply threatening the editor. —Preceding unsigned comment added by URINC (talk • contribs) 14:36, 30 January 2009 (UTC)[reply]

I made no threat of any sort, see WP:NPA. I consider it advertising because of extensive use of slogans and advertising terms in the introduction, see WP:PCK. The conflict of interest tag is CERTAINLY appropriate since your username now shows you to be affiliated with the subject of the article so I'm putting that back now, see WP:COI for further information on conflicts of interest. I won't restore the advert tag at the moment since you say you're actively working on the page, please give me an indication of how long you will be doing this. 82.32.73.70 (talk) 15:36, 30 January 2009 (UTC)[reply]

Clearly you are an expert in these matters. Your insight will be valuable. We will stick with the facts, and watch the use of slogans. We'll review the WP:PCK and make adjustments. Can you contact me at 0222@usa.net for further discussion? —Preceding unsigned comment added by 69.247.131.76 (talk) 15:53, 30 January 2009 (UTC)[reply]

Thanks for the reply, I hope we can work together to make this a good article, I prefer to keep discussions here on the talk page because my access to email is a little erratic at the moment. Obviously your company does fully deserve an article on wiki, we just need to keep it objective and neutral point of view. Let me know when you've completed the current rewrite and we can work together to make it a good article that doesn't need the advertising warning. 82.32.73.70 (talk) 16:04, 30 January 2009 (UTC)[reply]

P.S. you many want to apply either Template:Underconstruction or Template:Inuse to the main page, that would alert other editors to the fact you're working on the page and discourage them from tagging it with ((advert)) like I did till you're done. 82.32.73.70 (talk) 16:22, 30 January 2009 (UTC)[reply]

I've further reduced the use of advertising terms on this page and hence removed the advert tag. Although this page did start out as a blatant advert ( I initially considered listing it for speedy ) I do think we need to acknowledge that the contributors from UR. Inc. have made a serious and partially successfully attempt to write in a neutral tone. The COI tag however will most likely have to be permanent because no matter how NPOV the rewrite, a large chunk of the basic info was supplied by the article's subject. 82.32.73.70 (talk) 09:29, 3 February 2009 (UTC)[reply]

The COI tag will need to stand, agreed, unless someone can find some extended independent critical analysis, I suppose.

I don't know how familiar you are with the field, and with NASA, but the article is the replete with cant. They are taking pains to distance themselves from "classical artificial intelligence" (a phrase which has no meaning, except to draw a distinction between themselves and the weak reputation AI has with investors), and also from neural networks (ditto). Sentences such as this are straight, unfounded PR: "Efforts to apply Machine intelligence to industry, have met with only limited success." What does "limited success" mean? The implication (without proof) is that their new technology will have better success. Which is to say nothing at all: why would any project exist that didn't intend to do something better?

From this article, it's difficult to tell what (if anything) they've accomplished. Loading and unloading pallets? It's been done.

Statements such as "...NASA, where it has been the “brain” of their humanoid robot for four years" have several (incorrect) implications, particularly for readers who aren't familiar with NASA and research centers. The underlying fact is probably that there is a robot project that has been funded for four years. There are plenty of robots at NASA, operational and in development. Being "humanoid"...well...let me put it this way...if being humanoid was itself somehow critically important, NASA would have done that 25 years ago. In fact, many AI NASA types avoid "humanoid" as being a gimmick, rather than a legitimate hard science goal.

There are still a number of statements still that need qualification, retraction, justification or citation. Piano non troppo (talk) 16:35, 3 February 2009 (UTC)[reply]

OK you clearly have more specific knowledge of this field than I do, so if you think it's required put the advert tag back and we'll keep working on it. 82.32.73.70 (talk) 20:41, 3 February 2009 (UTC)[reply]

Well, I have written proposals for NASA, so the intent was pretty clear. However, I do not know the project in question. If it's been funded at any great level for four years, there probably is substance to what they're doing. It sounds as if, as so often happens, editor 0222 came to Wikipedia just for the purpose of adding this article. Without knowing the guideline and process, the project is probably rather taken aback by the response to what they considered a good faith edit. Let's see what they come up with in response? Piano non troppo (talk) 23:12, 3 February 2009 (UTC)[reply]

--Gary Bradski (talk) 01:35, 24 April 2009 (UTC)There are still spurious claims with this article. First of all, it touts a mysterious algorithm that is somehow animal like. A brief over view. Second of all, the claims against AI are either spurious or just wrong (comments in bold to an offending paragraph):[reply]

Efforts to apply Machine intelligence to industry, have met with only limited success Rather, there are limited but notable successes such as airline scheduling, automated inspection on up to Google. The problem stemmed from the belief that if enough facts where loaded into a sophisticated data base, a machine would become intelligent (i.e. the Artificial Intelligence approach).[5] You found a newspaper quote perhaps, but classic AI hoped that if you created enough rules, not facts that intelligence would emerge. Machine learning etc hopes that if you put in enough facts or Data that intelligence will emerge. This has limits, but things like Google are fairly intelligent in their domain. Since this approach was not working Working for what?? Certainly people in Google are making tons of money, your shampoo bottle rarely has a defective label due to this approach and robots in the Darpa grand challenges have driven miles across a desert and navigated traffic in situations that would fluster many humans. You can argue for a more flexible, scalable approach, but as worded this is just not true., a number of roboticists[who?] in academia came to believe that a machine could exhibit intelligent behavior only through physically manipulating the world and, through its own sensors, learn the immediate effects of its actions.[citation needed]You could cite, for example any of the reinforcement literature, or perhaps Q-learning Intelligence would emerge as the machine developed skills through its interactions with people and the world.[6] Brooks is a great guy and all, but reinforcement learning predates him. You might as well also cite his Heartland robotics company too since they are in the same space.

Advertisement