Talk:UniModal/archive

Archived 18:21, 26 June 2006 (UTC)

Article survived

The article has improved enough to reach a no-consensus vote for delete. Please can someone move it to a more suitable name, and I'll work on it a bit. Stephen B Streater 15:59, 23 May 2006 (UTC)

I will also try to work on it in the next few weeks. A Transportation Enthusiast 01:55, 24 May 2006 (UTC)

Woot! : ) Fresheneesz 21:01, 25 May 2006 (UTC)

Feasibility

Seeing as the debate on AfD centred around feasibility, I propose a new section: Feasibility. This would have two subsections: "Technical feasibility", and "Commercial/political feasibility". Under Technical feasibility, the various proposed features could be reported, possibly with Wikilinks to examples of them in other systems (eg voice recognition, headway), and under Commercial/political feasibility, sections on load averages and other aspects as applied to Unimodal. The disabled access point would be mentioned here too. Stephen B Streater 09:54, 25 May 2006 (UTC)

My feeling is that we should be looking at ways of trimming the article, not adding sections. The feasibility points certainly need to be mentioned, but perhaps they can be added directly inline with the other sections. I also think we should rely more heavily on the main PRT page for most of the system description, and only focus on the specific traits that distinguish SkyTran from other proposed designs. A Transportation Enthusiast 13:54, 25 May 2006 (UTC)

OK - I've merged some sections to make the article more concise but keep all information directly relevant to the topic. How do these look? Stephen B Streater 20:32, 25 May 2006 (UTC)

Most of those edits look fine, two stood out though:

• "but significantly cheaper than the cost of running a car when depreciation and tax is taken into account." - seems like someone's "pushing a barrow" or whatever. Hah, I think that needs to go for now. Although it may be true, and its actually much more significant than the actual number price, it could be constrewn as POV. I'll remove it now unless theres some outstanding reason to keep it.
• Concerns about the disabled - Why did you remove this section. I think its a significant point that isn't addressed very well by other sources. Its a common comment that SkyTran doesn't like disabled people But thats simply not true, and I think its a good point to make on this page. Fresheneesz 21:01, 25 May 2006 (UTC)

Here is my thinking:

• The cost of gasoline varies depending on tax. In Europe it is several times the US price. Typical car costs are $1/mile in the UK. I'll dig out a car mag for figures
• Disabled is not removed, but merged with hypothetical trip, as it is a special case of a trip. The article had too many short sections for my taste
• My next change may be to put Pod and Guideway under System Design - they seem a bit mixed up at the moment. Guideway comes as a section on its own and a subsection of System description

Stephen B Streater 21:59, 25 May 2006 (UTC)

Ah ok, merging the sections is fine. As for the gas prices thing, I suppose we could say "about a tenth of the accepted cost of owning a car in europe, or fifth in the US." You can change my revert if you want to, I just don't want someone with a mission to see it as a reason to camp over here and make a mess. Fresheneesz 04:34, 26 May 2006 (UTC)

I'll find the cost from a car mag. They all list cost per mile or hundreds of types of car. Stephen B Streater 08:02, 26 May 2006 (UTC)

Making this a real page

I was wondering at what point we should move this to its own *real* page, and have articles link here? Fresheneesz 21:03, 25 May 2006 (UTC)

Perhaps very soon? It's being edited now by multiple people, and will soon be sparkling new and shiny. Stephen B Streater 22:00, 25 May 2006 (UTC)

Multiple people? It looks like it just me and you : ) . Fresheneesz 04:35, 26 May 2006 (UTC)

Do you have a link for the Lawrence Livermore reference? Also the energy to overcome magnetic friction does not include air resistance - the comparable figure for cars is 0W, as they have no magnetic friction. Please could you clarify this? Stephen B Streater 08:32, 26 May 2006 (UTC)

Also, I couldn't find the reference to the $0.01/mile running cost or the $0.10/mile projected charge. Stephen B Streater 09:57, 26 May 2006 (UTC)

Linke for what reference? I already have a link to LL labs for ref 2, and so I find the 3rd reference unneccessary. As for the magnetic drag, that is compared to the tire drag (made by the tire hitting against the road, and bearing friction). But I don't have time to look for another reference at the moment. Fresheneesz 04:33, 27 May 2006 (UTC)

The prices can be found here and here. Fresheneesz 04:45, 27 May 2006 (UTC)

I can make a few edits while you're busy - that's the wonder of WP :-) [1], which mentions £0.10/mile, also says:

• "Malewicki and his associates have developed proposal that would link all the major communities and tourist attractions in Arizona for the same cost of the proposed 35 mile Valley Connection light rail system. Based on their projections, a Phoenix couple could climb aboard a SkyTran car after work on Friday and be at their cabin in Flagstaff in just 1 hour and 28 minutes for a cost of $13.80 cents a piece. And they could spend the entire trip relaxing, enjoying the scenery or taking a quick nap. Try that in your SUV."

This comes out at $0.40/mile. And the 200mpg is more than 1c/mile, and this covers only energy cost. I'll make suitable amendments. Stephen B Streater 08:59, 27 May 2006 (UTC)

Thanks for your input on this. When you have time, the last remaining reference is the one for 3000W/8000W for a car being for a comparable thing to the friction drag of the pod. It looks on the face of it as if the car power consumption includes air resistance, but the pod one doesn't. Stephen B Streater 09:10, 27 May 2006 (UTC)

Actually, I did the calculations and it comes out to 18.8 cents per mile (9.4 cents "a peice"). This seems odd to me, because that is twice what they were saying around that time (the article's dated 1998). It would make sense if they were traveling at 150 mph, because that costs twice as much (i'm not saying they would charge twice, but the power to overcome air resistance is twice what it is at 100 mph). But in any case, the time of 1 hour 30 minutes tells me that they were traveling at 100 mph (because pheonix to flagstaff is 145 miles).

Maybe I'll try to find a few more sources to verify the numbers. I'll also look for a reference for the car power thing. Fresheneesz 19:59, 27 May 2006 (UTC)

Using the equation found at rolling resistance, I calculated a 3000 pound (1360.7 kg) car at 65 mph (29.0576 m/s) as requiring between 4262 watts (coefficient of .011) to 11625 watts (coefficient of .03). Fresheneesz 23:38, 27 May 2006 (UTC)

I changed the wattage range to 3000 W to 13500 W. 3000 comes from calculating for a 2200 lb vehical with CRF of .011 and 13500 comes from calculating for a 3500 lb vehical with CRF of .03 - both at 65 mph. Fresheneesz 00:15, 28 May 2006 (UTC)

Energy is twice as expensive as it was then. Crude oil is three times the price. Electricity prices have not gone up as fast as oil, as there are other sources - an advantage I may mention. Stephen B Streater 06:07, 28 May 2006 (UTC)

Doesn't the energy cost per mile maximum have to assume there is one passenger only (eg passenger + bags)? Or even add in the cost of the empty pod reaching the passenger (though I'm assuming this will probably be small given that you will be getting the nearest pod). Stephen B Streater 08:14, 28 May 2006 (UTC)

(could you not interrupt people's comments, it makes it hard to read) Fresheneesz 10:45, 28 May 2006 (UTC)

I had assumed that the energy cost given was per pod-mile - but perhaps I was mistaken. I think we should write something clear like "energy cost per pod-mile" or "per seat-mile" (NOT "per passnagner-mile"), when we find that information out for sure. I suppose the magnetic drag depends on the weight in the vehical, but I would assume that not only is that difference negligible, but the cost rating is inherintly approximate. Fresheneesz 10:51, 28 May 2006 (UTC)

Good points. Stephen B Streater 11:32, 28 May 2006 (UTC)

Looking through a couple places for the per-mile figures, every time it seems like it means $0.01 per pod-mile, and charging a price of 10 cents per pod-mile. I think that probably the reference talking about "pheonix to flagstaff" calculated the price based on 150 mph, and the time based on 100 mph - but i guess thats just my opinion. Fresheneesz 01:00, 29 May 2006 (UTC)

I just noticed the comment about how traveling at 150 mph would cost 3 times more - and I realized my above statement is wrong. The drag equation implies that the power needed to increase ones speed by 1.5 times would be 3.375 times more power. I thought that was interesting. Fresheneesz 06:57, 30 May 2006 (UTC)

The cube law is only an approximation. The true formula is of the form a₀ + a₁x + a₂x² + a₃x³ + a₄x⁴ + ... with a_n typically much smaller than a_n-1. This means that for typical speeds only one power dominates. For very small speeds, the friction is about a₀ ie constant. So the point I am making is that the cube law is only approximate, and quoting so many digits is unreaslistic. That's why I changed it to "three". It's worth adding that the drag equation is an approximation, and if the air flow round an object is turbulant, it can make things a lot worse. Stephen B Streater 09:20, 30 May 2006 (UTC)

Hmm, even more interesting. The pages on drag (physics) and drag equation could use your input - since you obviously know more than I do about it. Fresheneesz 17:56, 30 May 2006 (UTC)

references superscript

I was wondering how the <ref> thing works? Is there a place to discuss that on wikipedia? I think its not very good style to have the reference number be a superscript because it breaks up paragraphs. Fresheneesz 23:43, 27 May 2006 (UTC)

I found it here. Fresheneesz 00:02, 28 May 2006 (UTC)

Thanks - I knew it was somewhere. Stephen B Streater 08:12, 28 May 2006 (UTC)

Moving this to real page

I'm going to request that UniModal be deleted so that I can move this page there. After that it should become a real page. Fresheneesz 09:40, 1 June 2006 (UTC)

OK. The AfD was ages ago, and we've done some work on the article since then. Stephen B Streater 12:46, 1 June 2006 (UTC)

Alright, moved it to a real page. Fresheneesz 06:40, 9 June 2006 (UTC)

Archived: 18:27, 26 June 2006 (UTC)

Adding section on pods

I would like to add this paragraph from the Pods section under "Unverified or questionable information":

Because SkyTran plans to operate at high speeds, the pods are designed to be aerodynamically streamlined. Vehicles with lower air drag use less energy and produce less noise. The pod is designed to carry two people in tandem (one behind the other) for decreased air drag. Resting pods would sit on small wheels, but when the pods accelerate to above around 2 mph, the Inductrack could provide enough levitation to lift the pod off the guideway. After slowing to below 2 mph, the wheels would act as landing gear to support the pod.

Reasons:

1. The information is sourced
2. How pods work is essential to understanding the concept
3. It does not duplicate any information from PRT
4. but it does provide a LOT of information about how SkyTran is different than other PRT ideas.

Comments? Fresheneesz 20:12, 10 June 2006 (UTC)

To keep the article short, I think this more concise version would be preferable:

• SkyTran pods would be aerodynamically streamlined (helped by the two passengers sitting one behind the other), to be quieter and save energy. Above two mph the Inductrack would lift the pod off the guideway; below two mph, pods would use wheels.

previously unsigned comment by User:Stephen B Streater 2006-06-10 21:38:13

I like Stephen's version; it's shorter and I think it sacrifices nothing. A Transportation Enthusiast 22:14, 10 June 2006 (UTC)

How bout this version:

*"SkyTran pods would be aerodynamically streamlined to be quieter and save energy (helped by the tandem design - one seat behind the other). Above two mph the Inductrack would lift the pod off the guideway; below two mph, wheels would support the pod. "

Fresheneesz 22:57, 10 June 2006 (UTC)

Even better. Stephen B Streater 06:11, 11 June 2006 (UTC)

I'll look at what we can take out to keep the article short. Stephen B Streater 06:12, 11 June 2006 (UTC)

I've added this in and removed a corresponding amount of less important text to keep article length. Stephen B Streater 06:23, 11 June 2006 (UTC)

In retrospect, I think the article before was a little too verbose. It was less about size than about wordiness (especially regarding points already covered in the main PRT article). Stephen and Fresheneesz, these changes really tighten it up, without losing the important points. Conciseness will make the article shorter and better. A Transportation Enthusiast 14:40, 11 June 2006 (UTC)

I've gone through the introduction section also making it more concise in a similar way to above, mostly by removing repetition which often builds up as an article is developed. I think the article structure is much clearer now: an introduction giving the bigger picture, and then salient features of the system itself. Stephen B Streater 10:13, 11 June 2006 (UTC)

PS I took out the bit about being suspended from the overhead track from the text because this is clear from the picture. Stephen B Streater 10:23, 11 June 2006 (UTC)

Stephen, you are doing good work - the article is informative without being misleading. Just zis Guy you know? 13:20, 11 June 2006 (UTC)

I agree, nice work. A Transportation Enthusiast 14:41, 11 June 2006 (UTC)

Thanks :-) Stephen B Streater 17:05, 11 June 2006 (UTC)

Those are excellent edits, they sacrificed nothing, and made the article much less wordy. Nice job. Fresheneesz 18:13, 11 June 2006 (UTC)

Actually, they sacrificed one thing: explanation about the 3D grid design - which doesn't seem to be explained on the PRT article (doesn't contain the strings "grid" or "3d"). I think it would be nice to re-add this explanatory bit:

"- much like freeway overpasses allow the cars on the freeway to continue non-stop."

Also, it sort of seems like "intersections" mean intersection roads or tracks, and the 3D grid thing specifically avoids *intersections*. Calling an overpass an intersection is a little misleading I think. Fresheneesz 18:20, 11 June 2006 (UTC)

As this feature is common to PRT designs, a detailed explanation should be added there, not for each PRT design. We can still make it clearer here. I don't like the freeway analogy because there are no freeways in England. I'd rather just say how it works. How about: The 3D grid design avoids intersections by allowing pods to pass above or below each other.? This is a bit shorter so you have a budget of two words for extras ;-) Stephen B Streater 18:36, 11 June 2006 (UTC)

Theres no freeways in england?!? Thats news to me. I'm truely surprised! Anyway, what bout:

"The 3D grid design avoids intersections by allowing perpendicular guideways to pass above or below each other." Fresheneesz 19:04, 11 June 2006 (UTC)

They're called "motorways" - and have some differences. Dinner's ready - I'll be back later to look at your suggestion. Stephen B Streater 19:07, 11 June 2006 (UTC)

I didn't understand your perpendicular bit before because I imagined the tracks to be a bit like railways, which don't cross at right angles, but meet at junctions and then split off again at other junctions. So what is not necessary in PRT is the crossing of tracks which may or may not be perpendicular. So what is important is the crossing of the tracks. The 3D grid design avoids intersections by allowing pods to cross above or below each other.. Or, using your version: "The 3D grid design avoids intersections by allowing guideways to cross above or below each other." What do you think? Dinner was great BTW :-)

I like the "guideway" version because the "pod" version could be confusing, as pods can't just cross above other pods at any time (they have to follow a guideway), but guideways can be build above or below other guideways whenever the developer wants. Fresheneesz 20:15, 11 June 2006 (UTC)

OK. I've added it in, starting with "A" rather than "The" because it is not referred to before in the article. Stephen B Streater 20:55, 11 June 2006 (UTC)

Ok looks good. Fresheneesz 22:29, 11 June 2006 (UTC)

I think you're working too hard to keep this article as small as it can be boiled down. But, I'm a little put off that you think it's a good idea to remove a key design feature from the text because it's clear from the picture. I have yet to find the software you use that interprets photographs for the visually impaired. For my nephew, the image explains nothing. Hopefully the same concisification doesn't hit the article on Saturn, lest he become unaware of the rings that are also clear from the picture.

Speaking of concisification, why is there so much concern about conciseness? I can understand a printed encyclopedia has space limitations that require sacrificing detail for some entries so that more significant topics can have the detail they deserve. But there is no such limitation with WP, and if there is, I'll donate more money to help the cause. I realize WP isn't the final authority or historical record on any subject. We shouldn't limit these articles based on character economy and sacrifice knowledge and understanding.

PPS. I don't know who made the edit, but my nephew also asked why did someone say that voice recognition is an "unproven" technology. He pointed out that he uses voice recognition on his computer and his cell phone, and many companies use voice recognition in the call routing and voice mail systems, and apparently he's even used an elevator that uses voice recognition, so voice recognition has even been integrated into some form of existing transport. I told him that's the editor's personal opinion. He just said, "Too bad." --JJLatWiki 17:46, 14 June 2006 (UTC)

Where does it say speach recognition is unproven? I've proved it to myself with software that came with my computer. And Acura, and many other car companies, are starting to incorporate speach recognition in their vehicals.

I have the same concern as you - while conciseness is excellent and helps people understand something quickly and easily, conciseness should not sacrifice information. Its simply that some powerful editors here think that what they think isn't important or notable enough justifies censorship. Fresheneesz 22:19, 14 June 2006 (UTC)

In the first paragraph: "The Unimodal system design (more below) includes a number of unproven technologies such as speech recognition to interact with riders,". I personally think "unproven" is POV. I've used voice recognition on my nephew's computer with great success. Maybe "to interact with riders" is a qualifier that the editor doesn't feel has been used sufficiently to be called proven. --JJLatWiki 15:24, 15 June 2006 (UTC)

Voice recognition works quite well in a quiet room when trained for a single user. However, it is unreliable when it has to recognise multiple accents in a noisy environment. Even 90% reliability would be useless when similar sounding places might be miles apart. Voice recognition is another marketing stunt - buttons with places on them would be much more practical. Stephen B Streater 16:51, 15 June 2006 (UTC)

Plus: name a public transit system which uses voice recognition. Can't name one? OK, so that makes voice recognition for destination sleection an unproven technology. Move along, now, nothing to see here...

Oh, and I have used voice recognition before - even after training to my voice (which of course a public transit system would not have) it was still not especially reliable. It sounds like a gimmick to me; a touch screen or some other standard input would surely be more reliable. Knocking the bugs out of too many new technologies at once is a great recipe for massive time and cost overruns, after all. Just zis Guy you know? 17:59, 15 June 2006 (UTC)

Actually, you all (including me when I talked about it) are talking about full speech recognition. SkyTran would use selective speech recognition that selects a speach pattern out of perhaps a hundred or a thousand different menu choices at a time, rather than the millions of word combinations that could occur within the buffer-space of a speech recognition program. Software you guys are talking about need to not only figure out what the sounds are, but where the separations of words are. Menu based slective speech recognition is many times more accurate. I think you can try it out on programs like Opera browser (although i'm sure there speech recognition tech isn't amazing - its still pretty reliable, and no training required [or possible] ).

In summary, having a menu of buttons for 5000 different destinations on a 2×1 surface would be "a fantasy" (or more accurate, a mess) and the buttons would be so small as to be "inaccurate" if you tried to press any one, but having 36 buttons that can be used to specify your destination would be much more feasible. That was a metaphor just to be clear. Fresheneesz 19:03, 15 June 2006 (UTC)

You could have a matrix of buttons, one for each letter of the alphabet, and people could press them in turn to spell out the name of the destination. Hey, that's such a great idea why don't I patent it? What could we call it? (yes, I know that was your point). So: why overcomplicate, when people are well-used to the keyboard? Just zis Guy you know? 19:41, 15 June 2006 (UTC)

Not sure, I asked Malawiki that myself. His reason was that speach was more natural for transportation (ie taxi). Personally I disagree with that. However, its very true that a microphone and speaker are much less expensive than a keyboard and a screen. My point, however, was that its not a fantasy - just a design choice. Fresheneesz 21:38, 15 June 2006 (UTC)

I've never seen it, but I've been told that there are elevators that use voice recognition. Obviously not a big noise issue and not very many options from which to choose, so it's no "proof of concept". But, I can imagine a system that uses it as its primary interface then falls back to touch screen input if it can't recognize the instruction. But, by the same token, a touchscreen interface wouldn't be very difficult to make or to use. One way would be to press letters to limit a list of available destinations...Press 'M' and the list of hundreds becomes, "Main St|Martini Way|Martin Luther King Drive|Mercy Hospital|Mockingbird Lane", then press 'O' and only "Mockingbird Lane" remains. Two taps on a screen for the first cross-street or street of the closest address. Not my problem though. Point is, it's doable and not very difficult. Maybe include Google Earth views as the destination tightens up. --JJLatWiki 00:10, 16 June 2006 (UTC)

It's a "design choice" that is unproven, like the article says. Just zis Guy you know? 07:15, 16 June 2006 (UTC)

The article doesn't say, unproven "design choice". It says unproven "technology". If you Google "unproven technology", I bet most uses of the term relate to some form of opposition. Opponents say, "unproven". Proponents say, "new" or "advanced". Maybe a more neutral statement would be: "The UniModal concept includes several design elements that have not yet been used in an automated transit system including, voice recognition to interact with riders and a passive form of Magnetic Levitation called Inductrack." --JJLatWiki 14:52, 16 June 2006 (UTC)

My understanding is that speech recognition is unreliable when multiple accents and background noise are present. So using it in this context is unproven technology. As this may have changed since I last looked in this area, do you have different information? Speech recognition: "The typical achievable recognition rate as of 2005 for large-vocabulary speaker-independent systems is about 80%-90% for a clear environment, but can be as low as 50% for scenarios like cellular phone with background noise." Stephen B Streater 16:34, 16 June 2006 (UTC)

I have no idea how successful voice recognition is in any context. I don't think a transit application would be considered "large vocabulary" though. But I still feel like "unproven technology" has a opposing POV connotation. It's not as if voice recognition is some impossible flight of fancy, as say, telepathy or mind reading, which would cast the whole proposal in a different light. How many current transit systems currently use LED touchscreens to choose from thousands of possible destination? If it's zero, is that also "unproven technology"? --JJLatWiki 18:57, 16 June 2006 (UTC)

In Wimbledon, the main line, underground and trams all use LCD interfaces to choose destination and ticket type. The point about speech recognition it that it has taken decades to get to where we are - which appears to be not far enough. One small transport company will not be able to push human technology another 5 decades forward in a reasonable timescale. Stephen B Streater 12:00, 17 June 2006 (UTC)

Yup. Unproven is factually correct, and is also the terminology which is used by engineers to describe - well, technologies which are unproven. The fact that unproven technologies have a poor track record, and hence being unproven carries negative connotations, is not really our problem, is it? Just zis Guy you know? 18:52, 17 June 2006 (UTC)

I would say that unproven brings along connotations of being unproven in all respects. Voice recognition and speach recognition have both been "proven" concepts - meaning they actually work. What has not been proven is if they would be reliable enough for public transit use. We all agree on the fact that the technology hasn't been used for this purpose before, but lets focus on the word "unproven" since that seems to be JJLatWiki and my only problem.

Does anyone agree with "...includes a number of new technologies that have not yet been found to be reliable in such situations, such as speech recognition...." or alternatively "...includes a number of new and developing technologies, such as speech recognition..." Fresheneesz 05:11, 18 June 2006 (UTC)

In the last month, I've used voice recognition at least twice - both times when I needed to call customer service using phones which are inherently low fidelity, and without any prior training to my voice or accent. Speech recognition is not just proven, it's in use everywhere. To make the blanket statement that it's "unproven" is just plain innacurate. It needs to be reworded.

If the concern is that speech recognition has not been proven in a transit system, then that might be more accurate. But, is it verifiable? Are we really sure that voice recognition hasn't been done before in a transit system? Has someone done a comprehensive survey? Personally, I don't think it's appropriate to label it unproven unless there is evidence of that. And it's already stated pretty clearly that SkyTran itself has no prototype, and as such the entire system is unproven. I say we remove the speech recognition clause and just say Inductrack is unproven (I assume there's no disputing that, right?) A Transportation Enthusiast 09:00, 18 June 2006 (UTC)

List all the transit systems which include this technology. None? Unproven it is, then. Just zis Guy you know? 09:13, 18 June 2006 (UTC)

You say "unproven" -- do you mean unproven technology or unproven application of technology? I say it can't be the former. As for the latter: if we make the statement that it's unproven in transit, we should have verifiable evidence that it's not been used in a transit system. Do we have such evidence? We already have one more verifiable example of an unproven technology in UniModal (Inductrack), and that one is much more vital to its operation than speech recognition. Therefore, I see no need to include the speech recognition example - it's superfluous and unverifiable. A Transportation Enthusiast 14:32, 18 June 2006 (UTC)

AFAIK, accurate speech recognition of the reasonable vocabulary required for the many possible PRT stops in a noisy environment without individual user accent and voice training is currently non-existent technology. I've quoted some accuracy figures from the appropriate WP article. You could imagine something might work, but realistically text messaging or LCD panels (possibly wth a map which you just click on) seem much more sensible. Unfortunately, the design seems to be vague and it is not clear which specific technology, if any, is planned. I suspect it's just a broad brush impression they are giving of the futuristic nature of the project. We could say something like: "are considering novel uses of speech recognition technology" which sort of implies it hasn't been tested. I would suggest leaving out the speech bit, but I think Fresheneesz quite liked it as it gives some flavour for the design. If anyone can find out anything about the speech recognition they are planning to use, or indeed can find any reliable untrained speech recognition technology, I'd be happy to reconsider the "unproven" bit. Stephen B Streater 21:50, 18 June 2006 (UTC)

"are considering novel uses of speech recognition technology" works much better than unproven. However, this leaves what kind of "novel uses" the system proposes a bit nebulous. How about we remove "unproven" and instead add a sentence after: "Current speech recognition technology may not yet be adequite for the purposes that UniModal proposes".

I very much disagree with the thinking that since speech recognition isn't used in transit yet, it is "unproven". Voice recognition may not be used in a baseball hat, but that doesn't mean that "baseball-hat speech-recognition" is some sort of new technology. Neither is "transit speech-recognition" new technology. It will be the same technology as all the other speech recognition technology.

What would be the most useful is specific quotes like Stephen gave. However, since it would be stupid design to impliment "large-vocabulary" speech recognition into SkyTran, I think the best thing would be to quote small-vocabulary reliability percentages - rather than throwing everything into the "unproven" dumpster. Fresheneesz 01:28, 19 June 2006 (UTC)

I found this sentence:

• "For example, in the Air Travel Information Service (ATIS) domain, word error rates of less than 3% has been reported for a vocabulary of nearly 2,000 words and a bigram language model with a perplexity of around 15." - [2].

Alternatively we could cite the large-vocabulary statistic, and note that medium-vocabulary systems would acheive better recognition rates. Fresheneesz 01:48, 19 June 2006 (UTC)

This information is interesting. It seems to look at words in pairs to gain context and improve the rate. If station names were carefully chosen to be long a distinct, the same trick would work on SkyTran. Do you know what the environment is for this - ie how noisy? 2,000 words would be enough. But with 10m journeys a day in London, 300,000 would be to the wrong place - a lot of annoyed customers! I think the destination reliability rate of a transport system as to be much more like 99.9% than 97%. Can you imagine missing a meeting once a month because you went to the wrong place? And how many tourists can pronounce "Gloucester"? Also, was this the designers figure, or a real world independent figure? Speech recognition works much better on a training set than the real world. So I'm still leaning towards "not invented yet" technology for SkyTran, though I may be fussy expecting not to go to the wrong place 3% of the time. Stephen B Streater 06:38, 19 June 2006 (UTC)

Well, I would guess that noone would end up in the wrong spot, because there would be feedback with confirmation - like in computerized telephone direction. But failure rate would definately be annoying. If the thing is wrong a consistant 3%, and error rate is entirely random - 3% would be very succesful, simply say the word twice and you're off.

• I found this page that I think is a manufacturors page, but gives numbers for a medium-vocabulary system.
• this one shows stats for untrained a small-vocabulary system in different levels of noise.
• this is same sort of thing, but less clear.

and thats all I got. "Glah chester" ?? Fresheneesz 08:53, 19 June 2006 (UTC)

I think it's much simpler than trying to establish a reliability figure by original research - as an encyclopaedia (rather than a research journal or technical publication) we simply note that it's unproven and move on. Just zis Guy you know? 08:58, 19 June 2006 (UTC)

The wording as it stands is inaccurate. Anyone who has used an automated speech recognition system (probably a large majority of readers these days) is going to be confused by that sentence. Speech recognition is becoming ubiquitous in society, and to make the blanket statement that it is unproven is incorrect. It's the application of this technology in a transit application that may be unproven. That's a lot of qualification that's missing from the article. A Transportation Enthusiast 10:05, 19 June 2006 (UTC)

Ubiquitous? Not really. It is finding a few niche applications, but nothing like this has been tried yet as far as I can tell. Proven, in the context of an engineering discussion, means tried and tested in similar applications. Just zis Guy you know? 10:48, 19 June 2006 (UTC)

We're not talking about speech recognition in a transit system, because the article doesn't say "in a transit system". It says that speech recognition is unproven, period, which is demonstrably false. I am personally aware of at least half a dozen customer service centers which use it, without voice-specific training and over low-fidelity communications links. The article as written is wrong. A Transportation Enthusiast 15:00, 19 June 2006 (UTC)

No, the article as written is right. In the context of a transit system, speech recognition is unproven. The context is clear from the article. The thing about unproven technologies is that they tend to take a lot of knocking into shape; in this case that's exactly what would have to happen. There is no comparable installation from which they can buy in a working product. Just zis Guy you know? 15:08, 19 June 2006 (UTC)

My understanding is that what is unproven is that the technology (whichever one is used) is not known to work well enough for this market. My feeling was that the error rates would be annoying enough to make it a standing joke. Speech recognition is not proven to be good enough for this application. Not just unproven, but untried. There is a market uncertainty and a technical uncertainty. What is unproven is that they meet in the middle. This is different from saying that all speech recognition is unproven, but consistent with saying something like it is unproven whether speech recognition will be acceptible quality for this application. Stephen B Streater 17:37, 19 June 2006 (UTC)

But, it doesn't say anything like "it is unproven whether speech recognition will be acceptible quality for this application". It says "unproven technologies like speech recognition". That implies that speech recognition is unproven in general, which is a ridiculous statement. All I'm saying is that the wording as it stands is completely false. It may be unproven in a transit setting -- the sentence should be changed to include those qualifications, or it should be removed. A Transportation Enthusiast 18:16, 19 June 2006 (UTC)

Like I said, JzG owns this article and the entire PRT family. Nothing is true until HE is satisfied. It would be nice to know however, some kind of criteria for softening "unproven technologies" (or allowing dollar estimates). Even though (I assume) JzG isn't the United Nations Transit Czar with his finger on the pulse of all things transit, he somehow knows with absolute certainty that no transit system in the world has ever used or tested voice recognition, at least he has repeatedly made such factual statements. Apparently, "proven" is the only way to remove "unproven" from the use of technologies that, although they may exist and are in use in other areas of everyday life, are not common in transit systems. So, until voice recoginition is the only form of user interface for all transits systems with unlimited vocabulary and 100% accuracy, I see no way of offering proof. Even then, if there are still no PRT installations in production, voice recognition will still be an unproven technology when combined with PRT. Likewise, if by some freak of nature, a large-scale PRT is built that uses voice recognition as its primary user interface but has to use a touchscreen for input 5% of the time due to unusual accents, speech impediments, or hardware malfunction, JzG can still claim it's unproven. So, since this article is owned by JzG and it is his quest to make sure all unproven technologies are described only in shadowy and suspicious language, I hereby surrender the debate to JzG's greater authority within WP. --JJLatWiki 19:31, 19 June 2006 (UTC)

I am not the one with no edits outside of this subject area. Just zis Guy you know? 22:28, 19 June 2006 (UTC)

Irrelevant. How much we edit elsewhere has nothing to do with the validity of our points here. In any event, Skybum and Fresheneesz do have edits elsewhere, so your argument doesn't hold water with them. A Transportation Enthusiast 01:26, 20 June 2006 (UTC)

ROFLMAO! Thanks JzG, I think that's the funniest defensive retort I've ever heard in any discussion, on WP or otherwise. I think that one is called Ad hominem. Not only is it a logical fallacy, the basis is unfounded. You apparently didn't look back far enough to see my contributions to other, totally unrelated topics. I don't often contribute and tend to make some minor contribution then never come back again, but they are there. I've also made anonymous edits prior to registering, but never since, simply because WP elitists use that to attack the messenger. Unless of course you weren't talking about me when you implied that the subject of the sentence has made edits only to the UniModal/PRT area. If so, then...nevermind. --JJLatWiki 22:51, 21 June 2006 (UTC)

No, the ad-hominem was accusing me of WP:OWN on an article which I rarely edit (and dramatically improved by comon consent in the recent past). Note that I have consistently referred to policy and guidelines in discussing content. But hey, let's not let facts get in the way of a determined POV push :-) Just zis Guy you know? 23:01, 21 June 2006 (UTC)

"rarely edit" - look at the histories and tell me how you can make that claim with a straight face.

"Common consent" ... who? I count 4 editors who have been opposed to many of your changes (me, Fresheneesz, Skybum, JJLatWiki). Are you saying that there were more than four who supported all of your edits? To claim common consent, I would think you'd have to point to at least 7 or 8 who supported you. I'd be interested in seeing that list.

"I have consistently referred to policy and guidelines in discussing content" - well, yes, except for all the times you didn't! How many unexplained mass reverts by you? I think there were close to a dozen. Not to mention the other reverts where your sole explanation was that we were supposed "proponents". If you're looking for evidence of a POV push, JzG, don't look at us. A Transportation Enthusiast 23:46, 21 June 2006 (UTC)

I favour keeping speech recognition in because it indicates to people who are familiar with it that the drive behind the design is futuristic rather than pragmatic. It could have used the minimal number of untried technologies, but instead includes many unnecessary risks in the basic design. This indiciates that if this design is built, it will probably be simplified to make it easier to get working. So we could say unproven for inductrack, and untested in a transit setting for the speech - it's not that they even tried it and haven't got the results yet. This would also remove any potential misunderstanding about this article expressing a view on speech recognition as a whole. The Unimodal system design includes electric linear propulsion using a form of passive magnetic levitation called Inductrack, an unproven technology, and speech recognition to interact with riders, a technology untested in this setting. Stephen B Streater 22:01, 19 June 2006 (UTC)

Applying the Streater "less is more" policy, why not "technologies which are unproven or untried in any comparable context"? Just zis Guy you know? 22:39, 19 June 2006 (UTC)

I changed the sentence to say that the technologies are "not proven in the area of transit". This is specific, and correct - so noone should be confused by what it means. Fresheneesz 23:03, 19 June 2006 (UTC)

I think this an improvement on the original too - I'd be happy with this. Stephen B Streater 19:07, 20 June 2006 (UTC)

Thumbnail size

Last night I edited the size of thumbnails in this article such that they use the size the user has specified in their preferences. Unless there is a very good reason otherwise, please do not change that back. Remember not everyone's equal - not everyone has the same massive monitor as you. Thanks/wangi 08:47, 12 June 2006 (UTC)

My preferences don't go that big ;-) Stephen B Streater 09:02, 12 June 2006 (UTC)

I was just wondering why we want user-defined sizes? I can see that its a nice idea to scale a picture based on user preferences, but frankly, I doubt that even 5% of wikipedians use that particular preference. Are there any ways we can make a default size, or tell pictures to scale based on screen resolution? The resized picture is tiny on my screen, and my concern as an editor is that its tiny on lots of people's screens. Comments? Fresheneesz 03:36, 13 June 2006 (UTC)

By definition a thumbnail isn't meant to be massive. I'd recommend you update your settings. There would be no easy way within the mediawiki software to scale based to screen size. Thanks/wangi 07:37, 13 June 2006 (UTC)

Would it be ok if it wasn't a thumbnail? After all, it is the main (and only) image for this topic. Fresheneesz 07:46, 13 June 2006 (UTC)

I preferred the original size. I think the picture encapsulates a lot about Unimodal and what the experience would be like, and the thumbnails don't go big enough. An alternative would be to get someone to add more size options to thumbnails, which would help the other 1 million articles too. Stephen B Streater 08:26, 13 June 2006 (UTC)

I don't think there is anything different with this photo compared the millions of others in millions of other Wikipedia articles. It's not special and it should just be a "standard" thumbnail size, a size the user specifies. /wangi 08:44, 13 June 2006 (UTC)

But Wikipedia has other options that thumbnails - not every picture *should* be a thumbnail. Its not a standard any more than equation formating is standard. Main pictures are usually larger, and thumbnails are usually for smaller and less significant items. The picture encapsulates what this articles is about - it is not secondary. Fresheneesz 22:12, 14 June 2006 (UTC)

What does "politicized" mean?

JzG, what is the point you are making with the word "politicized"? The concept itself is not political at all -- it's a technological proposal. The opposition is frequently political, but the concept itself is not. I think this needs to be clarified. A Transportation Enthusiast 14:55, 12 June 2006 (UTC)

Search for UniModal on Google. Just zis Guy you know? 22:49, 12 June 2006 (UTC)

I did... what's your point? A Transportation Enthusiast 00:41, 13 June 2006 (UTC)

SkyTran has "been around" in the media - it has not been "politicized", which I take to mean debated fervently in politics. There is media controversy, not political controversy. Politics doesn't seem to care at all about PRT or SkyTran. The whole section provides a reference to material it is critisizing - *not* a source of the critisism. That is clear POV and as of now, not verified and thus original research. I think I can safely remove that section since it violoates those three guidelines (POV, verifiability, and OR).

Also, JzG, please please don't just throw around vague and ambiguous statements like "seach google". Fresheneesz 03:26, 13 June 2006 (UTC)

The subject being politicized means that people commenting on the subject in the media do not have a NPOV, so we can't rely on them uncritically. The risk is that there are few reliable sources and neutral observers. However, my "concisifying" tendencies would limit how much space we give to caveats about the nature of the project. It must be clear but allow for the main reason why people will read the article - to find out about the design. Stephen B Streater 08:34, 13 June 2006 (UTC)

Yes, exactly. All engineering projects are politicised these days; the problem for PRT is that it has yet to get beyond that point to a stable large-scale implementation which can be assessed dispassionately. Just zis Guy you know? 16:20, 13 June 2006 (UTC)

Looking at the article as it stands right now, there is a short bit about the hypothetical nature of the system in the introduction, and a short section at the end which raises this different point about the dangers on relying on the available sources. Stephen B Streater 08:38, 13 June 2006 (UTC)

If I had to shorten the article, I would keep the new section but delete the last sentence in the introduction which refers to PRT in general. This can be found in the PRT article and such a short Unimodal article should stay focussed on specific Unimodal issues. Stephen B Streater 08:47, 13 June 2006 (UTC)

The "Controvery" section doesn't mention that the media in general can't be relied upon for NPOV material. It states something specific, and cites an article it specifically targets. Not to mention the allegation that the article in question is misleading people - its just a poor choice of words, like in any newspaper. They use rhetorical method to say "what if you could do this with some pods" "well these pods are already conceived of". The section itself is misleading and is POV.

In any case, the first sentence of the section is reasonable to have, but the part about "coverage" that "presents a false picture" is simply not verifiable, POV, original research that shouldn't be in this article. Is there something you can do to convince me that its not? Fresheneesz 20:26, 13 June 2006 (UTC)

"Some of the coverage of UniModal presents a false picture of its state of development ... " could become the more neutral and (you've guessed it, more concise) "Press reports are unreliable ... ". Are they intrinsically pro-biased or did they toss a coin? All press reports have to make a story and neutral is not so exciting. Stephen B Streater 20:45, 13 June 2006 (UTC)

I'm in full agreement with noting that press is not reliably neutral, especially on this topic. However, are you or are you not agreeing with the sentence alleging that the media has misinformed people that SkyTran is in operation? No source, just an article to point a finger at. Fresheneesz 22:13, 13 June 2006 (UTC)

Stephen's version works fine, it's supportable from the sources (although one source got deleted with the ref tags). It is verifiably true that the press coverage is unreliable. Just zis Guy you know? 17:53, 15 June 2006 (UTC)

adding pod pic

I was thinking about adding the picture of a pod back in. Now theres a section about pods, and the picture would be nice as a visual. Comments? Fresheneesz 03:43, 13 June 2006 (UTC)

Personally, I don't think it needs it because

• There is not enough substance in the article to warrant two pictures.
• It makes it look more like a brochure with more pictures than words.
• The main picture includes a picture of the pod already.
• The pods in the two pictures are a different colour, which is distracting.
• Having two design diagrams and no actual photos is too unbalanced. Stephen B Streater 08:24, 13 June 2006 (UTC)

Seems like reasonable arguements not to have it. Fresheneesz 20:32, 13 June 2006 (UTC)

Two deleted lines part 1

Why was "In the past, the company has estimated a possible $1 million per mile once in mass production." deleted? The cheap guideways were a large part of the interest that developed for the project. Fresheneesz 20:17, 13 June 2006 (UTC)

I just read JzG's edit remark "This is speculation. There is no currently demonstrated likelihood of mass-production.". OF COURSE its speculation. It was labeled as an *estimate*. The estimate was not about the likelyhood it would be made, it was saying IF it was mass produced, it was estimated to cost around 1 mil per mile.

I'm going to add in a slight change of wording to reflect that some people might entirely misinterpret the meaning of the previous sentence:

"In the past, the company has estimated a possible $1 million per mile if mass produced."

Fresheneesz 20:31, 13 June 2006 (UTC)

NASA estimated the Hubble Space Telescope would cost about 400 million dollars. Hubble's actual costs are estimated to be: US expenditure between 4.5 and 6 billion USD and Europe's financial contribution at 593 million Euros. So how accurate do you think Unimodal's estimate will be? Remember a low estimate is more likely to encourage funding. Stephen B Streater 20:53, 13 June 2006 (UTC)

Just so. The actual cost of a civil engineering project often runs well in excess of the tendered cost, and this is not even a tendered cost - if it cost 10c per mile or $10bn there would be no comeback, because this is not a formal cost estimate. Just zis Guy you know? 21:10, 13 June 2006 (UTC)

NASA's esimate appears on that page. I am well aware that cost estimates of any kind most often are underestimates, and I'm not sure how they calculated that cost. But I think it is important to note that they did estimate this cost, at least for historical reference. The low cost was and is one of the main attractions to the idea. I'm not picky about how the estimate is constrewn, but I think it important enough to note. Fresheneesz 22:11, 13 June 2006 (UTC)

Can I assume you're going to head over to the HST article and correct the line, "From its original total cost estimate of about 400 million dollars, the telescope had by now cost over US$2 billion to construct." to read simply, "The telescope had by now cost over US$2 billion to construct." The "400 million dollar" statement is uncited and seems to have been pure speculation. In my opinion, rather than nixing anything speculative in nature, because WP is rife with speculation, I suggest we trust that the reader can reasonably infer speculation from the use of words like, "proposed", "theoretical", "concept", or "speculate". So instead of removing a potentially useful fact, maybe it could still be useful as, "Company literature speculated mass production construction costs as low as $1 million per mile."? --JJLatWiki 22:21, 13 June 2006 (UTC)

I agree. What is the role of Wikipedia? To make the judgement that $1M/mile is speculative? Or to report that the $1M/mile number is claimed by company literature but those claims are unproven? Are we to be making judgements about the how speculative the cost estimates are? Shouldn't we be presenting the claims and let the reader decide? A Transportation Enthusiast 01:53, 14 June 2006 (UTC)

The information itself is not neutral or reliable. Is it so biased as to make it irrelevant? I think this misses the the key point which is not some random figure, but that the two man pods are light and this (the designers claim) makes the rails cheaper than alternatives - which may also have dubious pricing. So perhaps we could find a cite for the actual point here - small pods give cheap infrastructure. Stephen B Streater 08:30, 14 June 2006 (UTC)

There are theories from various engineers and scientists that say essentially, "small pods give cheap infrastructure". Ed Anderson's "Transit Systems Theory" says so in great detail. But is the purpose of the UniModal entry to make that case or to present the proposal? Is it not a fact that the designer of the UniModal proposal estimated construction costs as low as $1m/mile? It's more appropriate to put the theory of "small pods, cheap infrastructure" in the PRT article and put the specific estimate in the UniModal article, because, presumably, UniModal took that general theory and presented something specific. But, in the offchance that UniModal actually gets built AND turns out to cost $75m/mile in mass production, the departure will be a significant element of the revised article. Obviously the assertion is not necessarily a fact, but it is an unbiased and neutral fact that the assertion was made. And do we not have a citation for that fact? --JJLatWiki 15:48, 14 June 2006 (UTC)

And if he claimed that this was the best system of transport ever, would you put that in? In picking points out for our concise article, we must pick out credible claims. Having 2 man pods is credible, small pods being cheap is credible, $1m/mile is not credible as it is a politicised number for making a point and raising finance. Of course, if you can point to other similar projects where he has come in at or under budget, this will make it credible, as would an independent report which comes up with the same figure. Otherwise, it is just marketing. We are not here to market the concept, nor to accurately report the designer's marketing of it, but to report on the design itself. Stephen B Streater 17:18, 14 June 2006 (UTC)

One thing we could do is offset the claim of 1 million with the current claim of 10 million. Just an idea. Fresheneesz 17:59, 14 June 2006 (UTC)

WP should not say, "UniModal is the best system of transport ever." But a cost estimate is not pure marketing propaganda, just like NASA's wild-ass claims that tricked Congress into funding the Shuttle and we'll probably find the same wild claims in a couple decades about NASA's proposed new systems. WP would be biased and promoting UniModal if it said, "The cost of construction is expected to be $1 million per mile of track.". WP is not biased or promoting if it says, "Company literature speculated mass production construction costs as low as $1 million per mile." The former is speculation, and the latter is a verifiable statement of fact. By the same token, it would not be helpful to readers if WP said, "Company literature claimed UniModal will be the best system of transport ever." Reporting a speculated specific dollar amount is not the same thing as reporting an irrelavant opinion. By the way, can you point to any NASA projects that came in at or under budget that suggests any dollar estimate they make should be documented in WP? --JJLatWiki 19:20, 14 June 2006 (UTC)

This is not complety unreasonable, but there is a difference between NASA and Unimodal. NASA has a legal obligation to make reasonable estimates, whereas Unimodal can say anything they like. NASA has much experience making working systems in space, Unimodal has no proven expertise. So, given the innaccuracies in NASA's estimates, which were published and open to analysis by independent experts, what faith can we have that Unimodal have any real grasp of the ultimate cost of making a system. In the UK, travel rights would cost a fortune. Are these taken into account? What about cost of compensation for noise? Foundation costs would depend on the quality of the underlying ground. Allowances for increasing the pod weight in future (to improve brakes, for example, or aircon) could significantly increase the cost. I would be surprised if their figure was even within a factor of ten of the true cost if this system is ever built. NASA at least managed that, as did Eurotunnel who were the expected 2.5x out in their large civil engineering project. To me, their figures are just not credible. I think some might call them fantasy, though I prefer guesswork. Not up to my standards for WP then. However, I am still willing to concede an earlier point: whatever guide rails cost, lighter pods will make them cheaper than heavier pods, and this is an advantage of the Unimodal design. Stephen B Streater 22:48, 14 June 2006 (UTC)

As usual, I agree with Stephen. The fundamental problem with this estimate / guess / made-up figure / whatever is that it has received no impartial scrutiny, and no comparable system exists, so we have no basis for judging its accuracy. Just zis Guy you know? 23:10, 14 June 2006 (UTC)

Well, it's obviously your article so I'm not going to belabor this point much longer, but I see no need to judge the accuracy of the claim. Like the dollar claims for the Freedom Ship, or Branson has suggested an initial ticket price of US$200,000 for a suborbital tourist flight in Virgin Galactic, the fact that the estimate was made is verifiable. Or is Branson's claim allowable because 1) he has a history of making accurate estimates, 2) he's not a nut, and 3) his estimate has been submitted for peer review? --JJLatWiki 15:13, 15 June 2006 (UTC)

"it's your article" - exactly. JzG has assumed ownership of all the PRT articles and assumes absolute veto power over all changes. He generally doesn't consider the evidence we present when enforcing his policy that the articles should reflect his skeptical view. And he also believes that this overtly skeptical tone is a "general consensus", even though there are several of us who have expressed concerns.

Does anyone want to take this to a vote so we can show once and for all that several of us believe that the unanswered skepticism in the PRT article is unbalanced, and unsupportable with reliable sources? A Transportation Enthusiast 16:50, 15 June 2006 (UTC)

JzG is well aware of WP:OWN. This is the Unimodal article, not the PRT article. It is a small pod on a stormy sea, and one of the dangers of keeping it was that a there are only a handful of editors interested in it, and most of those are believers. There is some substance to this design, but also some hype. The references are inconsistent is a number of places, showing that the details are often just illustrations, not cast iron facts. If it becomes impossible to keep the article on the rails, it will probably be deleted again. I prefer a more concise and accurate article with links to the more speculative elements, such as pricing. PRT is more concrete and deserves a bigger article. I may help over there again when we have reached consensus on Unimodal. Stephen B Streater 17:03, 15 June 2006 (UTC)

Plus, polls are evil, especially when conducted among a group of editors so small as to make any result meaningless in terms of the project as a whole. Stephen, you have it exactly right. It's not that I am somehow "anti-PRT" but that I am strongly opposed to speculation in Wikipedia articles. I've seen this before of course: the supposed compromise being offered is the false centre fallacy - scepticism is the middle ground in this dispute, and to take an average between that and one extreme and call that the centre is clearly wrong. Just zis Guy you know? 17:48, 15 June 2006 (UTC)

You keep repeating the same tired arguments, JzG. Now, polls are evil, just like anything that could challenge your infallibility is evil. Here now we have four separate editors who have been making the same points about your attitudes on these pages, and yet you still treat it like your view is the consensus! Unbelievable! In any event, I think it's time for us to start making the changes in spite of your dictatorial actions here; if you disagree, then we'll just have to revert as a group, and if you continue to disagree, then you can take it to mediation. It's a shame it has to come to this, but I don't think I've ever encountered anyone as stubborn as you JzG, even when we've presented reams of evidence that your position is untenable. I'm through trying to negotiate this. A Transportation Enthusiast 18:12, 15 June 2006 (UTC)

For values of "tired arguments" that may encompass arguments I'm tired of having to make, because your refuse to accept that Wikipedia is not a soapbox so unverifiable speculation has no place here. The "polls are evil" page has a long and distinguished history, and I have had absolutely nothing to do with it at any point. You have not actually provided reams of evidence, you've provided a couple of references back to a single source and then reams of argufying, which is not the same thing at all. As Stephen has said in other contexts, if your evidence is rejected, instead of repeatedly restating it, bring better evidence. Just zis Guy you know? 18:36, 15 June 2006 (UTC)

I'm not the one on the soapbox, you are. Your view is that PRT is a Quixotic dream, and you will use Wikipedia to promote that view. Your assertion that the skepticism is the "majority view" on PRT is unsupportable. If skepticism is the majority view, then why can't you provide reliable sources of all this skepticism? Or debunk the verifiability of the sources that describe PRT? Or at least admit that your claims about the predominance of skepticism are baseless? A Transportation Enthusiast 19:31, 15 June 2006 (UTC)

This is not the PRT discussion, but the Unimodal one. And in the case of Unimodal one, no one has financed the project. This means that everyone with money is sceptical. Not a single entity has put up the money, and they've had many years to do so. This looks like the majority view to me. Stephen B Streater 20:33, 15 June 2006 (UTC)

I don't know where to begin with this new theory: "Funding is inversely proportional to scepticism" So since the vast majority of cities in the U.S. have NOT funded a Light Rail Transit system, the majority view of LRT is scepticism. The vast majority of cities have NOT funded office buildings taller than 20 stories, therefor the majority view of such buildings is scepticism. And doubting someone's claim is not the same as disputing someone's claim, which is what JzG's position appears to be. --JJLatWiki 21:12, 15 June 2006 (UTC)

But you have to agree it looks bad from the "majority is not sceptical" stand point when no one has funded Unimodal. Funding is related to scepticism. Given what we know about prevalence of cars in the US, I would say that the majority is sceptical about light rail there, but a few bodies are not, which is why there is some. If you want something where the majority is not sceptical, look at cars in the US. Or guns. Stephen B Streater 21:29, 15 June 2006 (UTC)

But I never said that the majority is not sceptical, and I wouldn't take that position. Anyone who says, "scepticism is the majority view" has made the positive assertion and therefore has the burden of proof. When a city declines to install a particular amenity, such as LRT or PRT or an HOV lane, that decision doesn't represent a majority view either. It means that at least a simple majority of the 9 people on the city council members were sceptical that their political futures could survive a vote to install said amenity. So at least 5 people in a city of 500,000. Not a clear majority. But, I have firsthand knowledge of one such city council member, who was part of a joint partnership that included a high-ranking police official and my uncle, who voted in favor of LRT in his city, and lambasted every PRT proponent who even uttered the sound 'P'. That joint partnership, through no small coincidence, happened to buy several large areas of land along the new LRT route shortly before the city council approved the LRT routes. Luckily, a lot of the land was in the poorest part of the city so it was purchased cheap and the value skyrocketed and man, did we ever have fun with that new ski boat and those snowmobiles. --JJLatWiki 23:30, 15 June 2006 (UTC)

Would they have made so much money if they had bought land along a Unimodal route and voted for Unimodal? That would have made some headlines. Stephen B Streater 06:18, 16 June 2006 (UTC)

It was their shared opinion that LRT presented the best option for personal profit. All the PRT designs that were put on posterboard for the board meetings made it a keynote that they would require very little dedicated land and would often utilize existing public land, rights of way, and only small easements. But they knew PRT didn't stand a chance of actually being built, so they didn't really care about it. And the anti-PRT voices were loudest and probably (in their opinion) accelerated the decision to build the LRT. --JJLatWiki 15:04, 16 June 2006 (UTC)

JzG, EVERY single estimate is speculation. Even if it comes in under the estimate, they still speculated. And I would bet that in a project the size of even a small transit project, they failed to account for everything you can think of. So you might as well remove all estimates from all articles and only include actual "to date" dollar amounts. Maybe if all those Angel investors had you as their advisor, the Dot-com bubble would not have happened, much less burst. But like I said, it's your article, so if you reject someone else's evidence, I guess it wasn't good enough. --JJLatWiki 19:35, 15 June 2006 (UTC)

In the UK, people making estimates when they raise money from the public are personally liable for reckless estimates, so the fear balances out the greed somewhat. On the other hand, there is no incentive for Unimodal to publish accurate estimates. They have already increased their cost estimate by ten times - on something the company controls entirely - the cost of their design. The dot coms mostly overestimated sales - an external feature, which they didn't control. So although they were wrong, this is less bad. Stephen B Streater 20:33, 15 June 2006 (UTC)

Two deleted lines part 2

I would argue that NASA has more experience creating fantasy estimates and they have more credibility because they are a government agency that is required to present their fantasy to another government body who themselves routinely create fantasy estimates in order to gain financing for their proposals. NASA's estimates should be documented for posterity as should the estimates presented for any other proposal. Maybe someday someone will read this entry after a UniModal PRT is built for $75m/mile and hold their feet to the fire and ask, "Whatever happened to the $1m/mile?". But, are you assuming UniModal based their estimate on the "lighter pod - cheaper guideway" theory, or is that verified? --JJLatWiki 23:17, 14 June 2006 (UTC)

I see both points, and It would be good to remember that any organization, goverment, private, or public, all have adgendas, and bias cannot be outlawed by law - despite many tries. Not to mention, the creator of UniModal isn't just some crazy person, he has years of real-world experience building machinery and transportation vehicals - including the California Commuter. Everyone has bias, but lets not say that UniModal has no credibility when its designer holds two official world records for efficient transportation.

Btw, what are "travel rights"? I tried looking them up, but didn't find any definitions. Is it like Rights of way? Fresheneesz 00:31, 15 June 2006 (UTC)

Yes, I was thinking of rights of way. The light pods -> cheap guideways is not quantitative, so more likely to be reliable. Stephen B Streater 06:33, 15 June 2006 (UTC)

Suppose Unimodal is approved. They may well insist on extra regulations - the guideway supports, many of which are along existing roads, must withstand a collision from a 40 ton HGV for example. Is this cost inlcluded in the Unimodal estimate? How long with the tracks last? If they have to be renewed every month, then this is much more expensive - cost per mile per year may be a more sensible figure. So I'm not clear what the figure given by Unimodal actually means in practice. Perhaps I'm almost saying it is meaningless, and that is why I'm happy without it. Stephen B Streater 08:18, 15 June 2006 (UTC)

Yes, precisely. We have no idea what the real costs will be because we don't even have a prototype - what happens if (like the famous bridge in London) some unforeseen harmonic vibration effect requires a complete redesign? What about the fatigue life of aluminium? There are so many questions which have yet to be asked, let alone credibly answered, that including speculative figures for costs is indefensible - we don't even have enough discussion of them in the technical press to be able to quote a range of informed estimates. As to whether Malewicki is credible, I refer the hon. gentleman to Eric Laithwaite, a man of no little reputation (inventor of the linear induction motor, on which inductrack is based, as it happens) who was nonetheless prone to championing quixotic and occasionally downright wrong ideas. To state that this system is credible because Malewicki has experience in other transportation projects is an appeal to authority, a logical fallacy. Just zis Guy you know? 10:43, 15 June 2006 (UTC)

So you dispute delete every dollar estimate in WP that you suspect hasn't taken all factors into account? Talk about quixotic. Suppose UniModal IS built, and it turns out to cost $50m/mile in material, equipment, and labor, and another $5m/mile for rights of way, environmental impact studies, construction insurance, and bribes. And then another $20m/mile to strengthen the guideways and supports because they didn't take into account the vibration and added fatigue of a pod that had to be heavier to accomodate motorized wheelchairs. Do you think the poor saps in the test city are going to stop using WP because this article said the company estimatesd $1m/mile? Maybe the financial canyon between the fantasy and reality will dissuade other cities from considering PRT. Then again, that doesn't deter congress from financing every NASA fantasy. --JJLatWiki 15:13, 15 June 2006 (UTC)

The figure is speculation. If you ever wear JzG down and there is a consensus to include a figure, perhaps "Unimodal speculates that costs will be ..." would be appropriate wording. However, my personal preference is to include links for those people who are happy to read through the unverified interviews for a full range of speculation. If the AfD had been a clear keep, I would support a larger article containing more of these views, but the article only survived at all by a whisker, and I think there is no harm in keeping it small. This is different from Hubble, which of course is up in the sky and working. Stephen B Streater 16:47, 15 June 2006 (UTC)

Yes. It's speculation whose authority we can't state, in fact, since it appears not have been discussed widely enough to say whether it is a guess, an estimate, a projection, a budgetary figure, or indeed just a figure Malewicki plucked from the air. That's the problem: we don't even have sources for how (in)accurate it is. Show your working, as they say in the test papers. Just zis Guy you know? 17:36, 15 June 2006 (UTC)

Two deleted lines part 3

To answer JJLatWiki's various points above, the HST estimate was presented to Congress as a basis on which to back the project, and the project was commenced on that basis. So: when UniModal has been ordered on the basis of Malewicki's estimates, then we include the cost estimates. And when it's built, we include the true cost and comment on the difference, just as with HST. Virgin Galactic is financed, has licensed intellectual property rights to a vehicle which is already in working prototype, and has a test schedule (which will likely slip, of course) - where is the protoype and test schedule for UniModal? And the finance? I can't speak to the accuracy of the estimate for Freedom Ship other than to say that cost estimates for conventional shipping can be expressed in dollars per metre length, to a reasonable approximation; the article clearly states, though, that current construction techniques are inadequate for a vessel of this size. It all seems consistent to me. Just zis Guy you know? 18:40, 15 June 2006 (UTC)

I belive the arguments against putting the figure in are: we mislead people by doing it, we elongate an article on a "non-notable" subject, and $1 mil might not be reliable. Arguments against those are: we don't mislead people if we say its an estimate or "speculation", one line isn't much elongation, and its just as much a reliable estimate as any other organization.

As for the crazy and wrong ideas some inventors have had, I think most inventors and most organizations in fact have had much the same track record of supporting outrageous ideas. NASA is working on propulsion breakthroughs using non-standard methods, they're working on going to mars and the moon again, many of their projects have failed. Any good organization will fail when they explore new areas. I'm not arguging that malawiki is credible just because hes gotten a few records and has lots of experience, i'm just saying hes as credible as NASA when it comes to crazy new ideas. Fresheneesz 18:45, 15 June 2006 (UTC)

JzG, is it safe to assume then, that you're going to delete Branson's $200,000 claim under Virgin Galactic? It was wild speculation made purely for the purpose of raising financing and there is no indication that he has considered regulatory requirements, if he's considered structural fatigue, insurance costs, etc. If you need any help finding more speculative dollar estimates elsewhere in WP, I'll be happy to help you in your quest. --JJLatWiki 19:14, 15 June 2006 (UTC)

(1) No, it's not "just as valid" because we have no basis for establishing its validity. (2) No, because they have a prototype in the air (see expanded comment whic was edit coflicted); I will, however, remove the figure form the Freedom Ship article if a credible source for it cannot be cited. Just zis Guy you know? 19:39, 15 June 2006 (UTC)

So now: 1) the Freedom Ship estimate is to be deleted because even though "cost estimates for conventional shipping can be expressed in dollars per metre length, to a reasonable approximation", you can't personally verify the estimate as accurate, 2) Branson's $200K claim is left in because initial development is financed and they have a working prototype therefor you assume they have factored in all the necessary elements required to establish a reasonably accurate estimate, 3) NASA's estimates for HST and STS and all future proposals are left in because, even though they have a history of wildly inaccurate estimates, their estimates are scrutized by...by...by the people who will syphon off some of that funding to build a bridge to nowhere, and 4) statements that UniModal has even made cost estimates are to be deleted because you think those estimates are unreliable. This ruleset is going to be hard to maintain. --JJLatWiki 23:04, 15 June 2006 (UTC)

Did noone like my idea of putting in both estimates of $1 and $10 mil? Here let me formally propose a line:

" In the past, UniModal has speculated that the guideway would cost $1 million per mile if mass-produced. Now UniModal is instead saying $10 million per mile."

Comments? Fresheneesz 18:45, 15 June 2006 (UTC)

JzG, is it safe to assume then, that you're going to delete Branson's $200,000 claim under Virgin Galactic? It was wild speculation made purely for the purpose of raising financing and there is no indication that he has considered regulatory requirements, if he's considered structural fatigue, insurance costs, etc. If you need any help finding more speculative dollar estimates elsewhere in WP, I'll be happy to help you in your quest. --JJLatWiki 19:14, 15 June 2006 (UTC)

I like that. Except I would say, "More recently, UniModal's speculated construction cost increased to $10 million per mile." Although I haven't seen them, I assume the UniModal or SkyTran web sites have stated those numbers in order for you reference them. --JJLatWiki 19:14, 15 June 2006 (UTC)

and I hate it because it compounds the fault of citing an unreliable figure by adding original research. Just zis Guy you know? 19:39, 15 June 2006 (UTC)

(1) No, it's not "just as valid" because we have no basis for establishing its validity. (2) No, because they have a prototype in the air (see expanded comment whic was edit coflicted); I will, however, remove the figure form the Freedom Ship article if a credible source for it cannot be cited. Just zis Guy you know? 19:39, 15 June 2006 (UTC)

This idea of quoting $10m/mile concerns me. A few days ago, the article said $1m/mile, a figure which I thought was speculation, and could easily be ten times out. Now it turns out that the designers themselves have increased their estimate - by ten times! And they haven't even started building it yet. This just reinforces my view that they are just not a credible source. Is the system design precise enough to put any figure on it? I suspect not, but if someone can point me to the source which says what assumptions this new figure is based on, it would be a start. And it would be good to know which assumptions changed between the two estimates. If we had a reliable source, or even an independent source, we wouldn't need to be so sceptical. For comparison, Virgin Galactic has numerous sources and wide coverage, allowing criticisms to surface. The HST had pages of specification when they made their bid for the cash. Stephen B Streater 20:22, 15 June 2006 (UTC)

The other thing that concerns me is the movement towards personal attacks. This is not appropriate here. Everyone wants the best article possible. Stephen B Streater 20:22, 15 June 2006 (UTC)

Finding sources on ones own is not the "original research" that NOR is against JzG... again and again, I don't understand why you insist on giving irrelevant one-liners meant to misinterpret the debate.

Malawiki has said that they now quote $10 mil/mile to be more credible - as people seem not to believe $1 mil/mile estimates. I don't advocate putting this in the article (anymore) because the only place I know of that I can source is my e-mail inbox.. but its just a side note. My personal guess is that 10 mil was chosen based on it being 10 times the original estimate, and nothing more. Fresheneesz 21:36, 15 June 2006 (UTC)

My mistake - it loked to me as if you were using the ten-times-as-much estimate suggested elsewhere. The fact that Malewicki has just changed the estimate by an order of magnitude does rather indicate that he doesn't have a solid basis for the figure, doesn't it? Seems we were right to be sceptical; let's wait until he has a firm figure which has been peer-reviewed and its accuracy established from reliable sources. Oh, and actually NOR does apply to theories advanced from a single source, even if that source is off-WP. That's the main reason we deleted Aetherometry. Just zis Guy you know? 07:13, 16 June 2006 (UTC)

Can we please place our comments at the end of the section? I find it very difficult to find new comments placed up there in the midst of old comments. Fresheneesz 21:36, 15 June 2006 (UTC)

Perhaps you could ask to look at the assumptions made in this calculation. Did he publish a proposal with this in? It seems slightly strange that if low cost was his selling point, he didn't just get independent verification of his estimate. Anyway, if we are letting this settle for a bit, are there any outstanding issues before we all move over to PRT and break the deadlock there. Stephen B Streater 21:48, 15 June 2006 (UTC)

I'll agree to let this issue take a raincheck. We can bring this up at another time, and I'll see if I can dig up more information on the estimates. Fresheneesz 22:41, 15 June 2006 (UTC)