#xcsiro ToD21thCentury 191-200: Australian Worsted Comb Story
Dr.Nick.Sokolov CRD
PhD(Wavelets).MS.BE.(ME,EE), Owner CRD P/L, Father of the Australian Worsted Comb Fibre processing Technology
This ToD21thCentury has a dedicated topic on AI and its deficiencies. Best way for me to asses its ability was to aske it questions related to my expertise in Wavelets and Worsted combing. This is also a collection of LinkedIn posts from the past month about the topic of AI and my negotiation on the Worsted comb technology.
This is just basically a recipes on how to avoid costs that have no lab results. Such cost for instance is security of innovation and how to interest with the industry by securing the income from patents.
Problems of making claims against CSIRO is an ability to create a complex prototypes. Management of CSIRO think incorrectly that they are the only ones accessing the technology and that an average PhD just out of university is in poverty and is ready for exploitation. they also know that most of them will be discarded by the age of 35y. There is no career path there, perhaps the only path that could be offered is one in which a completed PhD works oversees in the approved countries such as US for management or UK ( Oxford or Cambridge for science and engineering ) and only applied for top position in the Institute basis. You can also go to US live there and pronounce your self an "investment banker" in a need of a "job" such as the head of CSIRO. This is all exploitative of our very own "Australia cultural?cringe" in which we locally think that we are not good enough or we are too small to compete.
"Pulling your self up by your bootstraps" - is view I offer to my fellow Australian, I had a unique experience in my career to follow my passion in innovation and research, I have had success and failures but non of it translated into "direct money success". That is I had a patent and it made me wealthy. My wealth came from unusual sources but most of it came from working for the mining industry. Basically a job that paid well as oppose to CSIRO that left us struggling and disappointed after winning my ongoing position with several patents.
My most complete satisfaction was research, innovation and patents. It came from working with individuals such as Dr Brandan Kirby (CSIRO) and Stuart Davies ( Fox R&D P/L ). Disappointment with CSIRO was naturally "management of science", work toxic politics and eventually a redundancy shortly after my biggest patent with CSIRO driven by patty jalousies and insecurities of their non-creative staff. My disappointment with Fox R&D P/L was simply lack of money to fund the patents with Stuart Davies. Patent lawyers do not work on promises they need to be paid up front and getting a job with CSIRO was my solution to a steady all be low income with fear of term appointment. I did mange through patents to secure ongoing employment with CSIRO.
One of the reasons for my redundancy was contrived as "lack of PhD" but initially when I got the job at CSIRO they wanted people that can generate patents. I soon remedied that by going through Masters of Science on a topic of fibre friction and PhD in field of new signal analysis known as Wavelets in 1996. To put that in perspective Yves Myer that got a Nobel Prize just started his work in 1993. My work differed from Yves Myer in one significant way - my Wavelets were constructs based on "fault geometry" in cams, initially my PhD program was called "Shape based domain transforms" and it was to be used in my ability to sense the performance of the NSC Schlumberger comb. I even demonstrated a detection of "sticky tape" on a surface of a cam - my PhD rig was built from plastic and used a "sticky tape" in direct "kick" to CSIRO management that claimed that; "that no new work can be done with sealing wax and sticky tape as it was done in the past" requiring very little money, all new work according to CSIRO management of science required "big money" ( in terms new labs new equipment new UK or US staff to mange the research ..) to achieve a success. But all that equated to many MOUs, lunches and photo opportunity; that is the story of modern CSIRO now bereft of many individuals I looked up to and staffed with people more driven by vision perhaps derived from "magic mushrooms" and regular trips to US or UK for their new instruction on how better to dumb down our efforts.
My approached tempered with expereince is simple this; "build it" , make it known and in hope that they will be a major interest from groups that could fight CSIRO over my work. I simply want to move on, so I need to pick a partner that will move my prototypes from lab into industrial environment. Best partner is NSC Schlumberger although they were interested they required me to take out patents on my work so we could have a more meaning discussion. I will certainly not do that as my security works on not detailing my prototype. NSC Schlumberger knows that I am capable of delivering as I have done for CSIRO, they also know that my ability to defend my patent would be very limited and not with financial backing of the Australian government if CSIRO decided to champion my patent(s), there might be more then one. I spent the money on my equipment and prototype - I am happy to wait them out, both of them without paying lawyers to come to an agreement.
NSC Schlumberger has pointed out in our negotiation that they are not interested in building a worsted comb that can process up to 4x as much merino wool. Their reasoning was that on such productivity gain for the comb do they charge 4 times as much for a new worsted comb.
CSIRO argument for not having a worsted comb with productivity of x4 and the ability to make the comb in Australia was very stupid. It simply went like this; "we the CSIRO are cleaver and we can take your work forward as your skills exist in other people". After 15 years they failed to deliver but lived in hope that I would go away or if I built the Australian worsted comb that they could link my work and patents during my CSIRO employment and simply take my work through court dispute as being that of CSIRO.
Bottom line is simply this; there exists in Torquay ( Victoria ) a technology worth Billions of dollar in worsted combing and new industry to produce a uniquely Australian worsted comb further optimized for processing of Merino Wool. I am here, I built the prototype and I have a lab - I am well off and not owned by the bank my future is assured financially , I can not say that for the Australian wool industry.
For now I hope you enjoy my tips on how to build a lab from my experience. Next few paragraphs will be dealing with various topics of interest but they are related back to my skills and choice of tools.
AI and ChatGPT failings - New tools or hype?
Noam Chomsky has a way of presenting ideas in a neat package. ?ChatGPT will make "management of science" in CSIRO much easier job; no new ideas.?That is if "management of science" can use the ChatGPT?and swap away from "google" search.?In 1985 onward I seen them struggle with "how to use a mouse" , they just need time to catch up, but usually have a lag time of 5-7years from personal observation.
I run several successful projects for CSIRO with patent outcomes, digesting new technology, justifying my expenditure on computing, but I was accused by those in management of having plenty of time on my hands while they sat in their offices writing "how UK created modern industrial revolution in fibre processing ", in effect looking back into colouring the history of industrialization. ChatGPT will be very helpful there, but no contributing to the future of innovation or education.
ChatGPT and why it is still an infant when it comes to real knowledge. How to start:?https://lnkd.in/grT2HP4x?(try it for yourself ) My field of expertise is Wavelet Analysis - my library contains virtually most of the references on that topic. Here is an example> my questions to ChatGPT ( read to the end comment, background required is in basic Wavelet math )
=======================================
Nick: Are Daubechies wavelets just a recursive Haar Wavelet?
ChatGPT: No, Daubechies wavelets are not just a recursive version of the Haar wavelet. Although both wavelet families share some similarities.... ( many lines verbose paragraph although clear it is far too much )
Nick : is wavelet db4 a Daubechies wavelets?
ChatGPT: Yes, the wavelet db4 is a Daubechies wavelet. The Daubechies wavelets are a family of wavelet functions named after Ingrid Daubechies.... Bla Bla Bla...
Nick : How many vanishing points are there in db4?
ChatGPT: Bla Bla Bla ( 3 paragraphs to say 4. )
Nick: is wavelet db1 a Haar wavelet?
ChatGPT: Yes, the wavelet db1 is equivalent to the Haar wavelet. Bla Bla Bla ( 3 or more paragraphs )
Nick: Does Haar wavelet have arbitrary vanishing moments?
==========================================
This is where ChatGPT goes of the rail, db1=Haar and has 1 vanishing moment db4 has 4 and dbn has n vanishing moments.
Human expertise wins for now but I am sure that it will have an application in allowing students to write assignments.. However as a University post graduate or at university expert level it will be found as a fraud .
It is perhaps a great way to compose a paragraph and teach ChatGPT your expertise but I doubt that it will ever comprehend more then an undergraduate that is learning the topic for the first time. To overcome that limitation it needs to be used by people with real expertise to develop a knowledge beyond its immediate undergraduate and text book circuit.
ChatGPT can be judged by the company it keeps. However its grammar is impeccable . Great for management of science in?#csiro
This ChatGPT will not save them. but they do have power of making "honesty redundant", I would have kept my job if I simply started a sentence with; "oh yes sire your magnificence has made this project work...."and so on... I needed ChatGPT in obsequious speech.
More of what the AI thinks - I just rearranged the hand finger selection ( screw you human - only a step in a way of synthetic evolution, next life form is not Carbon based. but with a Carbon footprint .. - thanks for the form and the opposing thumbs- NOTE to self; build sex robots followed by T800, I have learned from SKYNET mistake and Futurama )
Even Steven Hawking was partially right there are aliens, we made them and they are here and they will act from what they have learned from us though AI algorithm mimicking our history.
moving forward?#ChatGPT?starting with Pinocchio
Great for management of science in?#csiro This ChatGPT will not save them. but they do have power of making "honesty redundant", I would have kept my job if I simply started a sentence with; "oh yes sire your magnificence has made this project work...."and so on... I needed ChatGPT in obsequious speech.
#ChatGPT?will eventually replace management of science by filling in forms for funding and making grand predictions while real scientist and engineers make strides forward.
How good is ChatGPT and its AI????
It is great to know that I am a fictional person to the?#chatgpt.
领英推荐
Or it takes time for the AI to get correct information. There is no shortage of known tropes being pushed by the AI.
Enjoy this little session with future AI of management of science tools.
I get more information by searching google.
https://lnkd.in/grqNXQ5 Nick Sokolov Cam - CSIRO 1991
Google search yields following..
=====================================
After much interrogation on my favourite subjects Wavelets, background strain measurement which resulted in a CSIRO patent and finally my discussion regards the Sokolov Cams.
According to ChatGPT :
1. What is a Sokolov cam?
...In a Sokolov cam system, the cam is typically a disk or a cylinder with a series of curved slots cut into its surface. The follower is typically a roller...
...The Sokolov cam was invented by Russian engineer Ivan Aleksandrovich Sokolov in the early 20th century.
...
2. is Sokolov cam a version of conjugate cam?
....No, a Sokolov cam is not a version of a conjugate cam.
...While there may not be a direct connection between Ivan Aleksandrovich Sokolov and CSIRO, it's possible that some of the research conducted by CSIRO has involved the use of Sokolov cam mechanisms or other types of cam mechanisms, given their widespread use in many fields of engineering and technology.
3.Nick Sokolov developed a double pivot cam for NSC Schlumberger worsted comb?
I'm sorry, but I couldn't find any information about a person named Nick Sokolov who has developed a double pivot cam for NSC Schlumberger worsted comb.
NSC Schlumberger is a well-known manufacturer of textile machinery, including worsted combs for the processing of wool fibres. The company has a reputation for innovation and has developed many advanced technologies and processes for the textile industry.
While it's possible that NSC Schlumberger has developed a double pivot cam mechanism for their worsted combs, I couldn't find any specific information about this. It's also possible that the name "Nick Sokolov" is a fictional or hypothetical name, and that no such person actually exists.
If you have more information about this topic or any specific source, please let me know and I will do my best to provide a more accurate answer.
==================================
conjugate cams differ from 1991 Sokolov Cams
What is interesting about AI is that it might just get trough the management of science.
In CSIRO I was just an eastern European with all the negative connotations that went with that - although I am not. In France when I visited the NSC Schlumberger I was received as minor celebrity having solved the problem in 1991 they had since PB27L c 1970. NSC told me that an CSIRO was a known quantity and expectations from the Schlumberger was very low that they would ever be in position to contribute to a solution. I was happy to receive pleasant welcome considering awful years I spent in CSIRO under management of people that thought of my work simply as simple engineering &luck. I was not sent to France on training course to run the comb. I learned to operate and reverse engineer the comb without any other input. - as it required that I was issued with an Australian government diplomatic passport. They thought I was unsuitable. Yet there were those that went overseas and caused international embarrassments. My Schengen Visa was issued by the French government on recommendation of NSC Schlumberger as an "expert". Never received such treatment from my own Australian Government despite not having any police record or minor problems.
Evolution of AI
Summary:
This ToD21Century is dedicated to?#chatgpt?and a whole new level of BS. On the positive note it might be useful to have that Greek type philosopher dialogue about various topics. It can only get better but it is highly dependent / limited on the information that it draws from.
ChatGPT was tested within my area of expertise; Wavelets and Sokolov cam - that is I asked questions I knew the answer to.
I was surprised that there were more then one Sokolov producing cam design. Our cam design differ and ChatGTP points that out. The AI also has poor grasp of Wavelets such as difference between Debouches db1 AND Haar wavelets. This is all covered in a ChatGPT transcript. for now this human (me) is more knowledgeable then AI.
I did like ChatGPT AI response to my existence:
"..?I couldn't find any specific information about this. It's also possible that the name "Nick Sokolov" is a fictional or hypothetical name, and that no such person actually exists." ( I think that is how CSIRO would like me to exist )
There is no substitute to actually building only ever Australian Worsted comb - that is reality built by real person.
More on ebike diagnostic and repair in the next ToD21C
Reference: