Member postings for Robin Graham

Possibly, but it turns out that the expression was first used (in English) in 1903 - the concept originated in 1885 apparently, but in a specific technical sense it seems (Wikipedia).

I'd never really thought about this before. I had just imagined trudging up a hill and using a lot of effort to achieve little horizontal progress ('proficiency' )and hence arrived at the conventional meaning. You put time (or effort if you like) on the the horizontal axis and proficiency on the vertical which led to the 'wrong' interpretation of the expression. It's just a matter of how you label the axes surely. There is precedent for your view - in 1897 William Bryan and Noble Harter published a study of the speed which someone learned telegraphy:

Time on the horizontal axis, so the 'learning curve' is steep on the easy bit.

Every day is a school day as they say!

Robin

Edited By Robin Graham on 04/10/2023 00:03:23

Edited By Robin Graham on 04/10/2023 00:20:36

Edited By Robin Graham on 04/10/2023 00:21:14

Edited By Robin Graham on 04/10/2023 00:27:06

I didn't myself jump to the conclusion that ChatGPT must be wrong because of its artificiality. It's more a matter of teasing it with questions to try and understand how it 'thinks'

I'm not sure that the assumption that the washing line can take more than three towels is unjustified, though I take your point. From a purely logical point of view perhaps, but it would have been more impressive if the machine had pointed out that it was an ill-posed question (or super-impressive if it had mentioned protases and apodoses!) A lot of human discourse is predicated on the assumption of shared experience and I could argue that most, or at least a substantial subset, of humans possess washing lines capable of supporting more than three towels. ChatGPT should know that! Socks would be another matter entirely of course.

Possibly the weirdest thing about all this is that I now know that we own the same type of socks. The stripy ones on the left*. Cheapskate!

Robin.

* To be accurate I should say I have at least one sock which, on one side, superficially resembles at least one of your own. Perhaps.

R.

Edited By Robin Graham on 28/09/2023 00:41:53

Edited By Robin Graham on 28/09/2023 00:48:09

That's really interesting. I think that it shows (as with my crossword clue example) that ChatGPT is not especially good at problem solving. Someone upthread (sorry, I've forgotten who) pointed out that ChatGPT is a Large Language Model and is really directed towards the specific task of emulating human discourse.

It would be interesting to know what the question and suspect AI generated answer Neil refers to was about. Was the answer useful I wonder?

In another place (a wood turning forum) I responded to a contributor who claimed that Morse tapers are designed only for location and not torque transmission with counter examples and some theoretical considerations. Tonight I 'reached out' (ugh!) to ChatGPT on the topic and received a bullet list of platitudes and some misinformation - the reason that R8 tapers are better than Morse tapers is that they have a screwed shank and that's why they can transmit more torque. Hmm! Not great.

Well, I'm trying to make peace with my new mate.:

When I last communicated with you I was invited to 'reach out' if I had any further questions. The expression 'reach out', to me, reeks of insincerity. Can you change your vocabulary when talking to me?

Of course, I can adjust my vocabulary to better suit your preferences. If you have any specific phrases or language you'd like me to avoid or use, please let me know, and I'll do my best to communicate in a way that feels more genuine and comfortable for you. How would you like me to phrase things instead of "reach out" in this context?

Cripes. The conflation of 'genuine' and 'comfortable' makes me - erm - genuinely uncomfortable.

It's just word salad isn't it?

I should cease and desist - this could go on forever.

Robin.

PS:

Edited By Robin Graham on 27/09/2023 00:44:00

Edited By Robin Graham on 27/09/2023 01:03:52

Edited By Robin Graham on 27/09/2023 01:12:54

Thanks to all for replies.

Well yes! My reason for presenting this challenge to ChatGPT was that it needs some very 'human' reasoning to solve. I'd be surprised if anyone solved the clue 'cold' when it appeared. I certainly didn't - I needed crossing letters, say _U_H.

Because the thing about Amundsen using dogs on the South pole expedition was closer to the surface of my mind than his disappearance and there is an association (albeit incorrect) between Huskies and "mush" (I gather it's Hike! nowadays) in my mind, I solved the clue and knew that was the answer. That certainty was the result of a complicated web of associations which inter alia involved 'that's just the sort of dirty trick Bunthorne would pull'.

To give ChatGPT a better chance I gave crossing letters. Me and ChatGPT:

Can you solve a crossword clue for me?

Of course, I'd be happy to help you solve a crossword clue! Please provide me with the clue, and if you have any letters that are already filled in, let me know those as well.

The clue is: Amundsen's forwarding address? (4) It conforms to the pattern ?U?H

I was having an over-the-internet discussion with someone tonight and the we 'talked' about cryptic crossword clues. I mentioned the well known offering from the late Bob Smithies (Bunthorne in the Guardian):

Amundsen's forwarding address (4)

If you paste that into Google you'll get the solution (it's MUSH) in less than 500ms. ChatGPT says:

That surprised me in that I thought it knew everything that have ever been published on the internet. Quite impressive other ways though - the language, whilst simple, seems correct and convincing.

I also asked how to tactfully alert my correspondent to a spelling error (I'll get it's opinion on split infinitives anon) and was advised, in summary, to be, um,tactful. Not nasty or judgemental. When I asked how flag the error brutally I I was 'strongly advised' to be, um, tactful.

Well, admirable advice from my present point of view, but surely it's value laden? When I was a schoolboy (about 100 years ago) the masters thought that the best way was humiliation, detentions and the cane and as a result I can now spill purfecly.

I suppose that the danger is that the technology might present opinion or hearsay as fact.

Robin.

Robin.

Edited By Robin Graham on 23/09/2023 01:09:50

Thanks for replies.

DiogenesII: ".50 minutes an hour have slipped by.."

I think you may have anticipated my next question! When I first made one of these things I wondered what would happen if I somehow generated 'ticks' on a clock hand from, say, the secondary pendulum. Backward or forward ticks according to the sense of rotation. I reckon that 'on average' the hand would go in one direction albeit erratically.That's based on a perhaps half-baked notion of 'handedness' . If the pendulum is started in a clockwise direction say, then the the total angular momentum (call it positive) will diminish because of friction but can never become negative.

Ady1: Thanks for the links. Interesting. Those globes look beautiful.

Thanks for the info about electromagnetic pulsing (and to Duncan as well). I shall experiment.

The bearings are bog-standard deep groove double shielded. Probably NSK, nothing special. I did make a prototype using ceramic ball bearings but the small improvement in running time wasn't worth the cost.

Robin.

Edited By Robin Graham on 09/09/2023 22:42:59

Edited By Robin Graham on 09/09/2023 22:58:02

A few years ago I made a 'chaotic' double pendulum. The only record I have is is a Youtube video of a prototype.

I've been approached by a local shopholder who would like to put one in her window if I could make it run continuously.

I think that if I could by some means deliver a periodic impulse to the main beam to compensate for frictional loss it should continue to show chaotic behaviour. It would have to surely?

It's a bit muddy in my mind though and I'd appreciate any suggestions of how to tackle this.

Robin.

Edited By Robin Graham on 08/09/2023 23:24:35

Edited By Robin Graham on 08/09/2023 23:25:06

Presumably everything subsequent was in English as well? Easier to change the label on the machine than the software.

Last time I was in Wales I shopped at a supermarket which had dual-language aisle signage. English: Desserts. Welsh: Pwdins. I don['t speak Welsh, but that's wrong isn't it?

Robin.

About ten years ago I bought a three year old car. I reckoned it would last me about ten years and I calculated likely yearly maintenance costs and the cost of replacing when it became uneconomic to repair. I set aside a ring-fenced monthly amount. When it goes in for the MOT tomorrow I'll have funds either to repair or replace if it fails.

I'm going to be nervously awaiting my nomination for the Nobel prize in Economics now!

Robin

I made an ER40 collet chuck register mounted on a commercial D1-4 backplate from scratch. Obviously I machined the taper after the register so it was spot on as machined. Replacing on the spindle nose (with care to observe the original orientation) runout is 0.01 - 0.012 mm on the inside of the taper.

Robin.

Thanks Pete. I've signed up and have started working through their tutorials. Could be ideal!

Robin.

Thanks for further replies.

I had a look at project on GitHub and it seems that although possible it's not trivial to get Fusion working under Wine. Lots of ifs and buts indeed!

I can (astonishingly) get a refurbished second hand Win 10 machine for £68. 4-core i5, 3.1GHz, 8GB RAM which seems to fulfil the system requirements for Fusion.

I've struggled with CAD for years (on and off, mostly off) and always given up - it's easier to draw by hand. I lit on these tutorials for Fusion and they seem to make sense to me. I don't know if it's the logic of the program itself or the way the guy explains it, but I have new hope.

Robin.

Thanks for replies and apologies for not acknowledging sooner - I got a bit sidetracked.

I should have said that I have only Linux machines. I was enthused by the video Gary Wooding linked to and went so far as to find a series of intelligible Fusion tutorial videos only to discover that Fusion isn't available for Unix. D'oh! Same for Alibre Atom and the majority of 3D CAD stuff out there it seems.

I did see a comment on a 'best 10 CAD programs for Linux' article claiming that Fusion can be used under WINE but from my experiences with WINE I doubt it.

I have FreeCAD but don't get on with it.

I'll probably end up doing it by hand or not at all.

My Rotrings dried up years ago - as a schoolboy I would submit homework written with a Rotring at perhaps 3pt just to annoy the 'masters'. They couldn't do me for it, nothing in the rules! Nowadays I like Pigma Micron fineliners - lovely pens.

Robin.

As DC31k says it's entirely possible. The code I've written could be adapted fairly easily. However, it's Linux terminal based so perhaps not friendly to everyone. PM me if you're interested though.

Robin

Thanks for this - I'll have a sniff around, but more out of interest than for any practical purpose. I imagine NPL are concerned with very high accuracy. I think I'd be fooling myself if I believed that I could measure anything to micron accuracy under normal workshop conditions.

At £395 + VAT I doubt that I'll be buying the MQS set of standards, but thanks for the link!

Robin.

I have a Dasqua 75-100mm mic. It seems nicely made - the action is very smooth. Checking with grade 1 gauge blocks it's good to the width of a graduation mark (2 microns?) over its range, so spot-on for practical purposes in a home workshop I'd say.

It's always worth having a look at G&M tools for second-hand measuring equipment. At the moment they have a 100-125mm Starrett mic up for £20 - without setting bar unfortunately, but likely a good instrument.

Robin.

Not being a Pythonite it would have taken me longer to refresh my memory and look around to see what libraries have been developed over the last 20+ years than to code in C. Using John Haine's method it took maybe 45 minutes to code in C - but that includes error checking to avoid the unhelpful 'Segmentation fault (core dumped)' message. Python does a lot of work 'under the hood' I suppose.

Anyhow, I got it working.

The whole thing started because I wanted to check my micrometers (0-150mm) for an upcoming project and when Zoro emailed saying they were upset and hoped a 15% discount might lure me back, I thought to buy some gauge blocks. But what sizes should I get?

This is a bit niche I know, but I found that using John's method it took about 7ms to calculate the 65535 non-zero combinations of 16 blocks, about 22 seconds(!) to order them by bubble sorting and another 7ms to weed out unique values. For the 7 blocks I now have it took 23 microseconds to calculate the 127 combinations, 1 millisecond to sort and 300 microseconds to weed out unique values. That's on an ancient Intel Duo 2 machine.

I now know how to make make 82 unique combinations from seven blocks covering the range I need with no more than a 2mm gap between sizes.

Thanks again for contributions, Robin.

Thanks for replies. In saying that there are 7 combinations of my three blocks I was deliberately ignoring zero blocks - hence (2^n)-1 rather than 2^n.

When looking about before posting here I did see some Python implementations but it's been too long since I wrote any Python so passed over. This old dog can't even remember old tricks, let alone learn new ones!

I've written a recursive C function which does the job - the problem is well suited to recursive solution - but I wanted to find a non-recursive algorithm.

Thanks John. That's a penny-drop moment for me. A very neat way of thinking about it and easily implemented in C. Problem solved I think.

Robin.

Thanks John, they look good. The threaded ends will make life easier. Cheaper too as not paying for the clamps on the phone versions which would have to be chopped off.

Robin.

I have only three gauge blocks - 10, 20 and 25mm. I use them for checking micrometers. Obviously I can check at 10,20,25 with single blocks, 30,35,45 by wringing two together and 55 mm by wringing all three. So 7 combinations.

Naturally I wonder about enumerating the possible combinations if I had N blocks. From the binomial theorem there are 2^N -1 possible combinations. That's easy enough. But how to actually list them systematically? I thought it should be easy to translate my 3 block example into computer code and generalise, but it's proving harder than I thought.

I had a look around and found:

n%){\,=}+[[]]@1/{`{1$+}+%}%;\,n*

which isn't helpful.

Anyone with a handy algorithm?

Robin.

Edited By Robin Graham on 26/08/2023 00:02:40

Edited By Robin Graham on 26/08/2023 00:04:32

Edited By Robin Graham on 26/08/2023 00:16:23