Wednesday, November 18, 2026

Waymo Complicated


Waymo has I believe arrived in London but has been pulled from freeways by the company after a series of incursions into roadworks. It's not the only time they've been off-piste, the footage above capturing:

(a)    uncomfortable dwell on a rail-road
(b)    quality-checking new asphalt
(c)    ignoring a rolling roadblock
(d)    mistaking water for asphalt
(e)    pulled over and unable to respond

Though when you think about it, there's nothing there we've haven't all done at the best of times after a beer or two.

AI models however appear to struggle most with the things you have to have lived. 

Like us they learn best from books and engaging with people, but it's mistakes that come later and which we dwell on in senescence that's yet to bother them.

Tuesday, November 17, 2026

Working Model #22


With skis loose at the front-end, I install the flap that serves three functions here:

(1)    It acts as a ram-wing to produce aerodynamic lift

(2)    It forms a fairing to shield battery-packs or fuel-tank

(3)    It stiffens up the pontoons to either side

This last function is particularly important because if you've set the skis by eye and they do not run parallel but diverge or converge along their length, then deflecting them to right or left to suit the prow will flex the pontoons off the vertical.

An element of this will cause no harm, but as in life it's a question of moderation in all things.

Isn't it, Gromit?

DC-X-Men


You may have marvelled at the capture of the second stage of a Starship rocket the past week, using a giant pair of arms to terminate its vertical descent. What you did not probably marvel at back in the 1990s was the DC-X, which was also ~ albeit at a third scale and as a prototype alone ~ able to take off and to land vertically.

Its origins date back to a visionary who first proposed the idea back in the 1960s and who would eventually persuade McDonnell Douglas to give it a try thirty years later and for a bargain-basement $60 million: peanuts in terms of space-flight. It would eventually be inherited by NASA, to whom it was the unwanted foster-child whilst they worked on their X-33; which ultimately proved a failure in commercial terms.

The DC-X did not, in so far as some among its engineers went on to do much the same for Jeff Bezos' Blue Origin space-ship.

Why though did the DC-X fly under the radar? Probably for three broad reasons:

(a)    There was simply less news around at the time

(b)    There was no global internet at the time

(c)    It was a project and not a commercial entity

Beside this, subsequent to the moon landings there was a loss of interest in space-flight generally among the public, the lunar lander itself technically a recovery by a similar means during its touchdown on the lunar surface. Due computer failure, by the way, that first landing had to be pursued manually by top-jock Neil Armstrong.

What Starships are doing by way of preparation for building on the Moon and travel to Mars is to regularise space-flight in the way flights by airliner feature machines in the air for much of the time, only pausing to refuel and to reload. Given the number of satellites SpaceX intend to launch, they need rocket-stages to travel to space and back on an equally frequent basis.

Hence the need for re-usable rockets, and the need for equipment able to provide for their rapid turnaround. What's with the 'chopsticks' thought to capture stages in mid-air? Well they effectively replace landing legs that add weight and complexity to any vehicle designed to return to Earth upright.

But still standing on the shoulders of giants, albeit third-scale giants. What it does for me here in bed, on the downside, is to make my own efforts look decidedly shit in comparison.

But we'll return to those efforts next.

Monday, November 16, 2026

Working Model #21


...and park the laths on top. They're more or less counterbalanced when pivoting at the forward edge of each pontoon, but I've added a modicum of ballast to secure as they stew in their own juices overnight. I've centred them by eye but you can do so too by feel ~ there's a width of about a first finger joint either side of the pontoon to the edges of each spar. The laths are of course released from the forward-most end of each spar during this part of the assembly.

Working Model #20


Pop a bead of adhesive on the tops (actually undersides) of the pontoons and tail-fins...

Working Model #19


I've pinned the forward end of this 4 x 46mm lath down and much as it pains me to cut down an eight-foot spar I've done so to around 60" (1524mm): taking the skis from head to tail.

Saturday, November 14, 2026

Working Model #18


In the event I shifted the fins outboard by around 10mm because it looked better, and we operate on an 'if it looks right it'll fly right' basis here at the ranch. These are not true aerofoils incidentally, and some of you may have noticed that those blunt edges might cause the airflow on the outboard of either side to separate?

Nonetheless this could still be expected to produce the desired result: an increased efflux from the left-hand propeller here (albeit the craft is inverted) would push the right side of the associated fin to the left, whilst low pressure caused by separation on the adjacent side would effectively such the same surface in the same direction.

All of which would create aerodynamic drag at the same time, but then life's a drag.