Boeing spaceship bug

Date: 2020-02-07 02:45 pm (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
People went to the Moon with software that was literally hardwired into the flight computer. The one time something went wrong, it wasn't the computer. Now we have two private companies competing to put people in low Earth orbit, and neither can seem to get basic safety right after years of attempts. So much for that.

Re: Boeing spaceship bug

Date: 2020-02-07 05:56 pm (UTC)
autopope: Me, myself, and I (Default)
From: [personal profile] autopope
I note that NASA now wants Boeing's spaceflight division (who built Starliner) to undertake the same kind of rigorous QA assessment that they insisted on for SpaceX some years ago.

Boeing are really not what they once were, in terms of corporate culture (not since they bought out MacDonnell-Douglas and the MacD-D board took over and inflicted their DoD-oriented practices on Boeing).

Sometimes a management reverse takeover works spectacularly well (see: Apple's "acquisition" of NeXT and the return of Steve Jobs) and sometimes it stinks. This is the latter.

Re: Boeing spaceship bug

Date: 2020-02-08 04:03 pm (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
Meanwhile, at NASA: https://www.inverse.com/science/nasa-brings-voyager-2-fully-back-online-11.5-billion-miles-from-earth — now those are engineers!
Edited (Added a thought.) Date: 2020-02-08 04:05 pm (UTC)

Re: Boeing spaceship bug

Date: 2020-02-08 06:06 pm (UTC)
armiphlage: Ukraine (Default)
From: [personal profile] armiphlage
Which brings up the point, what is all that extra code doing? Why add all the cruft that's not needed, when each extra byte of code is one more thing that can go wrong?

It's a universal problem, of curse. Notepad in Windows 10 uses more RAM than all of Windows 3.1, but has no features not in the Windows 3.1 Notepad.

Re: Boeing spaceship bug

Date: 2020-02-08 07:22 pm (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
A correct answer would be long and complex. Let me see if I can make a short one useful without going all snarky or ranty.


All that cruft has accumulated in layers over time, and younger programmers coming from behind were never taught to question it like they should have been. That's because many of these younger programmers have a superficial education at best, driven by one of two factors: either the industry convincing the academia to churn out code monkeys for them, trained to bang out enterprise Java code without really knowing what they're doing, or else YouTube "experts" trying to make a quick buck by teaching impressionable kids to bang out their first game or website in Python without having a clue how any of it works. Lazy old teachers jumping on bandwagons and leaving their students to struggle then end up asking for help online anyway only add fuel to the fire.

The big problems start when some of these kids end up working on software that runs dangerous stuff like literal space rockets. At that point they're doing the work of an engineer, but lack the *training* of an engineer, and more importantly lack the *mindset* of an engineer, not to mention the *ethics*. They have no clue about the importance of safety or reliability. They've heard of efficiency, but only in an economic context. And so they think it's perfectly okay to waste just because you can afford it. They can't even imagine a scenario in which you can't.


So much for brevity. Let's see if cuts work in comments. Hope this helps.

Re: Boeing spaceship bug

Date: 2020-02-11 07:03 pm (UTC)
jducoeur: (Default)
From: [personal profile] jducoeur
While I don't entirely disagree that software diploma mills are a problem, let's not forget the demand side, which is the real *source* of that problem: managers who believe that programmers are all fungible, whose jobs are incentivized to fill out their teams with warm bodies, and who believe that lines of code added is the correct metric for software development. They want More Programmers, Dammit, and aren't always picky about them; that produces ample incentive for the market to kick out quickly-trained, under-qualified developers.

(My current job does *not* have this problem, blessedly. But man, it is amazingly hard to winnow through all the crap applications to find the programmers who actually have passion and clue. And this conversation reminds me *way* too much of 1999, when the same problem was even more rampant.)

Re: Boeing spaceship bug

Date: 2020-02-12 04:45 am (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
Oh, absolutely. I was so focused on the "how" that I forgot to talk about the "why".

Derek MacKay

Date: 2020-02-07 03:37 pm (UTC)
danieldwilliam: (Default)
From: [personal profile] danieldwilliam
Much as I was sceptical about the timing of the Sun's story it doessn't seem like they've sat on this for days, let alone weeks or months.

Re: Derek MacKay

Date: 2020-02-07 05:52 pm (UTC)
autopope: Me, myself, and I (Default)
From: [personal profile] autopope
I'm kind of annoyed about the spin The Guardian put on it, which was along the lines of "wee Nicola was notified at 5:30pm on Wednesday but DIDN'T PUT OUT A PRESS STATEMENT UNTIL THURSDAY MORNING, AFTER MACKAY RESIGNED, WHAT IS SHE HIDING???"

Per her office she allowed time for MacKay to notify his family, and took legal advice on the party's position overnight. Which 'delayed' her statement by a whopping 14 hours. That's ... yeah, that's some cover-up.

Date: 2020-02-10 11:01 am (UTC)
danieldwilliam: (Default)
From: [personal profile] danieldwilliam
I have always thought that carbon capture and storage would not work. I don't think you can get round the conversion losses and the cost challenge from solar PV and wind.

My guess is that any time reducing carbon emissions looks difficult then solar PV will get a little bit cheaper and CCS will really struggle to find an economic role.

Date: 2020-02-12 04:51 am (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
It doesn't. Read this carefully: https://www.rte.ie/news/newslens/2019/0508/1048099-iceland-carbon-dioxide/ — in particular, pay attention to the numbers. It's a trap!

Date: 2020-02-12 09:39 am (UTC)
danieldwilliam: (Default)
From: [personal profile] danieldwilliam
That's an interesting article on Iceland but I'm not sure it says much about CCS generally.

Iceland's has unusual, perhaps unique, combinations of volcanic geothermal energy, geogprahy and a very small population.

The CCS project I have my eye on is whether Peterhead can economically pump all its CO2 back down the 40's pipeline.

Date: 2020-02-12 12:16 pm (UTC)
claudeb: A white cat in purple wizard robe and hat, carrying a staff with a pawprint symbol. (Default)
From: [personal profile] claudeb
Precisely! If Iceland, with their ideal conditions, has to spin the numbers like a roulette wheel to make them look at all viable, what's everyone else supposed to do?

Date: 2020-02-12 05:21 pm (UTC)
danieldwilliam: (Default)
From: [personal profile] danieldwilliam
Indeed, indeed.

February 2026

S M T W T F S
1 2 3 4 5 6 7
8 9 10 11 12 1314
15161718192021
22232425262728

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Feb. 13th, 2026 04:13 pm
Powered by Dreamwidth Studios