Scot Mcphee on Fri, 29 Mar 2019 03:33:50 +0100 (CET)


[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]

Re: <nettime> rage against the machine





On 29 March 2019 at 09:07:31, Morlock Elloi (morlockelloi@gmail.com) wrote:

Seemingly totally unrelated: 

1. flight recorders are brightly colored these days. The term "black 
box" originates in WW2, mostly because the first flight recorders, as 
all other "secret" electronics, was housed in metal boxes painted matte 
black. 

See 
https://web.archive.org/web/20171019110346/http://siiri.tampere.fi/displayObject.do?uri=http%3A%2F%2Fwww.profium.com%2Farchive%2FArchivedObject-8077CE76-2B43-6FAA-D11C-77AAFD6C72E8 

2. Schematic "black box", meaning circuitry/algorithm that is opaque and 
not supposed to be seen or understood, and only I/O is available also 
originates in WW2. 

It's hilarious that #1 and #2 overlap again these days, as most airlines 
have no capability of examining their own flight recorders, so we are 
back to black boxes: Ethiopian Airlines refused to hand over their black 
boxes to Americans, as they don't trust them. I


Hello, I want to weigh into this thread finally, after following it closely these past weeks. I haven’t had the time to respond, as I’ve been busy with my PhD dissertation, which I submit in just a few days. There have been so many good posts on this thread, particularly one from Felix timestamped 2019-03-28 0900 UTC where he talks about the rise of complexity in systems.

First, I want to qualify the position from which I speak: I am a technical expert who works for an airline. I’m not an aircraft engineer, but I design and build what are called ‘operational systems,’ which provide the inputs and outputs for the airline to safely fly it’s equipment. I’m not, therefore a strictly disinterested party in such discussions, but I do possess a certain amount of inside knowledge of airlines, and the often cryptic systems we use and the language we use to describe them.

First, I want to talk about the ‘black’ box. Yes it's bright orange. No that’s not why it’s spoken as a ‘black’ box. In fact the cybernetic explanation is the correct one.

The airline is not supposed to be able to read the contents of these boxes. That’s why they are ‘black’. The cockpit voice recordings and control data flow into them and are used for _safety_ investigations, by a _safety_ authority, like CASA, or the FAA.

Why isn’t the airline supposed to read the content of the boxes? Because the voice recorder is recording the pilots use of procedures which may be designed by the airlines. Yes, every airline has slightly different procedures, as long as these procedures are within the parameters of the aircraft designers, and the responsible aviation regulator, all is _supposed_ to be OK. But just as _technical_ failures, say with the pitot tube icing on AF447, can cause technical systems to malfunction (disconnecting the autopilot … although that’s not a ‘malfunction’ as such), and ‘human factors’ (cockpit design of Airbuses) the company procedures and culture can also cause or compound accidents. In the case of AF447, there was a toxic culture among the pilots. They did not co-operate smoothly. The senior pilot barged into the cockpit and basically bombarded the two pilots at the controls with theories and questions. __nobody__ thought to ask the most junior pilot onboard, who was sitting in the right hand seat at the time, if he had made any control inputs. He had … in fact he performed the __worst__ possible input when faced with a stall warning: he pulled the nose up. Anyway, the investigation found a cultural and training issue, which AF had to fix.

However, just because we can’t read the boxes, doesn’t mean we don’t monitor our aircraft. There are plenty of signals which the aircraft transmits during flight, and a ton more which are downloaded from it when it gets into the hands of the engineers. We have a entire department that analyses this information, offline. If they find issues, the pilots are asked to explain to the chief pilot (for the type) what happened: “why did you exceed the type’s maximum recommended descent rate for over 30 seconds last week flying VHxxx into YBBN on SMOKA 8 ALPHA to RWY 01LR between DAYBO and GORRI?”.

Anyway I don’t have any great theoretical insights at the moment but a lot of this discussion is interesting, and if someone has airline related questions I’m happy to answer them.

Scot



#  distributed via <nettime>: no commercial use without permission
#  <nettime>  is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: http://mx.kein.org/mailman/listinfo/nettime-l
#  archive: http://www.nettime.org contact: nettime@kein.org
#  @nettime_bot tweets mail w/ sender unless #ANON is in Subject: