What's gone wrong at the AAIB ?

[DaveReidUK]

The UK Air Accident Investigation Branch has for over 100 years enjoyed an unrivalled reputation for meticulous attention to detail in its accident investigations.

Should anyone doubt that, I'd recommend getting hold of a copy of the commemorative history that it published in 2015 to mark its centenary.

So why does an organisation that prides itself on accuracy have such a g*d-awful website that's riddled with errors ?

One favourite theme seems to be mis-identifying the aircraft involved in an accident by getting the tail number wrong (often, in such cases, that of a completely different aircraft). While we can be confident that AAIB investigators know the difference between, for example, a BAe146 and a Dash-7, or between a Shorts 360 and a Seneca, that's not the impression given by the website.

Another favourite is quoting a nonsensical aircraft type. While it's not too difficult to deduce what an "L-832" or a "Bell 22B" are supposed to be, others (for example a "Beat", or an "S310", etc) are a bit more of a challenge. Engines aren't immune, either - who knew that the O-320 was built by de Havilland, or that the M-14 radial is a turboprop ?

Even the dates shown on the website for accidents/incidents don't always correspond to when the events actually happened. While most of the discrepancies are only a day or so out, some are off by months or even years.

And lest we should think that all the howlers are trivial, serious researchers and aviation safety analysts can never be sure that a search for all the accidents for a given aircraft type will return a complete dataset. For example a search for Super Puma events fails to return at least two accidents, and yet they can be found by searching on date or tail number.

I know from conversations with AAIB's previous Chief Inspector that he was keen to remedy the deficiencies in the website, but I understand that following his sudden departure last year his successor is ex-military, so it may be that the concept of putting accurate information into the public domain isn't something that the new regime is familiar with, which is a pity.

That's the only logical conclusion that can be drawn from the fact that courteous offers of help (pro bono) to sort out some of the worst howlers don't get a response, or even an acknowledgement, from Farnborough.

[MacLaren1]

Probably the military approach...

..."Why do we need to tell anyone?" Rather than "We need to get this right because it could undermine everything else we do"...

[megan]

There is a saying, at least in Oz, which says, "Close enough for government work", perhaps they have adopted the saying as a standard.

[Daysleeper]

The UK Air Accident Investigation Branch has for over 100 years enjoyed an unrivalled reputation for meticulous attention to detail in its accident investigations.

Should anyone doubt that, I'd recommend getting hold of a copy of the commemorative history that it published in 2015 to mark its centenary.

So why does an organisation that prides itself on accuracy have such a g*d-awful website that's riddled with errors ?


The AAIB lost control of its website a few (four ish?) years ago during government cutbacks. It is now part of the gov.uk empire and probably costs more, achieves less and is almost impossible to influence.

[oggers]

ex-military, so it may be that the concept of putting accurate information into the public domain isn't something that the new regime is familiar with

...whereas the techlog forum is the correct one for whinging about websites and the military?

[Airbanda]

The AAIB lost control of its website a few (four ish?) years ago during government cutbacks. It is now part of the gov.uk empire and probably costs more, achieves less and is almost impossible to influence.

Spot on.

Review of govt websites c2011 under Martha Lane-Fox decided that government was, in commercial terms, one brand.

In practice of course, aside from differing identities and roles of main departments, there are many regulators, watchdogs and advisory bodies funded by but wholly independent of Wetminster/Whitehall. They fought long and hard to retain ownership of their own brand and website (I still have the scars). Eventually though pretty well all were swallowed up. Only the most glaringly obvious examples of independence - the Judiciary for example - retained control.

[DaveReidUK]

...whereas the techlog forum is the correct one for whinging about websites and the military?

I don't have any problem with the military, either current or former, only with the "close enough for government work" ethos that evidently prevails in this instance.

As for complaining about the poor quality of a website whose intended audience includes aviation safety professionals, I'd have thought that Tech Log was an entirely appropriate forum.

[john_tullamarine]

I'd have thought that Tech Log was an entirely appropriate forum.

If they can't get their own stuff on the mark - regardless of what the reason may or may not be - no reason why PPRuNe can't provide something in the way of a back up ... perhaps this thread can provide something in the way of a defacto information source ?

[oggers]

DaveReid

Accidents and Close Calls, Spectators Balcony, Computer and Internet Issues or Jet Blast are clearly more appropriate to discuss the AAIB's website inaccuracy. There is no way that the forum badged as "the very best in practical technical discussion" is the place to air your subjective opinion that

his successor is ex-military, so it may be that the concept of putting accurate information into the public domain isn't something that the new regime is familiar with, which is a pity.

:ok:

[DaveReidUK]

It’s a Short 360.

If you're going to be pedantic, get it right. :O

It's either a Shorts 360 (marketing designation) or a Short Brothers plc Model SD3-60 (Type Certificate).

[CloudHound]

Short 360

Short's 360

Shorts' 360

Super Shed?

:ok:

ps I agree that accuracy has suffered with their website.

[DaveReidUK]

I was correcting your shortened name for the type.

Except that you were trying to replace something that was already correct with something that isn't. :O

Short Brothers started using the contraction "Shorts" well before they launched their commuter aircraft and used the term in all the marketing for the SD3-30 aka "Shorts 330" and SD3-60 aka "Shorts 360" (like the sales brochure in my previous post - surely you're not trying to tell us the manufacturer got the name wrong!).

In fact you might be surprised how many manufacturers give their aircraft marketing designations that differ from what's on the Type Certificate: "Fokker 100", "CRJ 200", "ATR-72-600", etc.

But getting somewhat back on topic, since we happen to be discussing Belfast's finest it provides another good illustration of what I'm talking about: there are 24 accident/incident records for the type on the AAIB website, covering 19 separate events between 1986 and 2007.

But to find them all you need to do four separate searches, each of which will return a different batch of events, for:

"Shorts 360"
"SD360"
"SD3-60"
"SD 3-60"

(you need the quotes, otherwise the daft search algorithm will find dozens of unrelated events for airships, helicopters, light aircraft, etc)

No. I'm not kidding - you couldn't make this stuff up. :ugh:

[megan]

Dave, if it's of any comfort the Oz ATSB web site is equally terrible. Do a search and it invariably returns a big fat zero, I find it necessary usually to use google as the search engine.

[DaveReidUK]

Dave, if it's of any comfort the Oz ATSB web site is equally terrible. Do a search and it invariably returns a big fat zero, I find it necessary usually to use google as the search engine.

I take your point, but in this case even if the gov.uk domain dictates the platform and search engine that the AAIB is obliged to use, presumably Farnborough still retains responsibility for the accuracy and consistency of the content.

Leaving aside the howlers, typos and plain WTFs, it would be reasonable to expect that the industry should be able to do website searches based on one or more of the standard taxonomies (ECCAIRS, ADREP, ICAO ATDs, CICTT, etc).

But no, you have to second-guess how the investigator who wrote up each event decided to describe the aircraft type involved, which seems to depend on what they had for breakfast that morning.

A recent search for Fokker 100 accidents/incidents came up with 22 events, but only by combining the results from seven separate searches. Take your pick from:

"F28 Mark 0100"
"F28 Mark 100"
"F28 Mk 0100"
"F28 Mk 100"
"Fokker 100"
"Fokker F 100"
"Fokker F100"

An expression involving "fit for purpose" [not] springs to mind.

[Uplinker]

Well, the AAIB did not even invite me to an interview, (for recent AAIB inspector job), so I for one can't help them.

Shame really; I spent ages writing that application :rolleyes:

[DaveReidUK]

I used [ F28* ] (without the brackets) and found 23. Would the use of a * help your searches?

No, the presence or absence of an asterisk wildcard makes no difference to how many events are returned.

Either way, it will only find 16 Fokker 100 accidents/incidents (discounting duplicates, Fokker 70 and legacy F-28 events).

You still need to do additional searches to find the remaining 6 Fokker 100 events.

But cheer up, it could be worse. If you're a rotorhead looking for Enstrom F-28 accidents/incidents, searching for "F-28" will only find one of the 40 in the database.

Go figure.

[DaveReidUK]

Well I don’t if you are searching the AAIB’s website or the .GOV landing site, because for me ... Enstrom* on the AAIB’s own site retrieves 68 reports, F28* = 23 reports, BAe* =123 reports, Short* = 596 reports, as does Shorts*. And, 146* = 88 reports that include RJ too.

Other than the fact that you seem to have proved to yourself that the asterisk makes no difference, what's your point ?

You make it sound like it's a good thing that a search for Shorts*, for example, returns nearly 600 results.

That completely ignores the fact that more than 90% of those 596 results are accidents/incidents involving light aircraft, corporate jets, helicopters, jet airliners, etc with absolutely no relevance to your search criterion.

Could you explain how exactly that's supposed to be of any use to anyone ?

[DaveReidUK]

A simple “F28” search from the link page I posted returns 22 Fokker / F28/all marks reports.

You wrote that it requires multiple searches to find other reports for different aircraft types. You say it’s down to poor website design; I disagree, and suspect operator error.

End of.

You're not paying attention, are you ?

My original point, which you haven't attempted to address, is that multiple searches are required in order to retrieve the 22 unique Fokker 100 accidents/incidents in the AAIB's records. That's 22 Fokker 100 (aka Fokker F-28-0100) events - not Fokker 70s or legacy F-28s.

If you want to try to refute that, you are very welcome to - please just specify the search term you would use to retrieve those 22 Fokker 100 records (your search on F28 doesn't find them all, for the reasons I described in my previous post - you can easily verify that).

Make sure that your results include the events on 10 February, 20 March and 12 October (all in 1995).

When you find that they aren't included, feel free to do a date-based search for those dates and you will find them (G-UKFG, G-UKFF and G-UKFA, respectively).

I rest my case. :ugh:

[4EvahLearning]

Very possibly a sign of the times I'm afraid. How many "reputable" sites present with incorrect details, grammar, spelling or punctuation. All these things seem to be superfluous to just getting something on the page.

We are now in the age where the teachers don't have the skillsets to pass on. The checkers don't have the skillset to make the corrections or just find the task too boring or demeaning for them.

If it isn't the final official report it doesn't seem to be of importance, and here's hoping the "final" report is error free.

[hunterboy]

I’d say it is indicative of the calibre of person running these organisations nowadays. Look at how much they pay. Anybody with a bit of nous doesn’t go in aviation or government. As for piloting jobs, I reckon many of us are probably on the spectrum to some extent to be attracted to a job that relies on a particular skill set. That skill set isn’t necessarily the same skill set to be a good manager or executive.
Sure, there are always exceptions to the rule.

[parabellum]

Sounds as though the person responsible for actually inputting the information via the keyboard doesn't know the difference between an aeroplane and a London bus! Sounds as though their supervisor is of the same ilk and neither of them really care. Surely they have up to date copies of Jane's as reference?


Suggestions that an ex-military person may deliberately withhold information is a cheap shot.

[BARKINGMAD]

Maybe it confirms the suspicions of the grumpy old ex-aviators on this forum that the currant generation of snowflakes emploied by HM Gov had an inadekwate edukashun, thanks to modern skuling tekneeks imposed by pollytishians interfering in the honnored profession of teeching and imparting of nollege?