OpenBSD folklore and share/misc/airport
cambus.net"One can say anything about Lublin airport, but they sure do consider IATA codes are serious business. The tone was set right from the parking lot entrance."
The reason Lublin's IATA code is so prominently exposed is because the word "luz" in Polish means - among others - the state of being relaxed, chilled out. Just a bit of marketing on the city's part.
"LAX" also is short for "relax", but I don't think it's physically possible to do any such thing within LA city limits.
I belive a big "LAX" signboard exists, based on a Google image search.
Many airports around the world have a large signboard.
There's literally an LA rapper called "Lil Xan"
So, the opposite of the emotions experienced when flying and going through airports.
It was a lot less shitty pre TSA when it was named.
Ok, I'll ask the obvious. Why is OpenBSD trying to maintain a list of IATA airport codes?
(And before you ask why not, try to think of some answers yourself first. I can come up with a few drawbacks, though I'm nowhere close to a subject expert.)
There's also share/misc/birthtoken with a list of birth stones and flowers. That's actually much more weird than a list of airport codes.
Also a list of operator precedence, various phone area codes (country & local), an ASCII table.
I guess it's just a "list of databases that was useful/fun to someone, at some point". Harder to "just get it from the internet" back in the 80s.
Some things like the ASCII table have been there (or rather in the original location of /usr/pub) since near the dawn of UNIX itself: https://github.com/dspinellis/unix-history-repo/blob/8cf2a84...
(Also available as "man ascii" from almost as long ago https://github.com/dspinellis/unix-history-repo/blob/Researc... )
OpenBSD seems to like collecting extra items though
The phone codes must be terribly out of date and inaccurate those are so hard to maintain even with paid official data sources
I was always more of a Debian person but you're making me want to switch, this is some old internet style fun.
I think mac os also has the list of birthstones
From what I can tell, at least it's historically inherited. 4.4BSD had a list of airports in the same location.
Its inclusion in 4.4BSD seems a mystery though. No other files on the 4.4BSD distribution seem to reference it. The atc(6) game involves airports, but it doesn't seem to actually open this file, both in 4.4BSD and in OpenBSD.
It goes back a bit further than that, to 4.3BSD-Reno https://minnie.tuhs.org/cgi-bin/utree.pl?file=4.3BSD-Reno/sh...
The previous release, 4.3BSD-Tahoe, did not have /usr/share
It could just be a fun Easter egg to keep developers engaged.
History is so quickly forgotten: https://www.openbsd.org/hackathons.html
.. also: https://www.openbsd.org/events.html
Your reply doesn't seem to answer my question at all. I opened both links, searched for "airport" and "IATA" and nothing relevant came up.
Well then here's a few more keywords for you. 'jet-lag', 'travel'.
Hackathons tentatively take place in airport lounges.
Non-sequitur?
I'm reading "hackathons may occur in airports, therefore OpenBSD should maintain a list of all airport codes."
Suppose the list didn't already exist, how would you justify adding it?
And whatever the valid reason, why isn’t it an import of some official list?
It is an official list: a list of airports visited by OpenBSD developers.
It's just a bit of culture, don't worry about it so much.
So if you were doing a more serious program you wouldn't use this?
I'm not familiar with BSD but when I was using Kali if I was using a share it was to feed into something else, classic example being wordlists fed into hydra[1] or something.
Because that would require trusting some third party that claims (accurately? deceptively? who knows) that those airports exist. As maintainers of and contributors to a security-focused OS, OpenBSD devs want to verify the existence of each airport themselves before deciding it's fit for inclusion.
The official list is payware: https://www.iata.org/en/publications/store/airline-coding-di...
Alternatives are to scrape some other website (who presumably paid IATA for the data) or Wikipedia (https://en.wikipedia.org/wiki/Lists_of_airports_by_IATA_and_...) which may or may not be more complete/accurate than OpenBSD's own crowdsourced list.
Official lists are very often not permissively licensed (or openly at all).
Many weather apps use IATA codes; I've looked them up in that file before for that.
I think it’s nice to have a few small standard data sets that you can use in examples, tutorials, man pages and testing.
Not your everyday caveats section:
There seems to be a conflict here. The author of the linked article interprets the guideline, "New airports can only be added by OpenBSD developers who have visited an airport and thereby have verified its existence," as only requiring that you have been to the airport, not necessarily having flown from:
> Once again, the more astute reader will not have missed the fact that the rules do not stipulate any flying requirements. Neither did henning@, who not long after airport.7 was committed, added an entry for XFW, (the Airbus factory) which he had visited but not flown from.
Fair enough - he wrote the requirement, after all.
However, the Caveat in the man page seems to contradict this, and indicates that you do have to fly in:
> There are also railway stations with IATA codes. These may not be listed, except if someone landed there by plane and survived to update the file.
>New airports can only be added by OpenBSD developers who have visited an airport and thereby have verified its existence
This says nothing about rail stations. For a rail station to be added, you must land on it. Presumably, taking off from there does not count.
I think the caveat is just trying to say that it has to be considered an airport, so at least one person has to have flown in/out of it
I don't see a requirement that the OpenBSD developer needs to have flown into the railway station. Just that someone survived landing a plane at the railway station for it to count as an airport.
"survived to update the file" implies that the one to add it to the list must be the one to have landed there
> However, the Caveat in the man page seems to contradict this, and indicates that you do have to fly in
This is an attempt at humor, because railway stations are not airports.
Which makes you wonder why some railway stations would have a IATA code.
Up until very recently United had codeshares with Amtrak on the north-east corridor, and could even rebook onto Amtrak in the event of flight disruption. Lufthansa still offers connections to and from Deutsche Bahn on the same ticket (https://www.lufthansa.com/tr/en/rail-and-fly)
I hope to never see the day that air traffic is so disrupted that getting rebooked on Amtrak seems like a better option. For a variety of reasons, Amtrak's adherence to schedules is abysmal.
Has anybody here had this happen? 9-11 is the last occasion I could imagine where this could possibly be a win.
Amtrak's adherence to schedules varies by line -- the problem on many of them is getting trapped behind freight trains, but on the Northeast Corridor specifically, Amtrak itself owns most of the track, and the rest is shared with other passenger rail operations (various commuter rail), so they're about as reliable as airlines (which themselves are... not infrequently disrupted by weather).
the problem on many of them is getting trapped behind freight trains
I recently read in a railfan magazine that this is becoming worse because of the supply chain mess.
Fright train companies are running longer trains because of higher demand. Longer trains run slower, because there's only so many engines to go around.
Making things worse — slow freight trains would often pull onto sidings to let fast passenger trains pass. Now the freight trains are too long to fit into the sidings, so more and more the passenger trains have to wait.
Snowstorms in the north-east will frequently disrupt flights far more than trains. It's about 4 hours from Boston to New York, I'd take that over being delayed 2 days.
> Which makes you wonder why some railway stations would have a IATA code.
Such locations, which can also be bus stations and ferry ports, are known as "intermodal locations".
TL;DR its for ticketing purposes to enable flight and rail/bus/ferry legs to be issued on the same ticket.
Or at least that's the theory. To be honest its probably going be extinct in due course, given the complexities of integrating the airline world with the modern private railways etc. In most cases its just easier and more sensible to ticket seperateley. Plus some of the quaint old-fashioned things such as being able to check-through bags are becoming few and far between due to security and other aspects (e.g. disappearance of station porters and baggage cars).
Or sometimes it's just a bus stop, in a car park, with perhaps a small portaloo and a hot dog stand. Several times I've booked a through ticket from Salzburg to London Heathrow via Munich and the Salzburg-Munich leg was a bus ride that included a 'layover' at ZPR. According to [0] "Airport of Rosenheim is the most important airport of Rosenheim, Germany. It is modern and one of the largest airport of the Europe. Airport of Rosenheim is important for people and goverment of Germany." But actually it's just a lonely bus stop in a car park just off the motorway.
[0] https://airportsbase.org/Germany/all/Rosenheim/Airport_of_Ro...
Sigh, bot-generated content litters the whole internet. I was doing some address completion stuff and learned about virtual ZIP codes in my country, i.e. ZIP codes that are just for routing and don't have a geographical area, but if you look them up on the Internet of course there are sites that say "$ZIP is a city in the state of $STATE".
Hey Elon, can you pay $15 billion to get rid of bots on the Internet?
to be pendantic, all (US) ZIP codes are just for routing. Any correspondence to a geographical area is an artifact of this, and subject to change. Also, it is possible for two ZIP codes to overlap geographically.
In particular some numbered zones are entirely contained geographically within others. In some places in the US, there's a ZIP for delivery to private homes and businesses but another for delivery within the Postal Service network, including delivery to post office boxes. The one I remember most readily is Quincy, Illinois. The city is 62301 and the surrounding county 62305, but PO boxes are 62306, located within the post offices that sit within the 62301 zone.
Almost every single post office has a separate zip code (usually one higher than the city/area it is sitting in).
Some individual buildings have their own zip code, or 12345 for GE, or 20252 for Smokey Bear.
My post office box has its own ZIP+4.
+4 is cheating! Especially for post offices with less than 10000 boxes.
If memory serves me right, it's still used in Germany for trips outbound from FRA. You buy a ticket with a train somewhere → FRA leg, and an airplane FRA → somewhere-else.
It gets even better. One of the big Software Systems at Fraport has a „Flight“ class, which exposes the method „isTrain“… On another german airport the same class was extended by a „isTruck“ method for parcel trucks…
The disappearance of staff makes sense.
I’d have thought integrated security would be far more secure since the baggage goes and stays behind the secure boundary until the conclusion of the trip vs being retrieved, checked in, retrieved, checked in, etc
My favourite modern intermodal baggage transport is in Japan.
They have a small number of specialist baggage couriers called Takuhaibin (宅配便). These couriers will take your bags between any two destinations of your choice (Hotel-Hotel,Hotel-Airport etc.) for a reasonable fee. The pick-up and destination can (basically) be anything in Japan with a postal address.
Its not true check-through as you still have to pick up your bags from the counter landside at the airport and take them to the check-in counter but the distance is minimal since you'll already be in the right terminal and on the right floor.
(Technically for the pedants out there, Takuhaibin are also a general parcel company too. But unlike, say UPS or FedEx, a Takuhaibin will take your baggage, your furniture, your clothes or pretty much anything else as long as its safe and legal)
These couriers will take your bags between any two destinations of your choice (Hotel-Hotel,Hotel-Airport etc.) for a reasonable fee
We have these in America, too. They're call couriers. They'll even pick up your luggage that went to the wrong airport in a different city and bring it to you at your hotel in the right city. (Happened to me with a bicycle.)
unlike, say UPS or FedEx, a Takuhaibin will take your baggage, your furniture, your clothes or pretty much anything else as long as its safe and legal
This sounds exactly like UPS and FedEx. I shipped baggage, furniture, and clothing via UPS as recently as last year. I even had a service come over and pack it up for me.
UPS and FedEx will ship anything. How do you think elephants and whales get from zoo to zoo?
I think the key is that you can skip the "pack it up" step - UPS won't ship many things that would survive shipment just fine unless they're wrapped in a box.
UPS and FedEx freight or LTL (less than truckload) is another story entirely.
> I think the key is that you can skip the "pack it up" step - UPS won't ship many things that would survive shipment just fine unless they're wrapped in a box.
Indeed.
With a Takuhabin I can go pick up a freshly pressed wedding dress from a dry cleaners and hand it over just like that. It will arrive in exactly the same state it left the dry cleaners.
With UPS it would barely make it into the back of the collection driver's van before it looked like a second hand rag.
You can do true check-through with the Japanese carriers (JAL and ANA) AIUI.
Which makes you wonder why some railway stations would have a IATA code.
Because you could (still can?) book a trip that way.
Just like if you try to fly from Boise to Tokyo, you'll get a ticket set from BOI → SFO → NRT.
You could book a trip from ORD → PHL → whatever the code is for Penn Station in New York.
I don't think it's done that much anymore, but it used to be pretty common.
Here’s a copy of the current version of the file itself https://github.com/openbsd/src/blob/master/share/misc/airpor...
I'd argue that there's a bug. while TLV would appear to imply that the airport is in Tel Aviv, it is not. :)
And they list OSL as "all airports around Oslo, Norway". That's not the case, OSL is one particular airport (the main one, Gardermoen). They correctly list the second airport as TRF.
OSL is also the IATA metropolitan city code for Oslo, basically meaning "all airports around Oslo". Some booking engines support these for convenience - for example, ones using Sabre as a backend may support QSF (also listed in this file) for all airports around San Francisco. I flew to Oslo recently and know I saw OSL used as a metro code first on at least one booking engine. It doesn't really make sense to put these in a list of physically visited airports but I wouldn't expect most people to know the intricacies of IATA codes.
Why it's designed this way with namespace collisions, I have no idea. Technically QSF also is a tiny airport in Algeria.
I'll be pedantic and say that technically OSL as a metropolitan city code isn't an airport (i.e. should fall under the same caveat as the railway portion of the man page)
Wow, TIL, and I stand corrected! That's crazy!
See also: NYC for metropolitan New York airport, and HOU for Houston's airports, even though Houston Hobby Airport is also HOU.
> NYC for metropolitan New York airport
Sure, but NYC isn't also a specific New York City airport (JFK, EWR, LGA). The problem here is that OSL is apparently used both for the metropolitan area and for one specific airport (in the metropolitan area).
I guess you missed the second half of the sentence you quoted. Where I note that Houston, the city, is HOU; and the Houston airports include IAH, EFD, and HOU.
I had the same thought. I’ve seen some booking sites default to “OSLALL” when you select Oslo, which will show results for both OSL and TRF. Probably “OSLALL” is not a real IATA. Either way, to me OSL should mean Gardermoen and nothing else. And search results I am finding on Google seem to indicate the same. After all, it’d be pretty silly if you were traveling to Gardermoen and your luggage was tagged “OSL” but then your luggage somehow ended up on TRF instead.
So referring to OSL as “all airports around Oslo, Norway” is IMO a mistake on the part of the OpenBSD developers indeed. In spite of the point made by the sibling comment about ambiguity. Because the context is airports, I think OpenBSD should list OSL as “Oslo Gardermoen Airport, Norway” in their airports file.
See also https://en.wikipedia.org/wiki/Oslo_Airport,_Gardermoen which says:
> Oslo Airport (Norwegian: Oslo lufthavn; IATA: OSL, ICAO: ENGM), alternatively referred to as Oslo Gardermoen Airport or simply Gardermoen
Here is the complete file if you are curious: https://github.com/openbsd/src/blob/master/share/misc/airpor...
There are ~2,000 entries vs ~11,000 assigned IATA airport codes
And my airport is wrong.
My city has two airports which are like 50 km apart. One of them is for military only.
It shows the regular passenger airport IATA code mapped to the military airport name.
I wonder if I can correct it. I'm not a BSD developer at all.
I visited 4 airports that aren't in the list. Do you think I can add them to the list? Never committed before to OpenBSD so I do not know if I compute as a "OpenBSD developers" in "New airports can only be added by OpenBSD developers who have visited an airport and thereby have verified its existence."
Developers are categorized as people with commit access to the project. So contributing patches itself is not enough to add entries to this specific file.
I think maybe this could be workable if an OpenBSD developer were to visit you in person to verify your existence and thereby transitively verify the existence of the claimed "visited airport".
What if I know a guy who knows a guy?
I've checked and at least one airport is missing from the list that I've been too.
I thought this was going to be about https://www.openbsd.org/lyrics.html#43 where the author claims that Richard Stallman made a scene and had to be removed from a plane from Washington DC to New Orleans.
It sounds like a severe enough accusation that there should be some corroborating evidence.
Mr. Cambus is also the author of the excellent ansiweather script.