Talk:2009-07-06 49 10

Gripping report - congratulations on a successful expedition and staying alive by not getting lightningstruck. Me and my girlfriend were in a car when we got caught in a thunderstorm with massive rainfall a few hours after your expedition; around us frightening cloud formations looking like pyroclastic flow and a finger stretching down to the ground. I definitely would not have liked to be outside that moment. So, how was your BBQ? ;)

The weather forecast did say "rain and thunderstorms possible", but there had only been a few single drops during the day and the sun was shining when I set out. It really wasn't a nice experience. The BBQ was relocated to a balcony. - Danatar 10:02, 7 July 2009 (UTC)

"Auf Wiedersehen"

So that's what it translates to! Thanks! -- Jevanyn

The Wiki, Geohashing.info and Peeron say that the coordinates were at 49.0732875,10.5899034 near Lentersheim. However, the Wiki's map shows coordinates near Unterpleichfeld, and those coordinates were the coordinates which you visited. The same discrepancy arises in the other eastern expeditions on this day.
I wonder whether the Lentersheim or the Unterpleichfeld coordinates are correct. There may be some corrupted/inconsistent data, perhaps it has something to do with the Dow Holiday on the USA's independence day two days earlier. --Fippe (talk) 23:58, 24 October 2020 (UTC)

It should be easy enough to check, by doing the hash calculation from the comic [note: I didn't do that]... actually, what I suspect is going on is that for some reason different sources have different values for the Dow result. It happens that this Dow holiday was the first one since the invention of geohashing that is supposed to be observed on a date other than the actual holiday; this may be important.
My guess would be that either 1) the algorithms in use at the time wanted a value for the 2009-07-03 opening, did not get one because of the holiday, and used some kind of fallback that wasn't just the previous day's opening, or 2) for some internal reason the Dow did not have a holiday on 2009-07-03, but some of the algorithms assumed it did. I'm not sure which is more implausible. --January First-of-May (talk) 07:32, 25 October 2020 (UTC)

Okay, so I did some digging. You are correct, the USA's independence day fell on Saturday (2009-07-04), so the NYSE observed it on Friday (2009-07-03), so the last time it opened was Thursday (2009-07-02). That means that for Geohashing purposes, the Thursday value should be used for Friday (2009-07-03), Saturday (2009-07-04) and Sunday (2009-07-05). East of W30, it also should be used on Monday (2009-07-06). These are the values reported by different sources:

Date	Yahoo	WSJ	Crox	geohashing.info	peeron	map.geohashing.site	Wiki
Thu 2009-07-02	8503.00	8503.00	8503.00	8503.00	8503.00	8503.00	8503.00
Fri 2009-07-03	Holiday	Holiday	8503.00	8503.00	8503.00	8503.00	8503.00
Sat 2009-07-04	Weekend	Weekend	8503.00	8280.74	8280.74	8503.00	8280.74
Sun 2009-07-05	Weekend	Weekend	8503.00	8280.74	8280.74	8503.00	8280.74
Mon 2009-07-06	8280.74	8279.30	8279.30	8279.30	8279.30	8279.30	8279.30
Tue 2009-07-07	8324.95	8324.95	8324.95	8324.95	8324.95	8324.95	8324.95

So, everyone agrees that the Dow opened at 8503.00 on Thursday. With that, the eastern coordinates for Monday can be calculated.

echo -n '2009-07-06-8503.00' | md5sum = E1CE129B30D84EAF0BB61DE61109ABCE

0x0.E1CE129B30D84EAF,0x0.0BB61DE61109ABCE = 0.8820506695636549,0.0457476317106172

Those are indeed the coordinates for Unterpleichfeld, so Danatar visited the correct coordinates.

So why are coordinates different on the Wiki, Peeron and Geohashing.info? Well, it looks like for Saturday and Sunday, they used Yahoo's value for next Monday (8280.74). It is worth noting that this value is likely also not the correct value for Monday, since the WSJ and all Geohashing sources (not just Crox, but also the Wiki, Peeron and Geohashing.info) report a different value (8280.74).

Both the WSJ and Yahoo say that 8280.74 was the closing value for Thursday. For some reason, Yahoo wrongly used this value as the open for Monday, and the Wiki, Peeron and Geohashing.info wrongly used the value for the open on Saturday and Sunday. --Fippe (talk) 09:16, 25 October 2020 (UTC)

Incidentally, inspired by the discussions regarding Shotgun (...well, on Discord, anyway, couldn't find any on the wiki), I decided to check Shotgun's results for this hash, and it also gave the Unterpleichfeld coordinates. Which makes me wonder whether Shotgun's reported inconsistencies for pre-1960 retrohashes might actually involve the other sources being wrong... --January First-of-May (talk) 09:37, 25 October 2020 (UTC)

Geohashing.info now uses the 8503.00 value. Thanks for pointing it out. --Eupeodes (talk) 14:29, 25 October 2020 (UTC)