Osint Amateur Hour
secjuice.com
secjuice.com
But then in the next game you are placed right on midway island with no clues but sand and some birds...
Fun times
[0]: https://www.youtube.com/watch?v=M7w986ni7_g&list=PL_japiE6QK...
I feel like he could benefit from ten minutes to two hours looking at
https://en.wikipedia.org/wiki/Wikipedia:Language_recognition...
because it could greatly improve his speed and accuracy on that part. :-)
This usually brings up several possible matches. These can often be reduced by checking any interior pictures Including windows showing the immediate surroundings. Adding nearby buildings as features to look out for.
I just find this practice really annoying and I don't see how it helps sell a property.
See: https://www.bbc.com/news/technology-52978880
And: https://www.theguardian.com/film/2013/nov/12/spike-lee-sued-...
And, of course, reddit and the boston bomber.
Do this for fun, but don't get your pitchforks out
Identifying its SA is extremely easy with the Dutch text and newspaper names. There is no real hint this is Pretoria and not somewhere else. The satellite photos don’t have enough resolution to help find the stop lights (which appear not too common looking at street view).
You can kind of see a double roof in the building in the background, and the road arrangement, but it would take days to manually look at every city intersection.
I guess the easiest approach is to post the image on a forum / twitter and get locals to chime in :)
And as a South African, I would immediately know it was SA just by the look of the traffic lights and street signs.
I agree that there is no way to know which city it is in the first photo.
In the article there is the line: "Some quick Googling revealed several universities in Pretoria, and I decided to wing it and take a closer look at the Pretoria skyline."
That's glossing over a very important part of figuring it out. Because once you know which city it makes it much easier.
If you look closely, you can see stenciled markings on the yellow poles. These felt like official markings to me so I tried to figure out how to use the markings to narrow down the list of candidate cities.
According to Wikipedia, the _Burger_ newspaper is distributed only in three regions: Western, Eastern, and Northern Cape. There are not too many large-ish towns in these district.
So I made a list of all towns with >50k inhabitants. Looking at those cities on Street View one by one, I found there are only two towns where the yellow poles have the exact same typeface of stencil numbers on them: East London and Uitenhage.
So I took a closer look at either town. (After a few misses, you quickly get a feeling where to find the parts of towns where the roads actually have traffic lights). With a bit of luck, I actually managed to locate the first picture just by randomly clicking on corners. Its coordinates are 33.7648S/25.4012E.
This info wasn't enough for me to figure out the location, but it's something. If I was serious about the challenge or there was way to monetize this, I could write OpenStreetMaps / Google Street View scraper that would figure out the location in a matter of minutes.
I think a better approach on his second step would have been to search "<major city> skyline", preferably in order from largest population to smallest. South Africa isn't that big. I think it pays to be more methodical and work with the information you know, before speculating. Speculating is great if you're stuck, but I think he mostly just got lucky there.
1. Locating terrorist training camps by high voltage power lines visible in the background.
2. Shia LeBeouf's IRL Super Capture the Flag, "He Will Not Divide Us," located and vandalized no less than 5 times. The last one used astronavigation principles, and visible contrails from airplane traffic.
3. Identifying muggers in crowds based on nothing more than biking gear and facial hair.
The one thing these had in common, was a sustained call for effort. By keeping the limited original details available and obvious, people in every timezone and demographic could view them. This greatly increases the odds of specialist knowledge or community insiders being able to add information to the detail, which goes back to the general audience, forming an action feedback loop.
Amatuer hour indeed, but when you have 10000 random people you get results pretty quick.
Or am I missing something?