Lemme just say I want Python and RB back.. Grmbl…
“Hello,”
“World!”
ok
*sigh*
Via Karl, I have stumbled upon this excellent write-up: Wayfinding in Tokyo: Local Context and Direction Map Design by Chris Palmieri. While I know Tokyo well enough to navigate it – at least to go where I want to and not getting lost [but then again everyone who has walked a city with me knows that I have a GPS embedded in my brain] – I do relate to this article strongly, not only because of Tokyo, but of course because of Seoul.
I know Seoul, an immense megalopolis, better than my little 150,000 inhabitants hometown. There are not too many places I haven’t visited at least once. But this is, like Tokyo, a friggin’ large city without [mostly] street names and Napoleon-style house numbers – actually they have them now, but DON’T USE THEM! Idiots… So while you may know the area well, there will always be a back street, a new building that has served as landmark since… last week, that you don’t know about. As far as I can remember, meeting up with people has entailed a matryoshke-like approach. You point out the area, then define it a little more, down to the block of buildings where the person you are visiting is. And usually it ends up with a phone call or two.
Before mobile phones, coin-operated phones were available everywhere. And most of the time, if there was credit left from the previous call, the phone was left unhooked so that the remaining 60 or 20 won were available for the next person. Now that was social sense! You picked up a phone, dialed, and explained where you were. Someone came out to pick you up. Now with them gahdfershaken machines, aka mobile phones, you call ahead, even ask the cab driver to listen to the explanations, and usually make the phone companies richer than they should be. But 2006 or 1990, wayfinding hasn’t evolved much.
The problem is compounded by the fact that Koreans have close to no orientation skills. This is easily verifiable by first looking at the maps they produce – and at how they use them. I haven’t seen yet a map that was properly spatially arranged [you know, North on top, West on the left, that kind of thing]. Then, as Chris Palmieri mentions, the graphical quality of these maps leaves us wanting. Recently, during my last business trip, I took a vintner around Seoul to visit prospects. He had maps to go to each company, and he passed me the file. I had a look, and even for Seoul sub-standards, it was bad. Faxes, whose original had probably been sent hundreds of times and were by now illegible; very very sketchy details as for landmarks. The fact is, they expect people to find the place through a mixture of phone calls and taxi drivers. I’ll try to find some of them and post samples.
However, there are now very nice map services in Korea. Usually available only to IE/Windows, but there you go. My favorite – as far as I have a favorite – is Map@Paran. Properly oriented, lots of details, etc… But the experiences I’ve had with the cab drivers were unilaterally unimpressive, as they weren’t able to decipher them. Too many details, they were trying to orient the map according to the way we were going and our current location. But these maps were/are useful to *me*: as soon as I was dropped in the proper neighbourhood, the maps usually made it easy to find the office I was looking for. I once entered an office in a quite remote part of Seoul all on my own, and the office ladies almost freaked out. Where did that dude come from?
However, somehow, I like the old way of getting to a landmark and call for pickup. While it made things more complicated than necessary, there was an easy, sociable atmosphere about it that you don’t get much anymore. But then again, I’d really love it if they started to use the street names and building numbers…
Just so I don’t forget – or rather when I forget, just so I don’t lose and hour:
On the first machine [say my Linux box]
> erl -sname linux
Erlang (BEAM) emulator version 5.4.13 [source] [hipe]
Eshell V5.4.13 (abort with ^G)
(linux@toshibarf)1> erlang:get_cookie().
'KJTAQLETDTUEHWVNTHCE'
Now, on the second machine [my TiBook]:
> erl -sname tibook -setcookie 'KJTAQLETDTUEHWVNTHCE'
Erlang (BEAM) emulator version 5.4.3 [source]
Eshell V5.4.3 (abort with ^G)
(tibook@ddas-computer)1> Linux='linux@toshibarf'.
linux@toshibarf
(tibook@ddas-computer)2> net_adm:ping(Linux).
pong
(tibook@ddas-computer)3> nodes().
[linux@toshibarf]
Now, again on the Linux box:
(linux@toshibarf)2> nodes().
['one@ddas-computer']
Now both boxes are aware of each other. And say I have a third box, which I don’t so I’ll just start another erl session on my Mac. I’ll ping the Linux box, and watch what happens:
> erl -sname three -setcookie 'KJTAQLETDTUEHWVNTHCE'
Erlang (BEAM) emulator version 5.4.3 [source]
Eshell V5.4.3 (abort with ^G)
(three@ddas-computer)1> Linux='linux@toshibarf'.
linux@toshibarf
(three@ddas-computer)2> net_adm:ping(Linux).
pong
(three@ddas-computer)3> nodes().
[linux@toshibarf,'tibook@ddas-computer']
w00t! Since linux@toshibarf is aware of tibook@ddas-computer, then so is three@ddas-computer, without having to establish a separate connection… Very nifty.
Hmmph. PalmSecure™. Right.
/me slap bitch until barely conscious, drag her in front of ‘puter, swipe palm, all done. Or should I try to evoke a Snatchifying scene àla Boris The Blade, tea cozy and briefcase handcuffed to the arm? Chop Chop! Really? Methinks this won’t be necessary. I don’t have the answer to snarky questions like How are you going to make these biometrics gizmos check my bio integrity before going Open Sesame on my computer? but it sure feels necessary to have big brains work on it before touting these things as secure…
Joe Armstrong has started a blog. Don’t be put off by the rough edges, misspellings and other quirks, this is serious stuff. Joe’s one of the brains behind Erlang. Good stuff!
Let’s hope Joe posts regularly, it should be interesting…