Time |
Nickname |
Message |
07:44
🔗
|
|
pa has joined #urlteam |
08:15
🔗
|
|
pa has quit IRC (Quit: pa) |
08:34
🔗
|
|
pa has joined #urlteam |
09:00
🔗
|
|
Jonison has joined #urlteam |
11:15
🔗
|
|
pa has quit IRC (Quit: pa) |
11:46
🔗
|
|
pa has joined #urlteam |
11:52
🔗
|
|
pa has quit IRC (Quit: pa) |
11:54
🔗
|
|
pa has joined #urlteam |
14:06
🔗
|
|
pa has quit IRC (Quit: pa) |
14:06
🔗
|
|
figpucker has joined #urlteam |
14:11
🔗
|
|
pa has joined #urlteam |
15:13
🔗
|
|
Jonison has quit IRC (Read error: Connection reset by peer) |
15:24
🔗
|
|
pa has quit IRC (Quit: pa) |
15:25
🔗
|
|
jornane- has joined #urlteam |
15:26
🔗
|
|
jornane has quit IRC (Ping timeout: 255 seconds) |
15:27
🔗
|
|
pa has joined #urlteam |
15:32
🔗
|
|
pa has quit IRC (Client Quit) |
16:06
🔗
|
|
pa has joined #urlteam |
16:30
🔗
|
|
pa has quit IRC (Quit: pa) |
20:26
🔗
|
|
stevemcla has quit IRC (Quit: Page closed) |
21:30
🔗
|
JAA |
hub-me-a started. |
21:30
🔗
|
JAA |
I'm setting up hub-me-r as well, but I'll only start it once -a is done. |
21:42
🔗
|
JAA |
Aah, fucking spam filter... |
21:50
🔗
|
JAA |
So disq.us: There are three schemes: /CODE, /t/CODE, and /p/CODE. There's also /url?url=x, but that's a simple redirect with the full target URL encoded in the "url" parameter, so that's neither iterable nor worth archiving. |
21:52
🔗
|
JAA |
/CODE appears to be older. The codes are six characters long, case-insensitive, and seem to go up to 9?????. They redirect to either pages using Disqus or a specific comment. |
21:53
🔗
|
JAA |
/t/CODE and /p/CODE are newer. Codes are seven characters, only lowercase and not case insensitive (/CODE is also lowercase, but uppercase characters are accepted; /t/CODE and /p/CODE just return 404 if you uppercase any letters). |
21:54
🔗
|
JAA |
The codes for both /t/ and /p/ are still very "low", i.e. around 1?????? or 2??????. So it should be doable. |
21:54
🔗
|
JAA |
/t/CODE redirects to a discussion thread, /p/CODE goes directly to a specific comment. |
21:55
🔗
|
JAA |
Ballpark estimates: /CODE = 9 * 36^5 = 540M codes, /t/CODE and /p/CODE on the order of 36^6 = 2.2B codes. |
21:56
🔗
|
JAA |
So roughly 5 billion codes in total. Certainly a lot, but sounds doable. |
21:57
🔗
|
JAA |
Those lengths I mentioned are the upper limits/current maximum lengths; I've seen six-char codes for /p/, for example. |
23:12
🔗
|
|
dashcloud has joined #urlteam |