Re: [edict-jmdict] Re: Bulk Text-glossing activity.

[Jim Breen <jimbreen@*********>]

On 25 July 2015 at 22:22, Jim Breen <jimbreen@gmail.com> wrote:
> On 25 July 2015 at 16:43, lucerno@vmail.me [edict-jmdict]
> <edict-jmdict@yahoogroups.com> wrote:
>> You should also try logging all the HTTP headers that get sent. Perhaps
>> there's something distinguishing there.
>
> There is some interesting stuff coming from them, but we're still
> working on it.

The HTTP Client information is interesting. It gives the name
of the browser, or the app that is doing the call.

In the case of all these text-glossing requests, it is identifying itself
as "TRAG". The variants that have come in are:

TRAG/0.4.9.2.r173
TRAG/0.4.9.3.r174
TRAG/0.4.9.3.r175
TRAG/0.4.9.3.r176
TRAG/0.4.9.3.r177
TRAG/2015.04.29 r178
TRAG/2015.05.14 r179
TRAG/2015.07.24 r180

Does anyone recognize them? It probably supports Marcus's
suspicion that it's a game of some sort. It can't be driven by
humans - I put a block on 151.252.214.76 over an hour ago and
it's made over 1,000 calls since then.

I've put the same logging into the Monash site, and it's not
getting the same calls (what it's getting is a lot of other
automated stuff, which i should look into.)

Jim

-- 
Jim Breen
Adjunct Snr Research Fellow, Japanese Studies Centre, Monash University