mmmturkeybacon PandA HEAD request

Accepts a HIT with a PandA link by a HEAD request instead of a normal GET request. Add &MTB_HEAD_REQUEST to the end of a PandA URL to run. This is a very bare bones script and gives almost no feedback to the user, so keep an eye on your queue and edit REQUEST_DELAY to suit your needs.

< Feedback on mmmturkeybacon PandA HEAD request

Review: Good - script works

§
Posted: 2015-12-27

Great script. feature request: stop refresh on caught hit or capcha?

MMTB you have done it again! My data use is much less with several
p&a links over using c4c, even with checking Hits assigned every few minutes.
though, I'm a little worried about amazon throwing surprise capcha
at me every few hits.

maybe once change of TITLE is detected, it could stop refreshing?
I've thought about scripting something with cURL to do head requests.
Maybe a button could be put on the page to pause refresh, like after
a manual refresh and a hit is caught, not a good idea to auto refresh.
I can see there is a potential problem with capchas if they appear too.
Continuing to refresh other tabs is not really productive at that point
till the user is warned and the capcha is cleared.
Maybe a capcha warning popup like the timeout popup?
A bit of a pain to have to keep refreshing a preview link to check
if there is a pending capcha to clear.
I know it's a new script, so thanks for putting it out there.

§
Posted: 2015-12-29
Edited: 2015-12-29

This script sounds like an interesting tool for people who have received Masters and thus no longer have MTurk captchas, but I expect there's no practical way for its design to be compatible with captcha-detecting for non-Masters... there isn't anything in the HTTP HEAD related to the presence of a captcha. You'd probably need to run c4c-or-similar on a 'preview?' page of an undesirable HIT to keep an eye out for captchas.

> "maybe once change of TITLE is detected, it could stop refreshing?"

The vast majority of MTurk page types all have the same uninformative title, "Amazon Mechanical Turk". That's why I made this a while back, which creates meaningful substitute titles based on the page's body content... can't be done from just the HTTP HEAD, or even the html head element. (The existing useless title is in the html head element, which is not what this script requests, fyi... these examples may clarify what this script is getting for you.)

Post reply

Sign in to post a reply.