FanFare stats? August 12, 2016 10:20 AM Subscribe
I was looking through the trusty Infodumpster and realized it has never had FanFare added in. While I'm fuzzy on the specifics, I am aware in a vague way that the Infodumpster isn't an "official" Mefi thing, though it comes from official MeFi data or some such. In any case, will FF ever be added to that data and/or Infodumpster?
That all makes sense. Thanks for the answer.
In terms of potential FanFare stats, though, is there anything in particular folks would like to see, aside form the stuff already represented in the Infodump files?
That's a great question. Maybe we could ask things like "Which shows have the most posts/comments/favorites?"
posted by DirtyOldTown at 10:34 AM on August 12, 2016
In terms of potential FanFare stats, though, is there anything in particular folks would like to see, aside form the stuff already represented in the Infodump files?
That's a great question. Maybe we could ask things like "Which shows have the most posts/comments/favorites?"
posted by DirtyOldTown at 10:34 AM on August 12, 2016
Heh, I meant more like what kinds of raw data for the Infodump files, not to volunteer new knock-on Infodumpster interface work for CEL. Not that I'm assuming they'd mind the brainstorming, just don't want to write checks that they have to cash.
posted by cortex (staff) at 10:37 AM on August 12, 2016 [1 favorite]
posted by cortex (staff) at 10:37 AM on August 12, 2016 [1 favorite]
I realize that this is nuts to request, but any chance to open source some of the Fanfare database schema for us to play around with so that we could query up a way to pull the data out in a way to gel with the rest of the infodump?
posted by iamkimiam at 1:03 PM on August 12, 2016
posted by iamkimiam at 1:03 PM on August 12, 2016
What everyone REALLY wants to know is when the Corpus will be updated. The lurkers support me in email.
posted by Chrysostom at 1:32 PM on August 12, 2016 [3 favorites]
posted by Chrysostom at 1:32 PM on August 12, 2016 [3 favorites]
I realize that this is nuts to request, but any chance to open source some of the Fanfare database schema for us to play around with so that we could query up a way to pull the data out in a way to gel with the rest of the infodump?
It's a nice thought, but, yeah, probably a little overboard in practice. The difficulties lay more in the assumptions and general cruftiness of my perl code, rather than difficulties with the database schema itself; I'll need to revisit how and what I query to account for things that might be true for the union of the original big three subsites but not so for newer ones, add some new loops and conditional cases, etc.
Like cleaning out a spare room; it won't be too bad and won't take too long, I just need to make sure I have the energy and the extra (head-) space to actually get it done.
posted by cortex (staff) at 2:02 PM on August 12, 2016
It's a nice thought, but, yeah, probably a little overboard in practice. The difficulties lay more in the assumptions and general cruftiness of my perl code, rather than difficulties with the database schema itself; I'll need to revisit how and what I query to account for things that might be true for the union of the original big three subsites but not so for newer ones, add some new loops and conditional cases, etc.
Like cleaning out a spare room; it won't be too bad and won't take too long, I just need to make sure I have the energy and the extra (head-) space to actually get it done.
posted by cortex (staff) at 2:02 PM on August 12, 2016
Yeah, it's not the most inviting of tasks! /stares at own attic of side projects
posted by iamkimiam at 2:42 PM on August 12, 2016
posted by iamkimiam at 2:42 PM on August 12, 2016
Like cleaning out a spare room
See you're doing better than me already
posted by aubilenon at 11:01 PM on August 12, 2016
See you're doing better than me already
posted by aubilenon at 11:01 PM on August 12, 2016
cortex: "...and general cruftiness of my perl code,"
As awful as 2016 has been so far, one of the good things about it in comparison to say, 1994, is that we have these great scripting languages like Ruby and Python now. Gone are the days when writing a data migration script was to be as eagerly anticipated as a dentist visit for a root canal or getting hit in the head with a brick.
posted by double block and bleed at 1:52 AM on August 21, 2016
As awful as 2016 has been so far, one of the good things about it in comparison to say, 1994, is that we have these great scripting languages like Ruby and Python now. Gone are the days when writing a data migration script was to be as eagerly anticipated as a dentist visit for a root canal or getting hit in the head with a brick.
posted by double block and bleed at 1:52 AM on August 21, 2016
« Older I hate football, but...I love fantasy football.... | A ten year update... and an SF meet-up. Newer »
You are not logged in, either login or create an account to post comments
We don't generate any FanFare data right now; it'd be nice to add that at some point in the future, but it'll require setting aside some time to do a bunch of revisiting of the existing code and adapting it to deal with a different set of metadata.
That adaptation is the biggest speedbump; the subsites that the Infodump currently crunches are all more alike than different under the hood, whereas the absent subsites have more structural idiosyncrasies from one to the next in terms of what's stored in the db and how. Otherwise it'd be a quick-ish job. And FanFare is probably the most structurally experimental of all the missing subsites, and so likely the most work.
So at some point I'd like to set aside the time to sit down and hack through it, but I'm not sure when that would be.
In terms of potential FanFare stats, though, is there anything in particular folks would like to see, aside form the stuff already represented in the Infodump files?
posted by cortex (staff) at 10:32 AM on August 12, 2016 [2 favorites]