Show HN: Blanking all Wikipedia as SOPA Protest Live Stats | Hacker News Reader

Show HN: Blanking all Wikipedia as SOPA Protest Live Stats

hacksandthoughts.posterous.com

163 pointssbashyal23 comments

I was curious so I was reading over the comments, and I noticed that most people seem to back this with the caveat that it only be done in the United States since it's not an international affair. That was the major thing I took away from reading the comments (both support and non-support comments pointed this out).

Bug report: on inner iframe http://phonnel.com/projects/sopatally.php: Uncaught SyntaxError: Unexpected token ) (line 23)

    data.setValue(3, 1, );

Edit: and we're back!

Yep, I'm seeing that too, and a blank chart.

Me too -- maybe he's protesting SOPA? :)

Could you say what the results are roughly? I cannot see the graph either.

Support 55.7% + Strongly Support 30.3% = 86%

Oppose 9.9% + Strongly Oppose 4.1% = 14%

I completely forgot about the YQL rate limits and was being capped by YQL servers. I have brought the service back-up and am currently implementing cache to minimize YQL calls.

In case the creator is checking for comments here but not checking his own page: PHP errors, not currently working.

(But if you open the 4 URLs shown in the errors yourself, they do work fine and give you the count. Support is beating opposition by some margin right now.)

Fixed it! Thanks for the feedback.

Not sure if I'm overlooking something or if you have a bug, but the numbers in your chart don't seem to match the numbers that Yahoo is returning. And, based on the previous PHP errors, it doesn't look like the problem is caused by caching.

Right now the numbers Yahoo are giving me are 9, 34, 177. 56 (strongly oppose, oppose, support, strongly support) and the numbers in your chart are 9, 24, 124, 58. (And when I say the numbers from Yahoo, they're taken straight from the URLs you created that were showing up in your PHP errors.)

edit: Interestly, both sets of numbers given an almost identical % for support vs oppose, 84.42% or 84.65% supporting. In terms of % for individuals (i.e. without grouping support and strongly support together) the %s come out with a much greater difference.

Good observation but you are missing a little detail. "Strong Support" is a sub-set of "Support" because each "Strong Support" is also a match for "Support". I am doing this arithmetic in my PHP script: Support = Support - Strong Support

Oh of course, can't believe I didn't realise that after writing my "edit: interestingly.." - should have been obvious.

Incidentally just reloaded your page to see no graph, and the source showed:

        data.setValue(0, 0, 'Support');
        data.setValue(0, 1, -58);
        data.setValue(3, 0, 'Strongly Support');
        data.setValue(3, 1, 58);
        data.setValue(2, 0, 'Oppose');
        data.setValue(2, 1, 24);
        data.setValue(1, 0, 'Strongly Oppose');
        data.setValue(1, 1, 9);

Feature request: Any chance of getting a graph to show support over time? Might be able to scan through the edit history to get the stuff you've already missed.

I am now logging entries in a database for caching. So it would be possible to create a time-series. I don't see a purpose though!

I was thinking of linking particular events (post made by Jimbo, major news story) with features in the data, something like what Google Trends does. I always find that information interesting.

Fatal error: Call to undefined function apc_fetch() in D:\Hosting\8254397\html\sopatally.php on line 4

So excited to see it too!

> HTTP/1.0 500 999 This page is currently unavailable in /home/content/r/a/n/rangeenchara/html/projects/sopatally.php on line 5

I'm no PHP guy, but looks like server's getting a 500 on your REST query. The query itself works nicely from my browser though:

<Edit: query string apparently gets munged here, can't paste it...>

Sorry about that! It is back now though.

A fundamental misunderstanding of how decisions are made on the English Wikipedia.

The author of this particular straw poll is likely to know a bit more about the workings of the English Wikipedia than you do...

I was referring to this graph and the premise that it supports.

Also, Jimbo has very little to do with the running of the English Wikipedia. He stepped down as a leader of it because of how often his actions were sharply criticized by the community that he was out of touch with. Just because you co-founded something 10 years ago does not make you an expert on it.

The graph shows how many people support/oppose this concept. And finding that out was Jimmy's aim here. Not to make a decision based on the resulting figures.

Yes, though I think this exposes some scaling problems with wiki consensus when there are a ton of people interested and huge outside media interest as well. Straw polls aren't votes, but they function best as non-votes when there's, say, fewer than 50, maybe 100 people weighing in. Then the approach is something like: 1) call a straw poll; 2) people indicate their support or opposition and why, not just a "vote"; 3) eyeball the opinions to see if there's clear consensus for or against, and what the main reasons people have pro/con are; 4) try to revise the proposal to take into account the main opinions expressed in the straw poll; 5) iterate until either consensus, or you fall back on finally having a "real" vote among a few distilled options.

That all works remarkably less well when there are hundreds of people weighing in.

Reply on news.ycombinator.com