DISQUS

Hey everyone,

The problem is that in Prototype 1.6 the hash object is no longer API-compatible with an Object. That means you can't pass in a Prototype Hash where an Object was expected. To remedy this there is no a toObject() method.

I'll post a fix with documentation and the like, but for now you can fix it by adding ".toObject()" after ".merge(args)".

The call to Ajax.Updater should now be <pre>
new Ajax.Updater(pane, tabs[Event.element(e).id], $H({
asynchronous: true,
method: 'get'
}).merge(args).toObject());</pre>

belts,

If your site is going to be simple (read: only a few modules) and you don't have the programming skills, Drupal is fine.

um,

I'm open to all advice. Do you have any, or are you just complaining miscellaneously?

ken,

Yeah. I don't know what version what around when I wrote this article -- but that was 11 months ago. It'd also be unfortunate if Rails flipped out by default (i.e., required :id => false), since, AFAIK, Rails is supposed to be about "convention no configuration."

ken,

I think when I wrote this Rails would flip out if even the join table didn't have its own auto incrementing primary key. This might have changed, but it's also possible I was wrong.

grant,

You'd say there is a path from v₁ to v₂, but not that there is an edge between them. IOW, the edge relationship is not necessary transitive. In fact, the edge set doesn't even need to be represented by ordered pairs, so it's not even necessarily a "relation" in the strict mathematical sense of the word, e.g., you might have multiple edges between two vertices.

Most people mean "simple graph" when they say graph, so these "pathological" cases (i.e., loops, multiple edges, etc.) are excluded from the outset.

That said, if you have a graph G = (V,E) you could define another graph G' = (V,E') whose edge relationship is v ~ w iff there is a path between v and w in G.

I'm not sure this is a very interesting graph, though. You'd basically take the connected components of G and turn them into complete graphs. So G' is always a disjoint union of some number of complete graphs. I don't think it tells you anything particularly useful about G.

Hope that helps.

- Jesse

Jez,

Like so:
<pre>new TabbedPane('MyPane',
{
'pane1': '/downloads/pane1.html',
'pane2': '/downloads/pane2.html'
},
{
onClick: function(e) {
$('MyPane_overlay').show();
},
onSuccess: function(e) {
$('MyPane_overlay').hide();
},
evalScripts: true
});</pre>

Jez,

You shouldn't need to muck with the script itself. You can just set the appropriate fields in the <tt>args</tt> parameter — it's passed through to the Updater.

Mike,

X = f(W) is the random variable in question, if you want to make it even more formal. The specific properties of X, e.g., its CDF, are determined entirely by W.

Chris,

Yeah, I'm trying to get back into it. Ever since I sold Adonomics I've been working more than full-time on it. Between that and a desire for an active social life I don't have much time for blogging.

I'm going to try to write one article at least once every other week, though.

Jamie,

There's a way to use the invite UI with iframe apps. I forget how it works in detail, but it basically sends you off to an external page and then redirects back to your app with the results.

Look up the fb:request-form tag on the Facebook wiki. It should be documented there.

CT,

It's called Dean's Code Highlighter.

Saumitro,

In the general case we're not computing partitions of 26, but rather the value of Laundau's function. We only have to consider partitions of 26 because we're left with too few letters to fill each cycle.

Also, in the general case, I'm talking about cycles in permutations rather than partitions. Why I list 4 rather than 5 is a side-effect of the mathematical notation: cycles of length one are left out. Thus, up to conjugation, the permutation of interest is (1..4)(5..9)(10..16)(17..25). At least, that's how we'd write it in group theory.

So, IOW, if we have at least four distinct letters we can put one in each of those cycles, producing a game that lasts 1260 rounds. That the fifth, single-element cycle is left unfilled has no impact, as you said.

As for the name, I got it from Facebook. Blame them. :P

Prasad,

You're right, although that's not precisely what I said. If you have a callback URL of foo.mysite.com then facebook.com/myapp/ would go to foo.mysite.com.

They're 100% independent. I shouldn't have used the same URLs because it'd cause exactly that confusion.

Greg,

This system requires at least the prototype.js library. Are you loading both prototype and scriptaculous?

Coder,

I didn't mean connectivity in any technical sense, but yes, the vertex connectivity of that graph is 1.

I meant "connectivity" only as shorthand to talk about that nodes potential to transmit messages through the graph.

Andrey,

I only used Drupal because the company I (formerly) worked for used it. I'd fix it if I were paid to fix it, but I honestly don't care all that much.

I'd also add that I'm not anti-OpenSocial, but I think Facebook has solved a real problem, as evidenced by the fact that both their internal and third-party apps on the platform have taken off like a rocket.

If OpenSocial doesn't even address that problem then they're still two steps behind.

Matt,

The proof is in the pudding. My point is that it's not about the size of the networks (which is what the press says), the openness of the data (which is what O'Reilly) says, but distribution.

If OpenSocial can pull it of *then* we have a real fight on our hands. Right now it's a bunch of nothing.

patcito,

"The OpenSocial People data API hasn't been released yet; this document is a preview of the developer's guide that we'll publish when we release the data API. All of the details are subject to change, but this preview should give you a general idea of what the API will be like."

And as O'Reilly pointed out, this is like moving an inch in the direction of real data portability.

Daniel,

Yes, I understand Google is trying to outflank Facebook by playing the "meta" game. But, as O'Reilly pointed out, there's no data portability between networks so right now the most generous thing one could say about OpenSocial is that it's some sort of JVM for social platforms.

As far as developers are concerned they're going to be developing for verticals. Really, of course, this effectively means MySpace, since that's like 90% of OpenSocial's userbase.

We'll see how this plays out -- I'm not convinced.

Hone,

Lots of people say this, but the numbers don't lie. Facebook is actually growing faster than the application space, which itself is growing rapidly! Just look at the graph I included.

Dan,

You're right, but it'd be a pain in the butt to go back and correct it. The error will have to stand.

Jono,

View source and take it to your hearts content. :)

Vikas,

It depends on the size, obviously, but disk space and bandwidth shouldn't ever be a real concern. The bottlenecks will be your memory and your database.