sblom

That's not quite right. A choice quote from the kick-off call today helps explain more about what's going on here: "Some customers require Microsoft support in order to adopt a technology; some customers require community (non-Microsoft) support in order to adopt a technology."

You can think of the F# Software Foundation as the community side of that equation and, well, Microsoft as the Microsoft side of that equation.

What metrics do you want to use for purposes of this conversation? I.e. where do we look for vulnerability counts?

If your effort spent on tools speeds up your time-to-stuff, then the assumptions embodied in 1) are invalid.

Besides, with open source you have more than just the original developer on whom you can use cash to entice him to make your favorite changes.

I worked at Expedia while he was there. He and I didn't work directly together. Despite that he and I are only very loosely connected through 6 months as coworkers more than 10 years ago, it somehow feels like I just narrowly dodged a bullet.

I'm really saddened to hear about this. RIP Justin. And best regards to your survivors.

I've also seen an incredible amount of judgmentalism applied to Objective-C's message passing syntax. Does it look foreign to a speaker of classical C? Absolutely. But I found that pressing through the initial shock, I've come to realize that it's just another language syntax, and not particularly hard to either understand or type.

It has an incredible amount of elegant power for some things (see http://www.youtube.com/watch?v=a9xAKttWgP4 if you want to immediately fall in love).

But I've found that trying to use it for most things just isn't worthwhile. It feels like cheating every time I have to escape from its array programming core to the more familiar flow control structures that only came along in later versions of APL.

I think it makes a good puzzle language (i.e. the fun comes from figuring out _how_ to say something rather that just raw getting stuff done), but unfortunately, it's even more of a puzzle to figure out how code that someone else wrote works.

For full context: I've spent more time using J than APL itself.

This link got me the paper in PDF form with absolutely no hassle: http://web.eecs.utk.edu/~langston/courses/cs594-fall2003/HL....

(Appears to be the faculty web site of one of the paper's authors.)

If it's python3, the previously built-in reduce() has been relegated to the functools module. But I don't think this is python 3.

I'm not sure what it is, but I know I don't like "guess which subset of the language you get" being harder than actually solving the algorithms challenges.

One reason that I can think of is that a spam account here doesn't tend to get very far. Sure--they could spam the New Submissions list, but they'd have trouble getting upvoted and thus polluting the front page.

They could create a bunch of accounts that all upvote an article, but it's easy to detect articles that only get upvoted by low-reputation accounts as clear spam.

I guess the bottom line is that the reputation scores here provide a lot of high-quality signals for spam detection that a normal site wouldn't have.

I've tried it before on a previous project of mine, and the results were pretty bad. Many users walked away because it wasn't obvious why we needed access to any of their Facebook information--we weren't asking for any permissions until much later in the relationship, but users still balked.

There weren't any great ways to notify a user and it was unnecessarily hard to allow users to interact with people for whom they new an email address but weren't friends on Facebook.

I think your best bet is to encourage users to create an account on your site/app directly (perhaps with a Facebook/Google option), and then later link it to Facebook once they understand the benefit to them of doing so.

To me a developer unwilling to learn Windows: The Good Parts has always been like an IQ test, too.

I agree with peregrine on webforms--just not clean, hard to make feel like a modern web app oughtta. The good news is that ASP.NET MVC 3 makes it easy enough to completely avoid WebForms.

There are certainly aspects of this movement to which I'm sympathetic, and I see lots of ingredients in their campaign that give it a good shot of taking root. But I wonder if there's really enough passion in anonymous sympathizers or if it's mostly just entertained on-looking.

Magic Trackpad makes it just as awesome on a desktop machine.

Confusingly similar to "When banks compete, you win"? (A live trademark of Lendingtree, Inc.)

Seems to me that if a node can have multiple edges pointing to it, and an edge can point to multiple nodes, nodes and edges are effectively the same thing, or at the very least have identical properties. Am I missing something? I suppose it's a useful convention to still partition a hypergraph into nodes and edges.

That's really awesome. Who would've guessed MSDN magazine printed articles that out there? Might have to give it a try.

You mean "motion".

Is anybody else seeing strange, deterministic glitches where some of the video seems to have been snipped out?

I've published my python solutions for reference, kibitzing, critique, etc. on github: https://github.com/sblom/hyperpublic-challenge

Anyone else published solutions? I'd love to see one in Clojure.

Hmmm. Why the cryptic assembly language stuff? Surely this isn't (_very_) cleverly disguised malware...

It's a trap!

There's a shipped project that ties this in to OpenID. http://emailtoid.net

Some of the OpenID crowd were aware of it at the time, but it hasn't really caught on. It's kinda fun to use it as an OpenID provider to log in to itself. Very meta and clean.

I have a hunch that you could change its appearance substantially by tweaking the probability of connecting sets and connecting vertically.

Can you even tell us what types of startups exhibit this behavior? Might be helpful to the OP.

Just use IE. Unless silly corporate reasons have you on IE6 also... :-p

"Maybe the Wintel intentional software bloat to sell more hardware has reached a critical mass and can't be stopped."

Having worked on the performance team for Windows Vista (yes--it actually exists), I can tell you first hand that there's no conspiracy here.

Instead, there are thousands of developers each solving problems in their parts of the OS without a holistic view of the impact of their tiny little growth in memory use or CPU consumption or processes required at startup. Each team's impact is minuscule, but in aggregate, adds up to enough to fully use a new generation of hardware.

The perf team can help fix some of that, but only so much. And they only bother winning enough battles to make Windows run on the hardware that people willing to buy an OS upgrade are going to put it on. The vast majority of these buyers are tech freaks like us, or people buying a new machine from an OEM, so it's not malice so much as lazily chasing the easy dollars that causes commercial OSes to run best only on near-latest hardware.

Fwiw, Apple's similarly lazy...

Love it. Seems like the perfect compromise between no system and a too-rigid system.

I wouldn't call it "Getting Everything Done" so much as "Getting (Some) Things Done".

If you remember what a linked list is, how pointers (or references) work, and have written more than 10k real world lines of code in your favorite programming language, I would fully expect you to be able to write, not memorize, any of the linked list functions that you mention.

I certainly understand your frustration, but I really don't think either the tone or the content of this blog post is going to help land a job in this era where hiring managers ask Google what it knows about a candidate.