TimothyFitz

I love the graphs, but the article only presents data that would back up it's conclusion. I want to see the graph of access times and memory usage as a function of data size. Those are the two factors that are definitely going to be worse for CFArray over his naïve C arrays.

All major browsers will now do this for you, with data: urls. Take a look at the examples here: http://en.wikipedia.org/wiki/Data_URI_scheme

Here's the original source, which has the raw data with less editorializing: http://www.fidelity.com/inside-fidelity/individual-investing...

And here's the same survey from last year for comparison: http://www.fidelity.com/inside-fidelity/corporate/fidelity-s...

It might sound old school, but we're waiting until we can show success with a seed round before raising an A or B round!

Does Programming, Motherfucker really scale though? Every process listed is about teams of people Programming Together, Motherfucker. I've seen two people engaged in Programming, Motherfucker without Talking to Eachother, Motherfucker and the results were Disastrous, Motherfucker.

For any Hacker News readers at SXSW, Team Canvas will be around, drinking at other company's parties because our booze budget is still $0. Text me (http://TimothyFitz.com for my phone #) if you'd like to hang out.

Edit: In case you're wondering, in that photo Chris is working on Canvas via Skype chat.

Wow. It's dramatically better! Within about 20 seconds I was filled with that rush of curiosity / grandiose ambition that Civ creates so wonderfully. I immediately closed the window. Fuck. I don't have time for Civ. Must ship more code, startup crunch time, no time for playing I tell myself. I know I'll be back. Fuck.

I can't perceive the difference between domainzomg.com and instantdomainsearch.com. The latter is using plain old ajax. Theoretically I'm not even sure if I should see a difference. A plain old ajax call will get sent on an existing sleeping HTTP 1.1 connection, and should take 1 packet each for the request and response. WebSocket can't get any better than that.

tl;dr: Request/Response model on the web is already fast enough for small data. Use WebSocket where it matters, like a firehose of updates in either direction.

This is the exact reason why Chad Austin started working on ibb (I/O-Bound-Build): http://chadaustin.me/2010/03/your-version-control-and-build-...

Ship http://canv.as

The name makes me immediately think janky, not something I'd want to trust :)

Is the name some kind of reference I don't get (beyond Django, obviously)?

Related: Steve Blank's hypothesis that dysfunctional families breed people for the chaotic world of startups: http://steveblank.com/2009/05/18/founders-and-dysfunctional-...

This would be more of a "Big Deal" if it hadn't been available for many years in Mozilla via PyXPCOM: https://developer.mozilla.org/en/PyDOM

That being said, there's definitely room for competition. PyXPCOM is far from a joy to work with.

It seems paradoxical that one liners could dominate the comment section. They're only one line!

Of course, I completely agree with you.

Perhaps the real problem is that when a one liner gets upvoted it brings with it the one liner's replies. What's the only valid reply to a one liner? Another one liner, in response to the first, drifting further and further away from the original point of the conversation.

"Third parties can’t prevent or control your transactions." But they can devalue the currency by minting money and keeping it. Also the currently is worthless by definition as no one is willing to pay money for it.

Both of these are common problems among all new currencies, and the most common way to defeat both is to have the currency issuer back their currency with something of known worth (gold, a fiat currency, etc). If conversion to/from the backing currency is common, then most transactions become bound by the problems of both currencies, for example paypal is effectively a currency that is more or less worth USD * ##%, because of the currency switching costs.

I can't wait to see a crypto-based-currency jump these hurdles, but I can't yet imagine how they'll successfully do it (and perhaps some already have that I don't know about?).

In a nutshell, you're right. This looks like, on paper, a bad investment market, and it might be.

Indie Fund doesn't have to turn a profit and they'll be successful: they're fundamentally about funding the games that they want to see developed that otherwise can't for financial reasons.

Look at their submission setup. You won't find the word "market" there in any form. Not once do they ask about the potential upside. This is not an angel fund in any traditional sense.

I think it makes more sense to think of the "IndieFund" as funding Art and not businesses. They care about how the product will change them and the people around them.

A few weeks before the most recent WWDC. I believe he left for a stealth mode startup (which I don't know the name of, or anything else for that matter).

Pipelining is not enough even if the internet infrastructure was perfect. Why?

1. Slow requests stall the entire pipeline. 2. The initial request stalls the pipeline. 3. The pipeline is being pulled instead of pushed so your performance is still roundtrip-lag-bound if you have cascading references. For example an HTML file references a .css file which references PNG costs 3 round trips at least.

Google's SPDY proposal does about what you mean, and more: http://www.chromium.org/spdy/spdy-whitepaper

SPDY is definitely implemented in Chrome/Chromium, though I don't know if it's on by default. I believe at least some google services (google.com in particular) support it publicly as well.

This code is almost as old as the web itself. The debug loglevel stuff was in the original check-in in 1994, by Guido himself: http://svn.python.org/view/python/trunk/Lib/httplib.py?revis...

A simple find-replace would be enough to switch it over to python's built-in logging module.

There are dozens of things that would have to be broken to make this work, for instance sys.set_trace. Either tracing is broken, or tracing is rewritten with a new threading strategy (that's not this strategy) or you're back to GIL-land.

The pricing backlash is fascinating to me, because it appears that it's not that Twilio is overpriced but more that OpenVBX is appealing to consumers/prosumers that Twilio wasn't really reaching before.

So now I'm curious what's next: Does Twilio go after a non-b2b market? They could add a $X/mo unlimited usage plan with whatever caveats they need to make it financially affordable, or they could ignore the market and chase the simpler, more lucrative b2b side they're already good at.

Tangentially related: Four or five years ago I was an R&D intern working on replacing an ancient EOL'd embedded processor. We were targeting an FPGA with a soft processor (Alterra's Nios II), and found out they were powerful enough to run Linux directly. The device was doing realtime motion control, so we slapped two soft cores onto the FPGA and had them talk over drop-in bus. One did realtime, the other ran Python. It was a nifty way to hack on an HTTP status page, a test runner... not sure if it ever shipped.

I've been doing this with sshfs (Linux/Mac) or ExpanDrive (Windows) for years, with whatever my editor of choice is for the file in question.

If I'm reading the current status graphs correctly they're regularly seeing 1.5-2.5 second queries and 400ms direct fetches. That seems unbearably bad. Can anyone running mission critical apps on the platform comment?

Perhaps not so well known (which is to say, I didn't know this): Woz was a UC Berkeley dropout, but after Apple went public he went back and got a B.S. in EECS.

Agreed, this is such a common problem. Another really common manifestation is when you A/B test while looking at multiple variables. If you observe the results of 20 variables, you should expect one of them to have 95% confidence!

It's worth noting that splittester.com calculates 95% confidence that those two (identical) ads will have different long term response rates.

At first I thought: Like every other "static Python to compiled language X translator" this one will only translate some static-enough subset of the language, at which point it's not Python. Interesting as usual, but mostly academic.

But then I started reading the source code, and the amount of python features supported with regression tests is impressive! (Especially for something coming from the Haskell community which is notoriously anti-automated-tests).

http://github.com/bjpop/berp/tree/master/test/regression/fea...

Especially cool is the fact that you can use callCC!

Disagree. This is exactly why terms like "We're friends on Facebook" exist, which effectively means "I can remember seeing their face online".