dlg

In the economics world, tax incidence is a matter of serious study. There has been a decent amount of work on who actually pays oil taxes.

Hal Varian (who later became Chief Economist at Google) wrote a good piece[1] summarizing the basics in 2000 for the New York Times.

The summary: "A gasoline tax in a small country falls mostly on the residents of that country. The world price of oil is essentially independent of the taxing policies of most countries, since most countries consume only a small fraction of the amount of oil sold.

"But the United States consumes a lot of oil -- almost a quarter of the world's production. That means it has considerable market power: its tax policies have a major impact on the world price of oil, and economic analysis suggests that in the long run, a significant part of a gasoline tax increase would end up being paid by the producers of oil, not the consumers."

[1] http://people.ischool.berkeley.edu/~hal/people/hal/NYTimes/2...

Brookings, a center-left think-tank, has a pretty good summary of a lot of stats about income-mobility http://www.brookings.edu/~/media/Research/Files/Reports/2008...

(Also, a shorter note from the center-right AEI https://www.aei.org/publication/tracking-the-same-households...)

Sadly, this is not likely to have a good outcome--they drew anti-civil-liberties judges.

For example Michael Daly Hawkins had a dissented to argue that it's ok for a school to strip-search a crying thirteen year-old to search for ibuprofen. http://articles.latimes.com/2008/jul/12/nation/na-strip12

More generally, it would be cool if every account automatically got two free read-only SQS queues: one for events and one for upcoming events. Every known event (startup, terminate, permission change, network error, etc) could be published to the queue.

Not high-priority for us but the parent comment sparked the idea (I'm not in devops so maybe this exists via a different mechanism.)

At the time Facebook was building messages, Cassandra was still pretty immature/buggy (it's very solid in my experience nowadays).

Moreover, FB needed a lot of people who knew the tools. They acqui-hired a team of HBase experts. In 2010 I would have chosen HBase over Cassandra for what they were doing. In 2014, I'd choose C*. But as always, it depends on your application and what you know (or can hire for).

Cassandra has tunable consistency. A lot of people just think "eventually consistent" but it really lets you make tradeoffs yourself about performance, consistency, etc. Plus, the way they do read repair, etc even at CL==ONE makes it a lot better for messaging systems than you'd guess. I'm using it successfully with most stuff at One.

It's not available yet, but will be posted first at http://clerk.house.gov/cgi-bin/vote.asp?year=2013&rollnumber...

SF HNers: Rep. Pelosi was against the ammendment.

Anyone who works in interactive computing should be familiar with Doug Engelbart's work.

Years ago, at one of the first conferences on interactive computer, after people spent all day presenting their new work, Andy van Dam--builder of the first hypertext system with Ted Nelson who named it--stood up and said "you should all be ashamed that you don't know your history. Doug Engelbart invented almost everything presented here years ago." And he was right.

I hope a lot of you are watching The Mother of All Demos http://archive.org/details/XD300-23_68HighlightsAResearchCnt.... But it's worth reading and understanding the reasons why Doug was working on all of this.

Doug and his crew at SRI had the goal of "human augmentation". Everyone else at the forefront of the computer industry thought we'd have general AI by the 1970s. They instead believed that GAI wasn't within reach. They believed that the things we wanted to build and accomplish as a society weren't doable with the communication tools we had.

They had the idea that computers could be tools to help individuals work. Since computers were multi-million dollar calculating machines, the idea that people would have a computer at their desk and that they'd help us to communicate and manage information was beyond-out-there.

http://www.dougengelbart.org/pubs/augment-3906.html

But after leaving Engelbart's group at SRI, lots of his team joined PARC and built the modern GUI and networking.

For a history of the details, I highly recommend reading Markoff's What the Dormouse Said http://www.amazon.com/What-Dormouse-Said-Counterculture-Pers...

[Bonus: If you like Engelbart's MOAD, also watch Ivan Sutherland's Sketchpad Demo.]

I only tend to comment about once-a-year on HN, but it would be impossible not to say something about Doug and the impact he's had on our world.

Edit: Sutherland's Sketchpad http://www.youtube.com/watch?v=USyoT_Ha_bA

There are a few tech startups in Hollywood trying to help build a new model. I recommend this recent post for a sense of where things are going http://briannorgard.com/2013/06/14/rewiring-and-rethinking-h...

The scenario they're talking about with the SEC regulating secondary markets is the most plausible theory of and unwing I've heard yet.

This article was hugely important.

A radar operator named Doug Engelbart read this on the trip home from WWII. He realized that the computer was the tool that would make this possible. He went on and created everything from the mouse to lots of modern interactive computing. Many of you may have seen his "mother of all demos". If not, watch http://video.google.com/videoplay?docid=-8734787622017763097...

That demo, in turn, influenced a generation of young computer scientists to invent interactive computing. Notably Ted Nelson & Andries van Dam who created hypertext and Alan Kay who, while at PARC, invented the desktop GUI and the concepts for tablet computer.

I recommend reading "What the Dormouse Said" by Markoff for a history of that era.

(I consider this the "NLS thread" of computer history. The other major interactive thread is from PLATO to Ray Ozzie/NOTES to Mitch Kapor, et al.)

I'm an angel investor in Ad.ly, so I'm not 100% aware of everything going on there, but I at least thought I'd chime in.

First, Ad.ly wants to make sure the ads stay high-quality. That's why they have pretty strict rules that everything must be marked as an ad, nothing is written in the first person, everything must be approved by publishers, etc. It makes it harder--as Gabriel saw. They're working on ways to make publisher approval better (there is an email that is sent to handle approvals; I'll ask the ad.ly guys if there's an unusual error rate).

Second, I know they're working on lots of tools to make it easier to choose appropriate publishers. Everyone wants to use different metrics for how they choose, so it's a tricky problem, but the feedback is useful. Pricing is based on several measures of influence, and they tend to correlate well with CTR from what I understand.

As Gabriel noted, the CPC is usually significantly better than almost any other high-quality ad network. So, it's cost-effective on a click basis. I hope with the tools to get volume as well, they'll make it easy to get volume.

Be careful: If you're in the US, to do most angel investing, you must be an accredited investor, per SEC regulations. To qualify, you must have a net worth of at least $1M or have made >$200K in each of the last two years (>$300k if married). Almost any company with a good lawyer will require that you sign something certifying that you are accredited.

- US & State Government tax information. The tax code is amazingly complex. It would be good if you could figure out the tax implications of various events. Better if they simplified the tax code to where this was possible. Moreover, data.gov so far is a disappointment. Maybe they'll improve it. But I'd love to dig in to purchasing and personnel costs for every branch of government.

- Good stock market historical tick data (and streaming data). Opentick tried for a time, but mostly you need to contact expensive services. This makes it hard to mess around.

- A good local event API. Lots of companies have tried, few have good results.

- LinkedIn: They've been promising an open contact book API since 2007, but have kept it closed. If they had an open API (that lets you actually store data/invite), it would mean a lot of sites would be building on them.

- My dream: The big scientific journals would require everyone publishing a paper to upload all relevant datasets to a central repository which could be queried.

- Anyone you have an account with (Financial firms, banks, vendors) would have a standard commerce API. Sure, today you can export stuff in Quicken, etc formats, but Mint had to do a big deal with Yodlee to get the data in a uniform, queryable way.

And, mad props to Microsoft for opening the Bing API with pretty good terms. Google used to have a search API and it had horrible terms. Then they decommissioned it. Who would have guessed that MS would be more developer-friendly than G?

I'm on a plane that's about to close its doors so can't write more, but what a great guy. In school, I used his Randomized Algorithms books which was great.

I met him randomly, but he helped us meet the ultimate acquirers of our startup and gave us a bunch of great ideas about algorithms, software, and who to meet. Invaluable for two kids who had just moved to the valley.

Wish I could have worked with him more--was actually hoping to on the next startup.