Even in a normal economy, we will have about 4 million job openings. People naturally leave their jobs for various reasons creating job openings. Similarly, even in normal economy we will have about 4 million people getting a new job each month.
It was the ImageWriter dot-matrix printer that was released around the same time as the original mac. The pins in the printer cable were in the wrong place.
My father was a typewriter repairman, and his specialty was fixing the Selectric. It was quite lucrative -- these things were expensive enough to be worth repairing and were as important to companies as Internet is now. When I was 10, however, we were watching the Super Bowl and the infamous Apple 1984 advertisement came on TV. My dad immediately realized he was doomed. He went out and bought the Mac on the first day it was available. We brought it home. I plugged it in, turned it on, and my faster saw WYSIWYG for the first time in the text editor. The first words out of his mouth were "oh, shit." I went to print the document and it didn't work -- the initial batch of printer cables were incorrect. When we got the new printer cable from the store the next week, my dad decided to change professions and focused on real estate.
It's exactly like ice-nine. The universe exists in some sort of meta-stable state. The usual argument against this sort of thing happening is that we'd already be dead. Cosmic ray interactions happen all the time at very high energies, so these interactions have already been seen in the universe. The LHC, however, pushes the energy limits towards unexplored territory -- due the fact that it is a collider, the center of mass energy of the LHC may exceed the highest ever cosmic ray collisions.
All of my friends who have lived on both the east and west coast absolutely prefer west coast football times. I find it difficult to stay up until midnight or 1 AM watching Sunday or Monday night football and make it to work the next day. I feel sorry for those who attend the games. On Sunday mornings, there is no time to kill before the games start, and the local teams usually kick off at 1 PM.
The most important lesson I learned in graduate school came from a scientist at Los Alamos who told me "three sigma effects happen in physics much more often than they should"
Basically, I want to see if people can finish stuff. For software engineers, tasks tend to be easily defined and of shorter duration. At least at Bright, we have Data Scientists working on multiple simultaneous projects that take a few days to several weeks to complete. Knowing when you are done with a complicated, possibly open-ended, project is very difficult.
It you want to make an impact as a data scientist, it helps to prototype your ideas as a testable product. Many product managers and engineers ignore their data scientists' work because fighting fires almost always takes precedence. The recent HBR article (posted here) showed how the data scientist at LinkedIn was able to make a huge impact by running his own experiments against live users. At my company, we always provide a non-optimized prototype that can be run directly with our existing product. If the idea is good, we can optimize the code later.
Presumably the BBC outbid other British channels in a competitive bidding process. I suspect that broadcasters in the rest of the world know that 20 minutes of commercials per hour is unacceptable, so they don't bid as much as our networks. EDIT: The only source I could find indicates that the BBC paid about $100 million for broadcast rights to these games. That's 1/10th of what NBC paid, or about 1/2 of the American fees on a per capita basis.
I just watched the pirated BBC broadcast. I felt like something was missing, and then I figured it out. The stupid thing is that Americans are forced to watch the Olympics with 20 minutes of commercials per hour, while the rest of the world gets to watch it nearly commercial free. Why does the US citizen pay for the entire Olympics through the annoyance of commercials?
Here's the abstract for a nice review of search engine law: http://works.bepress.com/james_grimmelmann/13/ If I remember correctly, indexing a site that asks to not be indexed might be illegal as an illegal tresspass, but it is not settled law. The argument is that you are stealing resources (computer time) from the site owner.
I worked down there with IceCube. We had a few programmers in the project, but it was very competitive to get a slot at the Pole. Most of the people sent down were for physical drilling operations. I went down to do data acquisition software. IceCube is done with construction, so they don't need to send people any more. If you don't go with a science project, the other option is Raytheon Polar Services. Raytheon, however, pays their workers almost nothing because they know that people want to travel there. There were bus drivers with Ph.D.'s. The cook was a gourmet chef.
It's interesting that Microsoft is planning to charge more than Yahoo! Boss for access to the same data. BOSS charges $0.80 per 1000 queries, while Microsoft is priced at $2 per 1000 queries. BOSS is also pay-as-you-go with no monthly minimum. BOSS has less restrictive terms and conditions (re-ordering allowed, for instance). I've been using the Bing API instead of Yahoo! BOSS for my metasearch engine for two reasons: the feed is free, and the response time is faster. The Microsoft/BOSS prices are not unreasonable if you are actively monetizing your page with search ads, but if you are doing an experimental search engine it is very stressful. If I suddenly get a surge in traffic, I could be broke very quickly.
BOSS queries are $0.80 per thousand queries, so $0.0008 per search. A web search site that uses a premium sponsored link feed should be able to generate about $0.04 per query (50% share of the $0.09 per query that Google makes). Thus the revenue generated could be about 50 times the cost of the raw materials (the BOSS search results). BOSS is cheap in that regard. If you extrapolate to an entire year, DDG could be making about $15 million per year in revenue if they chose to show ads aggressively.
I was debugging a php memory issue yesterday, and noticed that at some point my get_memeory_usage() value became much, much smaller than the memory footprint recorded by top (20 MB in get_memory_usage(), 500 MB in top RES). It was a sudden jump during a loop execution according to top. Does top give accurate memory estimates for php scripts?
Errors like this have a life of their own -- there is no ill intent. Once a mistake is made in a seminal paper, most papers that build on that research will usually copy the error. Most authors do not test the statistical assumptions made in the literature and copy the statistical techniques used previously. I once traced an absurd FDA policy decision to a simple statistical mistake that had passed unchallenged through multiple papers and review committees.
If they were working on data center efficiency, that would be one thing. The RE<C program was an attempt to make solar power more efficient. This involves basic physics, material science, and manufacturing. Nothing that Google had done to that point would qualify them to work in that area.
I see they are shutting down their clean energy program. The renewable energy cheaper than coal initiative showed a unique level of arrogance. We, a software company, can solve a problem in which we have no experience better than people who have devoted their lives to the subject.
Most important I even learned in physics graduate school was "physics research has many more 3 sigma effects than it should." Part of it is that if you do 100 searches you are going to see one three sigma effect. Couple that with reporting bias (only the interesting stuff is reported).
Using Yahoo BOSS or the Microsoft search feed, anyone can build a fully functional, highly relevant "search engine" in a couple of weeks. "Search engine" is really a metasearch engine. Given enough traffic, you can syndicate a sponsored link feed and generate 3-4 cents per search. In principle, it is a great business. The place where all have failed is in the non-coerced acquisition of users -- everybody has to pay for users somehow, and by the time you've bought a one-time user you are forced to give them a terrible product (all ads, no quality).
What Gabriel has figured out and no one else has is how to build a search brand that people trust and are willing utilize. He seems to have a loyal core of users that come to his site without him paying on a per-search basis. Search quality, as perceived by the user, in as much about brand trust as it is about algorithmic quality. Too many companies focus on tech first and hope that the brand trust will come later (SearchMe, Blekko). Gabriel has actually invented a viable search business model. That's much more novel than his technology.
The interesting question is whether DuckDuckGo's organic user acquisition model can scale with additional resources. They put up one billboard in San Francisco and got a huge ROI -- will that work elsewhere?
An alternative is that neutrinos and photons travel a different distance because there are extra dimensions that affect the two types of particles differently. These are the so-called space-time foam models.
If I follow your workflow, my coworkers go ballistic. The problem is that the "git pull" messes up the revision history. I finally learned to use "git pull --rebase".
The trouble with this idea is that university researchers have very little cash and are loathe to part with it. Almost all of their grant money goes to paying the salaries of people in their lab (fixed costs). If they need to use a piece of equipment at another institution, the current method is to propose a collaboration with a researcher at that university. Additionally, any decent researcher already knows where the specialized equipment resides (they can figure it out by reading the papers in their field or chatting at conferences), so there is no need for a centralized database.
Are you sure you want to use your real name on this blog? You are advertising the fact that you are breaking the law in San Francisco (short term rental without a license) and violating your lease (subletting your apartment).
AdWords is a natural monopoly, so it is unlikely that someone will ever create a successful competing keyword driven advertising network. In order to generate high prices for the search ads, Google needs to have multiple bidders competing against each other for each keyword. Advertisers will only bid if there is lots of inventory to buy (lots of searches on those keywords). 10% of the search market does not generate 10% of search advertising revenue because of this lack of market participants -- on second tier pay-per-click advertising networks whole swaths of keywords are not competitively bid. The Bing-Yahoo search merger was an attempt to address this problem, but it hasn't worked so far according to the most recent financial reports. Google still generates 50% more revenue per search than Microsoft due to the network effects.
I did a similar study of traffic solely attributable to a single post on ReadWriteWeb. I wrote a technical paper with the results: http://www.stinkyteddy.com/documents/SocialMediaResponse.pdf The long tail traffic is well described by an exponential function with a "half life" of about 10 days. You should check your data to see if that number is universal.
Sharded search is not new. Solr, elastic-search, and Riak do it as well. The difference here is that we've built the Search on top of the BigCouch map-reduce view model. Views are calculated post commit so there are no data insertion locks. Multiple copies of each shard exist for fault-tolerance. Also, multiple map-reduce analytics passes can be used as input to the search.