izendejas

Thanks.

Read the literal sense of the phrase.

To use your line of reasoning, let's take another anomaly: Steve Jobs. The dude was adopted, was a community college dropout and did a ton of drugs. He woke up one day at 42 and found himself back at the CEO position of the company he co-founded and now he's considered by many the most successful CEO of our time. His company has a market cap of $300 billion+ and has created a whole lot of other intangibles by surrounding himself by other great people who love what they do.

Creativity and passion is not something you can teach your child--much less beat into them (this actually achieves the contrary). But yes, that doesn't mean some discipline and other blahs aren't important.

Ha.. I had a similar idea. Mine would also know when you were running out of item X (eg, based on weight and average delta) and order groceries for you based on this. I like your recipe idea better though. Mine involved getting planned meals for the week and using that information to also pre-order groceries.

I never implied otherwise (I agree), just spreading the word about appbrain.

FWIW, my girlfriend doesn't like playing Wii and absolutely dislikes hardcore games, but she was blown away by kinect's demo videos--especially the dancing ones and said she'd definitely get a kinect if within her budget at some point.

Anecdote aside, Microsoft really has to open up the platform as much as possible and embrace developer inovations. If they do this, they'll finally take over the living room. I have a feeling the media center division needs to get out of the way almost entirely for this to happen, also.

Android has several issues and I agree that the market app sucks, so I use appbrain. Give a try if you haven't already. It gives decent recommendations and you can filter apps by hotness among other things.

That last statement a key benefit of an app market. Tangentially, I've seen developers driven nuts by downvotes for bugs they fix in no time, but that can take days to get approved with native apps. Or on android, by fragmentation issues in which apps don't work well on certain phones.

As a developer trying to reach a wide audience but that doesn't want to go through the hassle of getting apps and subsequent updates approved, OpenAppMkt is a possible solution. Your saavy user will know OpenAppMkt just creates shortcuts, but how do you explain such concept to your iphone-using mom who doesn't know what a browser is and thinks google is the internet and much less that a web site is really just a (web) app?

And how do you create a shortcut to the khan academy app on yor home screen, anyway? At least on Android, I can think of a not so intuitive way that your average person won't easily stumble upon: 1) google such site/navigate to it directly; 2) then bookmark it on yor browser; and to create a shortcut (app button), 3) tap-hold an empty spot on your home screen; 4) select shortcuts; 5) select bookmarks and 6) finally, select the khan academy bookmark.

Using OpenAppMkt is just like using the regular app market. The only problem, potentially a killer one, is getting your avg user to download it. But here, I hope social recommendations and links just like the one contained in the page above (which take you right to the download page on the respective app marketplaces) will help.

I love that it's an html5 app--it's the way to go imo, unless you're developing a game. And I was even more glad to learn about OpenAppMkt. It solves key problems with html5 apps: discovery and easy "installation" (icon/bookmark setup) for avg users.

I'd say google IS an AI company. They do doc classification, nlp, speech to text, vision, etc. They may not be great at some parts of it, but their systems are constantly being trained and getting smarter as they release more products and acquire more data.

I've experienced two bugs.

1) a warning that two of my messages failed to send wouldn't go away after "clearing" them in the status bar. Had to go to settings and clear all data and force stop the app. Not critical, just very annoying. 2) when auto-completing a name and selecting such name, it actually adds the next person in your contact list as the adressee.

Thanks for sharing! This is a great set of resources.

@primgenus, @vladd. good luck to you and your teams!

Can someone please develop an awesome web-based IDE, so that we, developers, are not left out of the Chrome OS goodness? You can include seamless integration with github, or dropbox, for "cloud" storage. And of course, I'm assuming HTML5 will allow you to work offline.

Has anyone used Bespin and/or any existing web-based IDEs? Googling didn't return anything compelling.

You also have to realize that you're training Google's algorithms to better understand the things that matter to you, so that even though you spend less time reading emails, now they can target ads better, leading to more clicks (and higher bids due to better targetting).

Looks like they definitely just scanned some books. I downloaded a sample of "Information Theory and Statistical Learning" and the font was too pixelated and thus unreadable on my android device unless I zoomed in. Zooming in only evidences the pixelation even more.

An O'Reilly book seemed fine, so I want to imagine that technical/math-notation intensive books are hard to optimize for e-reading. Anyone have some knowledge about this? How hard is it to make a technical book's digital font easier to read if you don't already have the LaTeX/pdf?

That's a very important question to ask and I myself would like to get a response from some of the readers who're hacking away at it, instead of having you get downvoted.

How does one prevent a catastrophy that may result from "rogue algorithms" that are far worse than Infinium's? And is such risk truly worth it?

Funny, I was going to start and end my post with similar statements. My first statement was a quote "a person is smart but people are stupid" (due to herd behavior). And I was going to end with the fact that many leaders care more about maintaining their seats than losing any lives on their watch and so are ready scare the minority who actually are not afraid of terrorism much more than they are about religions, people and ways of life they ignore (most of the author's examples are in fact evidence of this).

But I edited my post because statistically the majority of us are not afraid. We care, and definitely vote for leaders who refuse to terrorize us a la Cheney, but we know we have to adapt our way of life to minimize the risks--which can be confused with fear, but is actually the rational thing to do, since we're dealing with irrational people.

I was too succinct, but yes the majority needs to do a better job of controlling the message we send to terrorists--that I should have conceded.

We're not really afraid/terrorized. We only require that leaders protect us and some are willing to forgo more liberties than others.

If we were terrorized, we would not fly by the millions... every day!

I won't get into the civil liberties topic. There are trade-offs we will all never agree on. One of those is security v. liberties. So everyone just tug their way and we'll see where we end up.

In the recommendation system space, it's called a sparsity problem (and/or a cold-start problem). You basically need to continuously feed it more data about your tastes for it to start doing a good job.

I assure you someone, more than likely Google, will one day develop this. Greplin could also position themselves in this space?

I ported it to Java a while back and was amazed at how a simple heuristic (looking for blocks of text with lots of commas) went a long way. Unfortunately, it has several drawbacks which have prevented my full use of it and came up with an algorithm which turned out to be similar to boilerpipe's (using decision trees) that worked better in some cases. I haven't gone back to this problem, but I have some ideas on ways that it can be improved.

I actually suggested that there be a <usercontent> tag to one of the people on the HTML5 committee because among other things comments sections throw off many of these web scraping algorithms. After 3 emails, he didn't buy my argument, so I gave up. So all web designers, please oh, pretty please, use the up and coming <article> tag correctly. Many data/text miners will thank you dearly!

The reason Obama, Bill Gates, et all promote college is because there's a huge difference between a Google and a Facebook--not to underestimate the latter. And notice that they more likely usually promote STEM (Science, Tech, Engineering and Math), not just any college.

Could Brin/Page or anyone, for that matter, have come up with a Google without finishing an undergrad degree in CS and some graduate Studies at a school like Stanford? Very unlikely. Google required a bit more than creativity. It required great hacking and also knowledge of the complexity in information retrieval problems which you didn't get back then with online lectures (what lectures?) on dial-up, etc.

Meanwhile, Facebook as great as a company that it is, mostly borrowed ideas and tweaked them a bit with "exclusivity," a clean design, etc. Not saying that isn't admirable, but yes, it doesn't require graduate level work, just creativity and a lot of hard work to get jumpstarted. And I bet Zuckerberg didn't start hiring high school dropouts/grads to help him scale his site, etc. Surely there were a few, but I doubt they accounted for many.

So if you want to build the next Facebook, you better be hard-working, smart, and creative.

If you want to build the next Google, you better be that and learn about the next challenges in machine learning, information retrieval, data parallelism, etc. And that more than likely requires that you learn it from the experts--ie, more than likely not your average college/high school dropout.

The key word there is if (in the first sentence). I doubt it.

Telcos/ISPs are headed the way of utility companies. They are natural oligopolies because the barrier to entry is too huge for there to be many competitors and it's in everyone's interest to have them functioning well and at least partially profitable, so don't be surprised if they get subsidized or even nationalized.

The companies that pick up one's garbage happen to be cities in most cases. So as stated above, here you might even be right--if by irrelevant you mean, that we take them for granted yet are still extremely important (eg, without electricity we would have little, without garbage services we would die sooner, etc...)

"Stupid" comments are one thing, "stupid" questions are another. With "stupid" questions you at least demonstrate a willingness to learn and not just spew political talking points, biases, etc.

I self-censor because of this and think we shouldn't mind. I end up writing comments I don't submit, but I consider the exercise worthwhile because forcing myself to write my thoughts out and then judging whether they're truly helpful/insightful teaches me what knowledge gaps I need to fill eventually and where I'm just falling back on preconceived notions/biases to form my opinions.

So if you think you'll be down-voted ask yourself why. If you don't know why, then re-formulate your post as question to understand others' reasoning.

Yes, and we all end up paying the indirect costs of this cheap labor because, as is well-publicized, most of these people won't have health insurance so premiums, medical costs, Medicaid costs, etc go up.

If legislators really wanted to reduce illegal immigration they would also go after the demand, implementing tougher penalties against those who hire illegal immigrants^^ or outsource to companies that do. Of course, this will never happen as the walmarts, mcdonalds, big tech corps^, etc will never allow it because surely their costs would go up and so would their prices.

Edit: ^ work at a big tech company? A decent percentage of the janitors and cooks are illegal (and, for the record, are not just Hispanic)

Edit2: ^^Including themselves--because they do.

edit3: (sorry I was on my smartphone before). If the supply of jobs for illegal immigrants goes down, so does their influx. It's true in the US--and it's true in Europe and it's true in Asia: http://www.eubusiness.com/news-eu/1247156221.99/ http://www.mcclatchydc.com/2010/05/01/93137/recession-enforc...

But then, of course, in a global economy, cutting the influx of illegal immigrants (and immigration in general) would lead to more jobs shipped north/south/overseas.

So in the end, many of these legislators are really just pandering and contributing to the fear-mongering here and everywhere as people are, in many cases, naturally xenophobic.

When will people learn that opting people in by default while pretending to be Mother Theresa out to save the world is going to backfire? If you want to offer a personalized experience that requires privacy-sensitive data gathering, give users a good reason to opt in and give them full control. Otherwise, how are you any different than spammers? If people don't opt in, then your "personalization" sucks and you need to rethink your strategy.

Were some of those citations made to prove those findings wrong? Dumb question, but would be a shame to not be thourough when attacking bad literature.