sketerpot

Considering that "Fred" deliberately burned down a client's factory during payment negotiations, I think that "asshole" is quite an understatement.

I was playing with FUSE, and had managed to mount a loopback filesystem. And then, for some ill-considered reason, I tried to delete the directory that I had mounted it inside.

I managed to Ctrl-C it before it ate any files in /home, but I still had some nasty cleanup work to get the computer back in working order again. Thank god for LiveCDs.

To anyone who learns the idiom, this code is fine, and perfectly readable. I don't even know Ruby yet, and it took me less than a minute to become fully comfortable with this kind of code. This isn't a huge barrier to "non-experts".

A periodic table will make the atomic number and element type easy to look up. You can get it on coffee mugs. It'll be fun.

This one isn't just a cache; it's a database with a simple key-value data model. It has very good clustering support. It can handle datasets too large for memory. It's fast, allegedly easy to set up (I haven't tried it), and you can communicate with it using existing memcached client libraries. I think it sounds like an interesting option.

I also think that, in a few months, Redis will probably have the same nice characteristics of Membase, but with a much nicer set of datatypes. Membase does everything I described right now, though, which is worth something.

To clients, it looks exactly like a memcached server.

Saying "probably" is a way of avoiding dishonestly over-representing his confidence in the marriage lasting. Obviously you're offended, but don't let that be an excuse to let your reasoning slip.

Why would any computer automatically run programs on USB sticks? It just seems like a really obvious security hole for negligible benefit.

Of course, it makes sense if you interpret it right -- but it looks at first glance like it's comparing absolute sales figures, rather than change from baseline for each.

I would actually really like for my search engine referrer headers to be blocked, even without the privacy concerns, for one simple reason: some web sites highlight the search terms they find in search engine referrer headers. That annoys the crap out of me, and I usually end up either closing the tab or going to the URL bar and adding and deleting a space in there, then reloading the page without referrer headers.

Web developers: please, please don't highlight search terms. What the hell is the point of that? Oh well; I guess it's soon to become moot.

I guess people will just have to use one of the many other BitTorrent search sites with similar features and torrent coverage. This sucks for the IsoHunt folks, but for the people actually sharing files, it's more of an inconvenience than a crippling blow.

Excellent basic guidelines. Some of these might overlap with yours:

* Give the user content that they can see at a glance. If you can glance at your web site without seeing content, fix it.

* Use readably large fonts. Even those of us who have good eyesight usually don't want to read small text, and there are a lot of people whose ability to read small fonts has deteriorated with age. You don't really need to cram as much text as possible onto one screen.

* Column width shouldn't fill the whole screen without a very good reason.

* Keep it simple!

I just wanted to point out, that is a really misleading graph. All the lines start at "100%", so the scales aren't comparable at all.

I remember the early 2000s. I devoutly hoped that Linux would get widely used in some way, even if it never really took off on the desktop. I wanted open source to be fairly common and well-accepted, and not just something that weird people wrote and talked about on Slashdot. And hacker-friendly open-source Linux-based smartphones would be great, but that was just too implausibly awesome to even hope for. I wanted pleasant languages like Python to get serious commercial use.

Now, all those things have happened or are happening. Want to get on board with cloud hosting in any way? The standard options involve Linux VM images. Languages like Ruby and Python are mainstream and well-supported. Android is disrupting the smartphone market, and has the kind of widespread commercial backing I never would have thought possible for an open, Linux-based thing. Microsoft is no longer an unstoppable leviathan; they've stopped being scary.

Occasionally I'll reflect on all this and chortle happily. I wasn't optimistic enough back in the early 2000s, and that's surprising.

A good book for this sort of thing is Toby Segaran's excellent Programming Collective Intelligence. It walks you through this sort of fascinating thing with easy examples and clear explanations. It's sprinkled with simple Python code.

http://oreilly.com/catalog/9780596529321

If you want a good introduction to Naive Bayesian classifiers, there was a pretty readable explanation in Artificial Intelligence: a Modern Approach. It's an expensive book, but I'm sure you can find a copy in any well-stocked university library.

http://aima.cs.berkeley.edu/

And there's the beauty of Aubrey de Gray's plan: it's not about finding a Cure for Aging. It's about finding a number of techniques that will allow us to extend people's life-spans long enough to find out how to extend their life-spans more, which will give more time for research, and so on.

It's a much more realistic approach than "finding a way around aging," which is how the problem is usually miscast.

This isn't surprising, but is is disappointing -- and perfectly understandable, considering the devout wishes of the people at Facebook to avoid being killed by religious nuts. The real solution here is to decentralize to the point where Facebook (and similar entities) don't really have much say in the matter. I really hope something comes of Diaspora; until then, there are still plenty of non-Facebook places to post Mohammed pictures.

I don't know what kind of books you're writing, but 100,000 pages is too many pages. Did you mean 100,000 words?

I've only ever written 50,000 word NaNoWriMo stuff, but I've had a lot of success with text files, simply marked-up, in emacs. One text file per chapter. A short Python script converts them into LaTeX and a makefile builds a PDF. All the notes are handled with org-mode. It's a really sweet setup.

I couldn't find any, but here's an explanation of the input format, with some example snippets:

http://code.google.com/apis/predict/docs/developer-guide.htm...

It's pretty straightforward.

Directed Edge, a promising YC startup, makes recommendation engines surprisingly easy:

http://www.directededge.com/

It's quite a bit higher-level than what Google is offering here, with all the benefits and drawbacks that entails.

There's also Weka, which can use almost exactly the same file format that Google is using, and do the same kind of things (though perhaps with different algorithms). It's pretty pleasant.

http://www.cs.waikato.ac.nz/ml/weka/

A fun thing about the VP8 patent situation: any company which attacks VP8 with a patent lawsuit has their license to use VP8 automatically terminated. From the license:

> If You or your agent or exclusive licensee institute or order or agree to the institution of patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that this implementation of VP8 or any code incorporated within this implementation of VP8 constitutes direct or contributory patent infringement, or inducement of patent infringement, then any rights granted to You under this License for this implementation of VP8 shall terminate as of the date such litigation is filed.

Naturally, the most insightful comment so far is a reply to a massively-downvoted "wake up sheeple" post.

Directed Edge is a lot higher-level than this, and more focused on making it easy for their users to make good recommendation engines. Prediction API is focused on more general tasks, and for that reason I predict that it'll have trouble competing with Directed Edge. Unless some startup comes along and competes with Directed Edge using Prediction API as a back-end -- but we can talk about that later, if it ever happens.

Prediction API is a nice broad tool, but Directed Edge still has a big advantage in their niche.

Download Weka. The file format is almost identical, and it has a nice GUI to get you started fast. Google's algorithms and infrastructure may be more sophisticated and scalable, but if you want something a lot like this for a side-project, Weka is easy to start with and it's available right now.

http://www.cs.waikato.ac.nz/ml/weka/

It can be used as a text classifier. It takes as input a collection of (output, input1, input2, ..., input_n) tuples, stored in the newly-announced Google Storage, and then uses a variety of machine learning algorithms (which I would bet includes some Bayesian stuff) to build a model which it can use to take (input1, input2, ..., input_n) tuples and predict the corresponding output.

So, that's the API. You can do a lot of the same things offline, with almost the same file format, using Weka:

http://en.wikipedia.org/wiki/Weka_(machine_learning)

So if you're interested in playing around with Google's Prediction API, you should probably download Weka and fiddle with it some. It's pretty easy to get started with, and it will definitely give you an idea for the sort of thing you can do here.

That's a dangerous road to go down, though. Nobody should try to be the inferior option that piggy-backs on other, more competitive products.

I wanted to check it out at their web site, but it insisted that I upgrade to the latest version of Silverlight before they would deign to show my any of their content. Some people spend hours trying to get page-glances turned into users with minimal impedance; some people apparently do the exact opposite.

http://www.modelingtheworld.com/

They usually have a lot more than 15 words to choose from, though.

As I said, I'm fine with a filter that has poor accuracy, as long as it makes twitter marginally better.