adamsmith

Update: thanks to rbritton it seems like http://freedns.afraid.org is a good alternative.

So, if you sign up you'll probably just get a link to http://freedns.afraid.org if/when EveryDNS changes some or all of its services to paid only.

I've been using EveryDNS for over five years now. Great service. Frankly I'm torn because David deserves lots of credit and to do with the service whatever he wants. I just hope this isn't a quick buy and turn for DynDns.

They are obviously a for-profit company, did an ROI calculation, and there are probably limited options for monetizing their acquisition without charging customers for things they used to get for free.

As Dharmesh would say, if I were a betting man, and I am, I'd give two to one odds they start charging for services that are currently free in the next eighteen months.

This, of course, would open up the opportunity for someone to fly in underneath them. They probably expect that, but of course in the short term there will be a win because it will take time for that person to get the word out.

So I whipped up a quick weebly site: http://everydnsforever.weebly.com/

Submit your email address there if you'd like to get one email if and only if (a) EveryDNS becomes less free than it is today, and (b) a credible, great, free service is created in EveryDNS's place.

(By making this site I am hoping to do my part to increase the cost to Dyn Inc. of changing EveryDNS. Thanks!)

(Edit: grammar.)

Sorry for the late reply.

Here are the two I can think of off the top of my head: the first is that decision trees (or even naive bayes) produce results that are intuitively understandable and human readable. In contrast the result of training an SVM is a huge vector of 100's of floats and there's no real way to explain to someone (e.g. someone buying your tech, a manager, etc) what's actually going on.

I didn't mention this before but SVM's do get you something that decision trees, naive bayes, and others don't: they will look at all kinds of combinations of attributes. This becomes critical for applications like machine vision where e.g. looking at any one single pixel as an attribute won't do anything for you and so you need something that will understand lots of different possible combinations of attributes.

Which is also the second weakness of SVMs: most data sets that aren't images don't express strong patterns across a wide variety of combinations of features. To take naive bayes as an example, it is a hugely popular algorithm and the 'naive' part (which tends to work well on my data sets) is the assumption that each attribute is statistically independent!

Thanks for the link!

I wrote a new section called "Possible Extensions" under that article, including a subsection for the "Plausible Deniability" angle that cookingrobot came up with!

This is a brilliant point!

I added your idea to an edit I made on the Monolith wikipedia page, which now describes three variations of the Monolith protocol: Pseudorandom Basis Files, Lawsuit Denial of Service, and Plausible Deniability.

http://en.wikipedia.org/wiki/Monolith_(computer_program)

Thanks!

This doesn't work because some of the bits the software downloads couple be "innocent" bits. E.g. the firefox installer or some random looking piece of data. Any bits can be brought into involvement by the system, hence the denial of service.

'Never give up' is also awesome.

Incognito already runs without cookies or sessions, fortunately.

They don't even allow cookies within an incognito window.

It'd be especially cool if it took both front and back radar into account. If nobody is behind you break more liberally but otherwise exercise more caution.

(It does seem like there would be room for such an "aggressiveness" parameter, since you can't predict the future breaking behavior of the car in front of you and it might have stronger or weaker breaks.)

Either way I hope all of this stuff works out to be a net win. We should know more in a year or two!

Nearest neighbor and k-means will return noise unless your data only has two or "at most" three dimensions, due to the curse of dimensionality. (http://en.wikipedia.org/wiki/Curse_of_dimensionality) This is the #1 trap that newbies fall into, since these algorithms seem deceptively intuitive but there's this huge unintuitive pitfall.

I agree wrt points #1 and #2. Especially #1.

My favorite supervised learning algorithm is decision trees. They have built in feature selection, and they are simple to understand and express in code.

Boosted decision stumps are my second favorite.

Both are "far better than" support vector machines for the real world problems I've ran into. Except the algorithm names don't sound nearly as cool as SVM.

Seems unrelated. The New Scientist article doesn't describe mate selection or reproduction.

In fact, it seems like they just bred fruit flies that over emphasize their past observations. They didn't make "clever" fruit flies.

I wonder how hard it would be for Facebook to allow dissidents to set up a second password that logs into a version which hides certain types of incriminating information. The trick would be making it indistinguishable from the real thing.

(This idea is very similar to TrueCrypt's hidden volumes.)

> how [does] making money from money contributes to something worthwhile to society?

I don't have any formal training for credentials, but here's how I've come to think about it...

The markets (of which the stock market is one, but the venture capital and private equity markets also count) are capital allocation machines.

This is easier to see in venture capital. The market for startups makes it possible for promising startups to raise capital. It also "encourages" "bad" startups to die.

It's capital allocation.

Same thing for the public markets. They give companies access to capital (either through debt or equity) at a cost that's proportional to how valuable the company is, or is likely to become.

So society, through the open markets, gives Apple and GE the opportunity to raise capital.

Dying public companies, on the other hand, have a really high cost of capital.

And my favorite way to think of capital is man-hours.

So if you were king, and you were responsible for allocating man-hours to different projects, this is the same problem as capital allocation, and it's the same problem that securities markets solve.

They also provide liquidity and at scale the operational overhead goes to zero.

They also provide opportunities for investors to hedge against inflation, participate in society's long term growth, etc.

But the most important function from a societal standpoint IMO is very efficient capital allocation.

Yes, seems like trying to stop free markets from developing is just like trying to hide information or keep a theatre free of ringing phones. It can be tried, and sometimes even successful, but gosh it's a lot of work.

I don't have any real data or direct experience to back this up, so I'm really just pulling this out of my ass, but the NSA seems like a very different entity than "law enforcement" when it comes to privacy.

The rules, laws, hiring, and budget at stake is very different at the NSA versus your local FBI or police office. Think ten or eleven figures versus six. Polygraphs and M16's. Football fields of data centers.

Frankly the NSA can read my email. Law enforcement is a whole different beast.

Yes it seems like a great idea! It's the equivalent of the computational challenge approach in antispam. : )

That's good advice. I'll likely give it a shot next time around!

In theory given enough time you could go to investors and raise cash (binding in the event that your bid wins) backing a bid. It's the job of investors to allocate capital, after all.

Of course, I realize that theory does not always match reality.

Very briefly -- http://www.xobni.com/about/opensource

Great idea, good artwork!

Minor amateur opinion: the artwork seems too large on the shirts / too overstated.

I think sachinag is trying to say that consulting companies are services companies, whereas startups and silicon valley seem to revolve around product companies. The semantics of the distinction are around scalability of the business. You could also look at revenue per employee or P/E ratios as other indicators.

Personally I agree that scalability is a better metric for enterprise attractiveness than raw size. Investors agree. CSC has a P/E ratio of 5.8, whereas Microsoft's is 12.9.

The list also just feels mistaken, but that might be my misguided sense of enthusiasm for shiny companies. Fiserv is a $6B company that I had never heard of. I'd probably rather go long on Facebook or Twitter, but a savvy investor might disagree.

The picture is just like of Leonardo Da Vinci's "The Last Supper"!

I would disconnect the laptop and Internet and keep a dumb phone.

In my mind social is the opposite of computers and isolation, so trying to remove both is going to push in the wrong direction.

"..the corporation must be engaged in a trade or business other than: one involving the performance of services in the fields of health, law, engineering, architecture, accounting, actuarial science, performing arts, consulting, athletics, financial services, brokerage services or any other trade or business where the principal asset of the trade or business is the reputation or skill of one or more employees; a banking, insurance, financing, leasing, investing or similar business, ..."

Looks like we're out, guys..

Especially if it includes a torrent client!

I've found that aggressively filtering resumes does a better job than a phone screen. E.g. the presence of "Three years at Microsoft as an SDE" or "Avid reader of Joel On Software and Coding Horror" give you better people than just those who know what a pointer is.

One could always add a random length of random bytes to the message so the size comparison method doesn't work.

This assumes that real ciphertext is indistinguishable from random data without the key.

This explanation rings very true in the lives of the post-college people I know. Indeed I think this is the strongest effect in the system: the spreads between the two financial groups are about 10-20% at all points EXCEPT FOR the spread for males in the 20-30 age group where the spread looks more like 50%.

It doesn't surprise me that females aged 20-30 don't show this spread.

I'd expect to see a similar dynamic in sex versus grades for college age males. It's the stereotypical example of partying versus studying on a Thursday or Friday night, with $$$ substituted for grades!

If anyone wants to save five minutes and use the PDF I put together to send to my kindle, email me at adam at viratech dot c-o-m!

It seems like speed is a feature for v2.0.

It's definitely all for the users.

But!, I think our engineering team has a secret love affair with the code. There's some pretty fun stuff going on under the covers. I'll give three quick examples..

1) Xobni's data store sits strictly underneath the sidebar code in the stack. It was originally built to support Xobni Analytics, our first product from 2006 that bombed. Fortunately we got to leverage the same data store when creating the Xobni sidebar.

It's very cool. When someone is building software leveraging our backend, say the sidebar, or the Invite Your Friends feature, the code ends up looking like this:

  foreach(var mail in new MailIterator()) {
    Console.WriteLine(mail.Subject);
  }

This code will print 10k subjects per second, from disk! And it's from a key-value store, so it's easy to add new data fields and types.

2) Not only that, but the data store is built to be client agnostic above "layer 1" where we interface with the mail client. So when we wanted to integrate Yahoo Mail all we had to do was build the adapter piece that knew how to speak Yahoo's language, and suddenly the mail floats all the way up the stack and appears in the sidebar right next to Outlook emails. :-)

3) The areas where we display information from Facebook, LinkedIn, etc are all little embedded instances of Internet Explorer. The code for those extensions is all just HTML and Javascript. When the user changes the current email we invoke a specific JS function called updatePerson(), and there's a callback object the JS can use to make HTTP calls and write lines to the log file. This architecture, which was invented by someone smarter than me, allows us to pump these babies out quickly and without much QA risk to the other parts of the program.

It just doesn't get any cooler than this!!!!

(Commercial: if you're a developer and are interested in being part of the team, send your resume to jobs@xobni.com!)