isomorphic

If it was leaked externally, it'd be more than just pissing off Google, it'd be justification for removing Symantec's CA certs from browsers. I.e., who knows what other domains Symantec might be "testing" that would go undiscovered because the owners of those domains don't have Google's resources.

But as the other poster says, this probably wasn't leaked at all.

You most certainly can use QR codes to store data, using Version 40 QR codes. I'm sure PaperBack and QR codes have much in common. Also, QR codes let you adjust the level of error-correction that you want.

Even if a QR code has a maximum size, you could use more than one per page.

https://en.wikipedia.org/wiki/QR_code#Design

I'm sure there are other 2D barcodes that could be used for this application. E.g., Data Matrix (also in the public domain).

https://en.wikipedia.org/wiki/Data_Matrix

Many of these sorts of apps reposition windows and scale to half or quarter of the screen. What I really want is intelligent repositioning so that none of my windows overlap. (xmonad, without resizing the windows.) Failing that, magnetic edges.

Unfortunately, what I want is really the job of a window manager, or should be done at the OS level. On the Mac, there is no true user-serviceable window manager that can be replaced.

While your point about content providers is well-seen, let's be clear about what we mean by "unbundling." I think the consumer expectation is true a-la-carte, i.e., channel-level granularity in package choice. What Verizon has implemented is packages of bundles of channels, plus a required minimum base package. While this is progress, it is not real a-la-carte.

Personally, the add-on fees and mandatory equipment rental accounted for 50% of the nominal price of my FiOS TV bill. Packages would have barely dented this. Therefore I canceled FiOS TV earlier this year.

By the time the cable providers actually implement a-la-carte, no one under the age of 40 will be subscribing.

This is not unlike how TVs are adjusted in retail stores. (Overbright, oversaturated TVs must sell better.)

Although many images benefit from some amount of histogram equalization.

This is off-topic and not the same thing, but you should look into xhyve:

https://github.com/mist64/xhyve

...and then:

https://github.com/coreos/coreos-xhyve

https://github.com/ailispaw/boot2docker-xhyve

Previous discussion here:

https://news.ycombinator.com/item?id=9690846

This is great. I'd like the ability to live-filter over and above the ability to search. Also, more than 10 hits per page. It'd be wonderful to have some sort of voting or popularity metric to sort by. And, for pie-in-the-sky features it would be nice to have filters based upon contrast / hues / etc. other than just "dark" and "light."

> Ports 25, 443, 993 are needed in order to have a fully functional Own-Mailbox.

Even if you flout your ISP's server policy, there's still DHCP to contend with (changing IPs).

> "it proves concretely that someone existed at a certain time and place, as verified by another certain group of people"

We've had the PGP/GPG web of trust for a while, too. People use that, but it's not exactly popular (mainstream popular).

In order for this to become popular, it needs foolproof-simple UI/UX and great marketing. Concluding with "look up BlockchainID on Github" is neither (well, except for people who are already on the PGP/GPG web of trust).

That's pretty much all one needs to know regarding BitLocker.

"The purpose of Newspeak was not only to provide a medium of expression for the world-view and mental habits proper to the devotees of IngSoc, but to make all other modes of thought impossible."

Qi themselves mention that their technology is related to inductive coupling used to charge early pacemakers. Given that, I'd stay well away from these until I heard an opinion from a doctor.

Write a trivial app in both. Use your own judgment to determine the best tool for the job. It is entirely possible that the best tool will be neither of these!

The website in question is written in Ruby on Rails. Since v1.0 (and Ruby 1.8). And it is now on current Rails (and Ruby!). Rails is a moving target, to put it mildly.

If I had to render an opinion, I'd say I love Ruby. Rails I tolerate. Some parts of Rails are great; others are tedious. By its own admission, Rails is opinionated software. You may or may not share its opinions.

I developed the site entirely. To be clear: The site is essentially the interface to the business. I won't say I developed it "myself," since I was in a team with non-technical people who knew what functions they wanted from the site. However I was the only technical person involved.

Now we have more developers, for which I am very thankful. Unfortunately, way too much knowledge is locked up in me, and I never have time to commit that knowledge to bits, paper, or oral history.

Believe me, buses and beer trucks have been discussed as possible ends to my consulting gig. It is my fondest hope that this business has key-man insurance on me. I'd be pleased to know that they have such insurance, but less interested in knowing how much the policy is for!

I think comp1927 means Knuth proves that one can get better with age. Which is a fine point, but Donald Knuth is a difficult lead to follow.

I'm over 40 and I've always worked as a consultant. Until recently I was the sole developer on a site doing eight figures in revenue (now I'm team lead). Without being arrogant here, if my skills match but someone doesn't want my level of experience, it's not a good sign for the someone.

On the flip side, I would love to throw out all of the languages and frameworks I've been using for the past few years and learn entirely new tech. That's part of the point of being in this business!

If you're a startup and have a bias against us greybeards, you're only hurting yourself. We'll learn your framework-du-jour and tell you how it bears an uncanny resemblance to something that didn't work so well in the past. Or maybe we've never seen anything like it, in which case we'll be delighted to cast aside all that old crap.

I grant your point, but that's a bit like saying Einstein was a good physicist.

No. Perfect advertising will convince you to buy the advertiser's product, regardless of what that is. This is the mindset described in the original article.

You're describing perfect advertising for you.

Even if the ads were limited to a topic you deemed relevant, the advertisers would still essentially lie (or elide truths). E.g., they'd show you good deals on hotels in Vietnam without bothering to mention that people are regularly knifed in those hotels, or that they have bedbugs, etc., etc.

Perfect advertising for the informed consumer would be a list of objective specs that can be compared between similar classes of products, allowing the consumer to choose the best product within their budget, needs, etc. If you're a company that produces an inferior product, how would you advertise it in that scenario? Reduce the selling price, or lie, lie, lie?

A couple of years back, Ars Technica started putting up banner ads targeting adblockers: "Ads pay for the lights, etc." I felt guilty enough that I unblocked ads on Ars.

Returning to the site one day, I see the technical equivalent of a "punch the monkey" flashing banner ad. Aaand, adblocking back on, plus I now block the anti-adblock ad. And I no longer feel guilty. You had your chance and you squandered it.

Ars Technica styles itself as a site for well-informed technology-oriented people. The ads should fit the site. Insulting your audience's intelligence might work on Facebook, but these days you don't deserve to make money from advertising (on what is essentially a niche site).

Two can play at that game. You can store data in JPEG that can even survive recompression with lower quality. If it's worth it to you to go to all that trouble, that is.

Years ago I tried configuring Asterisk and was defeated. Well, not completely defeated as I got it "working," but I couldn't claim to truly understand the configuration.

Asterisk's configuration system isn't the worst thing I've seen, but it reminds me of Sendmail, which (if you don't know) is not a polite thing to say.

Of course now there are probably glossy Asterisk distros you can run in a VM and configure via WUI. But I'm obstinate enough that I want to understand the text config.

There is also FreePBX, but I know nothing about it.

Even further off topic, I'd recommend the Obihai boxes as a cheap way to get started. No affiliation, I just have an OBi110. It's about as easy to configure as Asterisk itself. (Why is all this VoIP kit horrible to set up?) However, Obi has a through-the-web quick configuration that will hook it up to Google Voice with no other hardware/software. That alone is worth the price of the box (which occasionally goes on sale). ~US$40 flat and Google Voice gets you free phone calls in the USA.

As a greybeard who was using Internet email in the 1980s, I could wave my cane around and claim this newfangled web thing is just a passing fad.

Instead, I will point out that, besides email's asynchronous nature (which others have mentioned here), email's biggest strength is that it is federated and open. (The latter two are also true of the web.)

Sure, many people either get their email from or host it on Google / Microsoft / BigInternetCo, but there's nothing stopping anyone from registering a domain, setting up an email server, and actually communicating with the Internet.

Try that with Skype / iMessage / Hangouts / Messenger / etc.

From this report I conclude that horses are 50% as dangerous to scientists as unshielded plutonium cores.

It's worth noting that Homebrew has "gpg" and "gpg2" as aliases for the "gnupg" and "gnupg2" formulae, respectively. This might be confusing to some people wondering why there are two versions of the same thing.

Thank you as well!

Thank you!

  <url snipped> - Sorry, the Public links of my Dropbox
  account currently are suspended due to bandwidth limits.

Does anyone have a mirror?

To be pedantic, when I picked my handle back in the /. days, I thought it meant "a bijective homomorphism." Or, less opaquely, an invertible structure-preserving map (in the context of abstract algebra).

When I read about "isomorphic JavaScript" as, essentially, views rendered with the same code on the client or server (and yet a separate API), I can't help but think "You keep using that word. I do not think it means what you think it means."

Perhaps "symmetric JavaScript" would be better? But then the API is still an asymmetry...

The thing the article is missing is that algorithms will hopefully improve along with storage capacity. That is, Google may be able to infer what is "Precious" nearly as well as you can. The question is whether Google will offer us that capability, or keep it for itself.

If film actually lasts that long, then why not print a digitally-encoded film? (Think full-frame QR codes or how Dolby Digital used to encode digital audio between the sprocket holes.) Sure, the film would probably be longer than a human-visual print, but then the encoding could include forward error correction. Don't use inter-frame compression (so that one part of the movie is independent of another), and for that matter don't get creative with the intra-frame compression--just use RLE or something really easy to reconstruct.

Also, when printing such a digital film, be sure to write down the encoding algorithms on archival-quality paper, and store that along with the film. ("We sampled an 8192x4096 image in anamorphic ratio X, with three channels, 16 bits per sample per channel planar, over the color space Y...")

As with all such things, if people were serious about it (or money becomes involved), this would be a solvable problem (or at least a tractable problem).