jzila

There are those of us who look forward to exactly that: the ability to own our participation in the content we consume. When you pay for content, you are the customer. When someone else pays for the content, you are the inventory.

Yes, ad blockers work. But they are susceptible to the tragedy of the commons: it's always advantageous to install an ad blocker. Then, as they proliferate, companies must insert more and more intrusive ads. This encourages more people to get ad blockers.

Allowing a tiered economy for content will enable those who wish to pay to do so, and those who don't to see ads. The ads can become less intrusive because the companies are no longer playing a cat-and-mouse game with their inventory. That should decrease the incentive to use an ad blocker, allowing that revenue stream to stabilize.

Thanks for the feedback! We've recently implemented some pretty big performance improvements in the filesystem, including prefetching files, caching them encrypted to your local disk, and journaling for writes (allowing us to sync them over some period rather than blocking on your network connection). If you're still experiencing poor performance in the filesystem, please file a ticket on github.com/keybase/kbfs or contact me on Keybase chat (keybase.io/jzila).

Great question!

We give a little bit of detail here: https://keybase.io/docs/kbfs#frictionless-sharing

But the basic model is that when you share or chat with someone@twitter, that content is only encrypted to your devices. When that Twitter account posts a proof, and announces that proof on a Keybase account, Keybase's servers will notify your devices including a link to the tweet. Your device will independently verify that the cryptographic proof is validated by the keys of the Keybase account claiming it, after which it'll re-encrypt the keys to that data for the newly verified Keybase account.

This all happens seamlessly in the background.

First of all, congratulations on an excellent job with the design and UX of Blue Ocean! There's tremendous potential here.

I hate to use HN for bug reports, but since you asked about roadblocks (thank you for asking!):

For one, we use build triggers to ensure downstream dependencies are tested as well as the current project. I would have expected some kind of pipeline integration between those in BO (displaying the triggered build pipeline in the parent pipeline as a whole), but instead BO actually makes it more difficult to get to downstream builds.

Classic:

https://jzila.keybase.pub/classic_downstream_build.png

https://jzila.keybase.pub/classic_downstream_build_detail.pn...

Blue Ocean:

https://jzila.keybase.pub/bo_downstream_build.png

Note that in BO it isn't clickable, so I have to either manually type the URL or go back to Classic.

Some display bugs in the pipeline view:

- Issues with display of pipeline nodes in running builds that make what's happening pretty confusing: https://jzila.keybase.pub/running_build_display_bug.png

- Issues with display of pipeline nodes in finished builds (falsely reporting node failure): https://jzila.keybase.pub/finished_build_display_bug.png

Minor grievances:

- Nodes run in parallel don't have their hierarchy preserved: e.g. parallel(a: {}, b: parallel(c: {}, d: {})) all get collapsed into a, b, c, d parallel nodes in the BO pipeline view (as you can see in the bo_downstream_build screenshot above).

My contact info is at https://keybase.io/jzila if you need more details.

It's great and beautiful. However, it isn't yet as powerful as Jenkins Classic. So if you've been using Jenkins for a while, there's a good chance that Blue Ocean doesn't have the power/granularity you'll need.

It also switches your default GitHub links to Blue Ocean. If you only want to try it out, you'll have to tell your users to switch their Notification URL back to Jenkins Classic in their user configuration page. (If there's a way to do this for everyone, I'd love to hear it.)

I can't seem to enable indexing for my team's public repos, only my own. Other sites with GitHub login (e.g. Travis) allow you to login as your team rather than your individual user. Can this be enabled?

There has been at least one high-profile sex trafficking bust in the Seattle area recently [1]. That seems to correlate with an uptick in opinion articles decrying prostitution and all its problems.

I think most people can agree that sex trafficking is terrible, and we should pursue approaches to minimize it. Studies have shown that existing attempts at legalizing prostitution do not at all curb trafficking: in fact the increase in demand cancels out the increase in supply, keeping the market open for traffickers and abusers [2]. Given what we know so far, it's probably worth keeping prostitution illegal if we can't figure out a better way to keep sex trafficking down.

That said, there is a fascinating paper that analyzes the economics of the regulatory approaches that have been tried in various countries, and makes a recommendation for a legalization framework that might benefit all parties [3]. The authors note that while criminalizing supply (prostitutes) has no downward effect on trafficking, criminalizing demand does, and in fact trends toward 0 when the penalty is severe enough. They recommend a hybrid approach that regulates and licenses voluntary prostitution, while heavily criminalizing consumption of unlicensed prostitution.

If nothing else, this is a fascinating topic with plenty of room for further research. Conflating prostitution (transactional sex) with human trafficking doesn't serve to further that discussion, and they should be reasoned about separately.

[1] http://www.seattletimes.com/seattle-news/crime/online-site-w...

[2] http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1986065

[3] http://lsr.nellco.org/cgi/viewcontent.cgi?article=1303&conte...

I actually wrote something like that for Keybase before KBFS was announced: https://github.com/jzila/kb-login-ext

The approach you detailed is the way to go once the filesystem is in the wild. Such power!

From the article: "...hydrogen sulfide under a pressure of 1.5 million atmospheres exhibits phonon-mediated superconductivity at 203 K".

So, yes, dry-ice level temperature, but incredibly high pressure.

I think the point was to illustrate the difference in energy requirements for the rocket to impart to the payload.

That said, re-reading Elon's article, he did explicitly say _kinetic_ energy in the 120GJ figure, in which case 0 is the right number for New Shepard at 100km.

Sure, but it's energy that a rocket needs to impart nevertheless.

Potential energy is energy, so it's not quite correct that New Shepard had 0 joules at 100km. At 100km and 0 velocity it'll have about 1 gigajoule of energy per metric ton of mass.

This is beautiful. It's the first time I've been able to understand special relativity in terms of geometry.

Your visualization makes it clear how everything from time dilation to length contraction can be derived from simply knowing the Pythagorean theorem and the constancy of c.

Sam Altman talked about this in his Employee Equity post: http://blog.samaltman.com/employee-equity. "The idea is to grant options that are exercisable for 10 years from the grant date, which should cover nearly all cases (i.e. the company will probably either go public, get acquired, or die in that time frame, and so either the employee will have the liquidity to exercise or it won’t matter.)"

Of course, that isn't advice specific to your situation, but at least it shows you a general direction you can use to guide your thinking. There are also plenty of resources online that talk about the potential tax consequences of various exit scenarios.

Previous coverage: https://news.ycombinator.com/item?id=9622325

This is great. I especially love that candidates are given a choice, so if they prefer the traditional technical interview they can choose that.

Normalizing performance between the two interview types will be challenging, but I think the benefits to be reaped far outweigh the difficulty of the challenges.

So you're running all the containers on the initrd kernel?

While I like the approach of attempting to derive a moral framework axiomatically, the resulting philosophy experiences one of the main pitfalls that burdens authoritarian Leninism: it fails to motivate independent agents within the system.

You can accept the author's implication that discrimination based on intelligence, charisma, good looks, etc is immoral. You can even accept that business and commerce are inherently unfair because they discriminate on such characteristics. Despite all that, a business-based (capitalist) economy is the best way we know to align the interests of the individual with the interests of the society. The author's philosophy completely fails on that front.

There are other failures to this philosophy, but I think that one is the most egregious.

My favorite part of this is that it reduces the n*m engineers-to-employers search problem to an n+m problem (to an extent).

Right now, engineers search through many companies to find a job, and companies search through many engineers to fill a position. The traditional way this is done is an interview gauntlet, whose primary purpose is to verify technical ability. Since that's something that should only be done once per engineer, TripleByte seems to be providing that intermediary service. Then a much shorter set of interviews can be used to check fit, thus saving each candidate and company an incredible amount of time.

hired.com is another company trying to do something like this, albeit using a slightly different methodology.

edit: I accidentally a word.

If you'd like to try the extension, the README has an overview. However, if you don't want to peruse the code before trusting the extension with your private key, you can create a test Keybase.io account and try it out with that.

I'm privileged to be working with Jessica on a contract right now. She's not only the smartest person I've ever worked with, but she manages to elevate people to a level of ability they don't even realize they're capable of.

Her ability to process information and distill critical data is unparalleled: she'll find out everything she needs in 2-3 questions tops. I've seen her do this in front of both highly technical audiences and less technical ones: she perfectly tailors her speech to exactly the level that her audience needs. She sets a great example for how to communicate in a work context.

I think this is what makes her such a fantastic engineering leader: she uses her unique skillset to help remove as many barriers as possible. This enables everyone who works with her to be a better version of themselves, which pays huge dividends.

Carl Zimmer published a piece on this in the NYTimes almost a year ago (http://www.nytimes.com/2014/05/05/science/young-blood-may-ho...).

Here's the HN thread on that topic: https://news.ycombinator.com/item?id=7695621

This isn't _really_ news. It's been known for a while that cells can be made "immortal" by repairing telomeres [0]. In fact, for stem cells and cancer cells to keep their immortal properties they use various mechanisms to repair telomeres [1]. The major issue is that the more times a cell divides, the greater the likelihood that it will develop cancer. Some theories postulate that the Hayflick division limit on cells is actually a defense against cancer.

The Wikipedia articles on telomeres and telomerase have plenty of further resources[2][3].

[0] http://learn.genetics.utah.edu/content/chromosomes/telomeres...

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3370421/

[2] https://en.wikipedia.org/wiki/Telomere#Cancer

[3] https://en.wikipedia.org/wiki/Telomerase

edit: formatting

Why not just create a plugin for Jetbrains' IDEA platform? Unless you're building a featureset that isn't supported, but that doesn't appear to be the case here. It would be much cheaper to build, and you'd get the rich IDEA ecosystem for (almost) free.

From the repo the project creator linked in the README (https://github.com/yrutschle/sslh): "A typical use case is to allow serving several services on port 443 (e.g. to connect to SSH from inside a corporate firewall, which almost never block port 443) while still serving HTTPS on that port."

In the video he calls changes() on a query that has an orderBy and limit filter. In the description for Changefeeds (http://rethinkdb.com/docs/changefeeds/javascript/), it says this is not possible. Is this new functionality? If so, that's pretty great. Using the diff format makes that pretty intuitive (even if the client will be slightly harder to implement).

I sometimes wonder if this is just an elaborate money laundering scheme.

Are there any plans to patch this into OpenSSH? It looks like they have a great solution for the problem you highlighted.