Wow does "UC Davis Professor" ever bury the lede here; Phil Rogaway is one of the most famous working cryptographers. He's also notable for having patented the best known AEAD scheme and granting a free license for all nomilitary use.

For all the moaning and complaining I've seen on HN over the persistence of this topic, the conversation still hasn't gotten loud enough, and hasn't gone on long enough.

As much as MA rubs me the wrong way, I applaud him here. What an incredibly non-trivial and important use of his public platform with those guests.

This is not a standard terminal replacement, it's a Chrome extension providing quicker access to koding.com's vm service. Vague headline, cool product but not relevant to me right now.

Wow, that's a boring attitude. I think it's great this guy is interested in typography and crafting his own fonts, and that the results are quite nice, professional or not, with the right principles or not.

Just don't get why it's on the front page before release though. I guess for comments, but add a place where I can sign up to be notified when it's done or something...

No, it's not. Not even close, and this is really terrible reporting. It's unfortunate that so many young people are carrying large amounts of student loan debt, but there is no sign of a "bubble" bursting any time soon. I don't even understand how the notion of a "student loan bubble" makes sense, since people don't "speculate" on college like they speculate on real estate.

There are so many things wrong with the idea of a "bubble" popping. People can't walk away from student loan debt, like they can walk away from real estate debt. Interest rates are expected to rise, but only slightly with the 90-day T-bill yield. Not the insane rates 20% rates that banks were hitting people with in the subprime crisis.

People may find diminishing returns for college vs the rising costs, but that isn't the same thing as banks suckering people in with ARMs and then using leverage to make huge speculations on CDO's that lost 90% of their value.

Here's a madeup example from the subprime crisis. Someone puts $0 down and buys a $500k home with a loan from the bank. The bank expects to make $500k in interest over the lifetime of the loan due to ballooning interest. Now, let's say variable interest rates from the loan kicks in, and the person can no longer make his monthly payments.

In the past, when the real estate market was good, he could just sell the house for $600k, and keep a profit. But, lets say now the price of the home dropped to $300k. There is no way you are going to expect someone who put $0 down on a home he bought 2 years ago to cover $200k in debt for a home he doesn't even own any more due to foreclosure. So, now, instead of having something worth $500k in profit, the bank owns something that is a $200k loss.

Banks lost over a trillion dollars worth of assets in a matter of weeks.

Look at what was happening for years before the crisis though. In those times, it was very likely that a home could increase in value from $500k to $600k in a few years. How was that happening? A lot of it was because people figured out that they could put $10k down, and do this to make $100k, or 10x their investment in a few years. The true speculators are the people who just bought real estate only because they thought the price would keep going up. This in turn drives up the price without being tied to any kind of intrinsic value.

You simply don't have that kind of pure speculation in the college loan industry because you can't just buy and sell college degrees.

Also, people aren't just going to declare bankruptcy from their student loans in the same way as the subprime crisis because they legally can't. Wages can be garnished from student loans, and it's extremely hard to have the debt erased due to bankruptcy.

The "bubble" scenario is sensational, because everyone sees a trillion dollars evaporate in a few weeks. Instead, in the case of student loans, I think that we may see something just as bad for the economy, but it will happen over years, not weeks.

In a lot of ways, I think that's why this problem may be worse, because it may be just as bad, but much harder to notice than the subprime crisis. Journalists think that the only sign of economic failure is when a "bubble" pops though, which just isn't true. Slowed growth over 20 years could be just as bad if not worse.

That is exactly how criminal bullying succeeds whether it is organized crime moving into a neighborhood, or an authoritarian government rising to power.

There are plenty of people along the way who can put their foot down and say "NO" but all they have to do is threaten them and many people say, "I'd rather not get involved," or, "It's not my problem." But every person who turns a blind eye, emboldens the despot, bully, or mob boss.

Not saying every little 4-man start up should take the nuclear option when facing the NSA. But there are certainly many companies with clout who could easily turn the tide in public opinion against the rapidly expanding surveillance state. If Google or Yahoo or another big player stook a stand I think it could start to turn the tide.

I just feeli like if Ladar Levison can take a stand, I'd like to think there are others who believe strongly enough in what America stands for to do the same. America is worth it.

The problem is that we all have so much to lose and it makes this so easy for the government. I have a wife, I own a home, I need my job to keep these things and many people will do anything to keep their job and these things. If you're a founder and your options are give the government what they want to no pushback from your userbasr, or to not give the data and face harassment, audits, suspicious legal problems and things like that. Even as a pretty moral guy I have a hard time imagining putting my livelihood on the line for the very little reward if the alternative is possibly losing my home. The government knows how 95% of us NEED to work and keep an income going so they know they can do whatever they want because the leverage is 100% on their side. The only way I see this changing is if people on the Internet/citizens of our country start caring that the NSA is spying on all of us and realize that this harm is way worse than a plane crashing into a building, as horrible as that was.

Here's another reason that I wish was more prevalently discussed: it's not okay to violate Terms of Service anyway. Don't accept the terms? Don't use the service. We'd all have a field day if "big brother" violated the terms of an open-source license, but the freedom we enjoy in the community comes with a responsibility to honor our agreements.

As an architect, I appreciate many of the ideas expressed via the design - in particular seeking a reduced footprint.

However, while many US jurisdictions might not require a building permit for this structure due to its size, it would typically not be exempt from building codes and as a habitable structure and more importantly as a dwelling, the design does not meet fundamental principles embodied in modern building codes.

While the principle of occupant safety is fairly obvious and its susceptibility to liberatarian objections predicable, the code embodies a further less obvious principle, that of first responder safety. Doors and windows have size and operational requirements to allow fire-fighters to get in and back out. Structural systems have structural requirements and combustibility limits for the same reason. Buildings need to be anchored to keep them from blowing into neighboring structures in a storm.

Most people are urbanized, and the structural problems of housing is harder than can be solved by tents in the wilderness. Anyplace that this is a viable alternative for long term dwelling, it is likely that so is a used trailer or a building of recycled and scavenged materials.

Again, I appreciate the design and the aesthetic effort and the ideas it expresses. I just can't get carried away over an academic exercise.

Dear iDoneThis team - do NOT lower your prices [1].

As a fellow subscription revenue biz, we've found that lower priced plans invite customers whose support requirements are much greater. And you won't make it up in volume.

Real businesses that value their time will spent $50/month for a service that saves them time without thinking twice. Your early traction proves this out.

The graphic you had in your post about developer thinking makes me think you guys already know this, but I see the lower price advice a lot on HN, and almost always, I think it is the wrong advice.

[1] - This is obviously 100% opinion, and you know your business 1000x better than I do. BTW, our product's subscription price point is significantly higher $12k per annum at the low end, and we just moved this up from $7500 earlier this year. BEST DECISION WE EVER MADE.

Poor design = storing passwords in cleartext

Senseless acquisition = entering a fast-growing market like India by acquiring a 2nd or 3rd tier player like SoSasta.com(http://www.techinasia.com/groupon-india/). Instead, they should either have entered directly, or bought out a leading player.

Multiple u-turns in strategy = going from SoSasta.com to Crazeal.com to Groupon.in in the space of 2 years, without any traffic or revenue results to show for it.

I'll ignore your last question asking about my qualification to comment. Because it has no relevance at HN.

FWIW, Google doesn't particularly like it when people use APIs outside the terms, either. Having to take down an app is never fun or enjoyable for anyone involved. And the cases in which the terms aren't programmatically enforced right away isn't necessarily a choice; often it is difficult to differentiate at scale between legitimate usage and disallowed usage.

And while I don't know how every other company writes their developer Terms of Services, I do know that in Google's case we measure each word carefully and try hard to strike the right balance between offering as much as feasibly possible, while trying to create something long-term sustainable (so even if the usage grows far beyond our initial goals, we're still in a position to keep them available). We're hardly perfect at it yet, but we're certainly getting a lot better over time.

Also worth noting, often times these terms are designed to honor upstream data contracts, so if for no other reason, it's worth trying to respect them because it's the right thing to do for the other relatively small parties, too.

I don't think it's perfect, but I respect the attempt.

One quick note, though. One header in that write-up boldly states: "I am not a designer." Have some self-esteem! You're totally a designer. You're taking the steps that real designers [should] take when creating functional designs. You have design principles and goals and are aiming to satisfy those with a little ingenuity and creativity.

You're a designer. Own it!

If there is any project I'd be tempted to be snarky about, it's PhpMyAdmin (although last time I used it, it was 2006). From an administration perspective, it was often a hassle, especially in its beginnings. It is a very successful tool and for a while had the same reputation as wordpress: being a pretty remote shell.

But what PhpMyAdmin illustrates very well is how important tools can be to environments/communities. For a while, the presence of PhpMyAdmin was what gave PHP an edge over many other environments: here's a tool that gives you (and tech-savvy people in your company) insight into your database, allows you to do changes and some bit of backup visually instead of fighting with the command line. And, if you develop on PHP, you don't even have to install anything additional on your platform. It made running PHP hosting without shell access feasable.

How does it fair again strong winds in a storm and occasional hail? Is the isolation good enough to provide comfort in environment ranging from 35°C to -15°C ?

What is your solution for a bathroom? Where do you shower?

I'm pretty sure I've posted that in every pg 9.3 thread to date, but for application developers don't miss all the new diagnostic fields on PGResult[0]: you can now get (when applicable on the error) the relevant schema, table, column, column type and constraint[1] in order to translate error information back into your application domain and generate logging and error messages which make sense to developers & end users.

It's already available in psycopg2 (since 2.5)[2] and ruby-pg (in 0.16)[3], it may be available in other adapters (outside of libpq obviously).

[0] http://www.postgresql.org/docs/9.3/static/libpq-exec.html#LI...

[1] without having to perform string munging on localized error messages, which probably doesn't include half of them

[2] http://psycopg.lighthouseapp.com/projects/62710/tickets/149

[3] https://bitbucket.org/ged/ruby-pg/issue/161/add-support-for-...

Not enough people say this anymore, but congrats!

That first $1/$1,000/$1,000,000 is the hardest. Good luck.