badgar

My undergrad thesis advisor, Bill McKeeman, wrote his parsers in this fashion. I implemented a parser using this model and extended its existing lexer.

The token stream is an array of 32 bit integers, each of which is the token type bitmasked onto an index into the input file of the start of the token. If you need the token text, you reparse. Caches can be implemented as small hash maps from token index to cached value.

The canonical AST is the CFG parse tree with fixups to convert recursion to children of a node type for a variable number of children. It is stored as an array of integers. The node is a sequence of integers, with one integer for the root followed by each child in sequence. Each internal node's value is the index of the CFG rule evaluated to produce the node, and the children of the node correspond to the CFG rule's right hand side (minus keywords). Terminals are stored as integer indexes into the token stream. Nonterminals are the integer index of the child internal node.

Bill has been a big name in compilers for the better part of 5 decades now, and he said he's been using this pattern for almost as long. It's ridiculously fast, which is why he used it decades ago for DEC.

> Nope, it's bullshit, just like badgar's half ounce-a-week descent into ruin.

Excuse me? First of all, nothing I described was "descent into ruin." I said I'm having a hard time not smoking pot and it's affecting many aspects of my life. You have no idea what you're talking about, how dare you say I'm not struggling? Have you ever considered that people have a hard time quitting drugs? That pot might actually harm someone? No, your mind's made up: it's all a conspiracy by the drug companies!

> I have to doubt its legitimacy because the post reads like a plant by one invested in treatment facilities and private prisons.

Fuck you, you conspiracy-minded, condescending prick. I'm having a hard time over here but you think you're so important that treatment facilities and prisons are planting stories on a tiny technology startup forum.

> If not, then I truly sympathize, but it is entirely incorrect to link a behavioral addiction with a physical addiction such as one caused by opiates. They are in no way equivalent.

You don't sympathize, you're busy paranoid that prisons are astroturfing HN. And I thought paranoia was bad when you smoke on the streets. By the way, this entire subthread is talking about marijuana use and affecting career growth. So this entire point is irrelevant.

I fucking hate this site sometimes.

Nobody in this thread knows what they're talking about. You might as well have said "As entertainment, food is no worse or distracting than television."

Look, not everybody gets addicted to pot, in fact most folks don't. But it's psychologically addictive like everything chemically active you put in your body, and being constantly stoned because you're addicted is a miserable way to live.

I know, because I've been trying to quit all year. I've failed twice since February and I'm on my third go of it right now. And I only smoked a half-ounce a week for a year or so. That's probably a lot to the people reading this, but it's really not much as far as potheads go.

I'm sick of thinking so much slower than I know I can, being so much dumber than I used to be. I can't do my job nearly as well and I take pride in my work, so being too dumb to do my job has been really upsetting. I almost wish I were still addicted to cocaine, because at least then I kicked ass and took names when it came to thinking/analyzing/working in general.

I'm tired of being too lazy to leave my apartment all weekend, of the food I eat when I'm stoned, of not being present around my friends, family and co-workers, of needing to smoke when I wake up because it's so uncomfortable to be lucid. My motivation to pursue new ideas, to eat healthy and go to the gym, to meet new friends and pursue new women is all through the floor. My body desperately wants pot right now and I'm anxious and depressed as hell and my body is tugging at me to reach for any alternative, like alcohol or cigarettes.

Every drug taken to extreme can seriously hurt your life. Even weed. Stop trivializing it because you smoked a few joints in college.

Edit: and now I'm slowbanned, for sticking up for myself when someone kicked me while I was down. Time to roll a new account, I guess.

Yep, Galaxy Nexus here and it is completely unresponsive to any interaction whatsoever.

> Problem?

The problem is you aren't sticking with the narrative folks here are comfortable with. This particular Internet forum assumes Google has zero support, even though this is pointed out as false in literally every Google thread. This despite HN users genuinely believing they are smarter and have higher-minded debate than other forums.

What's up with Peter Norvig's comment at the bottom? Did he not RTFA? He suggests a solution that the article criticizes heavily halfway through.

The last I've heard of this was 2 years ago, an article in the New Yorker: http://www.newyorker.com/reporting/2011/01/24/110124fa_fact_...

At that point, it was ~40% of revenue and ~80% of profits from subscribers. Apparently they've gone nowhere, fast.

Random is incorrect.

That's because it's Google Now, not Google Planner. It's for giving you information relevant to right now. You don't plan to use Google Now for something, you check it when you want information relevant to you at the moment. If you stop trying to make Google Now into a personal organizer, you'll enjoy using it more.

> You can't be sure if a card will appear for flights

In my experience, Google Now finds my flights in my Gmail 100% of the time. For at least a dozen flights across the US. And for some airports (SFO is one, unsurprisingly) it even has the terminal and gate information ready! Google Now makes the "arriving at the airport" experience better every time for me.

You buy expensive makes then, I really hope you know that you're way above what most people spend on cars. Most cars don't have those chips yet. The only sort of mainstream car I've seen with a chip in the key is the prius, which starts in the low 20s new.

He had root, so he could have instead installed a rootkit, which can hide the existence of processes from all of userland. In a graduate OS class I took, we had an assignment to do hide a process live on OpenSolaris 8 using the kernel debugger (kdb). I wrote some assembly and overwrote some bytes in the syscall functions for process listing. We were on developer builds so you could just use the function symbols by name in kdb. I forgot to cover /proc/ though.

> uptime percentages are a useless measure

Uptime is only useless if you don't care about high availability. Many startups don't, but Google sure does.

Whenever I go to bed early, I wake up for the best hour or two of my life around midnight. Good to know I'm in working order.

When people don't argue in good faith, I think it's more important to make sure others following along see so and aren't misled by FUD and ignorance. I value that over being sensitive to the person who is participating in bad faith, as this user is.

Being pleasant and gentle to everyone is not the most important part of discourse, not even on HN.

As to the accusation of ad hominem: every single point I made was directed at - and based solely on - the arguments made by the other user. Once these arguments were ultimately found to be made in bad faith, I ended the discussion. That's not an ad hominem argument, that's ending an argument because of a negative evaluation of the other person's state of mind. Two very different things - almost unrelated, honestly.

IANAL, but cell phone calls are encrypted, and I'm pretty damn sure you need a warrant to legally decrypt a cell phone conversation over the air.

But I bet you wouldn't need a warrant if you wanted to just capture and show the existence of the emitted radio waves as evidence, ignoring their contents (if you were making some kind of a traffic analysis type of argument).

> Google offers support for exactly one product (adwords).

Since this a bald-faced lie, my unpleasantness is clearly warranted. You are clearly deriving arguments from a predetermined conclusion likely reached ages ago. At this point you apparently are trying to make inferences about demand without even trying to understand the existing supply, and the contortions you're going through to justify your existing beliefs are astonishing.

Generally, we consider a service being too expensive to be a good reason for a company not to offer that service. So not offering a service might be evidence that said service is too expensive to offer/operate. You can see this argument being made prominently in the Linode thread on the front page - someone complains that they don't offer a $10/month plan, and others point out it would be extremely expensive to support. Thus the lack of the $10/month plan is used as evidence that it is likely too expensive to offer.

You've gone the completely other way: you're saying the service not being offered is evidence that it is cheap. This is the opposite of the typical argument. So I'm reducing your idea here to its simplest logical form, trying to make it super clear to readers (and hopefully you) how ass-backwards it is.

No, logic doesn't work that way. Absence of X does not prove that X is cheap.

Considering how much heat gizmodo got for calling out the same kid two years ago, I'm not surprised.

> Its kind of depressing for me, because it seems to mean being showy, and having a good PR facade is worth far more than people doing the actual work.

Marketing is a huge part of the startup world. There's a reason people are told to stop building at MVP - what you build isn't really important, and you're really likely to throw away what you built and switch to something else.

What is the incentive to do so? BTC is expected to rise in value over time since it's a deflationary currency. So any sane businessperson should expect to be constantly decreasing the BTC price over time. It's a feature, not a bug.

> You likely lived in an MDU (multiple dwelling unit), i.e. an apartment.

Which is, like, nobody in new york city. Quit complaining everyone!

It's called a leak regardless. Often of it is on purpose we call it a "controlled leak," but it's still a leak. Not all leaks are bad for the party that does the leaking.

Sorry, evil (bribes in this case) is the only answer for when Google does something an HN commenter doesn't like.

Do you have any idea what construction work in NYC costs?

Is anyone here thinking any of their comments through past "X city would be great for fiber"?

I see. Money must not be explicitly trading hands in these commercial data-sharing agreements, because you're very serious about that one detail (and nothing else). I obviously don't have insight into these commercial data-sharing contracts - actually, you do. So I'll assume I'm wrong about dollars being exchanged directly.

But I'm not wrong that data is shared liberally and unnecessarily in these data-sharing agreements, which was actually the point. And I'm not wrong that those agreements are commercial in nature with Facebook receiving something in return. Or are you saying Facebook's data-sharing agreement system is a giant charity program. So I'm still okay with calling commercial data-sharing agreements "selling user data."

Undo a fork is at least an order of magnitude more difficult than making the decision to fork in the first place. Nobody has that on their minds right now.

Yes, yes it does. You'll need to refine your brief assertion to make it correct. It's all anonymized of course, but Facebook absolutely has many commercial agreements to share its user data with partner companies. A common part of these agreements, for example, is for the partner companies to also share their user data with Facebook.

> I find it odd that Facebook isn't scared of the same legal liability. Infact, they should be more scared.

You don't understand how partnerships work in this case. Facebook is paid to provide data according to a contract between the two companies. That's it. They have many, many such contracts to fulfill with many partners. These contracts don't say "we provide exactly the data you ask for, no more, no less" because that is a lot, lot more work and at the end of the day the contracts still get signed and FB still gets paid. A lot. Where do you see legal liability for FB so far? Their terms of service fully permit this data sharing, and the contracts are just between FB and partner companies.

Meanwhile, the partner companies use the shared data to produce value-added products which they provide to their customers under existing contracts and policies. They can't violate their customer's privacy or rights under any of the existing such policies, which are not identical to FB's. In fact, depending on the industry of the partner company, these policies can be extremely different. It is up to the partner company - not Facebook - to comply with the partner company's contracts/policies.

> I don't see any reason why Facebook would give away more information than they need to.

Facebook doesn't see any reason why they should waste time withholding more data than they need to. They still get paid the same amount, and the partner companies just filter out what they don't want. What's the point? There isn't even a big long-term concern here for Facebook, since extra social data isn't exactly very useful years later.

> They have the skill and resources to filter data themselves.

You say that, but every large technology company has to prioritize what they use engineering manpower for. Not everything a company can do or wants to do ends up getting done. For example, you might say "Facebook has the skill and resources to delete data from their CDN. I highly doubt they would throw away money on byte stored in Akamai" and yet they didn't do so for years and probably still don't. And that's a far easier problem with real measurable costs. Compare that to filtering the same enormous data set differently for each partner company when there's no compelling business reason to do so. The contracts are still going to be signed and Facebook will still get paid. Why should anyone care?

> I highly doubt that they would throw away personal and sensitive to other companies.

This is something FB is doing and have been doing for years. So your doubts are baseless.