deong

Yep. And there are the formal optimality gaps like "this algorithm provably finds a solution in polynomial time no less than twice the optimum tour length", and there are also the informal "I have no proof of anything, but this algorithm usually finds optimal tours or tours within 1% of optimality on problems of up to 10 million cities" type of gaps. Both are useful to know and understand in practice.

If my job were to solve TSP instances, I'd not bother trying to tell someone why it's hard. I'd through Iterated Lin-Kernighan or something similar at it and go get some lunch.

That's not the right analogy. It means you're not allowed to sue someone else for requiring an ID check in order to let you shop there.

I'm not a lawyer, but I'd be surprised if this defense was particularly robust anyway. It seems you'd just have to argue that there's some slight difference in the situations and that's why I'm doing something that guy shouldn't do. But either way, you're not taking away the correct message from his example.

I'm not sure it should really count as a "Lisp book", but The Art of the Metaobject Protocol" is one of my favorite technical books ever.

It takes some knocks for not really containing anything about how to use CLOS. Instead, it's how CLOS is built. At a higher level, it's "how to build an object-oriented language/object system from scratch" using Lisp as a vehicle.

Which 20 Republican senators do you imagine are going to stop him?

Oddly enough, I assume the Google advancement structure values anything but stability. You launch products to get promoted. Maintaining them is for fools and losers.

Granted, I have no inside information. That just fits the externally observable information and the tiny amount of corporate politics that occasionally makes its way public.

I think you're taking him a bit too literally. In these examples, you're not looking at simplifying anything. You're not getting rid of unneeded code. You're getting rid of debugged and stable code in favor of writing something new from scratch because fuck all knows why.

They used to as part of their quarterly earnings reports, if I remember correctly. They stopped during the latter part of WoD when by all accounts the subscriber numbers were very low (by WoW standards).

Even local optima are often pretty good in practice.

Also, it's a very good practice to iterate through the neighbors in a random order (if you're doing next descent, for steepest descent obviously it doesn't matter).

There's another issue as well. His mapping of difficulty reaching a key isn't really the whole picture. On a conventional QWERTY layout, neither the E nor the C keys are especially hard to reach, but if you had to frequently type them successively, that would be terrible. What matters is not just where a key is in relation to your hands, but also where two keys are in relation to one other.

In grad school, I did a lot of work on the Quadratic Assignment Problem (https://en.wikipedia.org/wiki/Quadratic_assignment_problem). If you model that "key reachability" metric as two-key sequence difficulty instead, you actually get a QAP instance to solve. I built a little toy project that would look at the source text of my dissertation in LaTeX as well as the accompanying C++ source code for bigram frequencies, and I mapped out a flow matrix manually in a similar way to the author here, but including that successive keypress difficulty, and ran my multiobjective QAP solver to generate a range of keyboards optimized for my specific dissertation work.

I did this as a fun little gimmick as part of a sort of "ignobels" within my department, so the results weren't really a thing you could use. To generate a usable keyboard layout, you need to also include a ton of heuristics for things like "put all the number keys together" and the idea that you can treat "a" and "A" as the same letter, but "2" and "@" may make sense to separate, none of which I really properly handled. But I think with some work one could actually formalize the problem of finding a real layout for normal people to use.

I have the previous xps 13 running linux now. I'd like to support them, but unfortunately, CostCo sold me the Windows version with mostly equivalent specs for like $400 cheaper than Dell had the Developer Edition at the time. I replaced the Broadcom WiFi chip for $30 and was good to go. I think I saw where the current model no longer has replaceable WiFi modules though.

I don't think he's at all concerned about the license. He's just saying most people won't and shouldn't do "real work" on a machine that could break at any point with no possibility for support.

I think if an x1 Carbon ran OS X, you'd see a massive number of people migrating from Macbooks over to it.

Sure, but all of those are different than an admin at hacker news doing an

    UPDATE posts SET text="I hate brown people" WHERE post_id=17560983

It's also important to remember just how far Git was ahead of most competitors in performance. I was using monotone quite a bit around this time, and while its UI was frankly not much better than git's, it at least worked acceptably quickly. Darcs was mostly defined in the public eye by the exponential merge issue. Hg was fine, for some definition of fine, but it was enough slower than git to be annoying -- maybe not deal-breaking, but noticeable on pretty much any task.

I remember installing cogito to act as a front-end for git because very early on it seemed obvious that (a) git was going to win, and (b) it was going to win in spite of its UI, which was saying something.

I believe the recommendation is to use something like Signal or iMessage -- an end-to-end encrypted channel that isn't email.

Almost every email client defaults to HTML. There are going to be people using both encryption and HTML.

I haven't been on a search committee in years, but in the mid 2000s, I think 3-500 was about the norm for a mid-range state school. Not all those people were qualified, but I think you can assume you'll have somewhere on the order of 2-300 qualified applicants for most jobs.

Tenure track positions are not at all hard to fill. There are 500 people applying for every one of them, and probably 350-400 of those people are completely qualified to do the job. But only ten of them are desirable, and a hundred other universities have those same ten people on the list.

Paying market rate won't help, because they're not leaving positions unfilled because they can't make an attractive offer. They're leaving positions unfilled because they're unwilling to live in a world where not every team gets Lebron James. It's better to do without than to not get your first choice, so that's what they do. If they were willing to pay $800k to fill a seat with whoever they picked as that "market rate" lecturer, they could have just hired the same guy for $120k -- he applied for the job and you rejected him.

As has been pointed out before, statistics like these "for every five open faculty positions, only one is filled" are incredibly misleading. The reason only one in five is filled is not because there's a shortage of qualified teachers. It's because the thousand faculty searches going on at any one time are only interested in competing for the same hundred people.

Universities aren't trying to hire someone who can teach undergraduate CS courses. They're trying to hire someone who can build a lab capable of bringing in a steady stream of seven-figure grants. You need to have a PhD from a top-five (ish) school, with an excellent publication record, and an existing network of collaborators to be competitive for the vast majority of open positions.

$125 an hour isn't crazy. I billed that doing some consulting for a company in Memphis maybe five years ago, and that was a no negotiation involved arrangement. I'd worked there before, so they knew me. I threw a number out there and they pretty much went, "alright, sounds reasonable enough".

Well yeah. I don't think anyone is under the impression that Apple made a laptop that, by design, fries peripherals when it's charging.

I think it's a general Apple failure mode. They're so wrapped up in the idea of making something completely unheard of, that they overly weight novelty. Most of these "clearly bad ideas" aren't all that clear at the time. I don't think I ever thought the touchbar was going to be a positive addition to a laptop, but when it was announced, there sure were a lot of people who seemed convinced it would. You can make an argument for why it's great, and if your company culture is predisposed towards valuing novelty, you'll anchor to those arguments, and there isn't going to be one guy in the room who can say "no".

That same culture has obvious upsides as well; I'm not trying to just knock Apple as incompetent. I just think the flip side of that coin makes things like the TouchBar more likely to survive scrutiny there.

First, "don't make mistakes" is a policy so idiotic in its conception that I don't think there exists a good-faith argument for advocating for it.

Second, a "cover-up" is when you hide evidence of your past misconduct. Pleading guilty in public is in every possible way the literal opposite of covering something up. If they'd just deleted the article with no commentary at all, that's covering up the mistake. Leaving the original URL up with a notice that they erred in publishing the original story isn't.

To attempt a sale is not a right, at least not one I am familiar with.

It's called "copyright". You have the right to control distribution of your creative work. A direct and unavoidable implication of that is the right to try to sell it.

Actually, in as much as that is included in the archival process, I disagree.

Archival doesn't produce a creative act by either common sense or legal interpretation. You can add value of course by writing backup programs or just doing the work of backing people's files up, so arguably I didn't choose my words carefully enough there. But the thing you created was the process of doing the archiving. You're entitled to control of and credit for that work, but not the actual files that your process created. Writing a program that saves the text of an ebook doesn't make me the author of the book.

Good thing Disney has nothing to do with it, otherwise copyright might stretch out to over a century.

I completely agree, but it's irrelevant to the discussion of whether or not copyright as a concept should exist.

The loss I've suffered is the loss of the right to attempt to sell your friend the file under the terms I want to set. I have the right to set those terms because I did the work of creating the artifact. You're completely allowed to find that a bizarre thing to protect, but you're not free to just declare that view the only reasonable one and expect no one to argue.

If I made my livelihood by designing homes, and I was good enough at that job that people demanded my services, then yes, copying one of my designs and distributing it without my consent is stealing something from me.

And I'm not conflating morally and legality. I'm really not. The two correlate pretty highly here (as most laws do for obvious reasons), so I guess maybe that's what's confusing you. But if I write a novel, and you put the original file on bittorrent, I created all the value here. You dragging a dropping a icon representing the bits on a hard drive isn't valuable work. And I believe that morally, the nearly infinitely greater amount of productive work I did to create that copy than what you did entitles me to more creative control. I believe that completely independently of whatever the legal system says. I also know that the legal system agrees with that determination and sets penalties for violating rules set up to enforce it. But I'm not using that as evidence for my moral position. It's not immoral because its illegal. The causation goes the other way around. It's illegal because the shared ethical framework of the people and society that drafted the constitution found it immoral.

And the correct lotto numbers tomorrow are free once you know them once, but that's not how prices work.

And what I'm saying is that you're simply declaring by fiat that there's no loss to the studio when you copy a file they own the rights to, and from there, building up to a conclusion that there's no loss to the studio when you copy a file they own the rights to.

It's a circular argument that falls apart because I reject the premise that I suffer no loss if you infringe on my right to profit from my own creation. You're not going to be able to argue me away from that by just saying the opposite is true with no further evidence.

Again, theft isn't depriving someone of property. If I go to your house and say, "ooh, cool guitar" and you give it to me, no theft has occurred. You've been deprived of that guitar, but I had the right to take it because you gave me permission.

Theft is committing an act of taking something that you don't have the right to take. The value of that thing doesn't determine whether or not it's theft, it determines damages. Regular competition or loaning a DVD to a friend don't consititute theft because you took no action you aren't allowed to take. The law allows you to make a competing product or to loan your books and movies to a friend for personal use. That's the whole reason why we don't refer to that as stealing something.

You can't steal something if you didn't steal something, regardless of the impact your action has on someone else. But in copyright infringement, you took something. You took the exclusive right to distribute a piece of content. That is an actual thing that is recognized by law and by common everyday economics as a real thing that has value. Depriving its rightful owner of that thing can be reasonably described as theft.

I haven't mentioned morality at all. I haven't even mentioned my personal stance on what I'd like to see IP law become. The only think I'm doing is arguing against this notion that it can't possibly be considered "theft" unless there's physical piece of plastic involved.

It's only ignorant to apply supply and demand to digital goods if you start from the premise that copyright shouldn't exist.

1. Copyright shouldn't exist. 2. If copyright didn't exist, anyone would be able to get any digital file easily without restriction. 3. If you can get any digital file without restriction, then supply is infinite and demand doesn't matter. 4. If supply is infinite, then supply and demand doesn't matter and you're ignorant for trying to apply it.

That's basically the argument you're making, and you're rather trivially just assuming the consequent.

On the second part, that's just a really faulty analogy. Competitors aren't "stealing" from one another for the simple reason that they haven't taken anything. Copyright infringement isn't "stealing" demand. It's stealing an item that has demand. It's the infringement that's the theft, not the consequences. The consequences are just the justification for having the law in the first place. To repeat, competitors aren't "stealing" when they lower demand for the stupidly obvious reason that they haven't stolen anything.

It's still going because that's the way people talk. You can actually "steal" an idea, even though it's not physically possible to remove the original from the person's mind. You can steal a kiss. You can steal second. Words mean what people use them to mean, and it's a perfectly acceptable use of the word "steal" to mean something like "adopted use of an artifact for the purpose of deriving a benefit rightfully belonging to its creator".

I'm personally surprised the "copyright infringement isn't theft" meme is still going strong.

This is exactly right.

When I was teaching machine learning classes, the biggest hurdle I had to get a lot of students over was their history and expectation that they can implement something, fix the compiler errors, fix the runtime crashes, and then just declare victory. Machine Learning is basically the study of getting a computer do give you an answer you don't know, so you have to be in this constant state of defensiveness, never quite trusting that things are correct.

When I was doing my PhD, I once resorted to turning on all logging, printing about 60 pages of the resulting output, taping them to the floor of the lab, and crawling around on all fours with a printout of the source code tracing the execution to try and prove to myself that a result I was seeing wasn't due to just a bug in the source code. In an odd way, it's maybe my fondest memory from my dissertation work.