The Wanderer’s Techdirt Profile


About The Wanderer

The Wanderer’s Comments comment rss

  • Aug 20th, 2018 @ 8:22am


    In my experience, it's actually more common to call them "telephone poles", and that name's implication that they're owned by the telephone company (which nowadays tends to be, or be owned by, a major ISP) is usually correct.

  • Aug 20th, 2018 @ 7:35am

    Re: Re:

    The logic behind "collect it all" is something like: we can only search hay when it's in a haystack, and if we leave too much hay out of the haystack we risk also leaving the needle out of the haystack, so that no matter how much searching of the haystack we do we'll never find the needle.

    Therefore, in order to guarantee that any needles which exist are in the haystack to be found, we need the haystack to include as much of the hay as possible.

    The logic is even sound, really. It just ignores A: the decrease in searching effectiveness when using the same techniques to search a much larger haystack, and B: the trade-offs in other areas that result from making the haystack bigger.

  • Aug 19th, 2018 @ 2:04pm


    On the off-chance you were serious, rather than trying to create an opening to make that joke: it stands for "Foreign Intelligence Surveillance Court". The name is derived from the legislation which establishes it and/or necessitates its existence: the Foreign Intelligence Surveillance Act of 1978, or FISA.

  • Aug 14th, 2018 @ 8:08pm

    Re: OPR meaning

    He did. From the paragraph immediately after the second quote from the 2013 decision:

    The appeals court asked the DOJ's Office of Professional Responsibility to investigate the US Attorney's office in North Carolina

  • Aug 11th, 2018 @ 6:11pm

    Re: Re: Re: Re: Re: Re: Re:

    Tell me why anyone works for anything if not for time or money or sex.


    Almost everyone finds sex to be fun. (Exceptions include the asexual, those with a physical or mental disability which detracts too much from the experience, and those too young to have the physical facility to enjoy it yet.)

    Some people find work itself to be fun. (For some specific types of work, different depending on the person.)

    Some people find "being appreciated by others" to be fun. (I think someone else in this discussion mentioned "fame" as a motivation.)

    Some people find spending money to be fun.

    Some people find just lying around doing nothing to be fun. (At least some of the time.)

    Some people find provoking angry or otherwise upset responses out of other people to be fun.

    Some people find helping people to be fun.

    There is almost nothing which is not considered to be fun, or otherwise enjoyable, by someone out there.

    There are basically only two motivations to do anything; omitting a long philosophical discussion, they can be roughly described as "in order to survive" and "for enjoyment". There are a lot of things which can be fed into by the desire to survive; there are a lot more things which can be fed by the desire to enjoy oneself.

    If you fail to consider "enjoyment" in your model of people's motivations, it should be no surprise that your model doesn't correspond to reality very well.

  • Aug 9th, 2018 @ 9:20am


    Re that first point: Not to mention that, IIRC, one or both of "indivisible" and "under God" was inserted later on, as a way to push a particular political position.

  • Aug 9th, 2018 @ 9:10am

    Re: Re: Re:


    I always thought "virtue signalling" referred to things like "Now, I hate Nazis as much as the next guy, but" - the disclaimer that "despite what the rest of this may make it look like, I'm really one of the good guys, just like the rest of you", particularly in cases where that disclaimer is not actually honest.

  • Aug 9th, 2018 @ 8:50am


    Got a link to that ruling? It rings a faint bell, but I can't dredge up any specifics...

    I'm interested in the reasoning, because of the possibility that it might line up with a chain of logic I noticed recently based on the First Amendment.

    When the First Amendment prohibits restrictions on "the freedom... of the press", the "press" it refers to is not "people engaged in the business of reporting the news", but "the means of publication" - which, at the time, meant access to a literal, physical printing press.

    Over the course of time, new means of publication have become available, and it is my understanding that jurisprudence has accepted them as being covered by the "freedom of the press" clause of the First Amendment.

    In the modern day, the most meaningful and prevalent means of publication - to an even greater extent than the physical printing press ever was - is the Internet.

    On that basis, any law prohibiting someone from accessing the Internet would seem to be unconstitutional.

    At the very least, that argument could be made in court, with some chance of success - unless it's already been shot down by existing precedent, of course.

    If the ruling you cite is based on similar logic, that would seem to strengthen the position. If it's not, that would be interesting as well.

  • Aug 5th, 2018 @ 10:03am

    Re: Re:

    I think you failed to grasp the difference between "white equalist" and "white supremacist".

    It's entirely possible to be proud of something without having to insist that it must reign supreme over all other things.

    It's also possible to not be proud of something without being ashamed of it either. I'm neither proud of nor ashamed of my eye color; it's just something that's there, worthy neither of pride nor of shame.

    It's also possible to not lump oneself in with other people just because one shares a superficial characteristic with those people.

    (That last point touches on the fact that both terms are still built on the core assumption which is the basis of racism, that being the very concept of race itself. We're a long, long way from being able to begin to abandon that concept, however.)

  • Aug 3rd, 2018 @ 5:53am


    As I understand matters, the idea of the "not clearly established" principle is that "you can't punish someone for doing something if the rule prohibiting it wasn't clear enough that they should have known it would be prohibited".

    So this isn't "ex post facto rights" - it's ex-post-facto liability again, on the officer's side.

    The manifestation of that in the slavery analogy would be saying that you can't punish former slaveowners for having owned slaves when slavery hadn't been declared illegal yet.

    Yes, the right always existed - but just because it existed doesn't mean that the fact that it existed was clear enough that the person should have known it existed.

  • Aug 2nd, 2018 @ 8:16pm

    (untitled comment)

    Where do you get "hostile" from?

    It seems to me that the fact that he stayed in the country for that long without incident would be fairly decent evidence that he's not hostile.

    Or at least that he wouldn't have been if he hadn't been treated with hostility by immigration authorities.

  • Aug 1st, 2018 @ 9:28am

    Re: Re: Normal isn't so normal anymore

    If my math is right, that works out to one in a hundred billion.

    While the numbers of law-enforcement officers worldwide who will ever personally encounter a terrorist are certainly low, I don't think they're quite that low.

  • Aug 1st, 2018 @ 6:24am

    Re: Re: Re: Re:

    I'm not sure that's going to produce a sufficiently good result.

    The first question is, proportional to what? The only answers I've been able to think of that make any sense are "party" and "candidate", and both have problems.

    If it's by party, how do you decide which parties get counted, and/or get to be on the ballot? (And this might also magnify the problem of how the parties decide who to pick for whatever seats they get.)

    If it's by candidate, how do you handle a case where the proportions of the vote for the candidates cannot be remotely evenly split into the available seats? (For a contrived example, something like 250 votes in a three-seat race where two candidates each get 120 votes and two other candidates get 5 votes apiece.)

    IMO, nothing short of ranked-preference voting is going to be a nearly optimal solution - and even most forms of that have their weaknesses; the one with the fewest that I know of is the Condorcet method, which has only one design weakness (the remote possibility of a true tie, which at that point can IMO legitimately be broken by random draw).

  • Jul 31st, 2018 @ 7:14am

    Re: Copywrong.

    As best I can parse it, Sony appears to be saying that all the other works derived from the original "Slenderman" piece by Knudsen were infringing Knudsen's copyright (and only got away with it because he didn't care to enforce his copyright), and that now that Sony owns the copyright to that work, all those existing works are still infringing what is now Sony's copyright (which Sony does now choose, at least in some cases, to enforce).

  • Jul 31st, 2018 @ 7:05am

    Re: *Remember the Dodge Salesman!*

    0-to-60 is acceleration rate, which is "quickest".

    The term "fastest" is reserved either for top speed or (to factor in handling, et cetera) for sustained-over-distance speed, as measured by time to cover e.g. a known-length reference course. I've never been entirely clear which, or what term is used for whichever is left over.

  • Jul 31st, 2018 @ 3:56am

    Re: Re: Re: Quite expensive...

    >They have like huge array of hash values,

    Which is far more than a simple "one if-statement in php".

    Even building the array (if so you choose to call it) in the software is more than that simple "if" statement, and actually populating it with meaningful data is well beyond even that.

    (And that's leaving out things like "detecting audio being played in the background behind other activities", where you can't just hash the audio stream because it includes so many things that aren't the copyrighted content, but where the copyrighted content is still present in the mix; you need to be able to do audio stream splitting and recognition, on a level that last I heard was at least partly still the domain of AI research. Not that having the radio on in the background while you do your video that should necessarily be considered a violation, but the RIAA et al. have certainly been willing to treat it as such in the past.)

  • Jul 30th, 2018 @ 8:04pm

    (untitled comment)


    I don't think I've noticed that, but I'm definitely going to keep an eye out for it now. I think I rather like that idea.

  • Jul 30th, 2018 @ 9:05am

    (untitled comment)


    I was hoping no one would respond to him (feeding the trolls never helps), and that if anyone did, they would at least have the courtesy to delete or replace the Subject line...

  • Jul 30th, 2018 @ 4:11am


    Would you be so kind as to stop spamming that comment into (apparently) every single article which so much as mentions a subject which that poster might be expected to object to?

    It doesn't add anything to the discussion, and at *best* it's an attempt to wind him up - and not even a particularly effective one, as he rarely if ever even responds to it, as far as I've seen.

  • Jul 30th, 2018 @ 4:01am

    Re: Quite expensive...

    I'm sorry, what "one if-statement in php"?

    If you seriously know that there is in fact a single "if" statement which handles this, please do quote it here.

    ContentID is certainly not a feature which "the php language already provides", or which "is already available in the php ecosystem"; it's a sizable, complex, continually developing system, consisting of both nontrivial amounts of code and (of at least equal importance) a large database of reference data, and - unless I'm much mistaken - is available only internally at Google.

    Developing such a system in the first place costs a lot of money; continuing to develop it costs a lot of money; maintaining and operating it costs a lot of money.

    The difficult part is not in filtering out the copyrighted works; that might indeed boil down to a single "if" statement if you arranged your code just right (although, in that case, a lot of detection complexity would be concealed within whatever functions set the variable being tested).

    The difficult part is in identifying which works need to be filtered out. (Which is not the same as "which works are copyrighted" - or, thanks to things like fair use and the public domain, even "which works are neither copyrighted by nor licensed to the uploader". Neither of which is easy to determine anyway, at least not without an authoritative database of copyrighted works and licensing terms and agreements, which the rightsholders have consistently refused to even try to make available.)

More comments from The Wanderer >>