> Rightsholders must not be allowed to control how works are preserved, else they can very easily steal from the eventual public domain
I was clapping my hands at the skilfulness of this satire -- the idea that a person could actually believe this near-perfect inversion of reality, where piracy is not merely acceptable but in fact noble -- and then I started to think that... you actually probably do believe this.
If you were in fact trolling all along, my hat is off, you got me.
In the time it took you to write that comment, you could have pirated something. Something that the original creator will doubtless now steal from future generations.
Sometimes sharpening the axe means breaking it completely for people still trying to cut down trees on WinXP, but you don't know that because you can't run those tests yourself, and grovelling through old logs shows nobody else has either since 2017 so it's probably no big deal.
Sometimes it's not clear which part is actually the cutting blade, and you spend a long time sharpening something else. (If you're really unlucky: the handle.)
Interesting! I read the underlying paper, and it looks to me like a variant of log-structured merge (LSM) trees, with at most 1 segment of each power-of-2 size, and all segments stored in a single array W. The 1-bits in the totalSize variable tell you exactly which of these segments are "active" (in use), which is clever. Note that the half-size B array is only used for temporary working space, so it could even be allocated afresh at the start of each modifying operation and thrown away afterwards.
The claim of O(log n) search time isn't quite true -- it holds only for inputs drawn from a uniformly random distribution (Thm 5, p. 9). Many inputs are like this in practice, but many are not, which will cause degradation to O(log^2 n) time -- though this is still very reasonable, and may still beat or be competitive with other truly O(log n) approaches due to lower constant factors (no pointer chasing).
One idea I had to speed up searching if you know you're going to be doing a lot of it with no modifications in between: Since the data structure is agnostic to the actual values within each active segment, whenever there are several adjacent active segments, that entire contiguous block of segments can be sorted without upsetting any invariants. From then until the next modification, instead of separate binary searches on each segment, a single binary search on the entire block can be done. This will halve the total number of binary searches needed in the worst case from log2(n) for 111111... to log2(n)/2 for 101010..., at the cost of these extra sorts -- but note that we don't actually need a full sort, just a (multiway) merge, since the individual segments are already in sorted order. Merging s segments can be done in O(n log s) time with a min-heap, which is not asymptotically better than a full sort (s could be as large as log2(n)) but will be faster in practice due to usually being lower, and lower constants.
In fact, the multiway merge idea could also be used to speed up insertions that trigger more than 1 merge: Instead of a series of s merges, each between a pair of equal-size segments, you could do 1 s-way merge. I would expect this to be faster as it does 1/s as many uncacheable writes (I'm not counting writes to the min-heap itself, which will fit in L1 for all practical n), but only profiling will tell.
The simplest answer is that sometimes other existing software that I need to use treats an empty directory (or, hopefully, a directory containing just an irrelevant file like .gitkeep) differently from an absent directory, and I want that software to behave in the first way instead of the second.
A more thorough answer would be: Filesystems can represent empty directories, so a technology that supports versioned filesystems should be able to as well. And if that technology can't quite support fully versioned filesystems -- perhaps because it was never designed with that goal in mind -- but can nevertheless support them well enough to cover a huge number of use cases that people actually have, then massaging it a bit to handle those rough edges still makes sense.
Legitimately asking, please share the name of software that expects/requires an empty directory and interprets .gitkeep in this way, but chokes on a README file.
Many filesystems cannot represent empty directories. Many archive formats also do not. I don't think this a problem in practice. I find this argument extremely weak.
> please share the name of software that expects/requires an empty directory and interprets .gitkeep in this way, but chokes on a README file.
Every mail server since postfix supports Maildir, in which every email is a file in one of 3 subdirectories (tmp, new or cur) of a user directory. If there's no jbloggs/tmp dir, postfix will think user jbloggs does not exist. So if you want to take a restorable snapshot of this with git, there needs to be a file there. I don't know if jbloggs/tmp/README would cause a problem, because I don't know how postfix will treat the presence of a file with a name that violates its expected syntax (which includes a timestamp and other metadata), but what I do know is that, after running `git clone any-repo`, I can safely delete every .gitkeep file to restore the system to its original state without having to understand that repo in detail -- while I cannot safely delete every README file. That's because the commonly understood semantics of .gitkeep is "I exist solely to shoehorn this empty dir into git", which does not apply to other files.
> Many filesystems cannot represent empty directories
I find it hard to take this objection seriously, since almost everything that isn't a physical commodity has some degree of "I don't know if I'm satisfied with this yet". Books and movies clearly do. But we expect to take a risk and occasionally pay for them, and it feels ordinary to do so -- so why not here?
I don't object at all to people not liking micropayments -- I don't like them either. But the reason I don't like them is because I'm accustomed to getting good quality content for free, and no other reason.
With books, movies, tv shows, music almost everyone is discovering based on recommendations or curation. Very few people are consuming much of that type of content with no outside input on its quality or interest. News is almost always a blind link with just a headline to work from.
The headline is a pretty big clue about what's going to be in the article, I think. (Though maybe headlines will become more coy, to entice readers to pay for the full article? We already see this with questions-as-headlines.)
Most of the news I read comes from the same handful of familiar sites, so I have a good idea of what I'm going to get, especially if they include an author byline and I recognise it. "Niche" news sites would do well to continue offering a small number of freebie articles to entice those who chafe at the thought of spending $0.40 on something they're Not Completely Certain Is Really Worth It.
It's a clue of what its about but it's not the same as a recommendation that its going to contain something you enjoy.
So every click is a decision on an unknown and that's where the problem comes in. You read the headline now you have a decision to make, is it worth spending money to continue. For longer form media your return on good stuff is going to be at least an hour of entertainment. For news you will almost certainly be done the article in 30 seconds on less. So you decide to click, you read the article, it's good, it's now 30 seconds later and you read the next headline and you have another decision to make, do I pay for this? Thats where the decision fatigue comes in and why it's so quick with this model. Even if it's just a penny you're not thinking well clicking this will only cost a penny, you're thinking do I really want to pay to see what the other side says?
And if you're at the point where you're spending a lot of time on a single source even if it costs more it generally makes more sense just to subscribe since then you are making a single decision once a month.
> That model is completely counter-intuitive and punitive to the consumer.
It's really not, as evidenced by the fact that paying for what you use is how almost every physical good works, and many professional services (a lawyer's time).
It's fine for you to dislike the model -- I dislike it too. I don't like that it makes me anxious about consuming the next small unit of <whatever>. But there's nothing inherently counterintuitive or punitive about it. It's the simplest and most defensible payment model possible.
Should people expect high quality journalism if revenue is based on the number of views?
Good journalism costs money, people should expect to pay.
Though I'd point out that publishing news is now cheaper than ever, and people were more than willing to pay for access before, so why shouldn't they be willing to pay less now?
Or perhaps more to the point, why are they _not_ willing to pay now? And is the reason something ad-based perhaps?
I interpreted your original post to mean that you found none of micropayments, ads or subscriptions to be acceptable. Now I have the impression that I misinterpreted you -- but I still can't tell what kind(s) of payment you would actually find acceptable.
What kind(s) of payment would you find acceptable?
My preference would be free, single payment, subscription. Probably in that order.
I don't mind micropayments as a _method_ to achieve any of those, but I don't like them as a replacement for ads. And I don't accept the premise that ads should be replaced with something similar.
Thanks. By "single payment", do you mean one payment for lifetime access? I'm aware of a personal cloud storage provider that offers this, but I don't think it could work for news.
An analogy for cryptocurrency that I like is lasers. I remember reading an Usborne book about lasers as a kid and thinking they were the coolest thing ever and would doubtless find their way into every technology because glowing beams of pure light energy, how could they not transform the world!
Lasers turn out to be useful for... eye surgery, and pointing at things, and reading bits off plastic discs, and probably a handful of other niche things. There just aren't that many places where what they can do is actually needed, or better than other more pedestrian ways of accomplishing the same thing. I think the same is true of crypto, including NFTs.
In 1982 I joined Leeds university as a computer operator/assistant. First job out of a CS degree in another uni. The department head was a laser physics specialist and I, and other snobs used to say "what does a laser man know about REAL computer science" and act outraged.
More fool us. More power to him. He was well ahead of the curve, him and his laser physics friends worldwide.
> Imagine the damage one bad kid could cause using deepfakes
Deepfakes are highly damaging right now because much of the world still doesn't realise that people can make deepfakes.
When everyone knows that a photo is no longer reliable evidence by itself, the harm that can be done with a deepfake will drop to a similar level as that of other unreliable forms of evidence, like spoken or written claims. (Which is not to say that they won't be harmful at all -- you can still damage someone's reputation by circulating completely fabricated rumours about them -- but people will no longer treat photorealistic images as gospel.)
I think it will be big, but I don't think it's bigger than the automation of manufacturing that began during the Industrial Revolution.
Think about the physical objects in the room you're in right now. How many of them were made from start to finish by human hands? Maybe your grandmother knitted the woollen jersey you're wearing -- made from wool shorn using electric shears. Maybe a clay bowl your kid made in a pottery class on the mantelpiece. Anything else?
I was clapping my hands at the skilfulness of this satire -- the idea that a person could actually believe this near-perfect inversion of reality, where piracy is not merely acceptable but in fact noble -- and then I started to think that... you actually probably do believe this.
If you were in fact trolling all along, my hat is off, you got me.
reply