Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
| sureanem |
Posted on 19-07-28, 19:47 in Games You Played Today REVENGEANCE
|
|
Stirrer of Shit
Post: #541 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
I haven't ever copy-pasted anything here. If my ideas are so ham-fisted, then why would I have bothered to plagiarize them? For that matter, what would be the harm in cross-posting great ideas? No, the engineering effort parts would still have to be done human. What I'm suggesting is to replace the part that's now being crowdsourced with an automatable part. What would be wrong with using it to do regression testing? My suggestion is essentially as follows: say you have an input movie of 60000 frames (1000 seconds). For each frame i, compute the hash and set a[i] = hash(framei). Then compute the Merkle hash of a. Repeat this process for the audio too. Now store (input movie hash, video frame Merkle hash, audio sample Merkle hash) somewhere as a test vector. Get the input movies from the fuzzer. So say you have 100% coverage for all the commercially released games. Now say you mess something up. Then one of two things should hold true: 1) The automated regression testing catches it 2) It doesn't affect the normal operation of any game in the slightest Of course, with 2) you miss something, and this is bad. But at least it doesn't outwardly break anything, and when the compound breakage causes inaccuracy you'll notice it. You could do this process for the hash of the complete system state to catch some cases of type 2. Although e.g. SM64 PU zero-division crash would probably still remain uncaught, and the implementation complexity increases by a lot. Is it perfect? No. Would it catch nearly all the regressions a human being playing could potentially be bothered by? Yes, assuming the fuzzer can get good enough a coverage. Whether it does that is of course an open question. But I wouldn't think it's that crazy an idea. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-28, 20:24 in Ubuntu: x86_32 is dead because WE SAY SO!
|
|
Stirrer of Shit
Post: #542 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by tomman Yeah but was this really an improvement? What you forget is that one of these things presumably caused the other. I mean, consider the Internet before and after 2008. All sorts of people started using it and brought the quality down. And all these supposed advantages of a "connected world" failed to materialize. News websites shut down their comment sections, websites moved into apps, and forums moved to Facebook; all to provide a more user-friendly experience for the new clientele. You see what I'm getting at here? User-friendly is not good, and should be avoided as far as possible. It's true that more user-unfriendly computers are less pleasant to use, but the discomfort you receive from it is far smaller than the new clientele's ditto. Go even further back, and then most users knew not only how to administer their computers but often how to program. It stands to reason there's been a decline in their quality since then and that this was precipitated by an increase in user-friendliness. It's like using complicated words. In theory, all it does is make things harder to understand and should be avoided. In practice, the effect is negligible for educated people and allows discussions of even fringe matters to proceed without much issue, while keeping the uninitiated from getting offended. Now you might object, that while you do get a lot more slag, the share of competent people ought to stay the same, and in the slags you should find some 'diamonds in the rough'. But this ignores the operational reality of things. If you have a project with 10 competent people doing whatever, then things presumably should go right on ahead pretty smoothly. Whereas, if you add in two utter buffoons who are preoccupied with adding in codes of conduct, displacing the competent people and putting their people in, and writing retarded "UX improvements," then it will all end up in the shitter. And you see this too - projects wasting lots of time going on utter wild goose chases because they weren't xenophobic enough with regard to the new clientele. Consider how many technological projects that greatly changed people's lives you got during the period 2000-2010. BitTorrent (2001, TPB 2003), Tor (2002, open source 2004), 4chan (2003), Bitcoin (2009). I'm sure you could name plenty more, depending on what stuff you're into. Now the 2010's are almost over. What did we get during them? Hardware stagnated, and the so-called experts claimed this would result in great strides in software instead. Where are they? You can add 5% for fairness, since we're still half a year from 2020. I can't think of hardly anything. Darknet markets, while utterly reprehensible, arguably form an improvement from the status quo. Silk Road was opened in 2011, so I guess we can count that in. We still didn't get assassination markets though, nor good betting sites. And the prediction markets are still all clearnet and with KYC out the wazoo. Any other takers? Because "it got slightly easier for people who frankly should not exist to order things they frankly should not order" really does not feel like a meaningful improvement to me. The those sorts of people I know still do that type of business like they would have used to, without relying on any high-tech encryption or the like. User-friendliness is like makeup. If it attracts good developers and results in the furtherance of the project, then, sure, add it, why not?, but if it's just to make things slightly more pleasant at the end cost of attracting useless people and thereby becoming useless, then what purpose does it serve? There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-28, 21:46 in I have yet to have never seen it all.
|
|
Stirrer of Shit
Post: #543 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Not the first time the EU kills some anime project, mind you. A lot of people in the West actually went to prison over cartoon girls doing lewd things, both in America and Europe, so I understand their choice. But why couldn't they just host it in Japan or {insert shithole with absolutely no standards here}? They don't tend to be overly concerned with such matters. Heck, CP was entirely legal in Russia up until a few years ago. (They might have pulled some GDPR move where an American living in America who had never set foot in one of their disgusting shitholes still was under their law because reasons, but I doubt even the EU would be that insane) Well, I suppose it'll fare just like that other project, in that you wait a few weeks and then everything is back to normal, just slightly worse but hopefully more secure. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-28, 21:57 in Games You Played Today REVENGEANCE
|
|
Stirrer of Shit
Post: #544 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
People complaining about derailed threads? On my BBoard? There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-30, 21:03 in Ubuntu: x86_32 is dead because WE SAY SO!
|
|
Stirrer of Shit
Post: #545 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by funkyass The ancient philosophers were right about a lot of things, far more than you'd think, and I'd argue on a lot of things we're now wrong on. In this case, you're probably talking about Plato/Socrates: Posted by Phaedrus I think he's got a fairly reasonable point here. Books then weren't what they are now, with the references and all. Anyhow, it doesn't feel all to relevant. My issue is that with user-friendliness, you attract incompetent people who should in the ideal case be kept far away from anything of importance. Not that user-friendly software is bad unto itself. If you would, for instance, ban stupid people from using semiconductors, there wouldn't be all of these problems. User-friendliness never caused any of the negatives you are complaining about, and the positives you are extolling never existed. Well, this seems like common sense. Surely, if you need less skill to operate computers, then less skilled people will use them? And if UX improvements have marginal returns, so that less skilled people get more out of them, which intuitively seems true - consider old people or kids - they couldn't find the start menu on a computer, but their phones they can play around with all day without much issue, whereas technologically skilled people tend to not have much trouble with either, and accomplish about as much (not to say less) with their phones as computers, despite the former being more user-friendly - then it stands to reason UX improvements get you more incompetent people. How didn't it cause said negatives? Of the people who used computers in the 80's, they sure were far more knowledgeable than the people who used them in the 90's, who in turn were far more knowledgeable than those who used them in the 00's, who in turn... you get the point. With lowering barriers to entry, you get a lower and lower quality clientele. That is all. My main point is NOT that software that's easier to use makes people dumber, although that might be the case too at the extremes. I mean, this goes for everything. Like with the case where Youtube made their pages load faster, and then they noticed average page load time was going up. How come? Because then they got people from rural Siberia who could suddenly use it, and they dragged down the average. And on a similar tone, go on any famous Instagram person's page and look at the comments. You know what I mean. Those people sure aren't experts with the computers, even if that's what they'd put on Quora. Posted by tomman Let's analyze the scenario in full: if it's indeed only used by people with degrees from (good) universities, then surely the average skill level should be through the roof. So that's not entirely a downside. At the other end... we get the smartphone generation, where a few "disruptive entrepreneurs" took user friendliness to the extreme, dumbing down to the point of uselessness, trying to cater to that mass of population to which computers and electronics are little more than tools to achieve something, without considering consequences or side effects. The extreme who got us diseases like social networks and advertising, and yes, that's another scenario some of us really hate. Well, this just continues the same development, it's not a flip side of anything, just the eventual progression it has to end up at if UX improvements continue. Extremes are bad, but that's not excuse to condemn user friendliness on principle. I will never use cellphone crapps beyond the basics, but that doesn't mean I'll ditch web browsers and move to Lynx/Elinks (I like watching my Well, let's put it this way. The share of fresh blood with rational ideas is going to be about the same as the share of young people who use computers who are any good at them. So if more stupid people can use them - the end result of any UX improvements - that'll go down. If it's at 0%, software will be K-selected - "large body size, long life expectancy, and the production of fewer offspring," and if it's at 100%, software will be r-selected - "high fecundity, small body size, early maturity onset, short generation time, and the ability to disperse offspring widely". At which end of the spectrum are we today, where were we ten, twenty, thirty years ago, and whither are we heading? The driving cause of this development is lowering barriers to entry, and they must be raised again. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-31, 12:22 in Ubuntu: x86_32 is dead because WE SAY SO!
|
|
Stirrer of Shit
Post: #546 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by funkyass That's a bit rude, to claim it's a logical fallacy to say anything at all has gone downhill, don't you think? I would think it is possible for things to have been better before, even if it's not always the case. The claim here is very simple, it's not just some "they don't make computers like they used to back then" tripe. 1) If computers are easier to use, more stupid people can and do use them. 2a) Programmers are drawn from the pool of users. 3a) Stupid programmers make bad software. or 2b) Programmers develop software in accordance with the wishes of the users, as determined by the market. (in economic terms, in accordance with the wishes of the market) 3b) Stupid people want bad (e.g. r-selected) software and furnish for the requisite financial incentives. (e.g. pay for it, either directly or indirectly) There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-07-31, 15:57 in Ubuntu: x86_32 is dead because WE SAY SO! (revision 1)
|
|
Stirrer of Shit
Post: #547 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by CaptainJistuce Well, that's a bit rude, but it's also an accurate characterization of my beliefs. I thought you weren't bullish on the smartphone/webapp trend. Et tu, capitanee? Posted by Wowfunhappy Aren't there one-size-fits-all no-Steam patches anyway? As long as you have the actual games preserved, you can sort out getting them to run properly later on. Not as if all copies of Windows XP will disappear any time soon, or as if the DRM is very strong. I mean, DRM isn't DRM. It's not like all of these games are shipping with Denuvo, or worse. And if they do, you can just wait a few years and then download them off of The Pirate Bay into your collection when someone eventually cracks it. Which I won't think should be possible for so long anymore with always-online DRM. But when it does, it won't be Steam who are to blame for the development. (It's curious why nobody did that yet - Denuvo made them mad stacks, and hardly anyone plays offline anyway, so if they merge DRM and anti-cheat in one then they'll kill two birds with one stone, and as a bonus get really nice facilities for A/B testing, analytics, and whatnot - it's also far cheaper and more effective) EDIT: were -> weren't There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 01:45 in Ubuntu: x86_32 is dead because WE SAY SO!
|
|
Stirrer of Shit
Post: #548 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by Kawa I don't mean offline as in by yourself, but as in actually lacking an Internet connection. I can think of a few reasons why you wouldn't have an Internet connection: 1) you live in a shithole without infrastructure (note, this could also mean "way out in the sticks") 2) you're visiting somewhere and don't have Internet there 3) your Internet is temporarily out 4) you don't feel any need to use computers Obviously, the people who don't use computers don't play video games on them and thus they don't purchase any, so we can count them right out. 2 and 3 are transient and is probably not a factor for most people (Christ, go read a book or something), and most of the people in category 1 have purchasing power equal to zero. So for most games, it should make perfect economic sense to use hard always-online DRM. Posted by tomman No, look, I'm not saying it's particularly nice or that they should do it, just that it'd be profitable and it's odd they're not fucking people over in this specific manner if it's profitable to do so. So much for the EMH, I guess. Alternate hypothesis: Since the desktop platform exists just to provide an illusion of control, and they're already redlining it with W10 and whatnot, too much cloud, especially in the safe space of Steam, home-built computers, and piracy, might be bad for business since it kills the platform's appeal, and considering kids today just use smartphones for their games, they're extra careful not to kill it off prematurely, since it can't really recover as it could. Alternate hypothesis 2: Desktop gaming just skijors off of console gaming, so whatever they do there gets done everywhere. For instance, Denuvo was explicitly designed around allowing offline play, so maybe they just want to get as "console-like" semantics without having to worry about anything else. Posted by CaptainJistuce What about the web-app trend then? There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 16:54 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #549 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by byuu What about using another GUI toolkit? Try nuklear or imgui, for instance. Cross-platform, minimalist, extensible, public domain. Then you could add in whatever controls you wanted to. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 17:33 in I have yet to have never seen it all.
|
|
Stirrer of Shit
Post: #550 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by CaptainJistuce The article says it runs at 20 MHz, make of it what you will. Posted by tomman Posted by CaptainJistuce Well maybe they just shouldn't have hired such code monkeys? Based on my experience with those people, this seems more like a case of "a bad carpenter blames his tools": Posted by https://www.bloomberg.com/news/articles/2019-06-28/boeing-s-737-max-software-outsourced-to-9-an-hour-engineers The middle management divisions have got to cool it with their credentialism. Three years of education is just not the same as three years of education, plain and simple. They can't claim that they have degrees so they should be as good as the usual engineers and everything's fine, because then this sort of stuff starts to happen. University is a giant fraud already, so just imagine how big a fraud it is with such contractors. How do you say it in English, wise for a penny and stupid for a pound? And they don't even seem repentant: In a statement, HCL said it “has a strong and long-standing business relationship with The Boeing Company, and we take pride in the work we do for all our customers. However, HCL does not comment on specific work we do for our customers. HCL is not associated with any ongoing issues with 737 Max.” They should have fired all those people along with whoever was responsible for hiring them and brought in competent developers whenever these sorts of debacles first started happening. (Of course, they claim they had their real engineers doing the parts that failed, but considering how low they already stooped in hiring them, do you really believe them?) There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 20:47 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #551 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Well, by my understanding, they do different things. His is a lowest-common-denominator mapping to GTK/Win32/Cocoa, which restricts you to using their widgets, but on the other hand gives you a native look. That is good for serious business use, but not so much for when you want fine-grained control. Whereas these are immediate-mode toolkits, which enable you to draw anything at all without restriction. I suppose another way might be to add a nuklear backend to hiro and then say amethyst should only ever be compiled using this backend. This is extremely ugly. But on the flip side, it would enable you to use bsnes/higan without having to start your X server. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 21:14 in I have yet to have never seen it all.
|
|
Stirrer of Shit
Post: #552 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Well, it's a bit misleading I reckon. You have the utterly disgusting but legally mandated glued-down tablet ("infotainment system") for parking monitor, GPS, radio, etc, which runs bog-standard Android on some low-end off-the-shelf chipset, and then you have the system that's responsible for all of the drive-by-wire stuff (e.g. ABS), as well as all the other sensitive functions (e.g. LDWS, fatigue warning, check engine light). This lives on some microcontroller and is probably very heavily regulated. I wouldn't think that microcontroller's code clocks in at 100 MLOC, but Android + web browser + GPS + other bloat doing so seems reasonable. The tablet will occasionally shit the bed, and nothing else will happen. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-01, 23:50 in I have yet to have never seen it all.
|
|
Stirrer of Shit
Post: #553 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
They can't, not more than that the infotainment system should be able to break without the car becoming unsafe to drive. I suppose the reverse/parking camera has to work, but that ought to be trivial to solve. Or perhaps that's not considered security-critical in the same sense - if it breaks, you'll notice immediately, whereas if your ABS breaks, you'll only notice it if something goes wrong. https://en.wikipedia.org/wiki/Electronic_control_unit - these run on microcontrollers and are rigorously tested https://en.wikipedia.org/wiki/In-car_entertainment - these run on ARM chipsets and are made by the lowest bidder You're right that the infotainment can interface with the ECU. Which is a bit of a shame if you download the wrong app on it and 1-2-3 "you" floor the accelerator and drift into oncoming traffic. Whoops. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-02, 00:19 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #554 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Posted by KawaPoint taken. Posted by tomman You've got to have some kind of standards. Writing substandard HTML is going to cause all sorts of issues later on down the line. It's like going too hard with the idiomatic C and invoking UB, and then claiming it's not UB because it compiles. Civilizations abandoned the law of the jungle thousands of years ago, when will technology follow suit? If the goal is just "ooga booga grug want text in browser fast," then why bother generating valid HTML at all? Then you could use a far simpler regex-based conversion. If you just write the elements you want and save as .html, Firefox/Chrome renders it just fine. There's no need to use the "real" tag names either. You can just make them up and use CSS to style them, and it will look as normal. No need for garbage like meta, head, body, or html, you can just start writing. Here, try pasting this into a HTML file and opening in your browser. It renders fine, so it must be valid HTML.
<h1>grug want big text</h1> <header>ooga booga</header> <p>Seriously, not every website has to follow "design rules" as long as the information is clearly laid out in a readable fashion. <span>KISS. byuu's websites are an <style> red { color:red} header{ font-size:2em; </style> excellent exhibit of this </p> <red>(his design choices may look weird in this era, but our eyes and bandwidth bills greatly thank sen<title>FAST WEB DEVELOPMENT - CONTACT SUREANEM NOW</title>sible designs like those - who cares if you're "Doing It Wrong™" with ancient HTML tags?!)</text> <html><!-- Granted, there is something to be said for this style of web development as a deliberate statement, but you should probably at least try to use the tags they give you as far as possible. Consider text browsers, for instance, or browsers without such sophisticated parsers. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-02, 14:05 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #555 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Using <header> to indicate a heading is arguably just as broken HTML. Neither example uses the elements as stipulated by the standards, both render correctly in modern browsers given adequate CSS and choke on standards-compliant but strict browsers. What's the difference?Posted by https://html.spec.whatwg.org/multipage/sections.html#the-header-element There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-02, 17:35 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #556 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Yes. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-02, 22:52 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #557 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Sure, but so does that abomination I just posted. That doesn't make it not broken, unless your idea of standards is "whatever Chrome does". There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-02, 23:22 in Retroarch's (controller) interface is an bad abstraction
|
|
Stirrer of Shit
Post: #558 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
The "all controllers are X360 or close enough" is an acceptable way of going at it, even if it has its negatives. Letting users configure individually, while nice, will be a lot of effort for the 20-odd consoles they support. I'd like to refield my suggestion: two layers. Each game gets assigned to a preset (FPS, racing, platformer, etc), and then that preset gets mapped to the controller in an idiomatic way. So for instance, in the FPS preset, you define the most common buttons: * movement (analog) * looking around (analog) * jump * crouch * reload * action * switch weapon 1 (e.g. to primary, or left) * switch weapon 2 (e.g. to secondary, or right) * melee * sprint * walk * equipment 1 (e.g. grenade slot 1) * equipment 2 (e.g. grenade slot 2) * aim * shoot * pause * scoreboard * wildcards 1-4 (e.g. c-buttons) And then, for each console, you have a base preset. And then some game-specific overrides, in case they decide to switch stuff around. Then, those presets have default mappings to the different controllers (respecting the applicable idioms). And players can override them. So for PC, this mapping might look like: * WASD * mouse * space * ctrl * R * E * mwheeldown * mwheelup * mouse3 * shift * caps * mouse4 * mouse5 * mouse2 * mouse1 * escape * tab * ZXCV Is it perfect? No, some games might do arcane stuff. Is it a lot of work? Yes, comparatively. But I would argue it'd be the best way of going at things. On the other hand, there is something beautiful about the homogeneity of fuck you, go buy a normal controller like a normal person. A consistent standard is always easier for everybody. On the same note, fuck 4K and your ridiculous DPI settings. A pixel is a pixel, period. People should use 1080p screens, and applications should take this into account (e.g. assume they do and invoke UB otherwise), unless they're also designed to be used on laptops. Terry was absolutely right on the macro level and catastrophically wrong on the micro level, as such often are. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-03, 00:30 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #559 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
The intermediate step between the law of the jungle and proper civilization, which is known as ...? There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
| sureanem |
Posted on 19-08-07, 20:44 in amethyst (text editor)
|
|
Stirrer of Shit
Post: #560 of 717 Since: 01-26-19 Last post: 1763 days Last view: 1761 days |
Just use <div> for everything. HTML is inconvenient and overly-complex. There was a certain photograph about which you had a hallucination. You believed that you had actually held it in your hands. It was a photograph something like this. |
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
