Feature request: if a link has NSFW and is being downvoted, please gray everything except the NSFW part (and translate it to those pure souls who don't even know what is NSFW)
Or, disable the link automatically for highly downvoted comments
Or at least make the link use the same color
(Or a bit easier, if it has NSFW, just remove it automatically)
> (Or a bit easier, if it has NSFW, just remove it automatically)
Why? NSFW doesn't always mean it's not interesting content. It's perfectly possible for there to be interesting technical discussion which happens to be regarding adult content (be it pictures, or large swear words, etc.)
At first glance I agreed with the downvoters, but... assuming this isn't updated manually, would be interested to see the code behind it, as it generates comic images which only replace the last frame with goatse. For example (and this is obviously NSWF and shows a goatse picture) http://goatkcd.com/strips/1170.jpg
At the risk of getting downvoted even more I will share with you that this is done automatically. The guy that made that site wrote a script that detects the last panel of an xkcd strip. I will see if I can get him to share more because it really is cool.
This was my initial reaction as well. However, I then realised that being able to see where the link leads to (assuming it's not a redirect) may very well be of more importance.
My suggestion on posts with as much downvotes as this one would be to remove the <a>-tags which, in addition to introducing the grey colour would also prevent unintended clicks and bring attention to the destination of the link as it requires copy-paste.
I really tried to be as clear as possible regarding the contents of the link, not sure how I could have made it clearer... maybe if I wrote "goatse/parody" instead of "goatse parody"? I think it's definitely relevant considering the technology behind it and the comic it's lampooning.
When I first clicked on it, all of the text besides the link was completely invisible, so I had no idea it was NSFW. I guess a lot of other people didn't realize that before they clicked on it at work.
Here is how the web site works btw, from the author:
"it flattens the image to 1-bit colour depth then looks in the bottom right corner for a ┛ shape. if it finds one sufficiently big it follows the edges up and left as far as possible to find out how big the panel is"
This is one of the coolest things about HN. There's a link there, and it crashes your browser. Instead of getting angry, you try to help the person whose creation crashed your browser.
A warning is fine, but I just hate it when a site doesn't even let me try in another browser. I use Firefox Nightly, maybe it fixed the issues that are breaking your app? If you're detecting via user agent, what if I use an odd user agent? What if I want to look at the web console to see just what breaks in case I know what the issue is?
> If it's using -webkit prefixed CSS, that's pretty unlikely.
Unsupported CSS features should not affect the functionality of a web app, only its appearance. Emphasis on "should"; far too many web apps end up fundamentally broken when this happens.
I think something like web-based SVG editor is pushing browser limits too far (some folks reported it crashed Chrome for them). CSS is not "just styling" here. It has functional meaning. For example invalid z-ordering, visibility or cursors can pretty much ruin the experience.
They should at least include the non-prefix css (or other prefixes) as well for when a feature becomes available on other browsers in case they forget to update it or if they're not aware of what other browsers support.
However, it does limit feedback for any of us not using Chrome. Author is probably cool with that though as it appears and I commend them for not making us jump through hoops to try it anyways by switching user agent.
The same thing can be achieved by displaying a simple warning that lets users bypass it ("oh, I was warned it might not work in my browser, and indeed it doesn't, I will take a look in Chrome").
Q: "I use Firefox Nightly, maybe it fixed the issues that are breaking your app?"
A: "If it's using -webkit prefixed CSS, that's pretty unlikely."
There's actually been talk of non-Webkit vendors supporting the webkit prefix. Opera already does, Microsoft and Mozilla have said they need to support it in at least some cases.
"Non-WebKit browser vendors first started talking about implementing the -webkit prefix [in February of 2012] during a CSS Working Group meeting. Microsoft, Mozilla and Opera all said they felt the need to support -webkit, lest their users be relegated to an inferior browsing experience (because so many sites are using only the -webkit prefix)."
Tantek Çelik of Mozilla: "At this point we're trying to figure out which and how many webkit prefix properties to actually implement support for in Mozilla. Currently we have zero. Zero is no longer an option for us."
Florian Rivoal (Opera), Sylvain Galineau (Microsoft): "Zero is not an option for us anymore either."
I hope browser vendors only implement support for webkit- prefixes for now-standard properties (eg. box-shadow), and not for properties that only WebKit implements.
It doesn't work under FF Nightly. I'm sorry this was a quick way how to deal with the issue. I don't want to deal with complains "it is broken / it doesn't work for me".
I'm going to push new version which will allow you to continue over that warning.
Thank you very much for the update! I apologize for being rude before, I don't want to bring you down on a very successful day on Hacker News. :)
For what it's worth, though the visual editor doesn't work in Nightly, the code editor + saving does work! It was really fun to mess around with the source and edit your sample comic. Great work!
The main problem with IE wasn't that it was an exclusive product incompatible with other browsers. The problem was that it was bad. I'm not a fan of people only supporting WebKit, but at least WebKit browsers are good.
I used IE briefly in 2000 and 2001, before Mozilla (0.9) became usable enough and Phoenix came out; back then (IE 5.5 to IE 6.0) it definitely was the best browser, by far.
It was the most forgiving browser, with a quirks mode that handled broken code like a champ. But Opera was the best browser, and if your code rendered correctly in it, it would probably render correctly everywhere else (except maybe Netscape 4, an abomination that took too long to die).
Opera was also commercial. At some point there was a free adware version, but I think that was later than 2001. After IE, most people stopped paying for web browsers.
Actually, for a while, IE was the best browser available. It was lighter weight, faster than Netscape, and had better web standards support. Mozilla, with its new Gecko engine, took a long time to mature, and was still fairly slow and bloated until Firefox (née Phoenix) came around. Remember, IE was where several features that now form the basis of the modern web came from. Heard of something called XMLHttpRequest? That came from IE.
The problem was that it was bad, in the ethics sense of the word.
IE was an attempt to aggressively eject the competition and kill the open web. When the web stubbornly refused to buy into IE-only technologies, it became Microsoft's dog in the manger, holding back the day that web apps made the OS irrelevant. It was good, technically, and then it was bad, technically, but it was always always evil.
You can run Chrome on Linux, Mac and Windows! And at least in theory you can have a go at trying to port the Chromium browser to other platforms, as well.
It works, fairly well. But I spent about 15 minutes doing one panel, trying to figure out the minor nuances. I kept thinking to myself "I could do a whole comic in about 5 minutes in photoshop" as I adjusted the person's location, arm, speech, etc (and that was just modifying the characters that were already in the scene). In the end it was just not worth it in my mind.
Also, imagine creating something like Randalls "Click and Drag" comic (xkcd.com/1110/) -- how would you create all these custom objects in the editor without changing it into something similar to an HTML5 web-photoshop, at which point you might as well do that?
Yes, this a prototype and definitely not for everyone. It has many downsides. I agree Photoshop gives you full pixel control.
I think it has some upsides too. You may treat comix source as html code and apply tools we have for coding today, eg. diffs, version control, collaboration via github. People may also collaborate on library of items, actors and their poses, so you don't have to re-create them or copy&pase all the time. Also localisation will be a snap. And it will possibly allow animations in the future.
Actually I built this for a bigger project. I want to explain Bitcoin technically in the form of XKCD comix. This won't be one-off comix strip. This will be something like wiki/book which will evolve and I hope it will attract other contributors (not necessarily graphic designers).
I agree it has some great upsides. I love the idea of localization, and version control is nice to easily revert changes, etc. Having a scalable comic that can change text on the fly is something that I've been hoping for for a long time.
But I think you're strongest element is the WYSIWYG editor, as most people aren't going to be coding from only the HTML view (unless they know what pose="-11,9|-5,117|-11,99" will look like in their heads). But the WYSIWIG interface is still pretty anemic, meaning I'm going to run into roadblocks faster (such as -- how do I create something like a tree? How can I add a hat to the man?).
I want to provide a nice blend between visual editing and source code editing (I'm a developer). Poses and transformations should be edited visually, but scene structure should be edited by hand via the code. This way you get total control over document structure. Creating good visual-only editor would require too much work and maybe it is not possible at all. Do you know any successful FrontPage-style HTML editor? Quite frankly, I don't need it myself, so that is the primary reason, I'm probably not going to build it at this point.
Actually the system already supports attach points, so you can attach items to hands, foots and necks (right now the bubble is attached by default to a head bone).
The last generic piece which is missing will be <drawing> element. Right now it just allows you to specify lines. In the future it will enable you to insert any SVG which is convertible down to paths. I will just resample them and convert them into XKCD-style lines while keeping other styling. This will enable you to insert arbitrary SVG drawings in the scene. I won't create web-based photoshop for SVGs, I will let you import SVG created in other tools as long it is convertible to paths. Imagine it as embedding bitmap images into HTML files. This will be similar idea.
I like this, but it's difficult to use and I expect that only the WYSIWYG will be a success. (Drag some control points, like foots or knees.)
But on another hand, some things are more complicated that one expects. Can you really draw well enough? Can you share an example of a xkcd-style comic?
Perhaps a higher-level comic description language that generates the html markup? A bit like the way that LaTeX generates TeX. It could have predefined 'styles' for character appearance, body posture, gestures, etc. Also the ability to embed lower-level html markup in the higher level description.
Perhaps I need to get back to doing some actual work...
Better idea, draw the comic by hand in 5 minutes (I think everyone can draw stick figures), photo it, desaturate and add high contrast in gimp, voila. No webkit required :)
This little comic I "created" some time ago is never going to be so relevant again so here it is [inmature humor ahead]: http://i.imgur.com/EnXc9e9.jpg
Unfortunately, the editor does not work for me. It makes Chrome crash the page :(
I'm running Chrome 24.0.1312.69 on one of those new Dell XPS 13 Ubuntu machines running 12.04.2 LTS. Is there any other information you'd want for debugging, or pages to try?
This is awesome! Okay,I agree with others that say that photoshop is easier, and that most people (ahem, like yours truly) would make crappy comics. But...c'mon...an xkcd editor?!?!?1!! hecka cool
Wait, there is WYSIWYG editor for tweaking "t" and "pose" attributes visually.
Also quite frankly my target audience is myself (a developer). I think I can use this tool to write fine comixes without being an artist. xkcd-style helps to hide this inability pretty well :)
I think its brilliant. I love the balance you've created between the visual editor and the markup.
Have you considered this as an educational tool? This is a fun and useful introduction to HTML and markup languages in general.
Would it be possible to highlight the relevant markup node when you've selected an object in the visual editor? That would be neat from a learning perspective.
115 comments
[ 0.23 ms ] story [ 187 ms ] thread[1] NSFW http://goatkcd.com : goatse parody
Or, disable the link automatically for highly downvoted comments
Or at least make the link use the same color
(Or a bit easier, if it has NSFW, just remove it automatically)
Edit: yes, I agree, ignore the last one
Why? NSFW doesn't always mean it's not interesting content. It's perfectly possible for there to be interesting technical discussion which happens to be regarding adult content (be it pictures, or large swear words, etc.)
http://eat-it-downvoters-you-cant-silence.me
My suggestion on posts with as much downvotes as this one would be to remove the <a>-tags which, in addition to introducing the grey colour would also prevent unintended clicks and bring attention to the destination of the link as it requires copy-paste.
Those who have seen it will know what I'm referring to.
WebKit - the new IE.
If it's using -webkit prefixed CSS, that's pretty unlikely.
> What if I want to look at the web console to see just what breaks in case I know what the issue is?
Anyone interested in doing that should be well aware of how to spoof a user agent.
Unsupported CSS features should not affect the functionality of a web app, only its appearance. Emphasis on "should"; far too many web apps end up fundamentally broken when this happens.
If you're building a fun hobby project, you can do whatever you like, including building it in a specific environment.
However, it does limit feedback for any of us not using Chrome. Author is probably cool with that though as it appears and I commend them for not making us jump through hoops to try it anyways by switching user agent.
https://github.com/LeaVerou/prefixfree
You are not answering the question. Why block a user-agent, if the check can be easily worked around?
Edit: I mean why block as opposed to display a simple warning ("this site is known to not work well with your browser").
A: "If it's using -webkit prefixed CSS, that's pretty unlikely."
There's actually been talk of non-Webkit vendors supporting the webkit prefix. Opera already does, Microsoft and Mozilla have said they need to support it in at least some cases.
"Non-WebKit browser vendors first started talking about implementing the -webkit prefix [in February of 2012] during a CSS Working Group meeting. Microsoft, Mozilla and Opera all said they felt the need to support -webkit, lest their users be relegated to an inferior browsing experience (because so many sites are using only the -webkit prefix)."
http://www.webmonkey.com/2012/04/opera-forges-ahead-with-pla...
Tantek Çelik of Mozilla: "At this point we're trying to figure out which and how many webkit prefix properties to actually implement support for in Mozilla. Currently we have zero. Zero is no longer an option for us."
Florian Rivoal (Opera), Sylvain Galineau (Microsoft): "Zero is not an option for us anymore either."
http://lists.w3.org/Archives/Public/www-style/2012Feb/0313.h...
I'm going to push new version which will allow you to continue over that warning.
For what it's worth, though the visual editor doesn't work in Nightly, the code editor + saving does work! It was really fun to mess around with the source and edit your sample comic. Great work!
IE was an attempt to aggressively eject the competition and kill the open web. When the web stubbornly refused to buy into IE-only technologies, it became Microsoft's dog in the manger, holding back the day that web apps made the OS irrelevant. It was good, technically, and then it was bad, technically, but it was always always evil.
We're doomed to repeat the same mistakes over and over again.
You can run Chrome on Linux, Mac and Windows! And at least in theory you can have a go at trying to port the Chromium browser to other platforms, as well.
It works, fairly well. But I spent about 15 minutes doing one panel, trying to figure out the minor nuances. I kept thinking to myself "I could do a whole comic in about 5 minutes in photoshop" as I adjusted the person's location, arm, speech, etc (and that was just modifying the characters that were already in the scene). In the end it was just not worth it in my mind.
Also, imagine creating something like Randalls "Click and Drag" comic (xkcd.com/1110/) -- how would you create all these custom objects in the editor without changing it into something similar to an HTML5 web-photoshop, at which point you might as well do that?
I think it has some upsides too. You may treat comix source as html code and apply tools we have for coding today, eg. diffs, version control, collaboration via github. People may also collaborate on library of items, actors and their poses, so you don't have to re-create them or copy&pase all the time. Also localisation will be a snap. And it will possibly allow animations in the future.
Actually I built this for a bigger project. I want to explain Bitcoin technically in the form of XKCD comix. This won't be one-off comix strip. This will be something like wiki/book which will evolve and I hope it will attract other contributors (not necessarily graphic designers).
But I think you're strongest element is the WYSIWYG editor, as most people aren't going to be coding from only the HTML view (unless they know what pose="-11,9|-5,117|-11,99" will look like in their heads). But the WYSIWIG interface is still pretty anemic, meaning I'm going to run into roadblocks faster (such as -- how do I create something like a tree? How can I add a hat to the man?).
Heads should be customizable in the future: https://github.com/darwin/cmx.js/blob/master/app/lib/cmx/ent...
Actually the system already supports attach points, so you can attach items to hands, foots and necks (right now the bubble is attached by default to a head bone).
The last generic piece which is missing will be <drawing> element. Right now it just allows you to specify lines. In the future it will enable you to insert any SVG which is convertible down to paths. I will just resample them and convert them into XKCD-style lines while keeping other styling. This will enable you to insert arbitrary SVG drawings in the scene. I won't create web-based photoshop for SVGs, I will let you import SVG created in other tools as long it is convertible to paths. Imagine it as embedding bitmap images into HTML files. This will be similar idea.
The WYSIWIG will come in it's million forms later and you shouldn't focus on it.
I just need comix.vim and I'm good.
I'm imagining real time comics. Chat comics. Comic discussion forums. It's endless! ;)
It'll be 1996 all over again: http://en.wikipedia.org/wiki/Microsoft_Comic_Chat
That was my first thought when I saw the markup. Brilliant work!
But on another hand, some things are more complicated that one expects. Can you really draw well enough? Can you share an example of a xkcd-style comic?
Perhaps I need to get back to doing some actual work...
http://yuml.me/e5d8e14e
http://yuml.me/diagram/scruffy;/usecase/edit/[Bob]-(This%20i...
T
Unfortunately, the editor does not work for me. It makes Chrome crash the page :(
I'm running Chrome 24.0.1312.69 on one of those new Dell XPS 13 Ubuntu machines running 12.04.2 LTS. Is there any other information you'd want for debugging, or pages to try?
Also quite frankly my target audience is myself (a developer). I think I can use this tool to write fine comixes without being an artist. xkcd-style helps to hide this inability pretty well :)
I just get an error message saying that the page script was taking too much memory and was stopped.
Have you considered this as an educational tool? This is a fun and useful introduction to HTML and markup languages in general.
Would it be possible to highlight the relevant markup node when you've selected an object in the visual editor? That would be neat from a learning perspective.