Hacker News new | past | comments | ask | show | jobs | submit | _xg6h's comments login

Can you link to the example? I've been interested in doing something like this, but as far as I can tell, you can only read messages using the Graph API; there's no method for sending messages.


Can I just add $$x^2$$ to my text and have KaTeX render it automatically (like MathJax does), or do I have to use katex.render() directly?


katex.render() if I recall.


You can have something like this in your .bashrc:

    multitail() {
        com="tail -f "

        for arg; do
            IFS=':' read -a array <<< "$arg"
            host="${array[0]}"
            file="${array[1]}"

            com="$com <(ssh -t $host 'tail -f $file')"
        done

        eval ${com}
    }
This will take the host:file syntax of remtail and turn it into Plugawy's example. I haven't tested it, but if you echo "$com" it'll spit out Plugawy's code.


nice solution! bash looping and arrays always seemed overly complex and confusing to me, but this is not nearly as complicated as I expected.


That's because the <title> element only contains "LKML: " on the thread, and "LKML.ORG - the Linux Kernel Mailing List Archive" on the main page.

Maybe they forgot to keep the "unofficial" in the title?


It's possible the false sense of security is worth more than people think. It lets them say "hey, we tried".


I think this is a very important point. These media companies have not been able to extract as much rent from their IP as they could ten years ago, and their shareholders are not happy about that. They absolutely have to try something.


Certain companies have been extraordinarily interested in implementing WebCrypto without mandating HTTPS. Which is to say, a man-in-the-middle attacker could trivially modify the JS that calls WebCrypto and cause different operations to be performed.

My suspicion is that they have contractual agreements with the non-technical folks in the studios that they have to "encrypt" content, and the technically-competent redistributor has no direct interest in the crypto being sound. If the API gave them 256-bit military-grade AES encryption, but only in ECB mode, they'd probably use it.


I'm not sure what the benefit of HTML5 video is if you still end up being forced to use some proprietary component. You ditch Flash but now you have to run some DRM software.

And there's so much more to HTML5 than video; I disagree that lack of DRM video support would set the rest of HTML5 back in any way.

EDIT: I guess the sandboxing aspect of EME is a significant improvement over Flash running as a plugin.


The benefits of html5 video are practical ones: it simply works better than flash does. It plays without stuttering, the audio works properly, it doesn't steal my mouse focus. Also, it doesn't get webcam permissions by default, or all the other crap that comes along with flash. Sure, its still DRM, but flash was terrible for plenty of reasons that aren't DRM.


Which is funny, because flash was also great for creating content... If you were building learning demos or content, flash is still better than having to hand-code an alternative with html+svg+js... Though one of my early hopes when Adobe bought macromedia was that they'd push the authoring tool to export an archive file (similar to xaml) that would contain a manifest along with svg and javascript as a bundle. That never really happened though... I still hope that now it can.

I know a lot of elearning content creators that really miss being able to use flash for most things... yeah, the player was horrible, and the formats sucked.. but the content creation experience was so nice. Creating data driven projects with Flex wasn't bad either.


HTML5 video works on phones and tablets


But with EME it will only work on those phones and tablets which are running on an architecture and platform that the EME CDM has chosen to bother supporting. So it's pretty much as bad as flash.


And blu-ray only works on blu-ray players, not Motorola StarTACs.

I don't get most of the arguments on this thread. And I'm very much on the fence about DRM as a producer of licensed content, and a consumer of it.


And video file formats are not physical objects you put in a physical disc player. False equivalence makes your point idiotic. Hard to believe how customers are willing to accept more and more of this bullshit, vendors did a great job at brainwashing them.


And you don't own those intangible things, you license them, so how is it you expect to be able to do MORE than the average consumer of a tangible product like a blu-ray?

It's not a false equivalence fallacy when the alternate argument only inverts the parameters while violating the same logical foundation. And it's insulting to suggest I (or anyone else) have been brainwashed because my world view is a bit more pragmatic than, "give me all the rights."

I'll eat the down votes for that, no problem, because this shouldn't be an adversarial debate between producers and consumers. There may be better ways to approach acknowledging producer rights and consumer usage, even if the consumers who feel their rights are violated are outnumbered by those who have no problem. Sure, the argument could be made, "if only they knew how their rights are being violated!" That argument could be made about any number of situations where one class feels like they need to lift up another one. And it's just as meaningless unless you do something to change the situation.

DRM is evil, blah blah blah, I get it, as a consumer, I do. I also get that there are arguments for DRM that are perfectly valid from the point of view of the producer, even if the consumer thinks that the producer is off his nut. Surely there are people out there on HN who aren't so polarized by this issue? Sometimes, it feels like the US Congress in here.


It looks like it's unstable. I can connect to it occasionally but most of the time Firefox is saying the connection was reset.


I think that probably falls under "launch other executable code".


You're wrong.

I just copied and pasted some Scheme from the net into the Gambit Scheme REPL. Worked fine. Apple did nothing to stop me, nor could they.


I don't know Perl, so I rewrote it in Python in an attempt to understand what's going on: [link redacted]

I'm not storing key,value pairs for elements that don't exist (not sure what the Perl code's doing), so it shouldn't be too bad in terms of space used.


I'm not entirely sure about this, but it looks more like Counting sort [1], than Pigeonhole sort.

[1]: https://en.wikipedia.org/wiki/Counting_sort


Oddly enough, if you search for "test", you do get results. Nothing else works, though.


Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: