Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[wierd]

I would say it is more likely to be a codec related issue.

Try encoding your upload with a different encoder before uploading. (Youtube is going to convert it to whatever the hell it is going to convert it to regardless, but giving it something less likely to get CPU stomped during re-encode, seems reasonable.)

[LordBaal]

Could be, why I suggested uploading videos not edited but the randomness of it seems to indicate something else.

[methylatedspirit]

(Youtube is going to convert it to whatever the hell it is going to convert it to regardless, but giving it something less likely to get CPU stomped during re-encode, seems reasonable.)

For unnecessary info, Youtube transcodes to the following codecs for general web use: H.264, VP9, and AV1. These are the main codecs served to devices if you run youtube-dl -F on a video. I know that YT uses at least parts of FFmpeg for video decoding-- I once uploaded a video in ffvhuff format, which is a codec specific to FFmpeg.

[delphonso]

Is there anyone on here with experience setting up VPNs that I could ask some questions?

[Schmaven]

Is there anyone on here with experience setting up VPNs that I could ask some questions?

I've set up over a dozen various kinds for work, always from the client side once the VPN has already been established by the site's IT department though.

Edit: Usually the IT department has an install guide of some kind.  There are some that require you to have a config file from them loaded into the VPN program.  Some have 3rd party revolving keys.  Some are just set up in the standard Windows network connections page.  I've only occasionally used the same type twice.  Only 1 I've I noticed blocks all internet connections on the client machine, most leave that open.

I suspect however that you're more curious about setting 1 up from scratch than connecting to an already existing VPN, in which case I have no experience there.

[methylatedspirit]

I'm trying to compile fortune-mod with MSYS2 on Windows 10, and it's not quite working as planned. CMake produces output, but the makefiles it generates produce this error:

Code: [Select]

[  1%] Linking C executable rot.exe
/C/msys64/usr/bin/../lib/gcc/x86_64-pc-msys/10.2.0/../../../../x86_64-pc-msys/bin/ld: cannot find -lintl
/C/msys64/usr/bin/../lib/gcc/x86_64-pc-msys/10.2.0/../../../../x86_64-pc-msys/bin/ld: cannot find -lregex
collect2: error: ld returned 1 exit status
make[2]: *** [CMakeFiles/rot.dir/build.make:101: rot.exe] Error 1
make[1]: *** [CMakeFiles/Makefile2:128: CMakeFiles/rot.dir/all] Error 2
make: *** [Makefile:156: all] Error 2

I've got rinutils compiled already. It's already linked, CMake is seeing it. (Now, I realize fortune isn't meant to run on Windows, but I'm trying to see if it even can.)

Am I missing a library? I'm using the CMake GUI, for reference.

[wierd]

missing lintl and lregex

The first is part of the gettext package.
The second is in libregex.

Use the package manager for msys to pull the missing dependencies.

[Starver]

So if someone had, say, securely compressed all my bitcoin wallets, what do you think I should try to do about it?

(j/k!)

Spoiler: ...also, good on you wierd, for answering meth (click to show/hide)

I was thinking I might be able to beg time to fiddle on a Win10 machine at the weekend (if nobody else had helped) but I totally forgot about it when I had the chance.

I know the (often linked) XScreenSaver is officially dis-supported for Windows by the author's request, but I know of no reason why the fortune-mod should not be trivially compilable upon that platform. At its core, I'm not even sure why it would ever need much more than the stdio library, actually, but I suppose it's easier[1] or a GUI-aware level of cruft. And while not exactly barest-of-bones ANSI C modules (well, Internationalisation has probably been revamped plenty of times since I first touched a C compiler, back when I only really worried about £s not looking like #s if they didn't have to) surely they're amongst the default-on optional extras in any decent compiler these days.

Been a while since I so directly used MSYS2/MingW(64?), though. Could have even been Minimal SYStem One back when I did. Too many alternate IDEs to flit between that do all the package-management near invisibly or do an 'install all' from the get go. And when I want to base myself or prototype something I retreat into good old (inefficient but lovable for it) mostly-cross-platform Perl scripting. 

...erm, anyway. This is all why I wanted a fresh install on the right platform to see what the problem might be. This was not actually supposed to such a long ramble about how I have no solution yet. Just to say that I (as well as wierd) was not ignoring the question, but I remain unable to help out directly. Or at all, at possibly. At least until my next opportunity to grab use of a Win10 machine.

Still, happy to further Bump the issue, or perhaps enrage others enough by my stupid comments that they make more valid contributions to counter them!


[1] If it weren't for multiline fortune-entries,  I could bash up a Batch File 'random line picker' for you. With a standard Perl port, I would have sufficient to regex multilines out, and probably wouldn't even need to activate advanced internationalisation options most of the time between ripping from source and shoving and shoving to console, though even then the codepageband multibyte conversions could do with some finessing. If I wanted to pop up things in a window, rather than console, then I'd add the Tkx package to the Perl header and be done with it. But I get the impression that 'getting fortunes displayed' is not the end result you seek, but just an exercise in getting something to work in a way you can understand.

[Naturegirl1999]

I wish there was a way to mod apps while on an iPad, maybe there’s an app that allows it? I wouldn’t need to be able to upload the results on the store, it would just be cool to see what code for various apps looks like

[wierd]

No, because virtualized code is FORBIDDEN on iDevices.

[Starver]

I wish there was a way to mod apps while on an iPad, maybe there’s an app that allows it? I wouldn’t need to be able to upload the results on the store, it would just be cool to see what code for various apps looks like

On the whole, I don't think you're going to get that ability for a number of reasons... Which is a complex list but probably boils down to: a) "(even if) free as in beer, not free as in speech", b) the need to sensibly decompile and/or bundle the source, c) [...]

[...yeah, ninjaed by an excellent single-point summary, I see. No need to go to into the detail I did.]

[Naturegirl1999]

No, because virtualized code is FORBIDDEN on iDevices.

why is it forbidden?

[LordBaal]

Thou shall not ask.

[wierd]

No, because virtualized code is FORBIDDEN on iDevices.

why is it forbidden?

Several reasons.

1) It would allow code not blessed by the Apple approval process to run freely on idevices.  This would be a circumvention of the apple store gatekeeper function, for the walled garden. Thus, forbidden.  (This is all in one stroke a 'security measure' (see below), AND an anticompetitive practice that reinforces the role of the gatekeeper, and its 30% cut of sales.)

2) Allowing that kind of code to function inside a sandbox, runs the risk of a vulnerability being exploited to escape the sandbox, and possibly be used to gain system access. This is a security risk for the device, and apple is real negative on that kind of thing. While outright forbidding it as a policy is like trying to outlaw global sea rise (as a 'solution' for climate change), and thus a form of security through obscurity-- the actual reduction in attack surface has real value in a security role. (Fewer people being able to get code to run in such a way, because of lack of tools, mean fewer exploits discovered, means 'safer' platform. The vulnerabilities are still there, but are harder to get access to. This is related to the reason apple is REALLY NEGATIVE on the practice of emulating an iDevice for security research. Such an emulator allows security researchers to look for, and experimentally hook those exploits in the virtual iDevice, without the obscuration of the apple store or the apple hardware. Apple is also big on being buddy-buddy with the copyright cartels, and uses the same tactics they do, so naturally the contested that their software was not LICENSED to be run in a virtual machine, and that this constituted an EULA violation. (And of course, they wont license it for use in that way, because, then people would find security vulns, and exploit them!))

3) Allowing virtualized code allows for content that violates apple's intellectual property rules-- Things like emulators for old consoles, where the rom images are of dubious, if not blatantly illegal nature, in terms of copyright, or platform licensing.  "Just Saying NO" means apple can have some legal protection against being an accessory to copyright infringement by knowingly permitting their platform to be used this way (because of how broken our legal system is in regard to this, and how greedy the copyright cartels are.)

The question really shouldn't be "Why does Apple forbid it?" and "How do open platforms manage to survive?"

[LordBaal]

In short it would be heresy for Apple, and they'll deploy lawyers in drop pods to your house, as it would hamper their effectiveness at milking you for every penny.