How to handle the takedown

[Cortys]

Hi everyone!

To keep things running. What are your ideas to keep this project available and continue the development?

[ideologysec]

DigitalOcean has VPS' for ~$5/mo, or we could set up a private repo somewhere like Beanstalk?

[The-Hoff]

Count me in wherever you go.

[tomchiverton]

Ho ho ho. So they've realised DRM can not possibly work and resorted to lawyers eh ? IDK if it's possible to contest the DMCA notice. It should be, but I've not heard anything from GitHub about my fork going down so I've emailed support to ask.

The problem of securing legal access to purchased products has already been well trodden. E.g. ebooks

In the mean time, yes, we need a private commit group with a non-public Git repo. Given the likely lack of usage I could host this but although I'm outside of the DMCAs reach, I'm covered by the EUCD so this may not be a very good solution. Amazon will let you run an instance free for a year iirc. I can set up a git repo. there with shared admin rights. Not sure if Amazon require giving a real address and credit card.

Secondly is distribution of the resulting .crx and it's updates. We can stick this on a random S3 share ? What seems to work well for ebook's is a public blog on wordpress e.g. https://apprenticealf.wordpress.com/ which distributes files via generic 'u send it' services. I think we could do that ?

[marvi1]

Did the lawyers contact you directly? On Apr 3, 2014 3:09 PM, "tomchiverton" notifications@github.com wrote:

Ho ho ho. So they've realised DRM can not possibly work and resorted to lawyers eh ? IDK if it's possible to contest the DMCA notice. It should be, but I've not heard anything from GitHub about my fork going down so I've emailed support to ask.

The problem of securing legal access to purchased products has already been well trodden. E.g.

In the mean time, yes, we need a private commit group with a non-public Git repo. Given the likely lack of usage I could host this but although I'm outside of the DMCAs reach, I'm covered by the EUCD so this may not be a very good solution. Amazon will let you run an instance free for a year iirc. I can set up a git repo. there with shared admin rights. Not sure if Amazon require giving a real address and credit card.

Secondly is distribution of the resulting .crx and it's updates. We can stick this on a random S3 share ? What seems to work well for ebook's is a public blog on wordpress e.g. https://apprenticealf.wordpress.com/ which distributes files via generic 'u send it' services. I think we could do that ?

Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-39492095 .

[tomchiverton]

No. I've heard nothing direct from GitHub or Cmxlgy (or their lawyers). First I knew was the invite here, when I checked my fork as well as the others were off, with a link to that.

[gu1dry]

On bitbucket.org, you can have up to 5 private repos for free.

[tomchiverton]

BitBucket seems a plan if there are less than five people who need commit access.

Needless to say, hopefully, you'd never want to connect to the public blog or upload services linked from without using something like Tor of course...

[roadworrier]

ehm... speaking of distribution... I never did acquire the latest............

[tjw01]

Another example of the DMCA system being abused. Correct me if I'm wrong, but at no point was any copyrighted code stolen and used in the Chrome extension, correct? In which case, since no Copyright infringement has been committed in the extension itself (and it's also debatable whether its use constitutes Copyright infringement either) then Comixology have no legal grounds to pull it down.

The funniest thing of all is you don't even need the extension to back comics up. You can just take screenshots instead if you really want, and there's no way they can block that. The extension just makes the job a lot easier.

[tomchiverton]

Which is basically what I told GitHub-give us a chance to come into compliance. Nothing over night though, and they have no interest in fighting for us.

I also think that future releases should include features to discourage pirates, like the Wii open source folks or Amazon MP3 do. This could be as easy as additional meta data and a small watermark on one page with the users name on. Cmxlgy obviously don't do this at their end for some reason.

Tom, sent from my HTC

----- Reply message ----- From: "tjw01" notifications@github.com To: "Cortys/cmxlgyTalkingThread" cmxlgyTalkingThread@noreply.github.com Cc: "tomchiverton" tom+github.com@falkensweb.com Subject: [cmxlgyTalkingThread] How to handle the takedown (#1) Date: Fri, Apr 4, 2014 01:24 Another example of the DMCA system being abused. Correct me if I'm wrong, but at no point was any copyrighted code stolen and used in the Chrome extension, correct? In which case, since no Copyright infringement has been committed in the extension itself (and it's also debatable whether its use constitutes Copyright infringement either) then Comixology have no legal grounds to pull it down.

The funniest thing of all is you don't even need the extension to back comics up. You can just take screenshots instead if you really want, and there's no way they can block that. The extension just makes the job a lot easier.

— Reply to this email directly or view it on GitHub.

[sidick]

I'd have no problem with login name being embedded in each image metadata, would make sense. I still think they'd come after the project but it would be easier to defend.

On 4 Apr 2014, at 07:08, tomchiverton notifications@github.com wrote:

Which is basically what I told GitHub-give us a chance to come into compliance. Nothing over night though, and they have no interest in fighting for us.

I also think that future releases should include features to discourage pirates, like the Wii open source folks or Amazon MP3 do. This could be as easy as additional meta data and a small watermark on one page with the users name on. Cmxlgy obviously don't do this at their end for some reason.

Tom, sent from my HTC

----- Reply message ----- From: "tjw01" notifications@github.com To: "Cortys/cmxlgyTalkingThread" cmxlgyTalkingThread@noreply.github.com Cc: "tomchiverton" tom+github.com@falkensweb.com Subject: [cmxlgyTalkingThread] How to handle the takedown (#1) Date: Fri, Apr 4, 2014 01:24 Another example of the DMCA system being abused. Correct me if I'm wrong, but at no point was any copyrighted code stolen and used in the Chrome extension, correct? In which case, since no Copyright infringement has been committed in the extension itself (and it's also debatable whether its use constitutes Copyright infringement either) then Comixology have no legal grounds to pull it down.

The funniest thing of all is you don't even need the extension to back comics up. You can just take screenshots instead if you really want, and there's no way they can block that. The extension just makes the job a lot easier.

— Reply to this email directly or view it on GitHub. — Reply to this email directly or view it on GitHub.

[stullz]

I'm fine with an unobtrusive type of watermark as well, and no matter what we do they will still continue an attempt to enforce their legal 'rights'. Whether we agree with the DMCA legislation or not it is currently law. As far as stolen code or anything like that, I believe their notice had mentioned that portions of their code or certain files had, in fact, been posted and available... so, if that is the case, I doubt there is much of a leg to stand on in terms of fighting the notice. Which is why I want to encourage everyone here (most of you probably already do) to join groups that attempt to stand up to flawed legislation like the DMCA or even the NSA spying scandals. (Demand Progress, watchdog.net are 2 good ones)

[ideologysec]

Whoops. Wrong button there - posting on mobile is hard.

Watermarking might help, but it seems a little silly. Anyone with the chops will just remove the bits in the extension that do the marking. Additionally, if we're backing up comics with the idea that at some point comixology might go down and we'd want to read what we paid for, watermarking would only serve to annoy us at that point.

Comixology seems to have made it clear that they're afraid or against the extension for whatever reason. I don't think that appeasing them so that the project can go back on github is the answer. Now we can focus on the technical aspects. Setting up a repo, and planning public distribution.

I vote a private bitbucket or beanstalk repo, and a tumblr with links to the extension hosted on Mega or some other non-US fileshare site.

[dmgoncalves]

I think the single watermark with metadata on each image is a good idea.

Generic watermark text suggestion:

"For personal backup only. Do not distribute. Your User ID is recorded in the image metadata."

My $0.02.

Thanks.

On Fri, Apr 4, 2014 at 2:35 AM, sidick notifications@github.com wrote:

I'd have no problem with login name being embedded in each image metadata, would make sense. I still think they'd come after the project but it would be easier to defend.

On 4 Apr 2014, at 07:08, tomchiverton notifications@github.com wrote:

Which is basically what I told GitHub-give us a chance to come into compliance. Nothing over night though, and they have no interest in fighting for us.

I also think that future releases should include features to discourage pirates, like the Wii open source folks or Amazon MP3 do. This could be as easy as additional meta data and a small watermark on one page with the users name on. Cmxlgy obviously don't do this at their end for some reason.

Tom, sent from my HTC

----- Reply message ----- From: "tjw01" notifications@github.com To: "Cortys/cmxlgyTalkingThread" cmxlgyTalkingThread@noreply.github.com

Cc: "tomchiverton" tom+github.com@falkensweb.com Subject: [cmxlgyTalkingThread] How to handle the takedown (#1) Date: Fri, Apr 4, 2014 01:24 Another example of the DMCA system being abused. Correct me if I'm wrong, but at no point was any copyrighted code stolen and used in the Chrome extension, correct? In which case, since no Copyright infringement has been committed in the extension itself (and it's also debatable whether its use constitutes Copyright infringement either) then Comixology have no legal grounds to pull it down.

The funniest thing of all is you don't even need the extension to back comics up. You can just take screenshots instead if you really want, and there's no way they can block that. The extension just makes the job a lot easier.

—

Reply to this email directly or view it on GitHub.

Reply to this email directly or view it on GitHub.

Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-39535705 .

[tomchiverton]

"Watermarking might help, but it seems a little silly." Yes, it does. Obviously it's easy to correct by unpacking the CBZ and digging out photoshop or what ever. That's not the point :-) I don't see how a small block of pixels (~ 8px X 6px X character count) is going to spoil anything, and it makes it very clear where the project stands on piracy Vs. backup.

[sidick]

I'd favour watermarking at the JPEG metadata level more than the graphics level as that's less in your face but easy to extract later for anyone who wants to.

On 4 Apr 2014, at 19:30, tomchiverton notifications@github.com wrote:

"Watermarking might help, but it seems a little silly." Yes, it does. Obviously it's easy to correct by unpacking the CBZ and digging out photoshop or what ever. That's not the point :-) I don't see how a small block of pixels (~ 8px X 6px X character count) is going to spoil anything, and it makes it very clear where the project stands on piracy Vs. backup.

— Reply to this email directly or view it on GitHub.

[ideologysec]

Well, to make an analogy, watermarking the image of each comic book page would be like inserting a one-second snippet of audio "Downloaded from Amazon by John Smith" into each song of the album that was deDRM'd. Watermarking a comic book affects the primary purpose of an image - viewing - the way that inserting audio into a song affects listening. ID3 tags on a song are there for anyone with eyes to look but don't affect listening.

Let me share my personal perspective - I don't like comixology's readers, especially on tablets. I prefer to use a third party app, and for that I have to use cbz/r files. I happily pay comixology or google books or whoever has what I need, and then want to convert their files into one I can use the way I want.

As many who fight against DRM argue, they are my files. I have no intent to distribute. I just want to use them how I want to use them.

I will happily accept watermarking, but ideally only on one page.

[ideologysec]

...or in the JPEG metadata, as suggested.

[Cortys]

I also think watermarking is a good thing. Hopefully that will calm the Comixology folks down a bit. Or they might take tumblr down... ;-P

How exactly should we do the watermarks? As @sidick said the metadata level is the way I'd prefer but I do not know any js-lib that can do that.

[The-Hoff]

I am against watermarking on the actual pages, but if that's the only way forward...... It is just that the comics I am collecting are almost complete and to have a watermark on the remaining 20% of them would bug the hell out of me and my ocd would kick in :)

[gu1dry]

I vote metadata watermark also opposed to a graphical watermark, if possible.

@The-Hoff if we're able to figure out a way to watermark via a metadata level, you won't see the watermark, unless you view the metadata. An example of metadata is the ID3 tags on mp3 (the artist, song title, etc).

[tomchiverton]

I was thinking of putting ID data in the CBZ, not the images, like you say I don't know of a JavaScript label to add that to JPEGs, where as drawing some text onto the bitmap data is easy enough.

[sidick]

Yeah I've found no libraries that can do it either after a quick google so it's not quite as easy as I thought :(

On 4 Apr 2014, at 20:08, tomchiverton notifications@github.com wrote:

I was thinking of putting ID data in the CBZ, not the images, like you say I don't know of a JavaScript label to add that to JPEGs, where as drawing some text onto the bitmap data is easy enough.

— Reply to this email directly or view it on GitHub.

[The-Hoff]

The current extension still works for me though. Is it just a matter of time till it gets blocked?

[marvi1]

Doesn't seem to be downloading pages for me. Gets stuck at 0%

On Fri, Apr 4, 2014 at 3:34 PM, The-Hoff notifications@github.com wrote:

The current extension still works for me though. Is it just a matter of time till it gets blocked?

Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-39602890 .

Jon Marvin

[Cortys]

Yup. The extension has been blocked again. They renamed "thumbnails" to "thumbnailz". Very creative.

I'm currently working on a more general solution to stop this cat and mouse game.

[The-Hoff]

Damn, you are right. Same here, stuck at 0%. It was working earlier today though.

[sperglord-enterprises]

Bummer.

[sperglord-enterprises]

My first idea would be to go to darkweb with this, but I think there is no cute collaboration system like github on torspace.

Also seriously guys, thanks for developing my original extension when I stopped having time to do it.

[tomchiverton]

My first idea would be to go to darkweb with this, but I think there is no cute collaboration system like github on torspace.

Did you try emailing them to appeal ? I've just received an email back about my fork, we may be able to get it resurrected.

[Cortys]

Hi guys. Just to let you know: I was busy last week so there is no fix yet.

I am currently working on one though. I added something like a "setup wizard" you can run every time cmxlgy changes their reader and it will try to adapt to the new DOM. It probably won't be perfect but definitely flexible enough to handle some class, id and also minor DOM tree changes.

[roadworrier]

Cortys, when you finish the new version, what's a good way for me to ask for its location? (I am following you on Twitter, so you can private message me there if you prefer non-public communication.)

[The-Hoff]

Nice Corty :)

[Cortys]

I've just finished my fix with the "smart" setup wizard, that scans the reader DOM. The scan is not fully featured yet so cmxlgy could still break the extension but it will take them more effort this time.

For the current scanner I wrote some DOM scanning methods we could reuse for a more flexible scanner. Even if we should find everything broken again in a few days, at least there is a solid background structure now to handle such things.

How should I share the new version with you?

[ideologysec]

MediaFire link?

[Cortys]

@Aktariel well. ok. http://bit.ly/1jJ1R5Z we should still find something more collaborative.

[sidick]

That's working well for me so far, congrats :)

On 16 April 2014 00:58, Clemens notifications@github.com wrote:

@Aktariel https://github.com/Aktariel well. ok. http://bit.ly/1jJ1R5Zwe should still find something more collaborative.

— Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-40548915 .

[The-Hoff]

Thanks Corty, I will download it this afternoon. On 16 Apr 2014 09:52, "sidick" notifications@github.com wrote:

That's working well for me so far, congrats :)

On 16 April 2014 00:58, Clemens notifications@github.com wrote:

@Aktariel https://github.com/Aktariel well. ok. http://bit.ly/1jJ1R5Zwe should still find something more collaborative.

— Reply to this email directly or view it on GitHub< https://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-40548915>

.

— Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-40576869 .

[Cortys]

I just created an update URL: http://cmxlgy.webng.com/data

You can add it in the extension options. It is pretty slow but as long as we have not solved the legal stuff I'd prefer not to host it on one of my private servers.

[ideologysec]

Cortys,

I appreciate all your work. It's working for me on this end, except it's still chopping up the pictures. Very strange. I think it has to do with the fact that I'm using a retina MacBook Pro - it seems to work fine on a normal (1x computer). Shame, though - I wonder if the images served to retina computers are higher quality.

As far as updates go, I'm not aware of anything collaborative and DMCA-ignorant, but I am still actively looking.

[The-Hoff]

Windows 8.1 here and all working perfectly. Thanks so much Corty, your work is also really appreciated.

[Cortys]

Just uploaded an update: http://bit.ly/1jJ1R5Z

@Aktariel It contains a fix for the retina display problem (which was awful to debug without a retina device).

And I finally added the "watermark" thing we discussed. Currently there are no JS libraries that can edit IPTC metadata, so I had to add the username differently. The username is inserted at two places:

• An invisible .meta.asc is added to each zip/cbz archive containing the username.
• The ASCII values of the username are added to each file as binary pixels. The last 8*n (n length of the username) are modified in their HSL lightness to represent the zeros and ones. This watermark is not invisible but if you do not stare to the bottom right corner of the page all the time that should be fine.

The new "originCheck.html" will print out the username for a given page so cmxlgy can check the origin of pirated comics fairly easy.

[sidick]

Nice, look forward to trying this out.

On 19 Apr 2014, at 12:31, Clemens notifications@github.com wrote:

Just uploaded an update: http://bit.ly/1jJ1R5Z

@Aktariel It contains a fix for the retina display problem (which was awful to debug without a retina device).

And I finally added the "watermark" thing we discussed. Currently there are no JS libraries that can edit IPTC metadata, so I had to add the username differently. The username is inserted at two places:

An invisible .meta.asc is added to each zip/cbz archive containing the username. The ASCII values of the username are added to each file as binary pixels. The last 8*n (n length of the username) are modified in their HSL lightness to represent the zeros and ones. This watermark is not invisible but if you do not stare to the bottom right corner of the page all the time that should be fine. The new "originCheck.html" will print out the username for a given page so cmxlgy can check the origin of pirated comics fairly easy.

— Reply to this email directly or view it on GitHub.

[tomchiverton]

That's some awesome work. Works well (Chromium/Linux)

[Cortys]

Two questions:

1. As @tomchiverton said before, we might be able to reupload the source code to GitHub if the comixology source code is removed and we stop posting screenshots of comic pages. Do you think this is worth trying?
2. I'm currently struggling with a good way of distribution. Posting about the extension on Reddit was not successful because the subreddits about comics do not want to have anything to do with piracy and DRM removal. How did you notice this extension and how/where could we address more potential users?

I'm looking forward to your ideas. I also created a wordpress blog as suggested before but I'm not sure if this will get any attention.

[sidick]

OK, I've subscribed to that blog. I found this after a lot of googling after I heard about that bash script which apparently worked ages back :)

On 19 April 2014 23:57, Clemens notifications@github.com wrote:

Two questions:

1.

As @tomchiverton https://github.com/tomchiverton said before, we might be able to reupload the source code to GitHub if the comixology source code is removed and we stop posting screenshots of comic pages. Do you think this is worth trying? 2.

I'm currently struggling with a good way of distribution. Posting about the extension on Reddit was not successful because the subreddits about comics do not want to have anything to do with piracy and DRM removal. How did you notice this extension and how/where could we access more potential users?

I'm looking forward to your ideas. I also created a wordpress bloghttp://cmxlgy.wordpress.com/as suggested before but I'm not sure if this will get any attention.

— Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-40882913 .

[tomchiverton]

Yeah, I found the original github project via Google too. So getting the extension anywhere where Google can see it is good. The working on the the blog is fine to me - it gives our stance clearly and makes it very clear you'd be stupid to pirate your comic using it.

[dmgoncalves]

This version solved my chopped images problem on my MacBook Pro. Clever watermark solution. Thank you!

On Sat, Apr 19, 2014 at 7:31 AM, Clemens notifications@github.com wrote:

Just uploaded an update: http://bit.ly/1jJ1R5Z

@Aktariel https://github.com/Aktariel It contains a fix for the retina display problem (which was awful to debug without a retina device).

And I finally added the "watermark" thing we discussed. Currently there are no JS libraries that can edit IPTC metadata, so I had to add the username differently. The username is inserted at two places:

• An invisible .meta.asc is added to each zip/cbz archive containing the username.
• The ASCII values of the username are added to each file as binary pixels. The last 8*n (n length of the username) are modified in their HSL lightness to represent the zeros and ones. This watermark is not invisible but if you do not stare to the bottom right corner of the page all the time that should be fine.

The new "originCheck.html" will print out the username for a given page so cmxlgy can check the origin of pirated comics fairly easy.

— Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-40867372 .

[sidick]

I think we'll need to add matching for comixology.co.uk now too as they've just opened their UK store which is now the only one I can apparently use, example url is: https://www.comixology.co.uk/comic-reader/7269/90450

Thanks in advance :)

On 22 April 2014 05:26, robotlordofjapan notifications@github.com wrote:

Hi. So it has been a while since I last posted, and I still seem to be having the same issue with the extension. I installed the unpacked extension, but I do not see how to use it. The extensions shows to be enabled, but I do not see the yellow bar going across asking me to backup the comic when I am in the reader. This is making me feel dumb.

— Reply to this email directly or view it on GitHubhttps://github.com/Cortys/cmxlgyTalkingThread/issues/1#issuecomment-41002831 .

[Cortys]

@sidick Fixed. http://bit.ly/PsZJ73