# Distorting PDFs to prevent character recognition **URL:** \ **Category:** Factual Questions\ **Created:** [March 17, 2013, 10:37pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206 "2013-03-17T22:37:28Z")\ **Posts on this page:** 20\ **Page:** 2
**Author:** ![kayaker](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/kayaker/32/441_2.png) [@kayaker](https://boards.straightdope.com/u/kayaker)\ **Post date:** [March 18, 2013, 11:34am UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/21 "2013-03-18T11:34:52Z")
[Maybe use one of these on the cover sheet or in the body of the paper?](https://www.google.com/search?q=fbi+piracy+warning&hl=en&client=safari&source=lnms&tbm=isch&sa=X&ei=8fpGUa2lDbGg4AO9k4CQCw&ved=0CAkQ_AUoAQ&biw=1024&bih=644) ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 1:44pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/22 "2013-03-18T13:44:20Z")
> [@Derleth](#): > > To be clear: **Neither of these things would work especially well against OCR. OCR does not work like a photocopier.** No, but OCR requires a scan, which is the same starting point as all copiers since about 1978, and the red-on-red or pale blue printing is resistant to clean scanning. (Tricks like filter sheets aside.) You’re talking about two different eras in imaging. Yes, today’s scanners can produce cleaned up text from almost any color combination. ---
**Author:** ![Terminus\_Est](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/terminus_est/32/3087_2.png) [@Terminus\_Est](https://boards.straightdope.com/u/Terminus_Est)\ **Post date:** [March 18, 2013, 2:02pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/23 "2013-03-18T14:02:40Z")
OP is talking about a PDF which is already electronic and hence is already a scan in effect. There’s no need to fiddle about with photocopier filters when the file is already there on your computer. ---
**Author:** ![GaryM](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/garym/32/241_2.png) [@GaryM](https://boards.straightdope.com/u/GaryM)\ **Post date:** [March 18, 2013, 2:14pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/24 "2013-03-18T14:14:05Z")
> [@scr4](#): > > But surely you could make a legible copy using a camera? Color film is designed to mimic the color response of human eyes, it should distinguish colors just as well. Possibly, but the idea was to make it relatively difficult to copy the floppy disk and copy the key sheet to pass along. I only remember this from one game, and now many years later, don’t remember which one. ☹ ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 2:27pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/25 "2013-03-18T14:27:50Z")
> [@Terminus\_Est](#): > > OP is talking about a PDF which is already electronic and hence is already a scan in effect. There’s no need to fiddle about with photocopier filters when the file is already there on your computer. Pee… Dee… Eff. Is that something new from them XeroX boys? A true PDF file is \*already \*characters and should not need any kind of optical recognition. What the OP is vaguely considering is making a blurry scan of a printed page and embedding that scan in a PDF file. The quaint historical discussion is about the various ways to put printed letters in front of someone but make them uncopyable using reasonable technology. Historically - back when we rode our dinosaurs down to the grocery store to put “coins” in an optical copier to duplicate things - there were some optical tricks that let us do that. In this day, when every mope has a digital camera and image manipulation software, it’s nearly impossible to protect printed material no matter what form, format or platform it uses. The protection is to subvert piracy at the start by not trying to sell printed material at a price point where a ripoff is a better deal than an original. ETA: Unless you can sell it under an enforceable NDA, signed by each purchaser. ---
**Author:** ![Terminus\_Est](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/terminus_est/32/3087_2.png) [@Terminus\_Est](https://boards.straightdope.com/u/Terminus_Est)\ **Post date:** [March 18, 2013, 3:01pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/26 "2013-03-18T15:01:10Z")
Do you need me to show you the onion in my belt, grandpa? My point is that there’s no need to resort to photocopier tricks when you can do it electronically. Equally, the end user can use electronic means to reverse or subvert whatever you put there in the first place. ---
**Author:** ![moriah](https://avatars.discourse-cdn.com/v4/letter/m/b2d939/32.png) [@moriah](https://boards.straightdope.com/u/moriah)\ **Post date:** [March 18, 2013, 5:11pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/27 "2013-03-18T17:11:45Z")
I can type about 50 wpm. And there are those who can easily do 70 wpm. If the document is still readable by a human being, how do you (the OP) intend to foil simple transcription that would be faster than any sort of digital re-enhancement and OCRing? Perhaps print black on black? ---
**Author:** ![njtt](https://avatars.discourse-cdn.com/v4/letter/n/ecd19e/32.png) [@njtt](https://boards.straightdope.com/u/njtt)\ **Post date:** [March 18, 2013, 5:25pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/28 "2013-03-18T17:25:03Z")
How about writing it out by hand, preferably in cursive, and preferably written by someone whose handwriting is not too tidy, and then scanning that into a PDF. I bet OCR would not do at all well with that, and people ought to be able to read it (though if the handwriting is poor, they might find it difficult and annoying still). May I point out to some of the others here that the OP shows that he is aware that there is not way to make copying impossible altogether. He only hopes to make it it a difficult, non-trivial task. ---
**Author:** ![Duckster](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/duckster/32/1244_2.png) [@Duckster](https://boards.straightdope.com/u/Duckster)\ **Post date:** [March 18, 2013, 5:41pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/29 "2013-03-18T17:41:44Z")
> [@RadicalPi](#): > > The reason I ask is that I have data that I want to present, but I don’t want anyone copying it. I know that that’s impossible, but I would at least like it to be difficult enough to copy that very few people would bother. That is, if they only way to copy the data would be to type it in manually, that would be good enough for me. What is it worth to you? If you plan to market this information the degree of difficulty for an honest user must be weighed against any possible loss by a dishonest user. If you set the bar too high, you honest users will just walk away. Your dishonest users will crack your system and redistribute it to others, probably for free (and deliberately so just to annoy you). Your reputation will suffer. What is that worth to you? **NitroPress’s** comment up thread about an appropriate “price point” should be heeded. Strike a balance between ease of use by honest users and not worth the effort by potential pirates. It doesn’t matter even if you give away the information, either. ---
**Author:** ![beowulff](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/beowulff/32/542_2.png) [@beowulff](https://boards.straightdope.com/u/beowulff)\ **Post date:** [March 18, 2013, 5:42pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/30 "2013-03-18T17:42:39Z")
Just have the guy who wrote the check in [this](http://boards.straightdope.com/sdmb/showthread.php?t=685269) thread transcribe your data. Apparently, even BofA can’t OCR his handwriting… ---
**Author:** ![ZipperJJ](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/zipperjj/32/211_2.png) [@ZipperJJ](https://boards.straightdope.com/u/ZipperJJ)\ **Post date:** [March 18, 2013, 6:04pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/31 "2013-03-18T18:04:42Z")
Are you looking to present this on the web, or sending people copies of PDFs? If it’s just the Web, you can use FlexPaper or any number of Web apps that display your PDF as a Flash file that doesn’t allow printing or downloading or copy/paste. ---
**Author:** ![drachillix](https://avatars.discourse-cdn.com/v4/letter/d/48db29/32.png) [@drachillix](https://boards.straightdope.com/u/drachillix)\ **Post date:** [March 18, 2013, 6:05pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/32 "2013-03-18T18:05:41Z")
> [@RadicalPi](#): > > The reason I ask is that I have data that I want to present, but I don’t want anyone copying it. I know that that’s impossible, but I would at least like it to be difficult enough to copy that very few people would bother. That is, if they only way to copy the data would be to type it in manually, that would be good enough for me. What is the format of the presentation? Are you giving folks handouts of some kind? Perhaps it would be helpful if we knew a little more of the details of the project. Maybe there is a different way to provide access to the info that is more conducive to your needs like subscription access to a web site built for the content… handouts have a simple sample but website has more detailed samples/methods ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 6:16pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/33 "2013-03-18T18:16:25Z")
> [@Terminus\_Est](#): > > My point is that there’s no need to resort to photocopier tricks when you can do it electronically. Equally, the end user can use electronic means to reverse or subvert whatever you put there in the first place. Obviously, but this discussion is apples and oranges and onions. If the text is in electronic character form, absolutely no amount of trickery with the formatting or colors will prevent someone from extracting the content, using freeware tools. If the content is images that are so distorted that OCR won’t work, humans will likely have trouble with them too. At which point we started discussing old optical tricks to prevent copying, which could work under limited conditions even in a PDF, if the content is images rather than character. In the end, though, even the most unbreakable imaging-based protection can be subverted by… simply copying the PDF file itself. ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 6:18pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/34 "2013-03-18T18:18:29Z")
> [@ZipperJJ](#): > > AIf it’s just the Web, you can use FlexPaper or any number of Web apps that display your PDF as a Flash file that doesn’t allow printing or downloading or copy/paste. There is absolutely no way to present info on the web, or any equivalent, and protect it against duplication. You can make it increasingly difficult at the expense of reader convenience and capability, which becomes the tradeoff. I agree, if the OP would share the specifics of what he’s trying to accomplish, a reasonable solution might present itself. ---
**Author:** ![RadicalPi](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/radicalpi/32/9806_2.png) [@RadicalPi](https://boards.straightdope.com/u/RadicalPi)\ **Post date:** [March 18, 2013, 8:40pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/35 "2013-03-18T20:40:20Z")
> [@beowulff](#): > > Just put a 25% - 40% screened background behind the text - that should hose any OCR software… This is an interesting idea. I will look into this. Thanks. > [@NitroPress](#): > > Your PDF documents should not cost more than they are worth to any new user. Price them accordingly, and you won’t see much piracy. There is no price. As I said before, I’m not selling anything. > [@Frumpy\_Jones](#): > > You can just disable OCR from your PDF when creating it… > > [http://forums.adobe.com/docs/DOC-3062](http://forums.adobe.com/docs/DOC-3062) > > However, that doesn’t stop someone from printing your pdf and scanning it, and OCR-ing the scan. This is another interesting idea. I appreciate it. > [@Frumpy\_Jones](#): > > If someone is determined enough, they’ll just type it out themselves. Your only choice in this matter, then, is to orally give your report, ban recording devices and then hit them with the flashy thing from Men in Black ehrn you finish… but then what is the point of the report in the first place if people are not allowed to remeber/use/regurgitate to knowledge delivered? Again, this would be going a bit too far. I want to just dissuade casual OCR. > [@t-bonham@scc.net](#): > > Yes, indeed. Given my age, and some health problems, it’s becoming harder for me to read many things. and I;m beginning to strongly resent webpages & other things that are hard to read. For better or for worse, my readers are in their late twenties. > [@Terminus\_Est](#): > > OP is talking about a PDF which is already electronic and hence is already a scan in effect. There’s no need to fiddle about with photocopier filters when the file is already there on your computer. This is correct. > [@moriah](#): > > I can type about 50 wpm. And there are those who can easily do 70 wpm. > > If the document is still readable by a human being, how do you (the OP) intend to foil simple transcription that would be faster than any sort of digital re-enhancement and OCRing? Again, I’m not trying to make copying impossible. I believe, however, that having to retype the whole thing would be too much of pain that most people wouldn’t bother. > [@NitroPress](#): > > If the content is images that are so distorted that OCR won’t work, humans will likely have trouble with them too. At which point we started discussing old optical tricks to prevent copying, which could work under limited conditions even in a PDF, if the content is images rather than character. > > In the end, though, even the most unbreakable imaging-based protection can be subverted by… simply copying the PDF file itself. Images don’t have to be all that distorted for OCR to not work. Ten levels of copies of copies is enough for the OCR that comes with Acrobat to garble about a third of them, which is more than enough to be useless. Copying the file itself is fine. > [@NitroPress](#): > > There is absolutely no way to present info on the web, or any equivalent, and protect it against duplication. You can make it increasingly difficult at the expense of reader convenience and capability, which becomes the tradeoff. I am aware of this. What I want is a level of protection equivalent to making copies of copies that doesn’t waste paper. ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 9:03pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/36 "2013-03-18T21:03:23Z")
Can you state exactly what your purpose and goals are? You’re giving it away free and they can copy the file all they want, but you don’t want them creating an editable text file of it. I’m having trouble coming up with a scenario where that’s necessary. Fill us in and we might have better suggestions for you. (I work daily with advanced publishing techniques and technology, and I’m not the only one here.) ---
**Author:** ![dstarfire](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/dstarfire/32/5762_2.png) [@dstarfire](https://boards.straightdope.com/u/dstarfire)\ **Post date:** [March 18, 2013, 9:37pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/37 "2013-03-18T21:37:20Z")
Check out this thread over on [stackoverflow.com](http://stackoverflow.com) . It talks about OCR-resistant fonts. [Is there a font that can't be recognised by an OCR? - Stack Overflow](http://stackoverflow.com/questions/13359491/is-there-a-font-that-cant-be-recognised-by-an-ocr) ---
**Author:** ![Amateur\_Barbarian](https://avatars.discourse-cdn.com/v4/letter/a/59ef9b/32.png) [@Amateur\_Barbarian](https://boards.straightdope.com/u/Amateur_Barbarian)\ **Post date:** [March 18, 2013, 9:40pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/38 "2013-03-18T21:40:06Z")
OCR can be trained, if necessary. If the fonts are consistently presented, even Jokerman or Giddyup or Ouch can be OCRed with high accuracy. Taking something like Kreepy and blurring it might be successful… but it would be successful in being unreadable by hoo-mans, too. ---
**Author:** ![moriah](https://avatars.discourse-cdn.com/v4/letter/m/b2d939/32.png) [@moriah](https://boards.straightdope.com/u/moriah)\ **Post date:** [March 18, 2013, 9:45pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/39 "2013-03-18T21:45:36Z")
> [@RadicalPi](#): > > Again, I’m not trying to make copying impossible. I believe, however, that having to retype the whole thing would be too much of pain that most people wouldn’t bother. Ah, but it only has to be done once and the result ‘published’ somewhere. If what you’re doing has any sort of popularity to it and enough people grouse over not having an editable version, they’ll transcribe it once and redistribute freely. ---
**Author:** ![Blakeyrat](https://avatars.discourse-cdn.com/v4/letter/b/ecd19e/32.png) [@Blakeyrat](https://boards.straightdope.com/u/Blakeyrat)\ **Post date:** [March 18, 2013, 10:07pm UTC](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206/40 "2013-03-18T22:07:25Z")
> [@moriah](#): > > I can type about 50 wpm. And there are those who can easily do 70 wpm. > > If the document is still readable by a human being, how do you (the OP) intend to foil simple transcription that would be faster than any sort of digital re-enhancement and OCRing? > > Perhaps print black on black? They don’t even need to OCR it. (Or retype.) Just scan it and distribute the scanned image files. It’s not like disks are tiny or internet connections are slow. [Previous page](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206.md?page=1) [Next page](https://boards.straightdope.com/t/distorting-pdfs-to-prevent-character-recognition/653206.md?page=3)