how to scan magazines?

[DiTBho]

Today a friend gave me 80 interesting magazines about Linux and C. They talk about php, ruby, networking, and kernel between 2.2 and 2.6, so I think they cover the tecnology between 1999 and 2008; a bit old and for sure outdate, but articles are very very interesting. Unfortunately it's a lot of paper and it takes a lot of space for my small flat, so I am really thinking about scanning them to make a digital copy so I could then give them away in their physical form on eBay or something.

But how to scan them in a short time? I don't even own a scanner, but it would be a good occasion to buy one.

[DiTBho]

For the digital format, I am thinking about a common pdf, but I am open to other formats.

[ataradov]

There are special scanners for books that are essentially cameras on a stand with a software that does cleanup/conversion/binding.

Using a traditional flatbed scanner would be a nightmare and take forever, so I would research those book scanners first.

Searching "Book Scanner" on Amazon shows quite a few options. They are not cheap, but unless your time is worth nothing, they are totally worth it if you plan to scan a significant amount of magazines.

[CJay]

If you're going to dispose of them by recycling then you could cut the spines and scan in a duplex scanner, many 'pro' all in one printer/scanner combos do duplex if you dig in the settings.

If not, then I've seen simple folded acrylic stands that holds the magazine/book like a tent but eliminates the sag so the 'correction' isn't difficult and there are apps which will do it quite well automatically.

[golden_labels]

DiTBho:
I do not care, but a warning must be given so you are aware of that: making a copy and selling the originals will likely be a copyright infringement. So would be publishing the copies.

From a person, who was archiving photos and some books, not electronics magazines with detailed schematics:

Don’t bother with a flatbed scanner unless you are going for physical 600dpi or higher and then making a very nice, aligned, high-quality package of that. Average photo camera nowadays with the ability to fix exposure and focus is likely to give you as good quality, in particular if you can evenly light the page. It will be faster too, if you can place it on a stable support and trigger from distance (aka bulb mode). The disadvantages: lens distortion and the need to build or obtain a fixture. The output will not look professional, but good enough for your own uses if you just want to just simply check something in the future.

If going for a scanner, check if there is no business around you with an offer to scan books. If it’s not too expensive to scan that amount of material, ask about a sample page and if the quality is good enough — use their services. 80 magazines will take ages and you may let other people do it. Be sure to scan some schematic with tiny letters and some photo to check if they are reproduced well. Sometimes they may also produce PDFs.

In either case skim over the copy before getting rid of the originals.

To give some example: see attachments. The upper part of “bookscan-sample.jpeg” is a scaled down version to see the overall image⁽¹⁾; the bottom two are of the actual size. The book was photograhed literally on my knee at about 10s/photo, two pages at once, in poor light conditions, at ISO-1600, with holding the camera in one shaky hand. The “me-0002.png” is a post-processed photo of a A4 page from a 1908 magazine. If the camera can zoom in at small distances, selected fragments can be copied at 1200dpi  — you literally see halftone patterns.

____
⁽¹⁾ Intentionally unreadable, as it’s not my purpose to give access to the content of the book.

[CJay]

DiTBho:
I do not care, but a warning must be given so you are aware of that: making a copy and selling the originals will likely be a copyright infringement. So would be publishing the copies.
In either case skim over the copy before getting rid of the originals.

Good points and I'd also add, check Archive.org to see if the job has already been done.

[bobcat2000]

Before you go nut scanning all the pages, have you tried to see if they have what you want in archive.org?

[DiTBho]

DiTBho:
I do not care, but a warning must be given so you are aware of that: making a copy and selling the originals will likely be a copyright infringement. So would be publishing the copies.

Umm, even if they are between 21 and 12 years old? 2020-1999=21 years, 2020-2008=12 years.
(the first magazine was published in 1999, the last magazine was published in 2008)

Dunno

[DiTBho]

Before you go nut scanning all the pages, have you tried to see if they have what you want in archive.org?

Yup, already looked there. Thanks 

[ataradov]

Yea, technically even if they were published in the 70s.

Obviously the further back you go, the less likely anyone would care to pursue this legally. Unless you cross someone the wrong way.

But even without destruction simple distribution is already problematic. And making archival copies for yourself is technically legal (or at least a bit of a gray area).

[DiTBho]

Searching "Book Scanner" on Amazon shows quite a few options. They are not cheap, but unless your time is worth nothing, they are totally worth it if you plan to scan a significant amount of magazines.

I am playing only to make a digital copy of these magazines and only because I don't really have the space to keep the paper. There are a lot of articles, some are about Snort, Ruby, Ruby on Rails, and Php, and these are what I am going to learn. But articles are split in many episodes, so you need to browse many magazine for a complete lecture.

I was thinking about selling the physical copies only for two reasons: to pay a part of the cost for the "book scanner", and because I think that destroying the paper would be a shame.

My shelves are already full of books.



Yup, most of them are about "how to make profit", but it's not the case about these magazines, and for now I will use the garage to keep the paper.



Thanks guys! 

[retiredfeline]

There are a lot of articles, some are about Snort, Ruby, Ruby on Rails, and Php, and these are what I am going to learn. But articles are split in many episodes, so you need to browse many magazine for a complete lecture.

Just be aware that software has moved on a lot so do some current reading also to pick up on all the changed and new features, particularly in RoR.

[ataradov]

Yes, I was going to say. This does not feel vintage enough to be in archives, but it is not recent enough to be of any practical value. They need to stew for ~20 more years.

So scan them, keep the PDFs, sell the originals to offset the cost. Many years later publish the PDFs. By that time they will get good enough vintage.

[sleemanj]

A couple years ago my dad asked me to scan his archive of his Phonograph Society's in-house magazine going back to the 60s, it took some doing but I got the job done with an old brother multifunction with an ADF, some BASH scripting, and the following tools

  * scanimage to scan the pages into ordered images
  * imagemagick to convert image formats and split pages (I didn't cut page pairs physically)
  * unpaper to straighten the page images and that sort of thing
  * tesseract to OCR the page images into hocr format
  * hocr-pdf to combine the hocr format into a pdf with selectable text
  * pdftotext to extract text out of the pdf

it worked pretty well but obviously took a day or two of work to get everything working together, and then quite a while sitting there processing them :-)

[cdev]

If you can use two cameras and use a 90 degree stand that lets the magazine or book be open and supported in a manner that the two cameras can alternate taking the pictures, its much faster. Its not necessary for both cameras to be identical , just decent quality and high enough resolution, youcan use any image processing with batch mode to clean them up afterward, the key innovation is just using two cameras not one and letting them rest against the 90 degree angle vastly improves the quality of the image simply because both pages lie flat. I would use greyscale or color but be aware that the color images are larger and slower to process. Once you have the exposure right you can just tear through them pretty quickly. A foot pedal speeds scanning up a lot by letting you just move the pages and press the pedal then turn again.

[David Hess]

If you have access to a scanner with an automatic document feeder, then you can remove the binding and feed the stacks of paper through in minutes.

[Haenk]

I am playing only to make a digital copy of these magazines and only because I don't really have the space to keep the paper. There are a lot of articles, some are about Snort, Ruby, Ruby on Rails, and Php, and these are what I am going to learn. But articles are split in many episodes, so you need to browse many magazine for a complete lecture.

a) 10 year old articles in these fast moving languages are almost always outdated. Unfortunately, there is no point in keeping these articles, nor books from that time.
b) Check with HumbleBundle. They often have nice (programming) book bundles for cheap. All supplied als PDF. I'm sure there are other book bundles as well.

(I have scanned a couple of 400-page phonebooks in the past. That's no fun, believe me...)

[RoGeorge]

First of all, search for the name and numbers of those magazines, most probably somebody else already scanned and uploaded them somewhere, maybe they are to be found on the wayback machine, maybe on some publisher's archive or some random enthusiast's blog.

If they were not scanned already, the simplest way to do is take photos instead of scanning them.  A snapshot is much faster than a scan, and current cameras (even the webcams) have enough resolution to catch all the details.

[Doctorandus_P]

If you want some magazines about linux thingies you can also start here:
https://magazine.odroid.com/

[golden_labels]

a) 10 year old articles in these fast moving languages are almost always outdated. Unfortunately, there is no point in keeping these articles, nor books from that time.

Archiving for historical reasons. While learning programming from a decade old book is not only useless, but actively hamful, such copies are a tresure to anyone wishing to know how things looked in the past.

As a curiosity: that page from a early 1900s page magazine I posted earlier is not very outdated after 110 years. It explains how you can connect zinc-carbon batteries in parallel and series to achieve higher voltage or to get longer operation time from a lightbulb. Though the language used is at least very interesting. Let me just tell you that “current leaks through a wire”.

[Doctorandus_P]

Hasn't nobody told you to use guppa to stop those current leaks?

[westfw]

How do you get the archive.org style scans, where they have what looks like the original scans,  AND they’ve done some sort of buried ocr, so that you can actually do text searches as well?

[sleemanj]

How do you get the archive.org style scans, where they have what looks like the original scans,  AND they’ve done some sort of buried ocr, so that you can actually do text searches as well?

See my post further up in the thread.  In short, scan and adjust to make better (straighten, de-curl crop, whatever), run tesseract over the images to do OCR and produce an hocr file, then hocr-pdf to convert that into a selectable (and more importantly, searchable) PDF.

Works surprisingly well.



Example: http://www.phonograph.org.nz/magazine/1975-Vol-11-Issue-1-Oct/1975-Vol-11-Issue-1-Oct.pdf

[wilfred]

A few years ago I bought an Epson workforce document scanner DS-520. I did about 300 magazines in a couple of weeks. Pretty full on though, loading the hopper from breakfast until late in the evening as I went about other things. A full hopper would do a magazine in under 20 minutes if I turned the OCR, straighten and various other post scan options. I think less than 8 minutes if only a plain non searchable PDF was the output.

What impressed me was the reliability of the paper feed, not a single page went through stuck to another page. It detected and stopped on every one. Usually it was because of a paper defect catching another page eg folded over corners or I didn't separate the pages properly. Which does mean I cut every magazine down the spine in a guillotine.

I found it easier to just restart the entire magazine scan rather than continue. Just because I would not feel the need to go back and verify I reloaded pages properly if I just restarted. Bear in mind I only used the thing for a few weeks for this project and learned only enough to get the job done.

The one issue I had was because of the old paper (1974 - 1994) there was a lot of paper dust as a result of the splitting of the stapled spine and I did a lot of cleaning of the scanner sensor strip (it is not a flatbed) to remove accumulated dust which would manifest itself as a strip on the scanned page. I used a paintbrush. Another issue was possibly a result of the ink used and the paper feed wheels would make a feint pair of stripes most visible on images with large areas of solid colour. ie. the ink in photos would come off and leave a slightly lighter area where it contacted the page. I tried cleaning the rubber tires with a damp soapy rag but it didn't fully clear it up. It wasn't noticible on most pages and was not objectionable.  It didn't happen with newly printed pages.

I'd recommend a document scanner. They must have some clever sensor to detect stuck pages because I was impressed.

[JohanH]

Check with your local library if they allow scanning larger quantities. At least over here, you can use scanners in the library for free. Some have book scanners that automatically scans a book. I don't know if they can also do magazines. Of course there are also companies that offer these services.