There are a lot of magazines in that database, some are quite obscure and hard to find. Some have been defunct for decades. I would love a full text search but it would end up being very limited in scope I fear. Keep in mind it would take more then just scanning.. they would have to be scannned then OMR applied to extract the text. This is not simple and its not cheap to do in quantity. My business is heavily concerned with extraction from image and its quite a process when we control the scannable.. never mind magazines of dubious origin.

While it only applies to MRR, my library has a full text search of major magazines. You can also download PDFs of articles for newer (say the last decade, give or take) issues. Its very handy.. but it just does not have the breadth of the model railroad index. Therein lies the rub.

I suggested to them that as an interim solution they create a static html page of each magazine TOC in the database. They already have that working in a way.. they could have an intern do it. That can be searched directly using full text engine (I did this with one called Veritas Verity). It could also be embedded in a sql database and searched using full text queries. Not perfect but an interim solution. Its something I have done myself in my web dev days.

I completely understand the security issues they are facing. Lets give them Kudos for keeping it going.. here is a publisher that put an index that contained competitor magazines.. how many magazines out there do you think would do that? I am sure all the guys at MRR want it to keep going. I really don't understand the partisan sniping at Kalmbach some posts have contained.

Chris

Chris, I don't see OCR as the bottleneck - even the latest version of Acrobat pro does a good job of it, and allows for batch and multiple file processing.

I did this before I left, as I wanted the knowledge in my collection, but not the collection.

Including slitting the pages of the magazine so it would fit in the sheet fed, duplex scanner, Pdf creation, and OCR ti took less than 15 minutes per magazine (i did remove "advertising only" pages).

I love being able to type a simple search string, and find pertinent documents, and I think others may as well.

I just find an index by itself is less useful than the plain text searching we're used too. Often, situations like this can present an opportunity. I don't blame Kalmbach for no longer supporting it, times are tight for everyone; seems unfair they should clover all the costs, with little direct benefit, another reason i'd like to see it out of the hands of a single for profit organization.

If there is a way to make that happen, it would be great!

I for one am greatly dissapointed by their decision.  I have the bulk of the past issues of most MR magazines that have been published in the last 40 years.  I used the index to locate articles that would not be possible to find if I had to go through each magazine.  I just don't have the time to do that.

I agree that it was sometimes difficult to use.  Years ago, before I discovered the original index, I was doing something similar with file cards. This was BC (before computers) and would never work now.  I basically cataloged articles of interest to me at the time.  Of course my interests changed as time passed, so it was of limited usefulness.

I agree that the best solution would be for some type of open source effort maybe similar to the JMRI software effort.  I don't doubt that it is not cost effective for Kalmbach to undertake a major effort to update it as they appear to already have some system for searching the magazines they publish.  It certainly would be a good PR move is they were do make the old database available for a follow-on effort.  I don't have the skills to accomplish the task, but find it hard to believe that the data can't be extracted.  There is a lot more that could be added to the existing information..  A summary of the article would be helpful, but may involve too much labor.

I look forward to learning what Joe finds out.

I have quite a few years of various model railroad magazines I am scanning into my PC. I shear the binder off and run the pages through a duplex Canon book scanner.

My stepson is converting many of his 3,000 books into digital format for reading with an iPad. Very slick device. He has a guillotine for shearing the binders and a Canon Imager Feeder duplex scanner with auto feed capability.

We are building a data base. At least I will be able to find info more readily in my magazines.

I am retired because of age and he is retired because of medical issues.

Rich

Very Nice equipment. I just have a yesteryear flatbed.. but its better then nothing.

Chris

 There's a whole site devoted making your own scanner for books and magazines http://www.diybookscanner.org/ 

But two points to pour a little cold water on this idea, even though I think it's a good one. Firstly we stared off talking about re-creating the index, now we seem to be thinking of making the articles available online as well, aren't we suffering from a little mission creep?

Secondly there is the problem that, even though the articles may not be obtainable, many will still be under copyright. Unless we can get a body like the NMRA  to persuade the publishers to release these  for the good of the hobby or something like the Google book agreement  we would be braking the law.

Copyright law has not kept up with the digital age unfortunatly.

I sent a private message to Kalmbach Customer Service inquiring about the Magazine Index and to inform them of my disappointment with their decision.  Their response was in part the following:

"Due to the design of the database and the language in which it was written, there are no plans to sell or contract the program to an outside party.  Again, we regret the disappointment this causes and thank you for expressing your concerns."

Though the Customer Service Staff might not know of any "negotiations" being conducted by Senior Staff, so things might change.  But for now, it looks like if there is going to be a database, it will have to be done from scratch.

There is a very large debate over whether the idea of scanning entire magazines and books and making them avaialble online in an index would violate copyright law or not.  I think it would be best to avoid the potential violation by keeping the index to just that, an index, telling a person where a particular article is located.  Doing the fully text scan would be useful, but it certainly would take more time, because we would have to identifiy each article and classify it still.  This project would be by no means easy, but not impossible.

Ken L.

personal use of the owner of the magazine is not a copyright violation.  If you bought the publications, you can do with it as you like for personal use.  What you can't do is share the publications with others.  The people who were nailed by the Napster scandel a few years ago for sharing music were not in violation of the law for converting miusic to digital formats.  It is only when you seek to distribute the digital files without the publisher's permission or post them online where they can be easily read or copied that you are in violation of copyright law.

Okay, after talking to the powers that be in the NMRA, they agree they want to resurrect the index and they will talk with Kalmbach about donating it to the NMRA. They will work on getting access to the digital assets.

However, from the sounds of it, this app is a home-grown, near machine-code level application that runs on MS DOS. The data structures use a custom written binary structure that will need reverse engineering to recover.

What I need is for volunteers to step forward who can help disassemble this app and recover the data. This is a call for highly technical volunteers who can assist with this once we get access to the app and data.

Please pass the word - this index is too valuable to just let it die - and the NMRA officials are very interested in taking ownership of this index if we can muster the technical savvy to resurrect it. I would say today's talks were a success and the wheels are now in motion to start pulling this together.

Hooray!

 I'll throw my offer  of help in the ring to start. I'm a programmer who started work writing assembaler programs for IBM 360's so I know a bit about churning though machine code.

I had contact Kalmbach myself about getting  the index to start an open source group to recover it but I'm quite happy to support the NMRA if they wish to run it.

On another line I've contacted someone who may be Jeff Scherb who originanly wrote the index. It's a bit of a long shot but I seem to remember he was a fairly senior technical man in a largish company. Now a while ago I read an article in a car mag while sitting in a doctor office in France about a Jeff Scherb who was former CTO who built a customised Jeep. Thought the name was familiar but could not place it at the time. Well having been reminded about where I know the name from, did a search on the web and found this site http://home.roadrunner.com/~jscherb/builds/builds.htm

I've emailed him and given a link to this topic.