Went to look up something on on the index today to find it’s gone.
The message says-
The advancement in Internet browser technology has created an insurmountable problem for the Model Train Index, which has been a fixture on Trains.com for many years.
The information in the Model Train Index and its integrated software originally were written in MS-DOS language 20 years ago by a train hobbyist on his PC as a personal project. It was not designed for the Internet and the software is unique to the index.
Kalmbach Publishing Co. purchased the index and its software many years ago. We created a website interface for the index and made the index a featured part of the Trains.com website
Today, the progression of technologies on the Internet and the antiquated software of the Model Train Index have intersected to the point that we can no longer offer the index to our website visitors because we can no longer keep the index free of viruses or other even more significant Internet security risks.
The index software cannot be rewritten to update its security level, nor is there any method of transferring the index text information to a more modern software platform since the design of the index software and the index text information are integral to each other.
Because of these issues, we have reluctantly removed the index from Trains.com.
I can understand their point, it’s always been a little temperamental and it is very old code, but it was so useful.
Why don’t they just write another index system with new technologies from todays methods. It’s not rocket science.
Internet technologies are constantly changing, trust me I’m in the tech field. If you don’t constantly update your systems to keep up with new technologies and methods, then you’re going to run into these issues.
This must have just happened today because I used it yesterday to find something I was looking for. Too bad. I guess that we’ll have to rely on people’s memory cells now. Ummmmm, what were we talking about?
Mmmm. Wouldn’t it have been possible (even if one doesn’t know the data structures of the files or database used) to extract quite a bit of useful information by using some kind of web crawler to crawl the “browse magazine content” links in the old app and just dump the data extracted from the web pages to a simple database.
Probably wouldn’t get you all the information from the old app, but it could potentially give you a good start for a data dump that could be placed in the public domain, allowing others to spend their time and/or money on building and offering a searchable index web app based on this partial data dump.
Well, time will tell what (if anything) happens. I will sorely miss that searchable index.
I have a hard time believing in 2010, where we are surrounded by databases, in fact this very forum is just a fancy front end for a database, that they couldn’t import the relevant data into MySql and get a php frontend written. If Popular Science can put every single issue online, then MR can upgrade the index.
Maybe we should contact Kalmbach and see if we can get them to release a copy of the database and if possible the source code of the program. While it may not be economically viable for them to update it, I’m sure there are enough of us here with a bit of programming and database knowledge to convert it into something more modern, maybe make an open-source project out of it.
Well firstly we’d need it back up to do this and secondly the program actually had code to stop you doing this.
I discovered this the hard way many years ago when I was using it fairly heavily updating my off-line paper index. I got locked out , and on inquiring why in the old forum I was told by the original author that if one IP address issued too many requests it was locked out. He said he did this to prevent people stealing his data !
Ah, that would explain why Kalmbach didn’t try that approach. Getting the server up again would have been pretty simple if it was taken down controlled (as opposed to suffering some kind of major crash). But that annoying copy right protection mechanism sounds like it would have been a major annoyance while trying to extract data.
The death of this database is a crime against all model railroaders. It needs to be sold or given to someone who will at the very least maintain the current information.
Since this covered other magazines in addition to Kalmbach publications, perhaps the NMRA should take it over. Without this index there is no way to search all that past content. The only way you’ll find this information on Google is if someone happened to ask about the very subject you are querying, in which case you’ll probably get a link to the thread in this forum,
That is absolutely the truth. At the very least they should open the data and the source to the program to the open source community. I’m sure there would be plenty of volunteers-- me included-- who would be happy and willing to undertake the restoration of the index. PHP and MySQL would indeed be fine. But of course there are also a number of existing frameworks available that could even improve its functionality. Another thing that you could do would be to give it “searchable urls” and then tell Google about it-- then you would get even better and more searchable index.
There is something decidedly odd about this decision. What came first, the discovery of these vulnerabilities or their increasing laxness about updating it? Let too much time go without updating and their investment in the index would also become more or less useless.
I regard the dropping of the Model Train Index from this site as a major mistake and loss to the model railroading community. For the more serious hobbyist the Index was a unique resource. Originating as a hardcopy publication, now long OOP, the elimination of this on-line version leaves a very large void in hobbyists’ choices for easily locating past articles in a wide variety of current and older hobby publications. To my knowledge there is nothing similar currently available to use as a substitute.
that’s the thing of it. No one thought to at least spin off the project to an open source group or even some sort of indexing firm to do this—one group that would be good for this would be Source Forge.
I’m wondering about this though. What if they just took the lists offsite? And are still keeping them? I can see them waiting until a demand for that index builds up then them selling it to the highest bidder[:-^]
It’s rather frustrating that one can go anywhere else and see all kinds of these indexes. Here, we see them not keeping that db?[banghead][soapbox]