After 36 years: Atari Educational System

[FULS]

Try these.

on tapes B and C, I could not get a good dump on either side.

 

 

 

 

 

 

Thanks,

Doug

[luckybuck]

Well, o.k. but is there any loss of information/data?

 

AtariWiki V3: Basic Electricity CX6016

 

AtariWiki V3: Great Classics CX6009

 

I could read in all tapes...

[tschak909]

If you see the Dorsett copyright notice, THAT'S THE BEGINNING OF THE TAPE.

 

-Thom

[luckybuck]

Hi Thom,

 

just some impressions from the Wiki:

 

 

[Kr0tki]

This tape was dumped as an 8bit file, not 16bit.

 

I figured a 35 year old analogue monotone tape will sound just about the same 8bit or 16bit with half the file size.

 

These WAV file sizes become crazy huge when you collect a few series of cassettes. Try it and see how it plays and sounds.

FULS, please record them in 16-bit. The size will be only 2 times larger (increasing storage size twice is not that large of an investment these days); the difference in fidelity is audible (well, it was audible _to me_ last time I tried); it may improve reliability of loading in case of low quality tapes; and lastly, 8-bit won't be enough for future attempts at digital processing of the files. Case in point: many Dorsett tapes have poor channel separation - the data track leaks into the audio track - so we might want to fix this issue with some digital signal processing. The 8-bit resolution might be not enough to achieve good quality result - you'll never know unless you try - so it's better to be on a safe side and store the tapes in the better quality format from the beginning.

 

 

on tapes B and C, I could not get a good dump on either side.

Unless they are truly worthless, with patches of complete silence, please record them as well. Isolates errors in the data track can be fixed manually (a labour-intensive task, no doubt, but I've done it a few times). Also, when another copy of the tape surfaces, your copy still might be useful - if the other copy also was incomplete, we might want to edit the two recordings digitally to achieve a fully-working dump.

 

Oh, and thank you for taking your time locating and preserving these gems.

Edited May 15, 2015 by Kr0tki

[tschak909]

Yes, I can't stress enough, the importance for 16-bit sampling .

 

If I may:

 

Digital audio principles 101:

 

Digital audio is the process of sampling analog audio (a continuous signal) at discrete points in time. The sampling happens as a function of samplings of amplitude over time.

 

The bit-rate of a sample indicates the number of discrete amplitude values between 0 (no signal) and MAX (the maximum signal that a DAC can reliably sample). It is this distinction which when fully extrapolated means that, with each addition of a bit onto the sampling rate, the signal to noise ratio of the signal increases. (in a positive way, the potential for a better signal improves exponentially, the potential for isolating and trapping noise, also improves inversely (logarithmic).

 

The sampling rate, determines the highest possible frequency that can be sampled, more accurately, the highest frequency that can be sampled, is described as the Nyquist frequency, which is described as half the sample rate. So for a sampling rate of 44100 Hz, you have a Nyquist frequency of 22050Hz. However, what can be sampled at this rate, is not guaranteed to be a sine wave. In fact, the quality of what can be sampled at this rate, decreases exponentially as the frequency approaches the Nyquist. How do you improve this? by increasing the bit depth.

 

So with the median frequencies being 1/4th of the Nyquist, you have sufficient bandwidth to avoid unnecessary noise errors (if anyone wants to see the math, let me know), and you have enough of a sample rate, to get enough points to accurately reproduce a sine wave.

 

As another note: Often, you'll see very wrong diagrams and descriptions of digital audio as a "series of stair steps" ... I mean, it does make sense on the surface, right? You're sampling discrete values at specific points in time, so you'd be compelled to draw each sample as a run of stair steps...

 

But you'd be wrong.

 

Since these samplings in time are just that, samplings in time, and not relative to the previous points, and thus they don't have any "length", a lollipop graph can more accurately plot the sampled points.

 

Due to various factors, ranging from the output stages of digital to analog converters, and the discontinuous nature of digital data, even at the lowest sampling rates, so long as there are enough points, a "stair stepped" waveform is actually rendered quite continously. Don't believe me? Run an oscilloscope on some audio coming out of a soundcard, alter the sample rate, alter the bit rate, you'll see, so long as you obey the rules of the nyquist, you'll see a sine wave rendered quite nicely.

 

Monty from Xiph.org has a nice series of videos on this very subject, which describe what I am saying in greater detail.

 

-Thom

[luckybuck]

Yes, fully agree. That is why I made post#29:

#29

Further, we have suggestions for creating PDF-files in the Wiki-Manifesto.

[luckybuck]

@Doug:

 

Thank you very much for the 2 tapes, now just a few tapes are missing, there is no package where all tapes are missing!

 

So cool you could help us.

 

All the best.

[cas]

If security is a reason, then the certificate issues should be resolved. If the certificate security warnings are not important for the people viewing the site, then there is no security reason for using https either. Just sayin'.

 

Hi Ken,

 

I'm the admin of "atariwiki.org" and responsible for the TLS cert. Currently, there are no free and trustworthy TLS certs available. The cert we currently use is issued by CACert http://cacert.org/

 

CACert is a community certificate authority. The CACert Root-Certificate is not shipped by default in common browsers. In order to resolve the issue, users need to import the CACert Root-Certificate into their browser. The CACert root-certificate can be found on their website.

 

Given that the common Internet-User is trusting all kinds of shady CAs that are shipped with the browser, adding CACert should not be an issue.

 

Later this year Mozilla and EFF will start issuing free TLS certs as part of their "let's encrypt" campaing https://letsencrypt.org/

Once these certs are available, I will switch. Mozilla has promised that the root-certs for "lets encrypt" will be in the browser default root-certs store.

 

A trustworthy TLS cert costs around 350 Euro per domain-name and year. For the Atariwiki-Site, we have 3 domains. The cost would be around 1.000 Euro per year. I'm open for donations.

 

Authentication of the TLS cert independent of the CACert root-cert can be done by installing the DNSSEC-Validator plugin: https://www.dnssec-validator.cz/

Edited May 17, 2015 by cas

[dmsc]

Hi,

 

Hi Ken,

 

I'm the admin of "atariwiki.org" and responsible for the TLS cert. Currently, there are no free and trustworthy TLS certs available. The cert we currently use is issued by CACert http://cacert.org/

 

CACert is a community certificate authority. The CACert Root-Certificate is not shipped by default in common browsers. In order to resolve the issue, users need to import the CACert Root-Certificate into their browser. The CACert root-certificate can be found on their website.

 

Given that the common Internet-User is trusting all kinds of shady CAs that are shipped with the browser, adding CACert should not be an issue.

 

Later this year Mozilla and EFF will start issuing free TLS certs as part of their "let's encrypt" campaing https://letsencrypt.org/

Once these certs are available, I will switch. Mozilla has promised that the root-certs for "lets encrypt" will be in the browser default root-certs store.

 

A trustworthy TLS cert costs around 350 Euro per domain-name and year. For the Atariwiki-Site, we have 3 domains. The cost would be around 1.000 Euro per year. I'm open for donations.

What about wosign https://buy.wosign.com/free/ or startssl https://www.startssl.com/?app=1

 

Haven't used them, but found the links in the Mozilla blog post about requiring https in the future.

[jacobus]

are you still looking for these?

 

[luckybuck]

Hi jacobus,

 

Yes!

 

Basic_Psychology_CX6011-Cassette_B-Side_1

Basic_Psychology_CX6011-Cassette_B-Side_2

 

Business_Communications_CX6010-Cassette_D-Side_1

Business_Communications_CX6010-Cassette_D-Side_2

 

 

Basic_Sociology_CX6005-Cassette_B both sides

Basic_Sociology_CX6005-Cassette_C both sides

But a re-recording of Basic_Sociology with 16 bit would be marvelous! :-)))

So great and wonderful, if you can complete it.

Many thanks in advance.

[jacobus]

 

Hi jacobus,

 

Yes!

 

Basic_Psychology_CX6011-Cassette_B-Side_1

Basic_Psychology_CX6011-Cassette_B-Side_2

 

Business_Communications_CX6010-Cassette_D-Side_1

Business_Communications_CX6010-Cassette_D-Side_2

 

 

Basic_Sociology_CX6005-Cassette_B both sides

Basic_Sociology_CX6005-Cassette_C both sides

But a re-recording of Basic_Sociology with 16 bit would be marvelous! :-)))

So great and wonderful, if you can complete it.

Many thanks in advance.

 

I don't have a cassette player anymore. PM me and we can make arrangements to send them to you (assuming of course, you'll send them back :-)

[jacobus]

Crap. I just had a closer look and my copy of Basic Psychology is a virgin. Is there anyone else who might possibly have a copy?

[luckybuck]

Hi jacobus, in case of that please just stand by. Just good to know, you have it. We are so close to 100 %.

[cas]

Hi,

 

What about wosign https://buy.wosign.com/free/ or startssl https://www.startssl.com/?app=1

 

Haven't used them, but found the links in the Mozilla blog post about requiring https in the future.

 

I know them and I know the Mozilla blog post.

 

I would trust them much less than CACert. Starting and running a certification authority, especially one that has the root-cert in the browsers, is not cheap. It is expensive. How can they offer for free?

 

Some people in the security business I've talked to have suspicion that these two companies are at least partly controlled by their governments/secret services (StartSSL = Israel, Wosign = China).

 

With StartSSL, the inital certificate is free, but any change/revoke in case of a security incident (such as Heartbleed) costs money

https://www.techdirt.com/articles/20140409/11442426859/shameful-security-startcom-charges-people-to-revoke-ssl-certs-vulnerable-to-heartbleed.shtml

 

I don't see these two companies as a trustworthy option for getting TLS certificates.

 

I meet (some) CACert people at open source conferences. In person. Their procedures might not be without flaws, but in my view much more trustworthy than WoSign or StartSSL.

 

Carsten

[+Eyvind Bernhardsen]

 

I meet (some) CACert people at open source conferences. In person. Their procedures might not be without flaws, but in my view much more trustworthy than WoSign or StartSSL.

 

This is getting pretty off-topic for AtariAge, but while I certainly agree that the SSL certificate trust model is broken, and I can see why you don't want to deal with those companies, asking people to add yet another root certificate to their browser or OS trust store borders on being rude. I don't want to invite another third party to verify SSL certificates for me; it would make my machine less secure (because it would trust CACert-signed certificates for any site, not just yours) with no real benefit to me.

 

If you really don't want to deal with the hassle of getting a pre-trusted certificate, you should put your certificate's signature in all your posts and ask people to add just your certificate to their trust store when they connect. They'll have to do it again when your certificate expires, but at least people who care about security will be able to verify for themselves that the certificate is correct, and nobody's security will be any worse.

[cas]

 

This is getting pretty off-topic for AtariAge, but while I certainly agree that the SSL certificate trust model is broken, and I can see why you don't want to deal with those companies, asking people to add yet another root certificate to their browser or OS trust store borders on being rude. I don't want to invite another third party to verify SSL certificates for me; it would make my machine less secure (because it would trust CACert-signed certificates for any site, not just yours) with no real benefit to me.

 

If you really don't want to deal with the hassle of getting a pre-trusted certificate, you should put your certificate's signature in all your posts and ask people to add just your certificate to their trust store when they connect. They'll have to do it again when your certificate expires, but at least people who care about security will be able to verify for themselves that the certificate is correct, and nobody's security will be any worse.

 

The certificate on the site changes every 6 month. Publishing a cert fingerprint in email or forum-posts does not scale, importing the sites certificate into the browser trust store also does not scale.

 

The real solution is certificate verification via DNSSEC secured DNS, aka DANE. Unfortunaly DANE is not supported natively in the browsers today and requires an extension plugin.

 

I guess we have to wait until the Mozilla/EFF certs are being available in summer.

[tschak909]

An Update from my side:

 

Now that the archaeology of the master cartridge disassembly and forensic analysis of the tape format is completed, and the encoder functions, I am currently writing up a book detailing the entire "dig", and will be releasing it to the public once complete.

 

It will consist of three major parts:

 

* Introduction

* Disassembling the Cartridge

* Forensic Analysis of the Cassette Format

 

The focus on this particular book is very much a "post mortem" of the Atari Educational System, with the intent of the book being viewed as a proper historical text deconstructing one of the earliest examples of both multimedia and computer based training.

I will then take the content of the book, and at the very least, take the reference bits and place onto AtariWiki, as this is the most logical place for it.

-Thom

[Bryan]

Is Loyd G Dorsett still around? It would be interesting to find out more about the project. He was given a patent for AV learning:

 

http://patents.justia.com/inventor/loyd-g-dorsett

 

and he wrote a book:

 

http://www.amazon.com/Audio-Visual-Teaching-Machines-Dorsett/dp/0877780099

 

Sounds like a job for Kevin Savetz, actually...

 

 

EDIT: Looks like he passed away in 1999. I wonder who else might have been involved. His name turns up a lot of interesting things.

 

http://newsok.com/dorsett-reports-loss/article/2263721

[tschak909]

Wow! I can't believe you found an edition of that book. I've been looking for it, for my research.

 

and yes, Kevin and I are talking. I am also talking to Joe Decuir and Liza Loop about this subject. The stuff I'm writing now, will at the very least, be source material for Joe's upcoming second system book, and I am writing it all out to be a complete stand alone book on its own... Liza is also interested in both the book material and the educational system and software for a virtual museum on the history of computers in education.

 

-Thom

[tschak909]

http://newsok.com/dorsett-systems-strengthens-position-with-2-new-contracts/article/1969101

[tschak909]

https://www.deepdyve.com/lp/american-association-of-physics-teachers/dorsett-educational-systems-inc-physics-t6f9aJs8aX

 

This seems to indicate by 1987, the programs were available for Atari, Apple, IBM, and Tandy machines. What's interesting is that the content's primary encoding is the Atari format, with different converter boards (basically bespoke tape interfaces) for the other machines.

 

-Thom

[tschak909]

Working on the book...

 

 

 

[luckybuck]

WOW Thom! That is such an outstanding work you do. Please go ahead. We really need your help and knowledge.

 

Thank you so much.