The Linux Foundation Projects
Skip to main content
I Am A MainframerPodcast

I am a Mainframer: Sam Golob

By September 18, 2024October 14th, 2024No Comments
In this episode of the I Am a Mainframer podcast, guest Sam Golob delves into the history of the CBT Tape, a vital collection of user-created software for mainframe systems programming. Golob shares how the CBT Tape began in the 1970s, evolving from a physical tape to an online repository. He also discusses his role in maintaining this key resource, which continues to serve as a hub for systems programmers to share tools and solutions within the mainframe community. Be sure to catch this insightful discussion!
Watch full episode here:
https://youtu.be/SVvlTjaO2fs

 

Or listen to and watch the episode here:

Transcript:

Announcer:

This is the I Am a Mainframer podcast, brought to you by the Linux Foundation’s Open Mainframe Project. Episodes explore the careers of mainframe professionals and offer insights into the industry and technology. Now, your host, Senior Analyst and Vice President of Sales and Business Development at Futurum Research, Steven Dickens.

Steven Dickens:

Hello and welcome. I’m Steven Dickens, your host here on the I’m a Mainframer podcast, and today I’ve got the brains behind CBT Tape, Sam Golob on the show. Hey Sam, welcome to the show.

Sam Golob:

Thank you very much, Steven.

Steven Dickens:

So let’s get the listeners and viewers orientated here. Tell us a little bit about you, what you do, and then we’ll jump off from there and get into CBT Tape.

Sam Golob:

Okay. I’m a MVS systems programmer now. It’s called z/OS by trade, and I’ve been at it for 40 years.

Steven Dickens:

So you started in kindergarten, Sam, is that what you’re trying to tell me?

Sam Golob:

Well, sort of, but I had to make a living, had a bunch of kids and I had a wife and kids and I had to make a living for them. So I went to computer school and started somewhere, but I was always very inventive and I tried to do some things that were a little different.

Steven Dickens:

Fantastic. Well, that’s going to make for an interesting podcast. So where are you a system programmer? Are you working for a particular company, are you contracting or freelancing at the moment? Where are you?

Sam Golob:

Well, right now, I’m not really working for a company right now, but I do manage the contents of the CBT Tape. Which is a tremendous collection of user written software for systems programmers concentrating on all levels of the MVS operating system back to the OS 360.

Steven Dickens:

So let’s start there. I’ve heard about CBT tape, the world thinks about open source. We’re obviously part of the Open Mainframe project, but I don’t think people give the mainframe the credit it deserves for the history that it’s had in the open source community and almost predating what we think of as open source with Linux and those types of open source communities today, CBT tape for me is one of the classic examples of a community coming together, sharing best practices. But tell us a little bit more about the history of CBT Tape and your role in maintaining that community.

Sam Golob:

Okay. Well, the thing is, there used to be a lot of collections like this. There was the Michigan tape and other organizations or people had collections because there was always a need to spread the wealth around. If somebody solved the problem for himself or herself, there needed to be a place where it could be given to other people. So there were a lot of collections. There were quite a few, problem with them is they all died. Either a person lost his job or they sold the bank, which is what happened with Arnie on the CBT Tape originally because he ran it for 15 years. You should know that the CBT Tape has been in existence for 49 years.

Steven Dickens:

Geez. This is what I mean about the open source community, what we think of as classic open source, and this is a Linux Foundation project. Those are kind of 20, 30 years old, Linux is 30 years old now. People kind of think, oh, this is a 30-year-old kind of phenomenon. Mentioning that CBT Tape predates that by 19 years for me is just crazy. So people were sharing best practices,

Sam Golob:

Of course.

Steven Dickens:

Yeah. I mean, that’s the story that’s interesting for me that doesn’t get told.

Sam Golob:

Well, the thing about this particular collection is that Arnie Casinghino, the guy who started it, was extremely devoted to it, and he was very finicky about details, and he wrote everything up in his documentation in such a clear and thorough way that when I got my first tape in July 1985, I took one look at this and I said, wow. And what happened was in the subsequent few years, Arnie’s bank, the Connecticut Bank and Trust Company, which gave CBT its name,

Steven Dickens:

That’s where it comes from? I’ve always wondered where CBT came from. So that’s the genesis story is it? Okay, that’s good to know.

Sam Golob:

The Connecticut Bank and Trust company was bought by another bank, Bank of Boston, which was bought by the Fleet Bank, and they announced their intention of dismantling Arnie’s data center in Hartford, Connecticut. And we had about a year and a half lead time on it. And at that point I was very, very friendly, with Arnie, and he was sending me all the recent tapes every time he made one, which was about 15 a year. And we were just very, very, very good friends and we had about a year and a half lead time to figure out what to do to keep this thing going.

So I figured out that, well, Arnie needed a lot of disc space to do it his way, but I was able to use some of the tools from the tape, and use a system called, how do you put this, Old Master Transaction Tape, New Master. So I was able to make a new tape with the new stuff in it and all the old stuff also in it. And then I gave it to somebody to duplicate. And since I used to work with an organization called NaSPA, I wrote a column for them in their magazine for almost 20 years, “MVS tools and tricks of the trade.” And by the way, all my columns from the magazine and other things that I’ve written are on the CBT Tape on two files.

Steven Dickens:

Oh, okay. So there’s an archive if people want to go back and see those, they can.

Sam Golob:

And it’s also online on our website. The articles are there too. So the reason why I wrote all of these things was because it was hard for me to learn the subject.

Steven Dickens:

And this is the thing I love about the z/OS, and MVS community, the annual conference, that was, what was it last week? I don’t know when this episode’s going to go out. It’s called SHARE and people don’t realize, but it’s because people used to share these tapes and share these best practices. That’s why it’s called SHARE.

Sam Golob:

Right? It isn’t just the name. It’s what we do.

Steven Dickens:

Yeah. And I think people don’t give, people don’t associate open source with mainframe. And John Mertic who runs the Open Mainframe project, and I went down this rabbit hole a few years ago, the mainframe community’s been doing open source for like 50 years.

So we’ve talked about CBT Tape, great to know where the name comes from, tell us a little bit, Sam, about what’s in CBT Tape. I get one of these, I’m assuming it’s no longer a physical tape, but I’m able to download this. What’s in there? What can I expect? I’m a systems programmer. I’m watching this podcast for the first time, I don’t know what’s in this thing. What is it? What can I expect when I download it, I suppose is the question.

Sam Golob:

Well, okay, now I just want to mention something about vendor products. If a shop is a rich shop, in other words, the first systems programming job I had was for a big bank and they could afford all kinds of vendor software in addition to what IBM gave out. And one of the things they had was something called Resolve. I think it’s produced by BMC now, but in any case, it’s a multi utility that can do all kinds of necessary tasks, that a systems programmer needs to do. And IBM doesn’t supply the real proper tools for it. They supply some tools for it, but they’re really not so adequate and they’re not so convenient for the simple reason that IBM is a software house as well, and their main thrust is producing the operating system. And yes, you need tools to do everything that you need to do when you’re maintaining the operating system, but they can’t produce tools, the way we can produce tools. If somebody needs a tool, at least they give you the macro libraries, they give you manuals, they give you information how you can find stuff out. There’s also stuff they withhold, but we can find it sometimes and we have to struggle to get the tools that we need. So now if somebody had a problem, it’s going to be for sure other people, it’s the same operating system, other people are going to have the same problem. So one person managed to write a tool to get around the problem, and if you can get this tool to other people, then they can solve the same problem without having to reinvent the wheel.

Steven Dickens:

So this is almost like predates Reddit where somebody goes on to Reddit to try and find the answer to something. Somebody’s fixed the problem, somebody’s recorded a YouTube video to say, Hey, I fixed my carburetor in my 1960s car, here’s a video of how to do it. This is almost that, but from mainframe systems programming, I think by the sound of what you’re telling me.

Sam Golob:

Well, in a sense it is. The thing is that we just produce the tools and there’s usually instructions that I try to make sure that there are instructions on how to properly install the tools. And as a normally trained systems programmer can put this on his own system if the company or the owner of the system will allow it. Sometimes they only want to have vendor software on there and they can afford to buy it because the vendor has a contract and they support the stuff. And these companies would like the reassurance that their stuff will be covered by a contract and supported. We can’t do that. And we do this stuff for love and for fun.

Steven Dickens:

That’s how communities grow though, Sam. That’s how communities grow. That’s the thing. I mean, people ask me why do I do this podcast? I was at the fortunate position of, I was the person who approached the Linux Foundation to set up the Open Mainframe project back in 2014. So I’ve kind of got this as my baby. Why do I do this podcast? I’m not a coder. I’ve never been a systems programmer, I’m not been hands on keyboard, but I want to help this community grow and expand. I can do podcasts, I can help with marketing, I can help with promotion. That’s what I can bring to the community. I have fun contributing my time. Can’t contribute technically, but I can contribute my time and my skills, and it’s fun for me. I get to chat to fun people like you who are genuinely passionate about the community. Our skill sets are so completely different, but if you look at what we’re doing, we’re both contributing our time, we’re both having fun, contributing our time, and we’re both moving the community on, me by giving you a platform to talk about your stuff so more people find out about it, you by doing great work to pull together that tape and provide those tools. This is what community’s all about for me.

Sam Golob:

Well, it’s the same for me.

Steven Dickens:

So we share a common bond, but our skill sets are completely different.

Sam Golob:

That’s what communities about. Well, this is the same with all the contributors of the tape because all our skill sets are different. It’s a large system. People usually have experience in what problems that their shop needed.

Steven Dickens:

Might be a db2 two guy, there might be a CICS lady, there might be a tape and backup person, but you can put all that together. You get that perspective.

Sam Golob:

Well, I don’t understand everything that everybody contributes, however. Well, Arnie’s precedent was what he said was, I just put everything up. Well, I try to understand a little more than he did.

Steven Dickens:

There’s a bit of curation, there’s a bit of understanding.

Sam Golob:

Yeah. Well, the thing is, sometimes a person will write something for himself or herself, and if you make a mistake in entering something, you need error correction. You need something to catch the error. And usually when somebody writes something for themselves, a lot of times, not always, but a lot of times when somebody writes something for himself or herself, they just assume that the correct things will be entered when you use the program. But what if they aren’t? So then the program can blow up. So I consider it part of my job, if I understand what’s going on, to check up on that and put some fixes in to make sure that other users who aren’t the one who wrote the program will be able to, if they do something wrong, it’s not going to blow them up.

Steven Dickens:

So Sam, tell me a little bit about, obviously this started out as a tape and used to get physically sent around, what’s the process now? Somebody listens to this podcast, where do they go? How do they find out information? How do they download this code? What do they do? Let’s spend just a couple of minutes there.

Sam Golob:

Our website is called www.cbttape.org. And It was created by my dear friend, Sam Knutson and he’s the one who pushed for it. It used to be that the tapes were duplicated by NaSPA and then we got the website this way and it was, okay, I don’t want to talk about politics here.

Steven Dickens:

Let’s move on, politics is never a good subject for a podcast.

Sam Golob:

I know. In any case, so NaSPA stayed on to help us along, but in any case, to make a long story short, the files on the tape are individually loadable if you have an FTP connection to the site. Now, one of our contributors, Lionel Dyck, created on file 43 I think it is, he created a system where if you have a z/OS system that has an FTP connection to our site, you can automatically download every file on the CBT Tape instantly. The time it takes for FTP to go grab it and unzip it. Most of the files are a zipped transmit format, and IBM created the transmit format back around 1985 for the purpose of sending files from one MVS system to another. And in doing so, they converted all the files, as many as they could, to the card image format, 80 bytes, six blocked records.

Steven Dickens:

So I can go onto cbttape.org, download those,

Sam Golob:

Yeah, you download the download the file and unzip it, and then you receive it on your system and you have usually a PDS, a partition dataset, which is like a library, and you could look at the members and it usually has documentation and it’ll show you what to do to install it. They’ll show you what you can do.

Steven Dickens:

Fantastic.

Sam Golob:

There’s also another thing that we have to help people out. I think it’s file 874, but I’m not sure. There’s a file of help members that are mostly in the format of IBM’s help members, but there’s a whole library of help members for tools that are on the CBT Tape. It’s not completely inclusive. There are maybe a small fraction of the tools that are on the CBT Tape, but there’s at least a hundred help members or something like that. That’s enough for people to get started with. And if they look through there and see what the different products do, then they might think of installing something for themselves. And each help member has tells you what file on the CBT Tape the product came from.

Steven Dickens:

Well, Sam, that’s fantastic. I think, I mean, the takeaway from this podcast for me is if you are a mainframe systems programmer and you don’t know about CBT Tape, go check it out. Absolutely. A treasure trove. What are we talking 49 years worth of best practice put into a website? Sounds like a fantastic place to sort of go and get resources. I could talk to you for hours Sam about this. It fascinates me, but I’ve got to keep us on track here. So we’ll pivot now for a moment. A couple of things I ask all the guests, and I’ve been fascinated to get your perspective on this. I ask these two questions of every guest. You’ve been on the mainframe for a while. You got the chance to go back chat to Sam, age 22 coming out of college. You’ve seen this mainframe platform over the decades. What advice would you give? We’ve got a lot of younger listeners to the show. What advice would you give to those listeners?

Sam Golob:

Well, it was different when I started. First of all, IBM was not so clear in the way they wrote their manual. And when my first systems programmer job, which was a couple of years after I started programming, it was an application programmer before. My first systems programming job, most of the people would tell me if I had a question, go read the manual. Yes, read the book. So now to me as a neophyte, there were 200 manuals and to me, it seemed like the correct information was contained in five of them, and I didn’t know which five. So that left me without really that much of a possibility of reading the book. So I really learned things one thing at a time. In answering your question, I think I had to go through that learning process originally, and I don’t think I would change too much.

Steven Dickens:

The process of the learning is what is the takeaway?

Sam Golob:

Yes, I had to go through the process of learning what I had to learn, and it was a long, long process and it’s still going on today.

Steven Dickens:

That’s been a consistent theme, Sam, as I’ve done this podcast, I don’t know, in five years or so now, is that consistency. Pivoting to my last question, and I’d love your perspective on this. You’ve got a completely different perspective of the mainframe of some of our listeners, and certainly from my perspective, I’m kind of tracking big market trends and tracking where the vendors are going. You are kind of in the engine room as a systems programmer. Where do you see the platform five years from now? If you’ve got a crystal ball and you’re able to look through the mist and the mists clear, where do you see the platform?

Sam Golob:

Well, it’s kind of hard.

Steven Dickens:

This question is my favorite question to ask.

Sam Golob:

The thing is like this. I don’t like to do that. What I like to do is to just take it one step at a time. When I’m thinking about it, we support all the levels of the operating system and the CBT Tape goes back a very long way. So if you have an old version of the operating system, say 25 years old and people are still running them, and you need something to help you at that level, well, we have the old tapes that were written for that level. So you can download the old tape, you have to do a lot of work to come through it, and look to see what you can find, but nevertheless, we support all of that.

Steven Dickens:

And you see that going forward, Sam, as you look five years out?

Sam Golob:

Sometimes you can’t predict what’s going to go on forward, but you can try to adjust to it as it arrives. That’s all I can say.

Steven Dickens:

Well, Sam, this has been fascinating. As I say, I could chat to you for hours. This is a huge topic. I think the key takeaway for me is go look at cbttape.org, let me get the URL right.

Fascinating conversation. You’ve been listening to another episode of the I’m a Mainframer podcast. I’m your host, as always, Steven Dickens. Please click and subscribe and do all those things to share this with your colleagues and friends, and we’ll see you next time for another episode. Thanks very much for watching.

Announcer:

Thank you for tuning in to I Am a Mainframer. Liked what you heard? Subscribe to get every episode or watch us online at openmainframeproject.org. Until next time, this is the I Am A Mainframer podcast, insights for today’s mainframe professionals.