t-kamal

Rapid Development Software Tool. Mature Solution To Make Changes And Put Them In Production.

Discussion created by t-kamal on Apr 9, 2018

I found this article in below link

From my perspective One of the fairest document about CA Gen

 

CA Gen Review By David Rothschild, President | IT Central Station 

Thanks David Rothschild

Valuable Features:

It's an extremely rapid development software tool with a lot of enterprise-level software maintenance functionality. It is similar to the software that was used in all of the impact analysis stuff that's really necessary in an enterprise world. In terms of impact analysis, this is an old tool. It actually came out originally as IEF (Information Engineering Facility) in the late 1980s or early 1990s. (The name was later changed to Cool.Gen and then to CA Gen.) A lot of big systems were built upon it. When you go to make changes in these systems, you need to know what you're going to impact so you can test it effectively and make sure you don't make a mistake because a bug in a production environment can frequently be a very high cost issue. You see that all the time. People make changes and their systems come down. The airline systems of late have been big in that. This tool facilitates you knowing where your impact is as opposed to you guessing where your impact is. It's lived long beyond what people thought was its life expectancy as a product, largely because it is such a quality tool that people can't replace. It would take, I would estimate, three times the number of developers to maintain the C code that it generates than it does to maintain the other codes.

CA Gen, at least where I work, is used in a way that causes predictable production implementations.  The tool supports extensive and predictable impact analysis. Other tools from partners capture changes and make promotions precise.

.Net is a great language and the Team Foundation tools are great but the merge process is not perfect. Backing out changes in an environment were multiple people can work on the same code can be problematic.

Improvements to My Organization:

The tool is so mature that the production implementation environments around it are very mature as well. I work in a shop that has some CA Gen and some .NET and there's always problems on the .NET side when they go to production. For instance, the day after production is always a nightmare. That's not really true with CA Gen stuff, generally speaking. There are exceptions to that. It is a mature tool that lets you feel confident in making changes and putting them in production.

Room for Improvement:

The dilemma is with people that are stuck in this old technology, and it really is kind of stuck. They don't really understand the new technology, and so it's a gap issue. Those people that would use the tool for more benefit don't really know enough about how to integrate it with the other technologies. The people in the other technologies don't care about your tool. You need that gap filled with CA expertise. CA is supposed to fill that gap and they don’t really do so. They don't really know how to do it, and we don't really know how to take it, to be honest with you. It needs to be figured out. The product is solid, and has been for years and years, and they dabble in new technology. Nobody knows what will be out there tomorrow. Yes, we'd like to address whatever comes out tomorrow and work with that.

They need to re-envision themselves not as a seller of new tool sets, but as a tool set enhancer. To enhance the usage of the tool sets they have. They get big maintenance money from this stuff, so it's not like they're not making any money. You don't need to sell new tool sets to make money. They need to enhance it in terms of connecting, or leveraging, with other technologies. As an example, I work on the State of Texas Medicaid system. A huge part of that's written in COOL:Gen/CA Gen. Using those assets in the portal world, in the Windows world, is really important to be able to leverage all the work you did over here and some of the new stuff you do, and we do, but we do it based on what somebody thinks might work, not on somebody sitting down and saying, "Okay, we understand the .NET, WinGen for .NET, and here's how you do .NET, and here's how WinGen does it, and here's how we expect you to use it. Here's how you can make it better, and work harder.”

Stability Issues:

The fact that it has been around for around 25 years says it all. CA bought the tool from a vendor that bought the tool from another vendor who bought the tool. CA typically does that kind of thing. That's their mode of operation. They buy tools and then ride them until they go away. They're not doing a whole lot of development in the tool which is kind of sad, but kind of expected as well. Personally, I think the tool has taken a turn for the worse. After it gets to be 25 years old and everybody that said that it was going to go away and it didn't, you have to kind of look at the tool a little differently and say, "Well, okay, it's going to be here. Let's figure how we can maximize it.” That's really where we need CA's help. It isn’t the case that they want to go sell a bunch of new tool sets and make a big new sale. I don’t think that's happening in today's world. You're getting that done for free, so why would you pay $3000 to buy a new tool? The other side of that equation is, if you can maximize the use of the tool sets that are already sold and show people how it can be leveraged in the modern world, then you may sell a tool set here or there, but more importantly, you've given the product a life and the people that are around it a life to look forward to, as opposed to a life of, I guess, "I'll be doing this for the rest of my life until I retire” mindset.

Pricing, Setup Cost and Licensing:

I think it is very cost effective. The issue has always been that CA charges for the tool. It's like Microsoft gives away .NET and CA doesn't, so you're swimming upstream against people that see the initial expense as a big problem and don't look beyond that. The tool has been around so long now, approximately 25 years. Several companies have tried to replace it, and insisted they were going to replace it, but they weren’t able to.

Other Solutions Considered:

We started using this in 1983 or 84. It was the latest and greatest tool, and I pushed to get involved in it and became a consultant with it, and have really enjoyed working with it over the years. In terms of alternative solutions, Microsoft is the classic one. I like .NET and I love CoolGen/CA Gen, but it is not a UI tool. So you wouldn't really want to use CA Gen to build a UI. On the other hand, I can build the backend data. (By the way, CoolGen is an older name, and CA Gen is the newer name for the same product.) I'd go back, but all of the backend stuff, the data management, the business rules and all the heavy stuff, can be done in CA Gen a lot faster than you can do it with a lot less code to maintain, and in a more stable maintenance environment. But again, you can't really do the UI. You need that gap filled with a clear direction of what to do with it as opposed to just guessing which is what happens.

Other Advice:

Get some experienced help. You can mess yourself up in the early days if you don't know where to go, but there are a lot of people out there who have built big systems in CA Gen and can help you do the right thing from the beginning and that will make a huge difference.

Outcomes