First, in my personal opinion, being a developer myself and working with web and mobile technologies, I think they're trying to cover all the sectors, and that's the problem. They want to be one toolbox for everything, but primarily, we are using xpa to develop desktop applications, and in that area they're lacking functionalities, flexibility, and modern stuff.

MSC is marketing xpa as a .NET-based solution, but their .NET integration is lousy. For example, in Visual Studio, you put a control, you right click, and you set everything. In xpa, they're using Visual Studio as a basis of the development tool, but you cannot do what I just described. There's also poor integration of third-party tools because, for example, to put something together using the very popular .NET framework and components framework, it takes me at least three times longer than it should. 

We started using this solution because it was fairly easy, and 10 years ago, the speed of development was incomparable to any other tools. My employees can develop and deploy something in a matter of hours. My clients buy from me because we can do everything very fast, but the applications we are currently developing with xpa are kind of outdated. Not the functionalities, because we can do almost everything, but the UI and UX and the mechanics of the application are outdated. The problem is that their grid functionalities are very bad in general. For example, in order to have the ribbon bar like the one you have in Word or Excel, we have to do all kinds of gimmicks and purchase external libraries. That's one of the problems, and that's something I would really like to change. 

I really don't care about the web integration with xpa. I don't need it to be a tool for the API, for the backend, for the frontend, or for mobile applications. I want xpa to be a very powerful tool for desktop applications.

It's a bit difficult to work with purely web-based applications to get the data and display it. There have been a few times when the connection was disconnected between the server and your browser. The connectivity on browser-built applications needs to be improved.

The mobile application development could be easier. They could include different external applications, like finger sensors. I'm not sure whether it's in version 3.8 or not.

There is room for improvement in Magic's marketing and licensing. If you talk to people about Magic, there is a pretty big chance that no one's ever heard about it. It's not mainstream like .NET or Java, and that's because of the marketing and licensing strategy of the company. They never put an emphasis on marketing, so now it's unknown. 

When programmers first came into contact with Magic, they saw the benefits, but they had to pay for a development license or a one-time license. I think that's the main disadvantage of Magic because that's a big threshold preventing Magic from being a mainstream competitor among Java and .NET.

As for new features, I would like to see more integration of web functionality. Client-side intelligence, rather than service-side, and done in an easy way. It's already possible, but there is still some work for you to do as the developer. The most standard functions could be automated a bit more.

• The configuration of the xpa RIA mobile environment is complex and a discouragement to new developers.
• Magic's documentation can be less than complete at times which leads to frustration for new developers. (I encourage new Magic developers to join the Magic Users Group).
• The Android environment is missing a number of functions for file/folder manipulation, sending receiving text messages (SMS) and the menuing options are limited. For now, it is left to the developer to write his/her own Java functions to include in the APK.
• It is missing basic charting tools for bar/pie/series charts. It is left to the developer to acquire and deploy charting tools or the customer to purchase a third-party reporting tool to produce charts.
• I would like to see a spell checker included with optional language support. Currently, this has to be purchased from a third-party.
• There is also an issue with table control sorting of data that does not result in an additional database call.
• The ability to display page up, page down, top and bottom buttons along the scroll bar would make my mouse-reliant customers happy.

Throughout my career, I've encountered difficulties when integrating new technologies with Magic xpa Application Platform. In particular, when attempting to incorporate features from other development languages into earlier versions of the solution called uniPaaS. I struggled to integrate .NET components due to the limited options available. This made the process more challenging and complicated. I find it challenging to create a more user-friendly experience for users who may be comparing the system to other systems they have used outside or within the company on different platforms.

Currently, I have encountered an issue as I am engaged in mobile development utilizing the Magic xpa Application Platform. The process of releasing on iOS is more complex compared to Android, and in future iterations of the solution, the integration of the iOS platform would be of great advantage. This is because when it comes to deployment, the procedure is more demanding in the case of iOS compared to Android.

Magic’s lack of documentation, in both the educational and technical fields, has been a thorn in my side since I can remember.

I worked for Magic at the US branch for several years as a consultant, as part of the tech support staff, and as the main instructor for both Magic xpa and Magic xpi.

I had to write my own study guide to be able to teach Magic v8.x, eDeveloper, uniPaaS and Magic xpa to people who were new to these tools. Magic’s own training material only consisted of either printed PowerPoint presentations inside a binder or a self-paced PDF document. The latter did not follow a clear programming methodology path or even instructions on what components to select during the first installation of the tool.

Magic has a tradition, when it adds new technologies/features to the Magic development tool, to provide either no documentation or documentation that does not provide an organized approach for bringing this new technology/feature to experienced Magic programmers.

I sometimes joke that while you could connect any two people on the planet through six degrees of separation, connecting any two Magic programmers would only need three degrees.

Most of the time new Magic programmers are the result of a company that already uses Magic as its programming platform hiring new developers because the old ones are retiring. At the annual Magic gathering, new faces are few and far between.

Magic has always struggled to bring new developers to its tools. The company has been working in the USA for more than 30 years and still cannot find a way to bring new people to its side of the fence.

The solution is complex and has a long learning curve. I am not referring to its initial setup, which is very simple, but its use. 

Support is very bad. The main focus is placed on their own customers, big banks and big insurance companies, so those, like us, who are independent software vendors, are ignored. 

While I have tentative, though not definite, plans to increase the usage, the solution suffers from certain weaknesses, such as its essential lack of an internet web interface, a feature I really want. 

I rate the solution as a nine out of ten, since everything about it is good, save for the internet connectivity.

Everything having to do with the internet is underperformed, which is not good. This is my main issue with the solution.

The licensing is too costly. The runtime cost is very expensive, especially in respect of a big server with many clients. So too, the development tool is expensive, as is the yearly maintenance cost. 

While the solution is perfect for one who wishes to make local, standalone solutions on a Windows network or computer with an Oracle database, it is not well suited for internet-or-cloud-based distributed systems.

In the past, my company didn't need to add product developments, but over the years, you have to know more about your systems, and there is also more demand. For example, my company uses .NET to meet all demands, and it would be great if the Magic xpa Application Platform could offer more security and special functions similar to .NET.

In the next version of the Magic xpa Application Platform, I want tables or small programs where I can directly add expressions. I can do it on SQL, but it would make life much easier if that specification were added to the Magic xpa Application Platform. My company has been asking for this for years, but it's still unavailable.

When you have several tasks, you open a screen in a subtask in developing mode, then you don't see the parent screens.

Debugging .NET snippets is not possible, so it lacks the effects to solve problems. First you have to write and test your .NET code  in VS. Then you have to copy the .NET code in Magic XPA because there is almost nothing to debug it. 

The company has now released version 4 for web development. But this new release is still pretty hard to use for mobile and web development because it was not improved as much as we expected in the areas of user-friendliness or development-friendliness. In general, the interface is okay, but to achieve a final result, it is not very friendly for developers.

It could be improved by adding services that can be automated. This will be a major improvement in functionality.