We had 3 such ISV’s in our section of the Expo all in the modernization field. We noticed a couple of missing ISV’s that are still in the space but did not attend the conference this year as exhibitors, we still saw their staff wandering the halls but no stand. We did have a quick conversation with one of them and they confirmed it was a short absence due to other activities and they hoped to be back next year.
One of the main reasons for Shield to attend is to network with our peers and have some face time with IBM about what they feel is happening in the IBM i space and any new announcements they can share. We were fortunate to spend some quality time with IBM staffers and have a good feeling about the future of IBM i and where Shield fits into that future.
Couple of points of note which we came away with. Open source is still front and center in the IBM i space, IBM is investing a lot of time and effort in bringing new features and products to the PASE environment which is good for the platform. However we are still lacking any emphasis in the Native IBM i OS area and ILE based applications, this is an area where we will continue to place our efforts in the hope of getting more people to join in and develop native IBM i applications which are open source.
I asked a person I admire and respect a question about the message we keep hearing that states RPG is the best language to use for business applications on the IBM i, I have always been intrigued as to why that statement is made so regularly especially when we discuss languages the IBM i supports. As RPG has become the defacto standard language used to develop applications on the IBM i it makes some sense to link it with the the IBM i, but why does that make it the best language for business applications? Apparently RPG has a number of attributes which make it easier to understand and program than most other languages on IBM i which has allowed many non developers to use the language to provide functionality to applications. I am not an RPG guru but some of the features such as string handling and known parameter sizing etc make a lot of sense especially when you look at the complexities you have to manage when developing in ‘C’ on IBM i.
One area where debugging code can be very interesting especially in ‘C’ is pointer management, keeping that from the users may limit some functionality but it does remove a lot of complexity so not having that support in RPG does make it a lot easier. RPG also has better string manipulation and handling but that comes at a cost somewhere. I agree that unless you really understand ‘C’ string functions you can end up with some pretty devious bugs which can be difficult to trace especially as ‘C’ lacks of bounds checking. So its easy to see why RPG is easier to use, however all of this can be negated by a programmer who knows the pitfalls and has the ability to code/test correctly.
In the end I am still of the view that while RPG is a great language, I still prefer ‘C’. Many of my applications are dependent on the use of pointers and complex string manipulation to provide the performance needed so it makes more sense to use it.
The new Power9 chip looks impressive, this is important as the world moves towards AI and the many other new technologies which are coming forward. IBM i gains from its increased performance but not sure how it will fit into the AI space as many of the AI developments we are aware of are aimed at the Linux space. Data is growing so being able to handle large data sets effectively and quickly could be where IBM i plays its part in the AI arena? We don’t have any specific plans for AI, but we are watching closely to see if its something we can work with to improve our products etc.
We saw a lot of attention to our HA4i replacement offer and a lot of interest has been expressed from the MSP community as they look for alternatives to existing HA options. Many discussed the need for migration tools which allow them to onboard clients without needing to purchase full HA licenses, something we will be looking closely at in the next few months. Others approached us because of the recent M&A activity and a fear that their existing solutions may become obsolete or priced out of reach so we will be following up with those to see what we can do to help.
We have been working on the Secure Sockets version of our HTTPGET program which will handle the retrieval of data from a secure (https) site on the web to the IBM i. What should have been a simple exercise turned into a nightmare as we explored the process and started testing. Its all been cleared up now so we will cover that more in the next post which should be coming out next week.
Until next time..