Tasting Tomorrow

You never know where the future might take you. Tomorrow is going to be different from today as today was different from the day before. Even in “Groundhog Day” the same day was different every time. They say that you cannot predict the future. They even say that you can’t go back. Both are true of course but some people have a knack for simulating both.

Citrix is not immune from this either. The trends and tides of the world carry the companies of the world with it. As self appointed soothsayer of Citrix’s future environment (just for fun… don’t quote me thank you) (second disclaimer – this is my own opinion and does not reflect the position of Citrix or what it believes will happen).

There are a number of factors at play here:

1. Bandwidth is going to increase greatly in the near future as latency is reduced
2. Raw CPU power is going to continue to increase mostly from the efforts of having more processors per box
3. Cost of exchanging data will continue to go down
4. More of the world will be wired or wireless as time goes by
5. Devices will continue to get smaller and more specialized as consumers find new uses for them
6. Storage will explode in capacity as it becomes possible to store data either internally or on the Internet
7. Web technologies will finally approach the interactivity and richness of typical Windows or Mac applications
8. Security will continue to get more intense and will require some big redesigns in existing systems
9. Users will want to know less about how things work and want their environment to be as seamless as possible
10. Distributed solutions will hit the mainstream and allow for hosting of components/applications from anywhere

This is just a sample of the kind of things that aren’t that hard to see within the next 10 years. Of course some are more obvious than others and some might not happen at all (like the full application level support on the web with web tools).

The point is that most of these are long term trends that will continue to expand.

The areas that I think would best help Citrix to address these items are:

1. OpenGL/DirectX/3D Graphics remote rendering to support highly demanding tasks like imaging and design
2. PortICA for other platforms to allow for a wider range of applications to deliver
3. Tighter integration with Virtualization platforms to guarantee performance and reliability of applications
4. Seamless integration of applications regardless of original platform (remote and local combined)
5. Unified Desktop model to allow the hosting of these disjunct set of applications
6. Abstracted server/client model which would allow an easier model to use on new/different platforms
7. Re-access the current product suite from the ground up to form a unified security model (potential redesign)
8. More tools included in Citrix software to access not only user counts but also the end-to-end management of sessions
9. Continue to evolve older Citrix products based on changing environments and customer requirements
10. Capitialize on the latest advances in computer technology and inject this intelligence into our products

This is not necessarily a comprehensive list but it is intended to stir up some ideas internally and externally.

One of the things that I have observed over the years is that it is often the work that is considered routine that is the most interesting to customers. Recently I have been reminded of this related to COM port redirection in CPS. It has been determined that with certain configurations it does not operate as best it could. If there is a relatively high latency (>200ms) over a WAN, it can take quite some time to get the port to open based on individual settings taking place after the open. The point is that COM port redirection was not implemented with that much distance (or time) between client and server. Normally it would work out okay even though it took so long. The user would just suffer the additional time. Unfortunately in this case it actually leads to a timeout at the software layer and the transaction is ended. In this case, this was found during a pilot at an undisclosed customer.

There is a possible solution but it will take time to design, develop and test. The most likely outcome is that the COM port redirection will be changed in the future to allow for buffering of requests so it will not need to acknowledge each individual transaction before continuing the next. It’s tempting to consider just trying stuff but long term this really does not make sense. Perhaps it would benefit the customer to see if this is a step in the right direction and also to show that we do care that this does not work as expected. However, it would still be better to try solving this with the bigger picture in mind for all the other customers that we have as well.

I’ve told you this story really just to prove two things. First of all, customers often care about what hasn’t been considered for awhile. Second, these issues are usually not rocket science and can be solved. The most daunting task related to this is evangelizing these kind of problems to get it solved more universally than just a fix for one customer.

Perhaps you can help me out there. I’m not advocating being a support person but I’d like to hear some feedback about the kind of things that you find lacking in the current Citrix product group. I’m looking for things that aren’t overly exciting (read new) but could potentially be solved in the not too distant future. I’m not expect tons of feedback but see this as your opportunity to share with the world what you wish Citrix would do better when it comes to the whole Citrix experience. I’ll probably be more receptive to issues related to CPS but if you want to tell me about other areas, go right ahead.

Thanks for that. I hope you believe that I really am interested. Because I am!

Please send email to my email address in the “About” page if you don’t want to submit comments here. Thanks again.

About Jeff Muir

Live near Brisbane, Australia. Software developer currently focused on iOS and Android. Avid Google Local Guide