vAMSYS 2023 Roadmap to v5

[Lukas Jankauskas]

Before we dive in, I want to give you a heads up: this is going to be a lengthy update. I began drafting this several days ago, and my intention is to outline the plans and goals for the upcoming iteration of vAMSYS, which we'll refer to as v5 for simplicity's sake. I also invite you to engage in discussion and share any suggestions you may have.

Prelude
Long-time customers will remember the hay we made about v4 not being a thing and how we will continiously improve the platform. Should you need a refresher - I will point you to this neat forum post here:

Back in January of last year, this approach seemed promising, and we were gradually making progress towards its implementation. However, the past year has brought about significant changes, particularly the departure of half our development team as they pursued new and exciting opportunities.

Undoubtedly, this setback has caused some delays. Nevertheless, we've managed to address numerous bugs through hundreds of commits and thousands of lines of code changes, with a particular focus on Orwell and Aeolus (our management tools for administering all VAs utilizing vAMSYS).

Considering the direction we're headed, the process of continuous improvement has become slow and arduous due to the inherent challenges we face. Over the past year, I have made it clear in our Discord channel just how tricky it is to maintain outdated code and an inflexible user interface that cannot be easily updated in today's landscape.

Therefore, I am pleased to announce that vAMSYS is undergoing a complete rewrite, and v5 is already in the pipeline. Naturally, we won't be starting from scratch and pretending that vAMSYS doesn't exist. Instead, I will incorporate relevant portions of the existing codebase wherever possible, making necessary modifications to better align with our goals of enhanced scalability and easier maintenance.

Back in January last year, this seemed to be a good idea and we were slowly inching on the right track to deliver that - however, a lot has happened over the last year - namely having the half of the development team depart for bigger and better things.

This has obviously set us back a little - nonetheless, there were 100s of commits and thousands of code line changes to fix bugs - with most changes focused on Orwell and Aeolus (out little management space to administer all VAs using vAMSYS).

The way things are progressing, due to the nature of the things we want to change - makes the entire process of continious improvement slow and very cumbersome - there are many reasons for that and I think over the last year I made it clear in Discord how tricky it is to maintain very old code and old UI which cannot be easily changed today.

With that in mind, vAMSYS is being re-writen and v5 is in the pipeline. Naturally, we are not starting from 0 and pretending vAMSYS does not exist - I will heavily borrow bits of code where and when I can and where needed re-write it to better suit our needs of better expandability and maintenance.

Let's start from the beginning:

There's one particular issue that has been bothering me, perhaps unnecessarily. It's the heavy reliance on the vamsys.io domain for every pilot's activity, from registration to daily usage. I believe we have reached a point where this can be changed, allowing vAMSYS to take a less prominent role and instead showcase the VAs themselves. For instance, instead of logging in at vamsys.io, why not allow pilots to go to pilots.yourdomain.com?

Please note that the vAMSYS.io login page and the familiar select page will still be available for now. However, with the re-write, I want to provide VAs with more opportunities to be independent and not tied exclusively to the vamsys.io domain.

This involves more than just configuring CNAME records for a domain. I will explore possibilities for modifying various email templates and language. I will also remove vAMSYS logos wherever possible and replace them with VA logos. This will require a significant amount of work, but I believe it will be well worth the effort.

I find the current feature set of vAMSYS to be highly appealing. The ability to customize CSS allows many VAs to distinguish themselves, and we will definitely retain this option in v5. We aim to make it even more customizable to better suit your needs. Although there are currently no plans to develop a web GUI for this feature, it may be considered for future implementation if there is a strong demand for it.

 Phoenix

I won't delve into specific ideas posted on the vAMSYS Vision, except for those that have been closed where we agreed they won't be implemented. I do want to outline the overall goals for the Phoenix side of the vAMSYS platform. I've always felt that the Phoenix dashboard is too focused on the Virtual Airline itself. While we've made significant changes to make it feel more like a pilot dashboard, I believe more can be done. The Pilot Profile page, which is seldom visited, should be the cornerstone of our design. When a pilot logs into your VA, they should see information about their flights, bookings, and past PIREPs. That should be the primary focus, along with features that enhance their engagement, such as prominently displaying events and a flight list that prioritizes friends instead of showing everyone currently flying in the VA. There are several ideas posted on Vision that align with this goal, and I'll make every effort to implement as many of them as possible.

Since we're doing a complete re-write, there will be significant changes to the flight booking routine. Your current needs are vastly different from when the page was initially developed, and it struggles to handle thousands of routes and destinations. Eliminating the pairs functionality will give you more granular control and reduce our code overhead. It may require more work on your end, but the new system will provide you with greater power to set up repositioning flights instead of relying on a couple of on/off switches. The same logic can be expanded to limit jumpseats only on specified routes. I also plan to bring back better filters to filer out destinations, maybe with some global pilot settings - the sky is the limit and even though I won't aim that high, I hope to make big improvements in this area. Additionally, I'll make general improvements and updates to how we handle routes, capturing and displaying information like service days and scheduled departure times to enhance pilot immersion. There will be more note fields and other enhancements that you've suggested over the years. Please hold me accountable to implementing these changes as they are crucial.

I'll also revisit the maps once again. It has always been my dream to unify vAMSYS behind a single map, as opposed to the current system of three separate areas: the dispatch map, flight map, and route map. Finding a solution to unify them is challenging, but if I or we can come up with something, I'll definitely give it a try.

Moreover, the long-awaited expansion of the event system will finally arrive with v5. Specifically, tours and repeating events will be introduced, addressing a critical and highly requested feature that has been sought after by nearly all VA owners.

Another goal I have is to provide better support for charter VAs. Currently, it's a nightmare to allow pilots to fly from any airport to any airport. Let me illustrate this with an example: let's say you are a charter VA serving 200 airports. To allow flights between every airport, you would need a staggering 19,900 routes. This not only creates a logistical challenge but will also pose difficulties in any VDS system as I see it. It's simply not feasible to handle and make usable such a vast number of routes. There is much to figure out in this regard, but one potential solution to consider is offering a route-less VA option. This means that no pre-defined routes provided by the VA would exist, and pilots would simply select their departure and arrival airports. It's an idea worth considering to address the challenges faced by charter VAs.

In addition to the goals mentioned earlier, I am also excited to explore the implementation of "pilot awards." This gamification feature would allow you to award pilots with badges based on their flying performance, participation in events, and other noteworthy achievements.

While this idea and the goals above are just the tip of the iceberg, I believe there is much more potential to uncover. We will have in-depth discussions at Team vAMSYS headquarters and actively seek your opinions on our Discord. Furthermore, we will always refer to the ideas posted on the vAMSYS Vision for inspiration and guidance as we develop this and other features.

Phoenix is almost as neglected in terms of codebase as VDS, and it's starting to show, especially with the increasing load due to the growing number of routes and pilots. vAMSYS v5 aims to deliver a better pilot experience, be more pilot-centric, and enable pilots to engage with your VA more effectively than the current version. I am very excited to start tackling this bit of the platform.

Orwell

A significant portion of Orwell will serve as the foundation for the new version. The changes I've made to Orwell over the past year have been invaluable in adapting to new practices and have served as a valuable testing ground. However, there are additional enhancements to be made, as I believe Orwell does not currently fulfill its intended purpose. Allow me to outline the goals for the new Orwell.

Just as Phoenix is designed for pilots, the new Orwell should be a staff-centric dashboard tailored for VA staff and owners. It will provide an overview of your Virtual Airline's status and offer in-depth insights into statistics. Currently, it is a blend of settings and PIREP review, but I intend to transform it.

In my vision, the new Orwell will be teeming with tables and statistics that delve into details, enough to make your head spin. We possess vast amounts of data, and it should be compiled in a meaningful way and presented to you. A modern and practical statistics feature is essential, even though I haven't finalized the specifics yet. I strongly believe that this is the area where Orwell requires the most improvement—to provide you with vital information about your pilots, their bookings, and your routes—because running a VA effectively without such data is, in my view challenging, if not impossible.

In the upcoming release, we aim to address the challenge of effectively managing your pilots. Specifically, we will focus on re-writing the Activity feature, particularly regarding pilots re-joining. Additionally, we recognize the importance for VAs to have the ability to ban their pilots from their VA without impacting other customers. We will also aim to provide a mechanism to flag such cases for vAMSYS review, enabling appropriate platform-wide bans, if necessary.

Lastly, I intend to incorporate the Vulcan Storage Service (VSS) into the equation, either during or shortly after the v5 release. For those who may have missed the Discord discussion, VSS is designed to replace forum downloads with a more robust implementation. It will grant you better control over file access, allowing you to manage who can download your files. More details about VSS will be shared as we embark on its development process.

Orwell is meant to be a tool rather than an immersive experience. With that in mind, my goal is to transform it into something vastly more useful and powerful in the upcoming v5 release.

Messaging & Communication

I believe it's important to address this topic separately, as it has been a subject of numerous discussions on Discord. The current system of alerts, notices, and notes is a patchwork of features added over time, without effectively addressing the challenge of pilot communication.

There are numerous ideas floating around, both in my mind and in the vAMSYS Vision, regarding how we should tackle pilot communication. This encompasses various aspects, such as implementing NOTAMs for rule changes, improving communication regarding invalidated PIREPs, and enhancing staff comments in general. These aspects span both Orwell and Phoenix and will receive significant attention and development time. I envision a superior NOTAMs system to replace the current notices, a fresh approach to alerts, and an improved tracking and acknowledgement system for failed and invalidated PIREPs from the pilot's perspective. I encourage you to continue sharing these ideas and ensure they are documented in the vAMSYS Vision for further consideration.

Infrastructure

There are no major changes planned for the next 365 days in terms of infrastructure. I am content with our current setup, and although there may be some improvements that could make administrative tasks easier, they are not a high priority at the moment. Managing the infrastructure is not a significant time drain, and it remains at the bottom of the to-do list for now.

VDS

I have created a dedicated forum topic about VDS and won't delve into it further here. Rest assured, it will receive the same level of attention and necessary improvements to make it a useful feature.

API

The exciting aspect of current development progress lies in the realm of API expansion. Although our statistics and map json are currently limited, the progress of v5 in my sandbox development area shows promising signs for enhanced API interaction. I can envision a future where a self-hosted vAMSYS utilizing our API can be a possibility, although it is not my primary objective. My focus lies in refining the backend logic and functions, which will eventually pave the way for a global API as a consequence.

In the meantime, I would appreciate your input and guidance regarding your specific requirements. I'm more than willing to expand API interactions and incorporate functionalities that meet your needs. Whether it's for your website, Discord bot, or even if you desire to create your own route importer. While expanded API may not be my top priority, establishing a robust backend infrastructure aligns with our ultimate goal and will enable better API access.

Pegasus

While not directly related to the v5 development, I'd like to briefly mention our future plans. Please note that these are early discussions, as our current priority with Pegasus is to upgrade it to the latest dependency versions and release a stable version with a functional upgrade routine.

For the v5 release, no significant changes are planned for the Scorer. All scorers will remain unchanged from v4.

Looking ahead, we are considering improvements such as continuous delivery of flight logs. This means that the log will be sent from Pegasus to vAMSYS in real-time as the flight progresses, rather than being sent in its entirety at the end. This opens up possibilities for more interactive real-time flight tracking and provides a better solution for incomplete flights where, for example, a simulator crash interrupts the flight and there is no way to resume it.

I am confident that Peter, our team member responsible for Pegasus, has many more ideas in store. We will share them with you at the appropriate time as our development progresses.

----

And there you have it, the broad set of goals for v5. We will work closely with all of you, aiming to fulfill as many feature requests from the vAMSYS Vision as possible. Additionally, I am considering providing regular 'Developer Diaries' to keep you updated on the development stages, give you insight into our progress, and provide an opportunity for feedback.

Now, you may be wondering about the timeline. There is no official or set date in mind; it will be completed when it's ready. However, my aim is to have something beta-testable before the end of the year.

If you've made it through this entire post, congratulations! Thank you for your time and for choosing vAMSYS as your VA Partner. We appreciate your support and look forward to delivering an even better vAMSYS experience with the upcoming v5 release.

If there is anything specific you wish to discuss - our Discord server and #staff-chat is the perfect place for that. I am also hosting vAMSYS Sunday tomorrow (16th July) at 12:00 BST if you want to join and say hello.

Till next time!