Hi Everyone,
We're in the final stages of beta now, and things are shaping up well! Thanks everyone, for your participation and great feedback. It's truly exciting to see dramatic new capabilities like Synchronized Application Control come to market, and the improvements in v17 as a whole, have been well received.
With every product release, Sophos internal IT participates in the testing process, by deploying the product into production environments. Some companies refer to this as "dogfooding". As in "Eating your own dogfood". It's a common expression that implies a fairly unpleasant experience, which often becomes true, when putting alpha or beta code into production environments. Through v16.5's lifespan, we've made big investments in quality, and while this beta is far from bug free, we can see the impact of that effort carrying forward into v17. Internal feedback has pegged this as the most successful firewall dogfoding exercise we've ever done, with no measurable service desk impact. This mirrors the feedback from many of you, that it's just worked for you. I'm very proud of our engineering teams, and the quality improvements they've delivered over the past year, and want to take this chance to recognize them for their work.
It's great to have success that can be celebrated, but I don't want to give the impression that we're ignoring the feedback from those of you who shared more critical feedback. Looking back through the feedback, there are a few general points I would like to address.
One bucket of feedback has been "Where is the feature I've been waiting for?". This is understandable, since most of those requests are relatively small, and XG has been in the wild now, for nearly two years. XG's goal has never been to add every feature that UTM9 has, but there are a number of feature gaps between UTM9 and XG that needed to be prioritized, after we launched v15. The features most commonly used were targeted first, and v16.5 closed off all high level feature gaps. v17 closed off a number of core capability gaps around email, firewall management, and troubleshooting. The list of gaps is down to a much smaller list of specific capabilities, such as several email management capabilities, and a couple IPv6 features. I know I'm glossing over a list of gaps that many of you have raised, but my point is that those of you who use UTM9 today, and have a feature reason that you're not yet switching to XG, don't worry. Your requests not forgotten or ignored, and the list of gaps remaining is really not very long.
In past, we've explained that XG version numbers would be year based, where the first digits are the year, and the second are the release within the year. This means that there isn't major and minor version numbering. The goal has been to have more, smaller releases. While the goal is still to have more, smaller releases, we've decided to go back to a simpler version numbering, and drop the idea of the first number being bound to a year. v17 is now the first of a family of releases, that we will release over the next year. Our plan is to do two or three feature releases before we get to v18, so look for 17.1, 17.2, and maybe 17.3 feature releases throughout the first few quarters of 2018. Each will have a number of improvements, but overall will each likely be fairly modest. Some features will be fairly significant and strategic, like new and improved SyncSec capabilities, others will likely be invisible to most of you, like OEM related improvements, but the majority will be to address user demand, and your feedback, like auth agent improvements, SSL VPN port customization and email allow/block lists. The list of features per version is not fully committed yet, but everything shared through these betas is under consideration.
In parallel, we are also working on v18. (In fact, we have already been for some time) v18 will deliver some core architectural improvements, that will address another class of feedback you've shared. Requests for behavioral improvements, like enabling/disabling interfaces, larger scale UX improvements through more areas of the product, renaming more objects, and more. v18 either solves these problems directly, or in a couple cases, lays the foundation to solve them properly. For example, renaming and disabling interfaces will be part of v18, but a fully responsive ui will become radically easier to build on what we deliver in v18. As it stands, v18 is too far out to go into any significant detail. v17.x releases will be the focus of discussion in the immediate future.
For those of you looking for more, v17 isn't done just yet. 17.0 is all but done, but you won't have to wait a year for more. Stay tuned, and we'll work to keep delivering features, at a higher standard of quality.
Thanks again to everyone who participated in this beta. It's not over yet, but the finish line is in sight!
