Xojo 2017 R2.1 and Updated Roadmap

I’ve had a hellacious travel schedule the past six months so I apologize that I’ve not been up to snuff on Xojo news. Last week Xojo 2017 Release 2.1 was made public and earlier this week Xojo CEO, Geoff Perlman, penned a blog post about the short-term roadmap for Xojo. Let’s dig in!

First, the 2.1 release has a mere 16 bug fixes in it. This dot release fixes some critical Linux, and Windows bugs. A few IDE bugs were addressed and the PostgreSQLDatabase plugin was updated. Many of these bugs won’t affect everyone but it’s definitely worth upgrading from R2 if you have not already done so.

Now on to Geoff’s short-term roadmap. You can read the entire blog at https://blog.xojo.com/2017/09/19/the-short-term-xojo-roadmap/. To summarize it in a nutshell: 64-bit is taking them longer than expected.

To this I say, ‘Duh’. Anyone who expected 64-bit to be done by this time hasn’t been around very long. Xojo has consistently blown their estimates for most (all?) of their big projects. The transition to Cocoa from Carbon took a lot longer than originally announced. iOS took longer than expected. The introduction of the Xojo IDE from the Real Studio IDE was also delayed (as a side note I’m still amused at the accusation that I personally caused a 6-month delay in the Xojo IDE due to a blog post).

The point isn’t, ‘Oh my gosh, Xojo has yet another delay’ it’s that Xojo is very bad at estimating this stuff. Some of this is on them but some of it is beyond their control. Supporting mac OS, Windows, Linux, iOS, Raspberry Pi, desktop, console, and web apps is a mammoth undertaking. I am constantly amazed at the magic Xojo performs without my knowledge. To me it ‘just works’.

Apple, bless their hearts, changes things on a whim every year at a minimum (sometimes even more often than that) and causes Xojo to play catch up. And they usually do this with little notice. The current change being that the iOS Simulator will no longer support 32-bit apps so guess who has to scramble to make it work now?

Linux distributions seem to change daily. Xojo has to scramble to catch up with those changes too. They recently updated Xojo to use GTK3 and that was a big deal. They sort of had to do this upgrade because GTK2 was pretty long in the tooth. I can only imagine the internal efforts they did for testing.

Windows doesn’t change much for better and worse. The technology from ten years ago still works but that’s not necessarily the best ‘modern’ approach for Windows. Many users complain about flicker when windows are resized. This is a valid complaint but because Windows does not double-buffer their windows like Mac and Linux the older Win32 controls flicker like mad under many circumstances.

Regardless, 64-bit is the roadblock for other things. If you were hoping for Android, interops, or plugins compiled in Xojo to be released in 2017 you’re out of luck. This is what happens with a small development team. There simply isn’t enough man-hours to work around delays in some projects. This happens in ALL software development products but with a small team this impacts everything.

The only bit of good news from Geoff’s blog post is that the Windows framework is getting a big update. This update will ‘dramatically’ reduce flicker according to Geoff. No word on what exactly they’re changing though a thread on the forums from Aug 9 might give a little clue. Changing transparency seems like a little thing but perhaps they plan on implementing a double-buffering scheme. Until we find out more it’s not worth speculating. It seems like this new scheme might make it by the end of this year. So either this has been in the works for a long time or it’s not huge from a coding standpoint.

Promising services based on a product feature that doesn’t yet exist is dangerous. If you were promising Cocoa, iOS, or 64-bit versions before they were actually released you did your customers a disservice. I remember one XDC where a database ORM (Object Relational Model), similar to ActiveRecord, was shown with a lot of interest by those attending. I’m still waiting on it. The lesson is to not depend on vaporware!

So what are your thoughts on R2.1 and the Short-term Roadmap?

6 thoughts on “Xojo 2017 R2.1 and Updated Roadmap

  1. As soon as I was done reading Geoff’s blog post I came here for the analysis.
    It’s great to hear that they’re addressing flickering on Windows. My project has a scrollbar that makes controls slide up and down because they can’t all fit in the window at the same time. It works good on macOS but not so great on Windows. I was thinking about redesigning everything but maybe now I won’t have to!

  2. Actually in the days before Geoff’s famous blog post of Sep 19, we as Windows builders were jelling for stability above new features. Lot’s of annoyances are outstanding much too long and we got the feeling not taken seriously when it came to the fact the lots of windows builders cannot get beyond 2016R3. This because of the decline of end-user experience with 2017Rx releases.
    Now, with the message of Geoff’s blog we’re just a awaiting what’s coming and when. Most big changes take a few releases to get stable, and it will not be the first time developers releasing stuff while not looking behind. Often, the next day they just continue with nice new features. Sometimes ignoring customers disturbing them because of missed bugs during alpha and beta phase.
    Developers should understand better that they can do this fantastic job because of customers paying for their living and hopefully more. And customers should understand that it’s hard to keep everything state of the are while resources are limited.

    • I can’t disagree with this. Stability is MORE important than new features. However, there is a balance between getting important new things in (64-bit, HiDPI, GTK3, etc) and stability.

      Regardless of what they do someone will be pissed off. If they delay new features until they are 100% tested for months we’d complain ‘what’s taking so long’. If they rush new features out (which tends to be what if ‘feels like’) we’ll scream about stability.

      I have advocated for the every other approach. Let R1 be a big new feature. Let R2 be bug fix/stability. R3 new features. R4 bug fix/stability. And so on.

      Xojo does fix a lot of bugs each release. Unfortunately, I think we are at point where the big things are REALLY big and affecting many other things. 64-bit is bigger than any of us can appreciate. The switch to HiDPI is much bigger than any of us can appreciate (because it required switch to Direct2D and GTK3). The more moving parts, the more chances of critical bugs showing up.

      This should probably be its own blog post….

Comments are closed.