At least in the Windows IDE. I moved the slider bar up to see what could be under it and I’ve noticed there is something wrong with the refresh of the pane underneath it. No biggie, but it may illustrate the refresh problems we have when we put controls on a canvas in Windows…
I just don’t see how 2015R2.1 could have received enough testing to be released as quickly as it was. It just doesn’t work right in Windows and can’t be used to create releasable software. I guess I’ll have to put off renewing my license for at least another 3 months and hope they get things fixed by Q3. This sucks!
I disagree, the point release addressed specific issues that were resolved.
Rather than being vague, could you cite anything specific? Are these things that we specifically broke or things that you wanted fixed which weren’t?
I have the same feeling, it’s just getting a bit frustrating that panes seems not to be refreshed properly since release 2015r2 and 2015r2.1. (Windows)
Although several issues are addressed it seems that some new have been created, or at least slipped thru.
I guess the majority of the testing is been done for OSX and not Windows and not Linux. Probably the right time to do a release that is just for fixing and stabilizing, not adding any new features at all.
[quote=182539:@Joost Rongen]Although several issues are addressed it seems that some new have been created, or at least slipped thru.
I guess the majority of the testing is been done for OSX and not Windows and not Linux. Probably the right time to do a release that is just for fixing and stabilizing, not adding any new features at all.[/quote]
There were 8 or 9 bugs reported just as we released r2 which we felt needed to be resolved before XDC, and that was the reason for 2.1.
We do try to test on all platforms, but we also rely on our beta testers to help us in that endeavor. 2015r2 had 4 weeks of beta testing, if I recall correctly, and we do our best to address regressions when they are reported. That said, the sooner a regression is reported during a beta cycle, the more likely we’ll have time to fix it. If testers wait until we reach final candidate to start testing, that doesn’t help us very much.
You’re right @Greg O’Lone , if all PRO users should at least spend some time testing their actual projects with the beta’s , a bunch of errors would be detected early. And as we all know, the earlier an issue is detected, the shorter the loop to fix it. (V-model).
I am to blame too by mostly waiting for the release.
But you’ve got me now Greg and I will start using beta’s to check up with some projects each time. Doing so I avoid surprises after the commissioning of a new release and help R&D team. Again, when every PRO does so, the whole range will be covered without spending too much time per developer.
[quote=182552:@Joost Rongen]You’re right @Greg O’Lone , if all PRO users should at least spend some time testing their actual projects with the beta’s , a bunch of errors would be detected early. And as we all know, the earlier an issue is detected, the shorter the loop to fix it. (V-model).
I am to blame too by mostly waiting for the release.
But you’ve got me now Greg and I will start using beta’s to check up with some projects each time. Doing so I avoid surprises after the commissioning of a new release and help R&D team. Again, when every PRO does so, the whole range will be covered without spending too much time per developer.[/quote]
Keep in mind… We DO understand that you guys have lives and businesses to run (heck, even we do), but if 20% of our pro users tested each beta cycle, a lot more of these bugs would probably be caught.
I usually switch all my development to the beta and whatever I find I report. And I do find quite a bit.
But even if more pros did the same, it is not a guarantee for bug free. I know my programming style and the kind of apps I do play a big role in the possibility that I find bugs that others may not, or, on the contrary, that I do not find them.
Every developer has his own pet topics and favorite controls and techniques, that will contrast certain things and tend to hide others.
One thing I find sad when some people go yelling in the forum and yet, would not touch feedback. Some even say they will not file report because it is not their job. Sure, anybody can act rude and refuse to show common courtesy, or consider themselves above the petty crowd of the Xojo community. But then, I suspect such individuals would probably not lift a finger if they saw an old lady have her bag snatched by a thief…
We get what we bring in. Sure, there are things in Xojo I dislike. But overall, there are many more I consider vital for my business and my personal programming enjoyment. Would not that be worth spending three minutes reporting a bug, would it be beta or not ?
You probably could increase bug reports in the beta phase, if you don’t limit your tester base to only the users who have the Xojo pro version.
Xojo Inc. is actually quite pliable with user concerns. If you have a true conviction to aide in beta testing, bring your concern to the Xojo team and they may be willing to meet half-way. As a side note, some bugs simply escape beta testing because they don’t rear their head in the RC version, or merely go unnoticed while developers spend time ensuring the new release “plays well” with their existing code within projects. With each new beta, I personally spend a good 2-4 hours within the environment a day, besides using the stable versions for development, usually side-by-side to test the same code and look for visual discrepancies; reporting any hindering flaws found. If you bring your concern to Xojo, they listen.
[quote=182558:@Michel Bujardet]
One thing I find sad when some people go yelling in the forum and yet, would not touch feedback. Some even say they will not file report because it is not their job. Sure, anybody can act rude and refuse to show common courtesy, or consider themselves above the petty crowd of the Xojo community. But then, I suspect such individuals would probably not lift a finger if they saw an old lady have her bag snatched by a thief…[/quote]
I have used this product since the days of CrossBasic, and I get an awful lot of fun out of it and like it very much. I have really been out of it for quite a time now, so my viewpoint is probably out of date.
But Real, when it was still that, always pushed and pushed new features to the extreme prejudice of bug fixing. Marketing before Engineering, as the paradigm goes. Way back I would report the odd issue on the NUG ( which now seems something verboten ) or email and bugger all would happen. Then they wanted formal bug logging and just refused to look at any bug not so logged. Even if you found a funny and told them, no effort made by them to log a bug. Hell, if someone tells me I have a bug, no matter how, I’ll look into it. Their way or the highway. Highway for me.
Then, release after release, used because of a required new feature, would break other things. I have spent probably 3 months of my life in the old days just trying to work around compiler bugs, trying this, trying that. Bugs not fixed for year after year. Lots of “new” features that were often just as buggy.
You’ve seen it before:
99 little bugs in the code
99 little bugs
Take one down, patch it around
137 little bugs in the code.
Now they are all snooty, and only “Pro” users get the betas these days, no more NUG it seems, only a managed and controlled forum.
So, while I would probably help an old duck having her bag snatched ( unless her assailant was pointing a Makarov ), I’m not going to bother logging bugs. I’ll just offer a prayer to St, Jude, the patron saint of lost causes, and hope for the best.
I’m going to qualify my statement before I even make it by saying that I really don’t know for sure what all ramifications there would be within Xojo to opening up the beta testing to a larger crowd (e.g., would it possibly make for too many inputs needing review by too few engineers?). But, having said that, it’s hard for me to understand why you wouldn’t want to open it up to some larger audience than just the Pro license holders. Here’s the two thoughts that occur to me (and bear in mind, I AM a Pro license holder myself and renew every year as such):
Does being a “Pro” license holder supposedly imply that you are a “Pro” programmer and therefore of a higher caliber than a “Non-Pro” license holder? Obviously not! I only have to look at my own elementary knowledge level compared to the likes of many of the contributors to this forum (many who don’t hold Pro licenses) that dwarf me in capability … that’s why I ask them questions and not the other way around.
Does Xojo view granting beta testing authority as a “perk” for rewarding people who go the whole way with buying the Pro license? I truly hope that’s not the case for all the obvious reasons.
Does Xojo feel that Pro users can supply broader feedback since they have the entire platform selection? I would hope not since most Pro users probably use one OS or another the vast majority of the time and therefore aren’t going to test something on all platforms anyways. And if that’s not enough of a reason, then refer back to item #1 above.
So, when you go “Pro” with Xojo, you normally do so because you either need cross platform with all the possible bells and whistles you can get to maximize your programming opportunities, or you just have the money and want the “top of the line” offering. I’m sure I’m leaving out some other possible motives there, but you get my point … None of that qualifies anyone as a beta tester.
I’m of the belief that Xojo ought to revisit their approach to beta testing to somehow maximize the return information they get more on a “knowledge” basis as opposed to a “license” basis … although I’m far from being Mr. Spock, it does seem logical.
And for the record, I do see the wisdom in using Feedback as the solitary tool for aggregating bug reports and feature requests … trying to capture any and all such inputs from multiple sources (e.g., this forum, NUG, S-O, and who knows where else, etc.etc.etc.) would be a small nightmare at the least and a boatload of extra work (that I’d rather have them using that time productively to fix the things that are wrong).
Back in the old days you had a lot more people in the beta program. Lots of downloads but little feedback. People were using it as a way to get the newest release. So they scaled it back and the easiest way to do that was to restrict it to “Pro” Licenses. It was an indicator that the developer was spending enough time in the product to spend the money for that license.
I do know that exceptions have been made. If you REALLY want to be a beta tester then you contact Xojo and convince them that you want to do testing then go for it.
Realize that being a good beta tester isn’t just running your projects through it. It’s creating good bug reports with reproducible steps and attaching example projects and then following up when the Xojo team has questions.
Lots of reports ? lots of actionable reports.
That’s one thing we clearly found when we used Fogbugz.
With respect, I don’t believe that this depends to users of the Xojo Pro version. If that would be the case, Porsche, Mercedes and Ferrari drivers were the best car drivers in the world.
It’s a personal attitude of how to do things properly. The above described bug could also be found from not so professionell developers.
Do the professional programmers that work around the clock have enough time or be willing to test, report any found back, give feedback and then answers possible questions from Xojo?
That’s in line with what I refer to as “knowledge-based” beta testing, Bob. Once again, everything you mention has nothing to do with “having a Pro license makes for better beta testing”. I know that opening it up to the “world” is not the right answer … but Xojo’s got some pretty smart people working for them and I’ve got to believe if they put their collective heads together, they would come up with something somewhere in-between the past and the present that doesn’t “give away the farm” yet yields more meaningful feedback and results. Like I said, in all fairness, I don’t know what goes on inside Xojo, but I do wonder things like, “Have efforts been made to benchmark those in the industry who may do it better? Have the approaches of other companies in other industries who have a reputation for solid releases been investigated to perhaps clone best practices?” Brand image can make or break a company … I would think that Xojo would see improvement in this area (beta testing) as a clear opportunity to build additional brand equity in that regard.
Throwing beta membership wide open was what we used to do.
And, as Don surmised, it didn’t work very well.
So we made changes to that and included Xojo Pro license holders.
We DO however also nominate individuals for inclusion in the beta program based on their history reporting bugs.
If you write good bug reports that can lead to us including you whether or not you have a Xojo Pro license.
And we do, from time to time, purge our beta membership lists and reset them so its not a permanent membership like it used to be.
Ever wonder why? People hoping the latest showstopping bug had been fixed.