This what I’ve made for myself in Purebasic 
Not for Web Applications but for small and solid JSON Web API Servers:
2 Likes
Many people are hesitating to leave while tons of sourcecode and knowledge are speaking their own language. I saw customers which invested years of work in Xojo to learn that they have lost all of it. That’s why iOS and Android are looking like nice targets and showing them the other way around that it is nearly impossible to switch. You’ll need to learn another language, that will cost you one year of work to be on the same level you where with Xojo. So people try to get the stuffs running and learning that it is not. We could debate about it. It will not change the facts, people having issues but also much money in it. Spening time for learning another language is helping to get out of it. Using languages like Java or C# are speeding you up to use the state of the art technologies. Learning javascript and python helping you to bring up state of the art web interfaces for the frontend and using of AI tools which mostly are written with Python. People can write with the right toolchain for even all targets like Desktop, Web and mobile.
2 Likes
I think the bottom line is the Xojo is not going to get better without hard work and dedication on their part. As history suggests they probably don’t have the fortitude, willpower, and necessary skillsets to accomplish this.
5 Likes
I’m sure @jay knows all this
I dont have an answer to the issues he’s encountered. I’ve not run into them since we moved everything to URLConnection (and some uses the Chilkat plugin)
Beyond that hope isnt a business plan
As so many have noted Geoff and team just do not seem to care and so it makes the lives of people like Jay hell
While painful moving to a different toolset that has way more users, more active development & bug fixing, and a more reliable release process is the only way I’ve found to make my life, and that of my clients, better
I’m truly glad that when something breaks its 100% MY code or my mistake and not the frameworks bugs or issues wrecking my project
YMMV
4 Likes
this!
3 Likes
I’m truly glad when the linter tells me I have a problem long before I try to compile. Half the time the only way I know Xojo has a problem (before compiling) is when AutoComplete stops working.
3 Likes
Yeah I love Rider as an IDE
It highlights the errors as I’m working
And it usually proposes the correct line of code
I just hit cmd+return a TON more than I thought I might to accept the proposed code
Yep
Just like this post makes it clear that figuring out what version of Xojo is safe for any app is hard if not impossible
The bigger your app the MORE critical knowing is
Amazing project! I was looking for a small wrapper on top of a MariaDB which routes HTTP requests to stored procedures. What could be a great starting point, I’ll definitely check it out. Thanks for the hard work!
I’ve been on and off the Xojo train for 20+ years. Whenever I have pressure to build something quickly and I don’t want to struggle with a huge stack of HTML,JS,CSS,React,Backend,Webpack and all the crap, I often catch the Xojo straw and can build a working prototype very quickly.
But then, I encouter bugs like every hour of coding in Xojo and spend another hour to properly report it just to see it will never be fixed. It’s sad to see that their focus on quality is so limited, which destroys all the potential Xojo as a product has.
I’ve two Xojo apps left, a web analytics dashboard and an internal Mac app that I use daily and don’t want to spend the effort to rewrite it in some other language. My lazyness will cost money next time I try to compile it without a renewed license.
IMHO Xojo should release at least part of their product as open source, so people can fix bugs on their own. Xojo could gain a free workforce and improve that for the product. I would estimate that they won’t loose a dime on licensing, quite the opposite.
It’s one thing I really appreciate when working with IntelliJ based IDE like Rider, RustRover, CLion, Pycharm and so on. Get your syntax bugs wide before compiling. And yes, it does it’s job. Was speeding me even up when I was learning GO with GoLand. Their Idea behind their IDE concept is brilliant.
Only thing I am missing is a usable UI Editor for Java Desktop. But with JavaFX I have a declarative way (fxml) and a UI Designer if I want to use one (Scenebuilder, even for free) or for a few bugs per year I can also have JFormdesigner for Java Swing.
At the end I can say: this kind of IDE is for me the best way writing Software. I was also working with Eclipse and Netbeans with Java, C and C++ but Jetbrains has the best implementation. And what I also like is that I can work with several languages from Java over Kotlin, Rust, C and C++ and even C# with the same IDE and the same IDE usability. Makes work faster and more controled. And at the end I get a better code quality while it comes with the needed tools for quality checking. And that even not for only one language but all languages we use in my company.
Something I will never understand is: they are not even thinking in this direction. The other way around. And I bet they are loosing customers and getting less new customers because of the big amount of bugs which never will get fixed. Their policy to do fast release and fixate themself to new features instead of concentrating on fixing bugs is not helping to improve the product or to impeorve the usability of the product. At the end you are getting more functionality and at the same moment more bugs.
On the other side we should not forget: Xojo is a really small company. Other Vendors having much bigger teams. For example GO has around 2000 contributors to their soucecode while Xojo has maybe 10 (last named list of developers was 6, maybe we count today 10 but I doubt that). With this extreme limitation comes the problem. Implementing new features costs man power, bug fixing costs man power. But also many other jobs inside of the development are costing man power. That’s making it complex to get the bugs out of it.
1 Like
Developers ?
Still a tiny crowd
Javier Menendez
Paul Lefebvre
Ricardo Cruz
Travis Hill
William Yu
The entire company has < 10 people now
That’s what I wanted to say. So let me say 10 versus more than 2000. And we are speaking about GOLang not about Java or C++. Python or even C#. That exactly is the problem. A too small team can not do all the jobs needed to be done for all the platforms and the big amount of Bugs. And they can’t grow up the company cause this means also: you may need for this many millions of Dollars. That said I realize: they can not get to a status a professional tool needs to have. One of the MUST HAVE for a Vendor like this is even a testing unit which would need to be at the mend minimum 30 times bigger than the entire company.
I can see: they try to do their best. But that is not enough. Only if you say: “okay, it is a citizen developer System not for the professional Software development with a limited functionality and a limited support capability” you may come to the conclusion that this will work.
Looking on the Xojo Customers I would like to say: they try to suggest the opposite: we are a full professional development system. That’s what comes out when a new customer is reading their forum and their marketing. Nobody can get out fast that it is so.
Also nobody can see how big or even not big is this company. That makes it nearly impossible to get before you started and bought the Software.
shorthanded is the only way to describe them
Oh Geoff could probably do it
but he wont
he’s to comfy
to risk averse
and I’d bet he honestly doesnt see why they might need to do it
They literally do NOT require unit test from the developers as part of their work
They could just start there
Their regression test suite hasn’t grown in years
That I’m certain of
1 Like
If they would test with a test department and invest manpower they would get out all their bugs before customers finding them as showstoppers
So simple
They arent even at that stage
they dont REQUIRE developers to create tests that show the bug (which many customers have done already) and then use those to show the bug is fixed as part of their internal testing
its so amateur and has gotten wore
at least when greg Joe & wi worked there we did that
not because it was required but because we knew it was a good thing to do to make sure bugs stayed fixed and new code didnt suddenly break something
the test suite every time we found & fixed a bug
thats stopped
no, he is just collecting his pension, no interest to invest into this product anymore.
How old is he? I am telling for years, Xojo will end like PowerBasic.
He doesn’t care about anything what we talk here. He rather would hire somebody for Marketing than a good Developer. For short profit with new customers, not for the long run with existing customers. it’s so clear.
3 Likes
The company I used to work for had a specifc testing plan
- The developer tested their own code and certified that it met the provided specification to the best of their knowledge
- The project manager then did the same
- finally the end-customer (in this case an internal dept) certfified it did what they wanted/needed
In some cases, where the code was external customer facing, there were other tests by the DBA etc.
1 Like
He’s never hired a marketing person worth their salt. He did have Matt Q a few years back, but that was at least 15 years ago, and Matt was focused more on direct corporate sales.
Having been intimately involved with the company for years, I can tell you that marketing was never anything, period. He even went as far as to once tell me, “Apple doesn’t market their products, they just sell a great product. That’s what we’re going to do.”
1 Like