XYplorer Beta Club

Is this the conversion software you used?

https://visualstudiomagazine.com/articl ... -core.aspx

No.

Hello,

I read this thread with a good deal of sympathy. I posted back in April? when you first started the foray into .Net. I have a couple of thoughts to share.

My migration from VB6 to VB.Net was hairy... it took me a couple of -months- at least to feel I really understood things better. The subsequent migration of that to C# a couple of years later was far less jarring, and in the end I simply like C#'s syntax and style much better, not to mention the community is massive around it compared to the VB.Net world.

When we converted our larger projects, we first identified pieces and parts which we wanted to experiment and run tests and just set up projects to test theories. After we had achieved success on those areas, only then did we start building things together. It ended up being a very good approach.

I'm really shocked to learn of performance issues; everything I've seen in the real world is a massive speed increase in most areas. If it's due to interface slowness then I have one thing to say about that: this is a terribly awkward time to attempt moving a desktop app to .Net Core. I assume you would be porting to WPF, considering the long-term viability of WinForms is not there, and UMP is a dead end. Plus, you get all the nice DPI awareness baked in. But it's so different... all the layouts, etc. have to be so carefully thought out. Anyway, .Net 5 might make things easier... .Net 6 (2021) would be what I would wait for, if you could. By that time things should have settled down from this multi-year .Net Core nightmare migration and things will be more solid. It's just a thought. It also gives you time to experiment with certain parts of your code between now and then and identify shortcomings or challenges.

On that note, JetBrains Rider is a very serious contender. We do a lot of work in it once interfaces are built, although Rider is getting better in that arena all the time. Rider is so much faster than VS, you'd be shocked... plus you get all the goodness of Resharper without the slowdowns it introduces in Visual Studio.

Considering the group replying here, perhaps if you ran across particular issues you could start another forum for help troubleshooting certain portions of code or performance problems. I'm not saying expose IP, but rather like you did earlier... "I have a problem with repainting XYZ... "

Anyway, love your product and have been using it for years. I'm sorry to see you despair a bit, but I do hope there's a light at the end of the tunnel for you (and us) and it's not an oncoming train! :-)

Best,
--J

Hi J,

thanks for the thoughts. (btw, I never tried to move to .Net Core.) Anyway, I've moved on and happily don't look back. If anybody feels like doing this job, make me an offer I cannot resist and I sell you the code.

Don

Hi Don,

I'm most definitely overstepping my boundaries here, but eh, it's a forum so I feel I may express this:
XYplorer has significantly changed my work life. I've taken the time to learn about 40% of the features, which I feel is already huge, and I'm getting so much productivity and happiness out of it.. and there's scripting and other advanced features yet to explore (hence the 40% I learned .)

It isn't so much about having purchased this software; given what I got, it's cheating you actually.
It's about this having become a tool I am using, on average, every few minutes and enjoying how well it works every damn time.

This tool can't die in 2-5 years because it may no longer be feasible to compile VB6 to target, or because 32-bit support fizzles out, or.. whatever.

I was jumping for joy to see the initial announcement. I get it.. it was enthusiastic and a little bit premature, but nevertheless.. I was SO happy. Then it quickly became very sobering to see the progress report.

And I understand exactly where you are coming from. I develop software. I have to maintain some 90's VB6 code, I also code in VB.NET, I do C programming, I really get what the challenges and difficulties are and how large a mountain of work and challenges this is.

But I implore you, don't give up on it just because of draw performance, or any single matter that isn't so elemental that XYplorer could not be implemented (which I challenge is probably nothing.) A .NET port right now may be DOG SLOW in draw performance. But by the Gods, I would love to help on a sub-forum for developers to maybe find solutions, give suggestions.

At the very least, a DOG SLOW port of this software would ideally retain its functionality and properly encapsulate the UI logic from business logic such that UI logic could be replaced as .NET improves or machines become so fast that it just doesn't matter anymore.

I bet it will take at least 18-24 months to even get most of your code ported and just work however slow. Then another year to start making use of more modern OS facilities, threading etc.

But in the end there'd be an imperfect, functional new breed of XYplorer that could be improved, perhaps with community involvement.

I doubt you'll be able to find anyone who'd pay for the source code to do all that work. It needs passion, and it needs intense knowledge of the inner workings. On a 125KLOC project with this much behind it, and the projected revenue and headache of having to support existing users, that'll be a tough sell.

So to repeat myself: is there some way you can open not not the XYplorer source, but issues like draw performance or whatever you run into, with small projects people can compile and run and try to find solutions for, just for the sake of making sure XYplorer will be alive in 5 to 10 years? Something, anything?

Truly yours!
hansmuff

Hi Hans,

thanks for the nice words!

Don

PS: well, it won't die because zombies don't die. ("The dead don't die")

PS2: sorry for the short reply, but anything else would just be a longer version of "no, thanks, I've seen enough". And I have some coding to do...

admin wrote: ↑16 Aug 2020 17:30 Coding for me has to be fun.

It's not just you This an interesting question why coding make people so passionate about it.

You know, I'm proud that I begin to master the art of "Indonesian Coding". That's my own term for what I saw 2012 in Sumatra and Java: DIY high-tech tools! A broken pocket calculator, some rusty wire, an empty Cola can, a bicycle dynamo, and some part of an electric guitar are connected with tape, rubber, and solder to make a fully functional cash register. Stuff like this. This is how I work under the hood of XY. Don't tell anybody.

Yes, a lot of things are created that way.

We all need a lot of

Be careful... I drank a lot of coffee for years, and it was a bad idea for health. From the October of 2019 I drink only a single cup right on the morning.

I drink exclusively Espresso. It's perfect for health.

Don, I just read this whole thread. I've been trying to ignore the forums lately and let you get your work done. But the way things stand now, I've got to put in my two bits:

1) You are an amazingly patient and gracious human being.
2) XY does what it does extremely well.
3) Don, have fun doing what you're doing! And thank you for the time you spend here. Keep it up!
4) To some posters: There are other options out there - you want something written in a different language, it's there. Buy it (and keep upgrading it) and leave Don to get some work done. I looked extremely carefully at those other options (you can guess the one I looked at hardest) before going with XY. I'm getting older, and I don't like learning something new that's soon to be no longer relevant. I don't see XY becoming un-useful before I'm no longer useful. Don's made some excellent usability choices that are simply not available elsewhere. I don't care what he programs in. I don't care if it's not multi-threaded and uses a separate program for file transfers; if there are a lot, I do my biggest ones with Teracopy started from XY, then his for smaller things while that's grinding away in the background. I like his dedicated scripting language - it suits its purpose, and he keeps incrementally improving it. I would so much rather see him sitting in his accustomed place, doing what he does best, than banging his head against a wall trying to re-create where he is already. And for heaven's sake be polite when you take his attention.
5) The only thing that keeps me from being better at using all of XY's features is how productive I am with the ones I know already - which all things considered, was a pretty painless process for what I'm getting out of it: I use this program daily, extensively - and so does my wife, and she's a bit intimidated by computers.

Mostly, thank you, Don, and feel supported. You've enhanced my life.

Thanks, pal!

admin wrote: ↑16 Aug 2020 17:30 Well, I tried hard and long, and I lost a lot of money and time, and now I can safely say: .NET and XY and me... it just doesn't work out. It would have taken years to create a .NET version of what we already have now. Years of total creative stagnation. Years of frustration for me because Visual Studio is so incredibly fat and slow, and I'm just not made to wait. Coding for me has to be fun. When it feels like filling out a tax form the fun is gone. Time to stop it.

Bye .NET, long live VB6!

Probably useless, but again I would suggest you have a look at Delphi or Lazarus. There's a free community edition of the former available that you could use to give it a try, and Lazarus is of course free, and Romain Petges actually changed his Attribute Changer from a C++/Delphi application to one built with Lazarus.

And as for me, I may have mentioned this before, I've got one particular program that I keep up-to-date in both Pascal/x86 assembler and PL/I, although those languages are probably (significantly) closer than VB and Pascal.

Forget it. It becomes clearer every day that my decision was right.

Hey folks...

Enough already. Don is the creator, author, father, and OWNER of XYPlorer. We have all done very well with his work.

(I happen to know developers whose programs written in VB6 are still running beautifully after more than 10 years. They are active applications where there's no need to change things.)

A lot has been said here and Don has been gracious enough to read and reply. Now let's end all this discussion--it has been a good one--and let him spend his time making XY even better...with the existing code base!

GUIguy wrote: ↑20 Sep 2020 21:30 Now let's end all this discussion

You're a new forum moderator, aren't you?

Nope. Just a man who can read and opine!