Thanks for clearing that up, and my take on it...

by Pete (no login)

Thanks for posting the QB64 reply that Galleon made and your reply to it. I need to post something in that thread, soon.

>In short, I'm making a 64-bit compatible version of QB64 for Linux users, but Windows users will benefit from it too.<

One thing I tried hard to stay away from is 2 versions of anything when it came to making big programs. Why? Updates and debugging take three times as long or more. Ideally, Galleon would be better off to make a way a pure 64-bit QB64 can run correctly on a 32-bit machine, instead of two versions. In other words, cough, either a compatibility mode of an automatic detection system that compiles programs for the correct system. So add a few more weeks to that two week estimate to make that so.

About the ten year guess. Let's suppose between 5-10 years. My argument would be since QB64 runs on all 64/64, 64/64, 32/32, and lower machines now and runs fast enough in most cases for what can be developed by BASIC programmers for now, why not table this idea until the GCC-32 bit compiler version is finished? Seriously, this is not some emergency life or death situation is it? I mean I see it more as adding another perk to an unfinished compiler associated with an unfinished system. In other words, I see that QB64 has changed from a hobby rather than a programming tool.

Again, from my personal experience, I would finish one of my office apps before starting a new one. Sure, I would get an idea for a new one, but I would memorize the idea and wait on it until I finished the app I was currently programming. Why? Because I had an office to run, not a hobby. I was able to bill with a skeleton program just a few months out of learning QBasic for the very first time. I added everything else over the years. The only thing I could not control was the bugs. Bugs in complicated apps happen when they happen. So even with one program finished after another, I needed to go back and debug each. But with finished apps, the debugs just hold you back a bit from completing the next project; they don't keep you from using the current one. Right now, there are a lot of things I can't check on QB64 because of the missing statements. If the Windows compiler were finished, I'd be able to check for bugs instead of doing extensive rewrites. And rewrites worry me, because if a bug happens, then I have to wonder if it is the compiler or did I rewrite something wrong.

So I am still of the finish as you go philosophy, and start new stuff before you are finished doesn't appeal to me. I bet if Galleon already had this project in schools, in other words it was being used in an environment where it had to work, I think he would use the finish first method rather than risk loss of credibility.

As for other systems like Linus, there are a lot of other BASIC compilers out there, Power Basic, True Basic, Pure Basic, etc., that already have a Linux version. It is a little hard to compete when your system is better underneath, but doesn't appear to be better on the surface. I still just wish the Linux version was made after the Windows one, but, and I don't know the answer to this, if it was considerably faster to do both systems at the same time, then I would agree that was the correct methodology to use.

And thanks for answering my question that QB64 will indeed run on a 64/64 bit system. Right now, they are the only ones I can find, except for netbooks.


Posted on Dec 9, 2009, 11:39 AM

Respond to this message   

Return to Index

 Copyright © 1999-2018 Network54. All rights reserved.   Terms of Use   Privacy Statement