PureBasic Forums - English

BorisTheOld wrote: The problem with third-party software is that it's a quick fix for an immediate problem. But it doesn't make the problem go away.

I agree, and the same can be said for PB be when using it just beyond the language.

The PB libraries are a great convenience, but when there is a bug in them you are stucked waiting for a fix.
And the time required is absolutely unpredictable. Realistically you can't even be sure it will be ever fixed, there is no guarantee about it.

I've read some time ago someone was advocating to open source the PB libs.
First the positive.
If the PB libraries were open source COULD be a nice thing. We already have sometimes bugs in the compiler, the debugger, the editor, etc. All that could be then treated with an higher priority by fred, while we, the user, could fix the libraries.
Not saying we should maintain or extend them. That would remain under fred control. We could just submit patches to him addressing the problems.

The negative.
All the above is just theory, I don't know how could turn out in practice.
For example someone could submit a patch thinking it will solve a bug when just creating another one... that's a problem.
fred could have to spend time evaluating the validity of the patches submitted instead of spending that time fixing something else.

The outcome it's not really assured.

If PB was open source then I'd never get any work done; I'd spend all of my time taking PB apart to see what made it tick! Just can't help myself.

BorisTheOld wrote:The problem with third-party software is that it's a quick fix for an immediate problem. But it doesn't make the problem go away. Eventually it has to be faced head on. Unfortunately, it's taken me a long time to figure that out.

This is very true. Although I am encouraging people to switch to PB, now that I am solely doing programming for fun and strictly as a hobby, I had planned to go back to that other language. In spite of that one person's actions, I did love his one tool and looked forward to exploring the limits of his GUI tool. So, I find myself in language limbo right now. But moving forward nonetheless.

luis wrote:

BorisTheOld wrote:The problem with third-party software is that it's a quick fix for an immediate problem. But it doesn't make the problem go away.

I agree, and the same can be said for PB be when using it just beyond the language.

The PB libraries are a great convenience, but when there is a bug in them you are stucked waiting for a fix.
And the time required is absolutely unpredictable. Realistically you can't even be sure it will be ever fixed, there is no guarantee about it.

This is a concern for any language and platform -- that they become obsolete, or finish up being bug-ridden and unsupported. But at least programming can't be classed as a dull profession, as we're always trying to keep ahead of total disaster.

Here's an example of self-sufficiency. When the IBM PC first came out in the early 1980s, I had a contract to write a manufacturing inventory/purchasing system in interpreted BASIC. I wrote a module for doing indexed file I/O, similar to ISAM that was in use on mainframes. Over the years the module has been written in different languages, but the file structure has remainded the same. I have some customers who've had files in continuous use for almost 30 years.

Over the years I've built up a library of modules that make me less reliant on third-party code. And by moving all my code to PB, and by using several hundred macros to hide the PB syntax, I'm hoping to make life easier in the future should I need to jump ship.

BorisTheOld wrote:Over the years I've built up a library of modules that make me less reliant on third-party code. And by moving all my code to PB, and by using several hundred macros to hide the PB syntax, I'm hoping to make life easier in the future should I need to jump ship.

Seems like you should have just chosen good old fashion C... It will never go away in your lifetime, and it's been ported to more platforms than any other language.

Tenaja wrote:

BorisTheOld wrote:Over the years I've built up a library of modules that make me less reliant on third-party code. And by moving all my code to PB, and by using several hundred macros to hide the PB syntax, I'm hoping to make life easier in the future should I need to jump ship.

Seems like you should have just chosen good old fashion C... It will never go away in your lifetime, and it's been ported to more platforms than any other language.

Teneja, that makes perfect sense from your perspective.
But tell me how could one choose something that did not exist?

BorisTheOld wrote: This is a concern for any language and platform -- that they become obsolete, or finish up being bug-ridden and unsupported.

Up to a certain point yes, but I was specifically talking about PB libraries here, and any other closed source library made by anyone else. If you just rely on the language, and then write all the rest by yourself, your level of dependency is a lot lower.
If the compiler work, you can manage the rest. And some languages are a lot more secure than others obviously.

If you don't care about Basic, as Tenaja said just by using using C you are practically 100% safe.
Not only its lifespan has been proven excellent, but you have tons of different compilers all compiling the same language.
If you don't depend on third part libraries you can happily survive even the death of your favorite compiler.

Obviously here, with a compiler still not perfect and the possibility to lose it all tomorrow, we face already a big risk it's wise to not compound by using third party libs. But I think this is pretty clear to anyone using a "indie" compiler and especially an "indie" language where only one compiler exists for it on the face of the Earth. Any kind of investment need to be weighted against that.
For hobbyists (is the right plural?) it can be acceptable.

heartbone wrote:

Tenaja wrote:

BorisTheOld wrote:Over the years I've built up a library of modules that make me less reliant on third-party code. And by moving all my code to PB, and by using several hundred macros to hide the PB syntax, I'm hoping to make life easier in the future should I need to jump ship.

Seems like you should have just chosen good old fashion C... It will never go away in your lifetime, and it's been ported to more platforms than any other language.

Teneja, that makes perfect sense from your perspective.
But tell me how could one choose something that did not exist?

As the quoting indicates, I am referring to Boris' decision to move to PB and use Macros to hide the PB syntax. C has existed long before PB. It was available in the 70's, even before his (unquoted) reference to the early 80's.

Tenaja wrote:As the quoting indicates, I am referring to Boris' decision to move to PB and use Macros to hide the PB syntax. C has existed long before PB. It was available in the 70's, even before his (unquoted) reference to the early 80's.

When I first started my software business, C had not even been released as a viable language. Most programmers were using Assembler, PL/I, COBOL, and FORTRAN. From 1977 onwards I was using mostly MASM and COBOL for my own use, and for developing business systems for mainframes and micros.

I got into BASIC when developing the above mentioned project -- it was a contractual requirement. But as far as C is concerned, I've always considered it to be a toy language not worthy of my attention. It's just Assembler with an inconvenient syntax, and should have been dumped years ago.

I don't want to waste my time worrying about low level stuff like API and interfacing with libraries. That's why I choose to write in high-level languages.

BorisTheOld wrote: It's just Assembler with an inconvenient syntax

Do you mean assembly ?
Well, a lot of empires were built on top of that "toy", its simplicity (even if a C compiler it's WAY more sophisticated than PB to put things in perspective) made possible to port it everywhere fast. The only problem was almost nothing really useful were inside its standard library, but again everything you built in the years with it was easily portable in time, and almost everything was/it's available with a C interface.

So you had the best of three worlds:

- it was low level enough to not miss assembly too much (at least compared to its peers)
- it was available everywhere and at any point in time from when it was born
- it was easy enough to transfer all your code between platforms, architectures, compilers and you can access decades of code written with it

Like any other language has its problems, but I would never dream to call the workhorse of the software industry a toy.

BorisTheOld wrote: and should have been dumped years ago.

and yet isn't happened, I don't wonder why.

BorisTheOld wrote: I don't want to waste my time worrying about low level stuff like API and interfacing with libraries. That's why I choose to write in high-level languages.

Understandable, but someone has to do it to give you the possibility to do so. And not everyone would consider that time wasted.
As always, it depends on what are you looking for when programming.

Tenaja wrote:

heartbone wrote:

Tenaja wrote:

BorisTheOld wrote:Over the years I've built up a library of modules that make me less reliant on third-party code. And by moving all my code to PB, and by using several hundred macros to hide the PB syntax, I'm hoping to make life easier in the future should I need to jump ship.

Seems like you should have just chosen good old fashion C... It will never go away in your lifetime, and it's been ported to more platforms than any other language.

Teneja, that makes perfect sense from your perspective.
But tell me how could one choose something that did not exist?

As the quoting indicates, I am referring to Boris' decision to move to PB and use Macros to hide the PB syntax. C has existed long before PB. It was available in the 70's, even before his (unquoted) reference to the early 80's.

BorisTheOld wrote:When I first started my software business, C had not even been released as a viable language. Most programmers were using Assembler, PL/I, COBOL, and FORTRAN. From 1977 onwards I was using mostly MASM and COBOL for my own use, and for developing business systems for mainframes and micros.

I got into BASIC when developing the above mentioned project -- it was a contractual requirement. But as far as C is concerned, I've always considered it to be a toy language not worthy of my attention. It's just Assembler with an inconvenient syntax, and should have been dumped years ago.

I don't want to waste my time worrying about low level stuff like API and interfacing with libraries. That's why I choose to write in high-level languages.

Thank you BorisTheOld for that beyond excellent validation of the point I was expressing to Tenaja.
Let me return the favor.

luis wrote:

BorisTheOld wrote: It's just Assembler with an inconvenient syntax

Do you mean assembly ?
Well, a lot of empires were built on top of that "toy", its simplicity (even if a C compiler it's WAY more sophisticated than PB to put things in perspective) made possible to port it everywhere fast. The only problem was almost nothing really useful were inside its standard library, but again everything you built in the years with it was easily portable in time, and almost everything was/it's available with a C interface.

So you had the best of three worlds:

- it was low level enough to not miss assembly too much (at least compared to its peers)
- it was available everywhere and at any point in time from when it was born
- it was easy enough to transfer all your code between platforms, architectures, compilers and you can access decades of code written with it

Like any other language has its problems, but I would never dream to call the workhorse of the software industry a toy.

BorisTheOld wrote: and should have been dumped years ago.

and yet isn't happened, I don't wonder why.

luis, I'm pretty sure that Boris was kidding about the dumped.

However C was not ready for prime time and many people did not like it.
This spawned the many unforunate object oriented programming paradigm implementations to attempt to remedy the shortfalls od C code development.
With hindsight a generation later, it may be easy to think that C was an obvious choice for a long time code library,
but let me assure you that it was not.
For example at the time one would need to rely on proprietary libraries to get anything significant done in C.
At the time I was a FORTRAN & PL/1 guy and absolutely loved the freedom that BASIC allowed.

Don't know, in the '80 for me it was quite obvious. Not saying it was obvious in the '70. But from the '80 is still a respectable 30+ years.
I fear to ask, but what's the meaning of the cartridge ?

Assembler vs Assembly

Boris was quite right to use "Assembler" and not "Assembly" as that was the name to the language.

Assembler is the program translating the assembly to machine language.
Assembl-er, compil-er.
http://en.wikipedia.org/wiki/Assembler

The funny thing is even the page hosting the image above tells the same.
http://www.pcmuseum.ca/details.asp?id=3 ... e=software

This assembler should also look familiar -> http://flatassembler.net/

I fear to ask, but what's the meaning of the cartridge ?

What cartridge? I can't find any reference to a cartridge.