Re: HLA Stdlib v2.2 is now available.

On Sat, 22 Sep 2007 18:06:58 +0200, rhyde@xxxxxxxxxx <rhyde@xxxxxxxxxx> wrote:

On Sep 21, 5:11 pm, //\\\\o//\\\\annabee <w...@xxxxx> wrote:
På Fri, 21 Sep 2007 20:27:13 +0100, skrev rh...@xxxxxxxxxx
<rh...@xxxxxxxxxx>:

> On Sep 21, 10:08 am, //\\\\o//\\\\annabee <w...@xxxxx> wrote:

>> Yes, of course. The problem is that newcommers are not told this.
>> most of the material is written with other languges in mind.
>> And that material is often way more difficult then asm tutorials.

> Please provide the proof of this statement. It runs completely
> contrary to every experience I've ever had.

Easy. You have no experience.

:-)
Whatever you say.
And just how many students have *you* taught such that you have the
experience to make such a claim?

I quit university when my programming teacher started talking nonsense.

For others, heres is a few links:http://betov.free.fr/Iczelion.zip


Funny, the iczelion tutorials do *not* teach assembly language
programming. And, in fact, it's considerable use of the high-level
language features of MASM run completely contrary to Rene's claims.
Care to try again?

It teaches the point. Asm is easy under windows.

I also saw a couple good and short tutorials for MASM.
(Yes, in fact, and some I download from Paul Hsien page)http://www.azillionmonkeys.com/qed/
Here is links to something.



Then there are the code written by Wolfgang.http://web.utanet.at/schw1285/KESYS/index.htm[codesnips]

then Erics tutshttp://www.quanta-it.com/easbell/

then Diego Fraga demoshttp://diegofraga.f2g.net/

Uh, dude, I think you missed the whole point (as you usually do in
your rants). How many *beginners* have learned assembly language using
these materials?

How would I know how many?

> And feel free to have people give testimonials of the RosAsm "12
> Lessons" if this approach is as fantastic as you're claiming.

I dont know.

Neither, apparently, does anyone else. That's the point I'm making.

You are making an ass of yourself.

They may be good on beginners, but I allways found the
_practical_ small demos and app much more usable and interessting.

Why yes, a sample size of one (and a bizarre sample case at that)
indicates complete and absolute truth.

You do realize that it doesn't really matter what *you* (or any other
single individual) thinks or finds "usable and interesting" What
matters is the *average* case.

Not in asm. NEVER in asm.

Furthermore, given the fact that you've
been doing assembly language for quite a while (What? ThreeYears now?)
and you've only achieved the proficiency level you have is not a good
advertisement for your preferred educational model.

Proof by example is false. Its a generalization of the whole asm community based on the results of a single man.

Yeohs
have a mountains of them, for instance, that all answers a spesfic task we
may want to know about. Very useful, very short, exceptionally easy to
make use of.

And they only cover specific tasks, rather than the general process of
teaching assembly language.
You just don't get it, do you?

No. You dont get it. You are saying that you cant learn english incrementally, for instance by reading books not about english. This is definitly wrong.



>> Having everything in one place, "monosource" makes the development, and
>> sourcenavigation possible to be instant.

> Yes, by wimpy IDEs that don't have the ability to manage multiple
> source file projects. You *do* realize things have gotten a bit better
> since the 1970s when we didn't have editors capable of dealing with
> multiple files, right?

That maybe true. I dont know. 3 years and 9 month ago, and the 10 years
before that, it was not even close to what RosAsm was at the time I
started with it.

No, such tools *did* exist. You simply weren't aware of them.

Yes, as I said. May be true.


> Just out of curiosity, have you *ever* used Visual Studio for a large
> project?

Exactly why would that have anything todo with a discussion about
assemblers?

I don't know? Why do you keep bringing up RosAsm's editor features
when talking about assemblers, then?

Because it is an assembler, and a very good one, showing the way for other spesific assemblers.

You *do* realize there's a difference between an editor and an
assembler, right?

As they both are needed for doing programming, I see no reason to seperate them. Espesially not when this makes asm programming a true hell, like with so many of the other assemblers. Intergrated tools helps a lot.

>> Yes this is a very useful thing,
>> in everyday practical context.

> Not really.
> For team programming, it is a complete disaster.

It works well.

No, it does not.

Yes it does. Why dont you go into the reason you think otherwise.
This is another baseless claim.

How can you keep claiming thoose visious lies, without an
argument?

What "visious" [sic] lies are these?
Exactly how many "large teams" or even "medium sized teams" have you
worked on so that you have the experience to make such claims?

I havent made any claim. Its just simple logic. RosAsm has TITLE to break down a project into parts for each programmer. A project needs a manager. It need planning. It need coordination. The issue of solving a project well, is not only about the tools, it is about having a good plan first. A good tool like RosAsm will help a lot. But surely it will not help a lot at solving the problems the wrong way. RosAsm has allready been subject to team efforts. Ludvig made the debugger. That is an awsome debugger.

Its just an empty claim, no substance.

Try picking up a book on software engineering some time.

Maybe good for flexing a muscle or two. But yea, no thanks. I _have_ beeing reading theese toilett papers, and what I have gathers so far is that the composition of a project is determined by the requirements of the project, the various facts, and not by some preexisting philosophy. When I do OOA in asm, this is not because I want to claim this is the only one way. I am just showing how simple it can be done in asm. It can be made even simpler. But we do not need this kind of thing in asm. It add overhead there is no reason to have. The flexibility of asm can create how many such "ways" as we want.

> It's fine for tiny projects with one hacker working on the code; it
> completely fails for just about any other development model.

Another empty claim.

Try picking up a book on software engineering some time.

Is this you argument? "Try picking up a book on software engineering "??
Is this the gist of your argument?
Guess who I would be tempted to hit on the head with it...




>> Whereas it may take months to years
>> developing an application.... cutting away the dead code, takes an hour.

> I call "bull***".
> We've seen your code.
> You never bother stripping out the cruft you've left behind.

How can you say that? I just posted the same app in 3 versions.
one was 400k, the second 90k and the last was 16k.
HLAFucker :D You may have hear about it?

Different issue.

"Ho hum, pigs bum" :D


Here is the links to each.

http://www.szmyggenpv.com/downloads/HLAFucker.Zip
http://www.szmyggenpv.com/downloads/HLAFuckerClipped.Zip
http://www.szmyggenpv.com/downloads/HLAFuckerv196.Zip
http://www.szmyggenpv.com/downloads/HLAFuckerv197.Zip
http://www.szmyggenpv.com/downloads/HLAFuckerMini.Zip

:D

I said : At THE END of the devs. I am not even halfway done.

That's your problem.
You're never done. Therefore, you never get to the point where you
spend that "one hour" at the end of the project to clean things up.

Allready did many times for testing.

I
suspect when that day actually arrives, you'll discover that it takes
a *lot* more time than an hour to clean up the mess you've made.

Even a day would justify it completly. Even 3 days.

Come back and talk to us about software engineering issues when you've
actually *completed* several projects and (realistically) evaluated
your productivity on them.

Ah, and what "projects" have you completed then? HLA?
I have at least completed HLAFucker :D

> And once again, though you *might* be able to make this claim for a
> tiny project that is never worked on by more than a single hacker, it
> completely misses the mark for large projects involving teams of
> programmers.

Basesless claim.

Pick up any book on software engineering.

HLAFucker! HLAFucker! HLAFucker! HLAFucker!


> If someone is not intimately familiar with the code in
> question, deleting obsolete code is a very dangerous thing to do
> because they'll never really know if the code is completely dead or
> still used under certain circumstance.

Wrong.

You've never worked on a real team project, have you?

HLAFucker!

For the code they write themselfes, it is clear that they will know.

Reread my statement.

ok. Done. Yes.... You are wrong.
Still .. dont clip code when you dont know what it is.

Dont clip code when you dont know :D

And therein lies the problem.
In a *team* project, which pretty much describes most real-world
software projects, you have no choice but to work on code you don't
know. You have to learn it.

ah. Yes. So dont clip it, until you know it.

That's why it's important to learn how to
write *good* code -- because someone else is going to have to look at
your mess and deal with it sooner or later. This is something you've,
apparently, never figured out.

Something you seem to not understand is that the overhead comming with structured programming is hurting the eyes over some level even in languages like Delphi.
If you doubt it, take a look at the interface section of the translation of the DirectX libraries.

To be serious. Once some code grow over a certain complexity, it doesnt matter what high level language you use. They all make the complexity look hostile. Object principles does not work for codebases large enough. No principle works then. I dont recommend any books. Here you must have experience and also be somewhat versed in the art of uncovering deception. Lucky for you, uncovering this as a fact is a complex as writing two or ten or more application from beginning to end in two languages, of similar complexity. This is the problem here. We may disagree, or agree, but we cannot just point to the facts like you do. A book on software engineering is _NOT_ facts. Those are attempts at wrapping up the limited experience of the last decades, in HLLs!!!! in an area that make massive changes every year. If you had a minimum level of intellectual honesty and experience with programming you would be able to see this.

You claims have no base.

Pick up any book on software engineering.

This is not a basis. This sound more like the statement of a dement.

What is more funny is that you think that HLLs are able to cut code
"automatically".

I have no idea what you're talking about. I realize that you're not a
native English speaker, but you should make a small attempt to clearly
state what you mean when you post comments to this newsgroup.

I think you belive that using object modules, and a compiler helps at clipping redundant code. Isnt that what you ment?

Shows how much experience you have, NULL.

Oh well.
At least I *have* worked on team projects.

Well just about any programmer has. So it doesnt say anything about you real experience.
I have worked 9 month as a carpenter. I have in addition helped building several houses for friends. This doent mean I am verses in carpenting. At all. I would needed to have built many diffrent houses, (3-4 a year), on my own, over many years to become verses in something. And this would still not say anything about my talents, which may have been relativly weak even then. But it would give me enough experience to make resonably well founded arguments in a debate about carpenting. Enough to know when someone tried to bull*** me on the basic prinisiples. And thats what you are doing here.

In most projects,

Really?
Care to provide references that show your claims to be true for "most
projects?"

The once I have seen. Maybe 1000s.
I provide references, and or disclaimers when I see you rising above the level of a twoyear old in you argumentations style. You have not supplied any basis of any of your claims, why should I provide more on my part.

And I'm talking about most projects in the real world, not most "Eric
projects."

I am talking about the real software that most of us use every day.

you will find
than 99% of all the code builds or makes use of all the other code.

This statement is absolutely meaningless.
Perhaps you would care to restate exactly what you mean?

I do not mean that all code use all other code, but that most code used, is somehow tied to most of the code in the libraries in a big project.



And
the 1%, if really exists, just dont count in practice.




This is why Delphi 7.0 gave you 400kb executeable (may have been larger, I
dont remember anynore) without writing a single line of code.

What does this have to do with anything?

It means a simple window cost 400K in delphi. Before writing any code. While that cost 4K in assembly. And that again means that without writing a single line of code, you link in 400kb using Delphi.

Is that hard?

And why every windows DLL links to damn near every other DLL in the entire
os :D :D
THAT is the evidence of where "modular" programming takes you.

Delphi's VCL has a huge dependency graph for the windowing objects
(which, btw, has little to do with "modular programming" -- indeed, an
object-oriented system like the VCL tends to violate many of the
precepts of "modular programming")

Indeed.

What does Delphi's VCL have do to do with the fact that you never
clean up your own code? Is the fact that the Delphi VCL has a huge
dependency graph simply an excuse you use to write bloated code that
you never clean up?

I dont see the link here. I can clean the code in an hour. I have done it many times for testing, and this is fantistically simple. As I showed with HLAFucker. I use RosAsm, the most complete devtool for windows on this planet. And this is a mere fraction of what it cant do. A projectmanager could do this with a team of programmers and a projector, in a couple of hours with RosAsm. And still not touch or destroy the development code, which at anytime can make reuse of the now unued, but later maybe useful code left in the original sources.

>> So I gladly trade that hour, at the very end of a project, for the
>> significant efficiency of RosAsm source navigation.

> How come every project of your's that I've looked at has remained in
> desparate need of refactoring, then? You make the typical hacker
> claim of "I'll get around to that someday" and "someday" never comes.

In 6 years, _and_ if you have completed your 10+ years "prototype" of HLA
_this_ year.

Oh, and let us not forget that I've written six books, worked full-
time, maintained a huge website on assembly language programming, and
written a bunch of other code during that time, too.

:D You making exuses based on you state of you personallife.
You think you're the only guy on the planet with problems?
Would you allow the same for someone else?

And if you'd like, I'll be more than happy to share with you all the
old versions of the HLA source code so you can go in and verify that,

:D Nothanks.

yes, I have done a *lot* of code refactoring on the source code over
the years. That's why I laugh at your "one hour" estimate. I *do* have
a little bit of experience (despite your claims) with respect to
cleaning up code.

Of course, you do have some experience. But you are a C coder. If you ever wore an asm coder, it is now a long time ago. You GAVE UP on asm, remember? I GAVE UP on HLLs.
This is two REVERSE situations.

We will have "hacked" away at each our project the same
amount of time. Only you wouldnt yet have finished anything but a
prototype.

Granted.
Now exactly what is it that *you've* produced in all this time?

To early to tell really.

With the massive experience you _pretend_ to have, given that
you claim doing asm since 1785 vs my 3 years of assembly programming ....

The sad part is that just about anyone with three years of assembly
programming behind them is *far* more advanced than you have. While
that's not absolute proof that learning with RosAsm is a dismal
failure (still just a sample size of one), it's not a good sign...

No. Using one man as an example to deem a whole assembler, is such an absurd lie, it should get your fired instantly, if you had a job.

:)) I am going to loose a LOT of sleep over this :D

As well you should.
You should carefully consider how you've wasted away your past three
years and only gotten as far as you have.

Pathetic. I wonder what your story for the ferryman will be.

- I wrote HLA.
- What?
- A magnificent multipurpose portable assembler.
- Never heard of it.
- It is as good as C. Its a High level assembler.
- Ah C, yes I heard about C....
- But it is an assembler.
- hmm... dont know that much about the topic. whats an assembler?
- Its a tool for writing computerapps at the level of the machine.
- You mean like lowlevel?
- Yes.
- So why do you call it High level then?
- It frees the programmer from using lowlevel statements.
- But doesnt that make it not-an assembler?

That should tell you that you need to reconsider your educational
approach.

:)

hLater,
Randy Hyde

.

Loading