Rene's Revised History of Assembly Language



Betov wrote:
>
> Liar. In 1999, your started the noise about HLA,
> the exact same way you have started the noise about
> HLA2, that is to say, long before having anything
> up and going.

Hmmm...
Here's the post with the original announcement:
http://groups-beta.google.com/group/comp.lang.asm.x86/browse_frm/thread/70b3f0b07460413f/c1b9cc53d6d85923?q=HLA+v1.1&rnum=1&hl=en#c1b9cc53d6d85923

You'll notice I gave a download link (webster). You'll also notice
there weren't any follow-ups complaining that there was no code to
download at that link.

It's really fun to watch you call me a liar while making claims that
are so easily refuted. Care to try again?


> >
> > And the disassembler will still be incapable of automatically
> > disassembling a substantial amount of code out there. That's why
> > people who've spent a heck of a lot more time that you researching the
> > problem have developed interactive disassemblers. That's why books on
> > the subject of disassembly and reverse engineering pretty much tell
> > you that automatic disassemblers are toys.
>
> There are people whO write book, and there are people
> who do the job. These are not usually the same ones.

In the case of IDAPro, there are members of the team who *have* written
several respected journal papers (at least, this was true when the
papers were written, maybe they've moved on, I don't know).

You, OTOH, have yet to do the job and you've not written anything. So
why should we believe you?


>>
> > But keep up your dellusion! The more time you waste tilting at these
> > windmills, the longer your product will fail to be competitive, even
> > with the other "hobby-level" assemblers that exist.
>
> Well, it is now evident that it will beat all other
> Disassemblers hands down, but you say what you like
> to say...

We're still waiting. What has it been, now? A year and a half. I
remember back when you were promising an automatic disassembler in just
six months (search for yourself in this very newsgroup).

When you produce a disassembler that someone outside the RosAsm inner
circle feels is far superior to IDAPro, I will be *very* impressed.
Good luck.

>
>
> >> The very last of the Recognitions
> >> Methods (that are the Short Jumps probabilities) are
> >> over, and the very first version with a Map File,
> >
> > Ooooh!
> >
> >> for
> >> introducing a bit of Interactivity is ready, and fine,
> >
> > OHMYGOD! Do I read this right? Are you finally admitting that an
> > automatic disassembler isn't going to cut it and you're going to have
> > to make it interactive? This can't be the great genious Betov
> > admitting that he was wrong all along, could it?
>
> ???!!!...

As in, you've made claims in the past that interactivity was nonsense
and unnecessary in a disassembler. Now you're adding it to your
disassembler. That, to me, seems to indicate that you've finally
realized that an automatic disassembly will *not* produce satisfactory
results (despite your previous claims) and that interactivity is
necessary if you intend to produce a usable disassembler. Something
*many* people around here have told you from the start.

>
>
> > Note: You've got a long way to go to catch up with IDAPro. I still
> > don't buy the argument that this disassembler is somehow going to
> > change the world. You've made that same argument with the IDE. With
> > the Wizards. With the "preparsers". Nothing else has seemed to set the
> > world on fire.
>
> For your information, IDA does not even _try_ to
> recover the Resources,

Your point?
RosAsm does not even try to disassemble Z80 code. How's that for
throwing out a non-sequitor? (btw, IDAPro *does* disassemble Z80 code).


> and has no plan for, ever,
> being able to re-Compile any PE in two Clicks,

Yes, IDAPro is a disassembler. It doesn't include an assembler. So the
whole concept of reassembly is a moot point. Of course, it's also the
case that only a tiny fraction of people use a disassembler for some
sort of brain-damaged source code recovery system. Disassemblers are
*really* used for program analysis, threat analysis, and reverse
engineering. You don't really need the ability to reassemble the source
code for those purposes. So Data Rescue has wisely chosen to produce
code that is easier to analyze rather than worrying about emitting code
that is syntactically compatible with any particular assembler.


> the
> way RosAsm already does with some pretty success
> on a significative number of small Applications,
> and Demos.

Sure. The fact that you cannot disassemble an arbitrary file
automatically does not imply that you cannot disassemble *any* file
automatically. But just keep in mind that when you announce that the
RosAsm disassembler is complete and does the best possible job, I'll
still be able to break it's automatic disassembly in a few minutes with
only a few lines of code.

And given that disassemblers find their main use in code analysis
(particularly, in threatening code analysis, which is usually written
to confound hackers with disassemblers), the tool you're going to wind
up with will not be suitable for most jobs for which a disassembler
would be put to serious use. Then again, why should your disassembler
be any different than your assembler, eh?

>
> > You may very well wind up with a decent disassembler, I don't know.
>
> Oh, yes: "I may". :)) :)) :))

We're still waiting.


>
>
> > What I do know is that the number of people that tool is going to
> > attract to RosAsm is sufficiently low that it can be ignored. After
> > all, the free version of IDAPro has been out there for years. If
> > having a decent disassembler is what was needed to fuel the "assembly
> > rebirth", it would have happened years ago.
>
> What makes the RosAsm case a very particular one is
> its level of Integration, that enable completly new
> things, like the "Two-Clicks-Disassembler-ReAssembler".,
> that is a never seen Tool, up to now.

That's right. Up to now we've never seen a disassembler that produces
RosAsm compatible code. You've got an amazingly unique product there!
The assembly rebirth has begun, people are switching to RosAsm in
droves. Data Rescue is going out of business. Hell has frozen over.
Etc., etc.


>
> An massive advantage that you cannot understand.

That's for sure.
The only problem is, I'm not alone in "cannot understand"ing the
massive advantage of this two clicks disassembly/reassembly. Let's
suppose the operation was 100% perfect. What have I achieved by
disassembling the code and reassembling it with only two clicks. Why
now I can *run* the executable. But wait! I could have done that
*before* I disassembled the code! What was the advantage to this again?
I'm missing something here.

The bottom line is that a *lot* of *manual* analysis needs to be done
once the code is disassembled. First, to handle all the mistakes in the
code that the disassembler produced. Second, to extract the routines
that you want to swipe (legally or otherwise) from existing code and
incorporate into your own programs. Third, to clean up the code and
make it presentable for people who are reading your source code. Being
able to click on a menu item and reassemble the code immediately after
disassembly buys you absolutely nothing. And this scenerio, of course,
assumes that the whole purpose of disassembly is to lift code from an
existing executable and put it into another (which is not what most
disassemblers are used for). In fact, most people use disassemblers to
take apart viruses, trojans, and the like to research them.
Reassembling such code is of little interest (i.e., why would you want
to produce an executable image of a virus? Particularly if you already
had the original code?)

So what was the massive advantage again? I think I missed it.

>
>
> > But go on fooling yourself. While you're playing around with the
> > disassembler, work on features that could actually make RosAsm
> > competitive just isn't getting done.
>
> As you like. But, if you are really convinced of
> these bullshits, why did you fight so hard against
> the implementation of a "Two-Clicks-Disassembler-
> ReAssembler", in RosAsm, since day one of this work?
>

Because, as usual, you started making outrageous claims about what it
would be capable of. And it's fun to point out how full of crap you are
with respect to such issues.

Personally, I don't care if you get the disassembler working. More
power to you. I hope you succeed in producing a more powerful tool than
IDAPro. Then, at least, the RosAsm system will have at least *one* tool
that is worth using. I'm not holding my breath though.




> > Guga seems to have "been gone" since January. How long does it take to
> > get moved to another town? If I didn't know better (and actually, I
> > don't), I'd think that he abandoned you.
>
> Sweet dreams Master Pdf ?... :)
>
> No, Guga is not dead, but he has some problems with
> re-installing (family, office, home... all problems
> that you do not have to know about...).

You're right. I don't know. What I do know is that it's been a *long*
time since the last major update to your disassembler, and the last
time I asked about it you made reference to the fact that you were
waiting for Guga to come back. Now, you claim the contrary. Apparently,
your statements in the past weren't exactly truthful, eh?
Cheers,
Randy Hyde

.

Quantcast