From: Peter Olcott on
On 5/23/2010 11:42 AM, Joseph M. Newcomer wrote:
> See below...
> On Sun, 23 May 2010 08:21:29 -0500, Peter Olcott<NoSpam(a)OCR4Screen.com> wrote:
>

>> I have found that bot to be the case. Nothing else may be 100%
>> completely definitive but these performance estimates are accurate
>> enough to matter very much.
> ****
> What part of "if you have MEASURED it you know NOTHING" have you failed to understand?

I completely understand that such over exaggerations tend to be wrong
most of the time. By the way you said it backwards.

From: Hector Santos on
On May 23, 11:33 pm, Peter Olcott <NoS...(a)OCR4Screen.com> wrote:

> The link that you provided superficially seemed to be very high quality
> work. I would guess that this guy's code may beat my code by a little
> bit. I would guess that this little bit would be something like 50% faster.

Superfically? A little bit? Lets repeat the link, because u seem to
wish to hide it.

http://bjoern.hoehrmann.de/utf-8/decoder/dfa

There is no std::vector, there is no switch statements, sweet and
elegant and more importantly, prior art.
From: Hector Santos on
Please stop the over quoting with .05% input and please stop the
quoting of signatures.

> >> And peter, just because you are limited in understanding prior art,
> >> does not mean that when you get an AH-HA, that it becomes "My IDEA",
> >> "My Proposal," etc.
>
> It is my idea in the sense that I thought of it now, and am proposing it
> now. It is not my idea in the sense that no one else has ever thought of
> it.

It is always great when a student learns, even teachers get a smile.
But it should not be celebrated as your idea and begin to show disdain
and a high degree of ignorance of the experience of experts, prior and
existing work. All this is already on the net. Easily found on the
net. I did a simple search: UTF-8 DFA and found the link I provided:

http://bjoern.hoehrmann.de/utf-8/decoder/dfa

--
HLS