Roger Brown wrote:Adam Hair wrote:
SNIP
You mentioned one point that I was unaware of. I know that several
amateur engine authors use the CCRL to indicate the approximate
strength of their engine. I did not know that at least one commercial
engine website did the same. Call it naivety or stupidity on my part,
but I honestly did not know that the CCRL, whether directly or indirectly,
could be considered advertisement for commercial engines. Or at least
I have never thought of it that way before. That gives me some food for
thought.
Thanks again,
Adam
Hello Adam,
A bit of mischief if I might be allowed.
Wouldn't permission be required to use the list in this way?
Later.
If you allow me to state my personal opinion and the reason for that
opinion.
My focus has been on freely available chess engines. Most of the
authors of free engines are hobbyist, the same as myself. I feel lucky
that they make their work available to the members of the computer
chess community. So if they want to link to a rating list that I am
involved with, or if their placement on such a list helps to draw
attention to their work, then I feel that it is fair compensation for what
they have given me.
I have had to give some thought in regards to commercial engines.
When I ran my own rating list, I made the decision to not test commercial
engines. It wasn't due to any animosity towards commercial authors.
I admire their ability and hold no grudge if they want to try to make some
extra money by selling their engine/program to the public. For me, the
reason I chose to not test commercial engines is that I could not buy
every one. If I were to test one commercial engine, then I would need to
test all of them. Quite honestly, that was not and is not feasible for me.
What if I could receive some of them for free? I would still choose to not
test them, unless I could purchase the rest of them. I would not want to
exclude any of them, regardless of whether or not a particular author
really did not give a damn if their engine was not on my list.
How do I connect the paragraph above to your question? I feel that if
every commercial engine is represented on a rating list and that list
is made public, then it is fair for a particular author to use that list as
advertisement. After all, the information is already being shared with
the public. I don't see where permission is needed.
I hope you find this a satisfactory, if long-winded, response to your
question.
Now, after my statements above, I know I will have to clarify at least
one thing. "
I would not want to exclude any of them". That
statement could be used to apply pressure to include Ippolit engines
and/or Houdini. After all, Rybka is on the CCRL lists and it seems to
have a cloud of illegitimacy surrounding it. There is one difference ( if
no other ) between Rybka and the other engines. Though I have only
been a part of this community for close to two years now, I am aware
that Rybka was considered a legitimate engine by most people for
several years. There is a certain amount of inertia due to that fact.
It is not unexpected that some people would want abundant, undeniable
evidence before they changed their opinion about Rybka. Yes, some
experts have repeatedly stated that there is no doubt that Rybka
contains code from Fruit. However, other experts have stated their
doubts. When you mix in the appearance of an agenda by people on
both sides of that debate ( whether or not anybody actually has had an
agenda ), it does not surprise me that some people have not changed
their stance regarding Rybka. I would not expect them to, though if
Fabian at some point says that Rybka contains Fruit code, then they
should.
On the other hand, there has been a cloud of illegitimacy surrounding
the Ippolit engines and Houdini from the start. Take for example Ippolit.
I don't believe we could define it as a clone or even a derivative of Rybka,
at least not using the widely perceived definitions of those terms. Yet,
BB himself stated: "I would find it wholly plausible that IPPOLIT is
a "reverse-engineered Rybka with many changes", and indeed I would
assert that the IPPOLIT maker(s) certainly knew much of the internal
workings of Rybka." He is not the only expert with that opinion ( reverse
engineered), though I am not certain he still holds that opinion. Throw
into the mix the manner that Ippolit's author(s) have presented the engine
and themselves, I do not blame anybody that decides not to test Ippolit
et al. As for Houdini, there has been enough circumstantial evidence
presented to cast some doubt over its relationship to the Ippolit family.
Certainly enough to warrant the same scrutiny that Rybka has received.
It is easy to see parallels between the origin of Rybka and the origin of
Houdini. I could be wrong in my speculation, but it seems that the
exact relationship between Houdini and the Ippolit family is being hidden.
Perhaps the stigma that would be attached to a modified version of
Robbolito or Ivanhoe is something Robert would like to avoid. I find that
to be sad, for if Houdini was a modified Robbolito/Ivanhoe and Robert
stated that publicly, I would be compelled to defend him. This of course
is supposition on my part; I have no proof otherwise and lack the ability
to determine the truth. I could only generate more circumstantial
evidence. Yet, this opinion of mine seems to be widely shared. And I
do not want to test an engine that I have doubts about the author's
honesty concerning his/her engine. I have done so a few times already
and wish to avoid that. Again, this is simply my opinion in regards to
Houdini. I reserve some right to choose the engines I test.
Roger, I'm sorry that I have used this simple question of yours as a
platform to voice my opinions. But I know that my answer to your
question leaves the door open to more questions by others and
I simply wish to close that door. I doubt that I have succeeded.
Adam
