81 replies to this topic

[Alan Perkins]

Sounds a bit like most of science, Alan, certainly on both the macro- and micro-levels, if not also on most in between.

Sadly though, Ron, it's nothing like science. Science is about trying to make sense of the great mysteries of nature - not trying to reverse engineer the algorithms of man.

[Ron Carnell]

Some might say the difference between great mysteries and algorithms is only one of perception, Alan. I think it was Arthur C. Clarke who remarked, "Any sufficiently advanced technology is indistinguishable from magic." Seems to me it's equally so that any sufficiently complex algorithm will seem like a great mystery to most.

Of course, the real difference isn't in the complexity, but rather in the source of the mysteries. Pasteur, Einstein, and Hawking revealed to us secrets previously unknown to any man, not secrets devised by man and merely hidden from the view of the many. And, yea, I'll grant you that is one mighty big difference. Something tells me, though, that each of those great scientists spent a lot of years using the scientific method to deduce what was already known by others long before they were capable of using it to explore what was known to none.

The real point, however, wasn't a question of whether one "should" unravel man-made algorithms, inevitable as that might be, or even really whether one can make those determinations for others. The real point, rather, was that not being able to look under the hood, not being able to ever really "know" if one is right or wrong, is pretty much par for the course in most human endeavors. Real knowledge, I think, is about probabilities, about what is "more" or "most" likely to be true based on what we've seen so far. Certainty, after all, is just another word for faith.

But then again, hey -- you're talking to a guy who spent every available hour for six solid months in 1980 struggling to solve the great mysteries of a Rubik's Cube.

[Jill]

But then again, hey -- you're talking to a guy who spent every available hour for six solid months in 1980 struggling to solve the great mysteries of a Rubik's Cube.

Did you get it? I did it a few times, but never quite had a solution that I could come back and do it every time. Can get mighly close, even today!

[Karri]

Some might say the difference between great mysteries and algorithms is only one of perception, Alan. I think it was Arthur C. Clarke who remarked, "Any sufficiently advanced technology is indistinguishable from magic."...

...Real knowledge, I think, is about probabilities, about what is "more" or "most" likely to be true based on what we've seen so far. Certainty, after all, is just another word for faith...

...But then again, hey -- you're talking to a guy who spent every available hour for six solid months in 1980 struggling to solve the great mysteries of a Rubik's Cube.

First of all, your Rubik's Cube flashback is freakin' hilarious. Sigh ... the 80s are a scientific mystery unto itself.

But I cannot resist disagreeing a wee bit with one of your suggestions (I'm a little rusty, so bear with me): People like Einstein unraveled some natural mysteries that--at least in this strange dimension we call life on earth--had not been articulated or understood as such prior. Same goes for many other scientists before Einstein (you know, like the guy who was going to be hung or beheaded or something because he said the earth wasn't the center of all that is.) These were scientists in the truest sense of the word. They were plain old ballsy and pursued the truth for no other reason than to improve our understanding of where we fit into this crazy scheme of things called the 'universe.'

Now, I would say that all of the discussion around what the bleep Google is up to today, yesterday, tomorrow, or in 6 months is more of a political debate teetering on philosophy and pretty darn close to an exercise in reverse engineering. But science? Hmmm, not in IMHO. Ten years from now no one will care much what the 2005 Algo of the Year was. But in a thousand years, people or martians or whoever will STILL be talking about the theory of relativity, E=mc2 and what such discoveries meant to the history of the world.

Google is a technological enigma created by people trying to make a living and maybe having a wee bit of fun along the way (aren't we all?). And if reverse engineering the latest algorithm change makes for a good brain teaser (and ok, ticked off a few SEOs), then that just makes life all the more interesting!

Thank goodness for common sense. Too bad it ain't vogue anymore because it sure is reliable!

Karri

[qwerty]

Well, I know it's off topic, but what the hey.

The scientific method involves looking at what's there, forming a hypothesis about why things are the way they are, or how they'll change under certain circumstances, testing that hypothesis, and coming up with proof.

What Einstein did was to a great extent theoretical. A number of his ideas (including relativity) were proven later.

[Ron Carnell]

I suspect had I not finally solved the Cube, Jill, I'd probably STILL be trying. And, yes, since I took fairly methodical notes (we are talking about science here, remember), I was able to do it very consistently. Perhaps the most telling thing about that incident is that I wouldn't LET anyone give me hints or cheats because that would have utterly spoiled the challenge. Makes me wonder . . . if Google mailed me all their source code, would I *really* want to open it?

And perhaps the answer to that question lies in the fact that I haven't touched a Rubik's Cube in over two decades?

Kari, the guy you're thinking of, I suspect, was Galileo, who didn't invent the Copernican theory of an Earth circling its sun but made the horrible mistake of supporting it and the worse mistake of being a very visible target. He was convicted of heresy by the Inquisition and spent the rest of his life under house arrest.

And, yea, I certainly agree that ten years from now no one will care about a search engine algorithm used in 2005, though I would probably have to add there's not a lot of excitement about Galileo's pendulum clock today either. But again, I don't think that's the point. There are, I believe, many good arguments against trying to reverse engineer a search engine algorithm -- but being unable to look under the hood or to ever know anything with certainty isn't one of those reasons.

Bob, I was with you every step of the way right up until the "coming up with proof" part. The hallmark of the scientific method is predictability and repeatability, flip sides of the same coin. If I do A then I can expect B (predictability) and every time I do A the result will be B (repeatability). Put another way, an hypothesis that doesn't make a prediction can never be tested, and an experiment that can't be repeated at will is useless. Proof isn't necessary (or common) and, indeed, the Theory of Relativity is still called a theory because it has never been conclusively proven. Relativity made predictions (many, in fact) that have been tested and, so far, no observation has ever contradicted the theory. However, Newton passed that same test for some 300 years before a contradiction in calculating Mercury's orbit arose -- which ultimately led to Einstein's Theory of Relativity.

It's interesting how many different directions this thread has taken of late, but I honestly think everything discussed -- from the challenge of the Rubik's Cube to the persecution of Galileo to the vagaries of scientific proof -- remain very much on-topic.

[Jill]

I suspect had I not finally solved the Cube, Jill, I'd probably STILL be trying. And, yes, since I took fairly methodical notes (we are talking about science here, remember), I was able to do it very consistently. Perhaps the most telling thing about that incident is that I wouldn't LET anyone give me hints or cheats because that would have utterly spoiled the challenge. Makes me wonder . . . if Google mailed me all their source code, would I *really* want to open it?

This is very interesting to me too because it illustrates the differences between us.

I too would have wanted to smack someone had they told me the answer to the cube and would not have bothered with it. My sister, in fact, was able to solve it before I did, which of course gave me more reason to want to do it.

But more interesting is that you took notes so you would have a complete solution, where I just did it, and did it, and did it until it eventually popped into place. I could still do it at other times, but not like clockwork because I never quite knew exactly what I did to make it work. I'm sure I could still do it today, but it would probably take me a few hours. (Where as you could probably do it in a few minutes.)

I think this may translate into my SEO style too! I just keep doing stuff until it works, but never have an exact formula.

However, the important thing here is that for SEO this is actually much more suited, because there isn't and will never be just one solution. Since the algos are constantly changing, and there are so many ways to get the same results, my way of just doing stuff till it finally "pops" appears to be the perfect method!

[Karri]

Jill & Ron: I think you two are the perfect illustration of how art & science are often not as different as we think.

So, if the "hallmark of the scientific method is predictability," where does that leave SEO? Can SEO results be repeated over and over? Probably not in an entirely precise way. If SEO were toooo scientific, and thus too predictable, it might not be as interesting (or fun!).

Art is the journey. Science is the methodology we use to add a sense of order to the chaos.
Karri

[Randy]

That's one of the reasons why around here we like to say that there is no such thing as cookie cutter SEO Karri.

Are there some things you can do over and over again and end up with halfway decent rankings? Yes there are. Thinking specifically about ranking and not at all about actually selling, I think most here who have been doing it for years could lay out a purely scientific plan to get a site in 99% of the markets out there into the top 50. Do all of the things on a checklist and you're there with today's search engines.

The problems become

• The site may not (probably won't) convert visitors into customers taking a purely scientific approach.
• Each market and each site is different, so while you may be able to get it into the top 20-50 you may not get much qualified traffic.
• The search engines themselves are always changing and becoming smarter. They're always looking for ways to bring more Human Experience into the mix.
• It's no fun at all! There is no challenge.

At the end of the story, in order to make a site the best it can be and maximize its potential you need to use some science and a lot of art.

For good or ill, the two are totally different disciplines.

Moral of the story? I was told when I was young back in the prehistoric Internet days that I was weird because my main interests in school were Mathematics and Music. Secentifically proven equations and art. I can't even count how many counsellors, teachers and professors told me that my interest in the two disciplines just wouldn't work because they had nothing in common. I could do both, but they would never intersect.

They were all wrong.

[Jill]

It's no fun at all! There is no challenge.

That's why I would never have written down my process on the rubick's cube. Cuz I'd never want to do it again.

What fun would it be to have the SE algos all figured out? (Aside from the obvious monetary benefits!)

Funny, Randy about the math and music. Sounds like my son!

[Ron Carnell]

So, if the "hallmark of the scientific method is predictability," where does that leave SEO?

I wish I could liken SEO to the weather, a science that falls within Chaos Theory, where the number of variables is simply too large to support absolute predictability. I wish I could, but I can't.

I think SEO is more like one of the soft sciences, such as psychology, where the variables aren't just astronomically huge but are also often unique to each person. Alternatively, we could say SEO is like one of those board games, such as Battleship, where your goal is not only hidden from you, but is also constantly being modified. Whether we compare it to psychology or a board game, though, the key feature is that there's another human being sitting across the table and THAT has historically defied prediction.

Doesn't mean we'll ever stop trying, though, as anyone in a relationship or playing poker will quickly attest.

That's why I would never have written down my process on the rubick's cube. Cuz I'd never want to do it again.

Which is likely why I haven't touched a Cube in over twenty years.

And, Randy, in my opinion your counselors were full of crap. Math and Music are intricately interwoven, and it's not at all uncommon to find people with interests in both. I think it is only at the most rudimentary levels they might appear to be different, but beyond that level, both Math and Music rely on the recognition of patterns and the weaving of those patterns in a creative way.

Anecdotally, I think I have known far more programmers who played musical instruments than not, a fact that never really surprised me. What I did find puzzling, over the years, was the number of programmers I met who also flew airplanes?

[Randy]

They always frowned when I tried to explain to them that they were full of it Ron. Dunno, the connections always made sense to me. As does programmers being musicians.

The programmer/pilots probably has to do with the attention to detail that both take. Who knows though.

[lyn]

I can't even count how many counsellors, teachers and professors told me that my interest in the two disciplines just wouldn't work because they had nothing in common.  I could do both, but they would never intersect.

They should have looked a little further, Randy! The correlation between the mathematically talented and the musically gifted is actually very very high.
You're not alone!
When the Ontario government had the hatchet out to kill non-economic school programs, I was involved with a group defending the music programs. There's an amazing amount of documentation on how music instruction and practice leads to higher achievement in math and computer sciences. There's also quite a history of musical mathemeticians. And vv...

Hope this never got in the wayof your vocational choices!

L.

[lyn]

Certainty, after all, is just another word for faith.

Thanks for this, Ron. Yours?
It's a keeper...

L.

[qwerty]

For me, I guess it depends on your definition of the word "faith." If it's

confident belief in the truth, value, or trustworthiness of a person, idea, or thing

I'm with you, but if it's

belief that does not rest on logical proof or material evidence

...sorry, no can do. Sure, some people can be certain about something like that, but not everyone (and I'm talking about more than just religion).