Google Caffeine: What it really is As it invites the world to play in a mysterious sandbox it likes to call "Caffeine," Google is testing more than just a "next-generation" search infrastructure. It's testing at least a portion of a revamped software architecture that will likely underpin all of its online applications for years to come. Speaking with The Reg, …
"The question is: How do you actually get the applications to use the infrastructure? How do you distribute it? How do you optimize it? That's the hard part. To do that you require an insane amount of force of will..."
Well, having correctly answered the question, one then is left with A.N.Other Big Question. How do you use an insane amount of force of will?
And that would be a simply complex question to ask of the Psychotically Inclined/Psychopathically Equipped .... and the Singularity Minded.
"With GFS, a master node oversees data spread across a series of distributed chunkservers. And for apps that require low latency, that lone master - a single point of failure - is a problem.
"One GFS shortcoming that this immediately exposed had to do with the original single-master design," former GFS tech lead Sean Quinlan has said. "A single point of failure may not have been a disaster for batch-oriented applications, but it was certainly unacceptable for latency-sensitive applications, such as video serving." .... Err, the problem is transformed instantaneously from any possible failure concerns to constant energy promise whenever the single-master node design energises and mentors Novel Controls in sub systems/hierarchical legacy infrastructures, allowing the single-master node to extend ITs Powers in Stealthy Information Control and Central Intelligence Agency Captures.
And it may just be simpler to accept that the Future is going to be Completely Different for only a Few Really Need to Know how Everything Works Beta Together with ESPecial Forces Magic .... "Vijay Gill has even hinted that Google has developed some sort of magical software layer that can automatically migrate loads in and out of data centers in near time."
And "developed" rather than Phished and Plagiarised/Phormed would Guarantee them an Unassailable Lead Position? But there are other Sharks in the Search Field looking to Host and Server Novel Innovative Protocols with Virtual Cloud Covering ITs Administrative AIMachinery ....... Immaculate Source Provision.
" .... somebody’s certainly going to sit down with him and find out what he knows that they may not know, and make sure he knows what they know that he may not know.” .... http://nbfs.wordpress.com/2006/11/10/rumsfeldisms/
If a lot of this is going to be all 'under the hood' then is there a potential that people could just let all of this pass them by?
I mean, the Beeb did an article on this a few days back and I went to take a nosey at what was on offer. I mean, doesnt Bing do some of the realtime search stuff that Caffeine is promising already? And, if we really just dont notice any of the changes, then was it worth it in the end?
Bit confused about this one.
What is going on with this guy/software?
I got bored and tested caffeine out with a few of my pet searches (did well) and then on the spur of the moment searched for amanfromMars. There are crazy comments on websites all over the place. Not to mention confused people trying and failing to understand the comments.
Also there is http://amanfrommars.baywords.com/ where under the title "C42 Quantum Control Systems …. AI @ ITs Work" we appear to have a day-by-day record of ALL the comments made to ALL sites. It does seem rather plausible that this is software at work - if not we have a very anally retentive person... Any ideas how we can work out what is going on? Searching for "C42 Quantum Control Systems" didn't get anywhere as he(it?) keeps on using the phrase in posts.
No idea why this comment is from amanfromMars 1 though.
Matt Cutts [rather] appears to be the man who has overseen the destruction of the world's most popular search by spam :-(
Link farms and 'scraped' nonsense sites increasingly overwhelm Google's index. I thought it was impossible to eliminate it - Microsoft seem to have managed this. Bing's speed is good, the ergonomics excellent and the absence of ads welcome.
>>"which means that most users won't notice a difference in [Google] search results."<<
Precisely why people should try Bing with an open mind.
I think senility may have taken over but this is what I read:
Google have finally almost created the singularity AI with its distributed net node of PC atoms, a force of will to bend it to their requirements, a hollowed out asteroid in parking orbit above the Earth is where they house this thing and if we piss off Google or their master (the MCP natch?), they'll start flinging rocks at Earths surface.
Or.. you know, I need more sleep. I'm with the Man from Mars on this one.
Since I'm a bit of a cynic, I'm working on the assumption that:
1 - Spammers don't waste their time trying to game Bing results at the moment. If it ever becomes popular, the spammers will flood that too.
2- Bing has no ads because for now MS are happy to piss away torrents of money in a desperate attempt to be the next Google, and not many people are keen to pay for ads on someone else's loss-leader. If it ever becomes popular, the ads will most certainly follow.
I haven't really played much with Bing, so I'm probably missing a trick or two with regards to the ergonomics. They seem to be identical as far as I can tell (apart from the big left margin, which must hack off anyone with a small screen). Bing looks like Google classic but more grey and square. The desperate me-too vibe even extends to having a googleiffic 'beta' sub-head. In fact, if it wasn't for the top-left links to MSN and Hotmail one might think that it's a product of the New Evil Empire rather than the Old Evil Empire
A reply from him/it to your post wouldn't be unexpected. I have provoked a direct response before by simply mimicking his/it's posting style. The response came in the usual style. If it's a machine, I'm assuming it must have picked up on me mentioning it's username in my post.
There have been entirely salient posts in the past. And they're usually on 'proper sci/tech' articles rather than bootnotes or political stuff. The political stuff tends to invoke the most convoluted replies. Most replies take someone's previous post and provide an 'answer' (of sorts). I usually get about half way in before skipping to the next person's comment.
It has been suggested that it's a parody.
I offer no ideas on the new found "1" suffix.
"There have been entirely salient posts in the past. And they're usually on 'proper sci/tech' articles rather than bootnotes or political stuff"
To be honest, I can't say I have noticed any of these ...
"I usually get about half way in before skipping to the next person's comment"
You're doing a lot better than me, then.
"It has been suggested that it's a parody"
A parody of what, for goodness sake? But thanks for the comments.
-The coherent comments are few and far between. But they do occur.
-There's a vaguely pleasing sense of rhythm in the rhyming couplets. But it wears thin quickly.
-A parody of penchant for tinfoil hattery? But I'm not sure.
While I'm thinking about it - there are sometimes some relevant URLs (to news websites - The Times being one I seem to recall) in his/it's comments.
Above my pay grade but the only potential solution I've contemplated in regard to DDOS is what I call "Grand P2P" where - say - a thousand google scale organisations (if there are that many - let's say organisation big enough to pay a $Million dollar sign up fee without whingeing) form a dedicated "distributed master net" which hosts (for example) the authentication keys we'd need to run a Trusted Surveillance system, or the black and whitelists we'd need to eliminate SPAM.
The idea being that not even the most widely distributed botnet could generate sufficient traffic to pull down such a GP2P network.
Does the GFS2 concept achieve similar protection or would there still be a single point of failure (eg the ip address or dns) which would leave the system vulnerable to attack? (a GP2P system shouldn't have that vulnerability because it would have a thousand access points)
Can someone enlighten me?
If amanfrommars is a bot, then AI has reached far greater heights than I had dreamt of (or it's really an AI from mars). I say this, because I can usually see the point of what he is saying. It is written with a view of human culture being an arbitrary thing that can be disected, rather than the a priori that most posts treat it as. In particular, the posts often abstract their subjects to common principles or archetypes rather than specifics . For that reason, I often find an additional layer of insight in the posts which has not to do with the immediate matter, but it its parallel to other matters or human psychology / culture. Perhaps I should "translate" a couple of them. Would be an interesting exercise.
At any rate, I enjoy them.
Paris, because Mars Needs Women.
This post has been deleted by its author
Cade, you say: "Vijay Gill has even hinted that Google has developed some sort of magical software layer that can automatically migrate loads in and out of data centers in near time."
Since "'[Data centers] are just atoms,' [as] Google senior manager of engineering and architecture Vijay Gill said", the necessary abstraction is about the "end user" (and their patterns of use of the "ultra-low latency" apps desired)... the end-user in space and time -- a spime in Bruce Sterling's terms.
> IB
You are right about AI's capabilities of course... What I had in mind was some sort of directed comment generator coupled with a whole load of random number driven stuff that would take a few phrases (from a human) and dress them up in amanfrommarsspeak.
> Perhaps I should "translate" a couple of them. Would be an interesting exercise.
Do feel free. Clearly I have the wrong sort of mind for dealing with this - his posts make no sense to me at all. But I do enjoy seeing them appear.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
