The other cloners had 10 years to ride up on Google’s panties. Indian coders, Chinese coders, morons with a few thousand bucks that lost it all on elance, morons with a few million bucks that lost IT on valley devs and college kids. You name it. They all fscked up. They all tried to scale the wall, and they all died.
1 kernel developer is 200k per year and up.
You CAN NOT ride up on Google without one. You CAN’T. You can’t have developers that think that PHP is the computer, and that their code gets processed as instructions on the CPU. You can’t have Kool aid. Kool aid doesn’t work in Google land. Google is hard core.
Zoho came close, but they missed the most important part, the SEARCH ENGINE DUH? People don’t visit Google for peripheral services.
Some portals think that if they just build everything “around” the search then eventually it will “all work out” and they will get the 200k per year kernel devs and be able to finish it off.
It STARTS with the search engine and the other sh1t builds AROUND it.
*** I’m now speaking to a hypothetical arrogant VC who throws money at Tech problems for a living ***
It’s too expensive?
You can’t find interested kernel devs? (shock)
You hired people on elance?
Well you just lost all your investment didn’t you?
There’s a floor in this game and the floor is higher than the highest high you can think of in that little head of yours.
Your little money pile isn’t SH1T to the people with the knowledge to carry this type of thing to fruition.
I am having too many problems with the Fedora setup against kernel compilations. I am only going to use Fedora for the mirror servers running lighttpd and the various daemons that sync across the distributed computing system.
The compiler and all other backend daemons will be running on a new x86_64 compiled version of my own GNU based Linux distribution, Deity Linux.
This version will not be made public, nor will the sources. As it is not distributed as software, it does not fall under the clause of the GPL that would require to make the sources available.
b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License
3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:
This program is not being publicly distributed, so like Google and IBM did, we are going to use GPL code internally without publishing the sources.
“All machines run on a stripped-down Linux kernel. The distribution is Red Hat, but Hoelzle said Google doesn’t use much of the distro. Moreover, Google has created its own patches for things that haven’t been fixed in the original kernel.”
So that’s that. I knew making my own Linux Distribution would come in handy some day. Also it will be updated to today’s packages and re-vamped.
But YOU will never see it, as it will be powering OUR search engine!
I felt a bunch of warm fuzzies when I saw the bright yellow Deity Linux boot screen text again! I forgot how awesome it is.
When Peeplr becomes huge and some fruit loop in India tells you they can clone it, you tell them.
You CAN’T. Because it runs on a closed Linux distribution like Google and IBM. You can’t clone it you little b1tch3s. Then watch them run under the bridge.
As if the low people that do that sh1t could hack the kernel anyway or do any serious speed optimizations, or any distributed computing architecture. People that do that stuff are only good at UI. At faking the functionality. That’s all they are.
There is the business plan. Of course this will only work if the search is cool, accurate and motivating enough for people to use.
It will be.
Yup, I’m still working on this old kernel hack. It’s really a HUGE pain to work on the kernel as opposed to any other piece of software on Linux. The code base is so huge that when you rebuild it takes forever. There is no integrated debugger, so you have to printk() debug, and it totally sucks.
I also have a night job to do for some hosting company to which I sold a social networking license for our social networking software last summer that we haven’t updated in 2 years since 2005. They always ask for this or that little change when they license the thing and it always goes on for a few months. I usually tell people to just hire somebody locally but they never, ever take my advice.
I tell them I can’t hire somebody for a week or 2 to edit files, but they insist. The $40/hour I get is a pitance, and I wish they would actually just hire somebody. They don’t understand that they can’t possibly succeed without a staff. You can’t deal with ignorance so I digress and just do it after work anyway. This is the last time though, next time it’s total fsck off land. Even this time it’s fsck off land because I’m not doing it all. I hate telling people to fsck off, but sometimes you have to be reasonable where they are not. If you want to invest in something like that it costs tens of thousands of dollars(for starters). You CAN’T cheat and succeed. You have to hire people at a reasonable salary. You CAN NOT automate to success. They still think you can win the lottery on the web and you can’t.
So yeah, I’m in rotten land. Plus today’s whether is actually FAR WORSE than yesterday’s in Quebiria. I’m expecting to drive by a gulag any day now. The taxes here feel about right.
PS- Before you think something like “hey, isn’t cloning google’s search functionality like a lotto ticket attempt”
My answer to that hypothetical question, is yes, it is, but I have the key ingredient to make it work. I have sustained human resources with superiour programming skills. The ones that die off do not. They either have no programmers, or programmers that are good at PHP which is very, extremely week as a skillset. What they call Kool-aid drinkers. The kind of people that would be lost editing kernel code, or doing any type of systems optimization or distributed computing.