Beating G.Panda

I've been wanting to mass create websites, blogs, and forums depending on a specific keywords and grabbing the content from all over the web. The biggest hurdle now as I understand it is the G.Panda update, which penalized such websites.

However, I am wondering if automatically regenerating/rewording "stolen" content to make it unique will beat Panda? Technically, this regenerated content will be unique and will be the only one of its kind, even though it does not have to be semantically correct for human reading. Or does the big G actually have some algorithm to detect non-semantic content, thus making content auto-regeneration not working?

For those still doing autoblogs and grabbing content from the net, what is your take on this whole G.Panda issue and how do you beat it?

Thanks.

I am looking for same answer.. But I think G will know..

I didn't have any issues with Panda and I have a lot of niche auto blogs with 100% dup content. I honestly think it was not as big of deal as some people made it out to be. I know some of the really big boys like ezine and ac got singled out and there was undoubtedly an algo change but I think its just a factor, if you still have strong backlinks and are pulling regular content you'll still be OK it just may take a little extra work, just my personal observations though.

there will be a new project called titan panda, maybe by the end of 2011 to 2012. and it has to do with digital licencing from google, it will be an icon placed in each page thats the license, in any corner of the page, and their spiders will look for this code first. you can only be index with the acusition of this digital licence, the bots take your article stores it in their data base, then give you this digital license.

So if someone else has the same words or fairly close enough article, then the original content that got license is considered genuine, and the other dupcon, hence getting deindex. BIG G has an issue, with this right now, because it doenst know whether to award this digital licensing to be able to be read by their bots, because people who owns the originals might get left out for something they created.

Right now they know there will be alot of legit webmaster casualties, so the implementation of the second stage of panda aka "titan panda" is in early stages of deciding wheather to implement or not.

i have a family member that works in googles research and development DEPT. panda algo, searches the first letter of everyword in the article if even you manage to change some words with syn, it will stil detect dupcon, the article needs to be under 30% not the ame as far as first letter of each word in each sentence in the article, in order to be considered genuine.

now what i found out that is very effective is to translate the content to another language, then i manually retranslate the content back to inglish. also i have one domain name that i do all my auto searches for dupcon, then on my other sites i dont have anybots running, only on the site that is supposed to collect articles, this one blog site, is not allowed to publish, only to collect. all my other blogs are hosted elsewhere, and not on the one that collects articles, i do a normal copy and past to a translating software to spanish, then i manually in my own words translate it to english and adding about 10% my own lines. and so far i have no issues i have high ranking blogs that get anywhere from 3 to 4 articles a day each article les than 300 words.

I generate lots of money this way. The bots only serve me for getting content which i rearrange. You cant expect to be serve everything in the mouth. some of these subject specific blogs i have are very keyword dense ,and subject specific, and i have a good 4 blogs having about 50 to 60 article each. i retranslate manually to inglish about 1 article per subject blog. Excuse grammar errors, typing in a hurry, no time to edit.