Skip to content

GSA Hand-Spinning Platform Coming Soon?

I wonder how much interest this title will create!! ;) 

Why? Because I KNOW many would like more options... :|

I am unhappy with the software I'm using! The interface looks great. However, the software NEVER (or seldom) works perfectly. And the (few) updates haven't solved the (many) issues. :( 

I think @Shaun's recent experiences show us hand-spinning may be more important in the future, more so than auto-generated, auto-spun content.

@Sven, I know you had no intention of writing this a few months ago, but maybe in the future, we'll see a **SICK** spinning platform from GSA?! I can only hope...I guess this is an undertaking, really, since there are all the different languages that would have to be done.

Since this is just imagining, why not say what features I'd like?

-Many spin options. I would like it so ten users could end up with ten different sets of finished spintax (or spun articles) by using different settings. Various spinning engines also.  (As you know, there are different ways to do this.) The idea: the ability to create highly variable output. Otherwise these spinners are USELESS---!

-The ability to truly spin EVERY available word, as well.

-TRUE part of speech recognition.

-REAL grammar: True matching articles with single/plural nouns.

-A reliable platform that auto-saves, has amazing (and rather simple!) editing features rarely found in text editors anymore, and WORKS!!!!!!


  • SvenSven
    hmm you have wordAI for that already unless I didn'T understand your request here.
    creating a good spin-db is not easy.
  • DeeeeeeeeDeeeeeeee the Americas
    edited January 2018
    I didn't get WordAI yet. 

    I know it's been mentioned before, but I still haven't looked into it very deeply. I think I remember someone on here saying it doesn't have a hand-spinning platform, just really good spun output?

    I have the license for another product, trying to be respectful to them and NOT mention names.

    "creating a good spin-db is not easy" -Sven

    True; the more I thought about it, the more I realized even if the program were coded perfectly, the rules and DB have to ALSO be assembled carefully for each language.  And that's a lot of words and associations!!!

    Definitely a time-consuming task.

    I did think about "training", letting the program self-assemble rules and word associations by going through properly written texts in each language. Check via probability of past occurrences. You could also train with feeding it lists of common grammatical errors.

    (pattern matching this way would be ridiculously sloooow when working, tho, but more complete, I guess? So limited in being OK to generate spintax as a platform to work on, but not so OK if a server-side utility to use with API with GSA-SER to spin scraped content at-the-moment, unless you cache the articles and use a cron job to keep the article caches filled to a predetermined level. Hmm...How could this part even work? I guess every time it finds (if also scraping)/is fed a new un-spun article it found to spin and send back and post, the spinner would actually send  another article that's already been spun and finally ready, from scraped content retrieved before. Wow. I've edited this post a record number of times! lol Sorry, reader for making your head spin. lol)

    But I see it's far more than that. 

    There's also a need for synonym lists with graded replacements as very good to very poor, and multiple definitions ascertainable by context.  It seems live people need to do some of this because such resources do not already exist, as far as I know.

    There's also letting users assemble the databases. (But then I guess you can't sell it without a database in place, only once it's put together--maybe for development period let users assemble it for a cheaper license, reduced by X price per contribution?)

    The program I have claims to recognize parts of speech. I don't see it happening, honestly. :/ The program I've been using creates spintax/spun articles with terribly obvious grammatical errors. Toggling the setting does nothing I can see, lol

    In any case, Sven, I know you hardly need business, or any other kind of,  advice! :) These are just my musings.

  • shaunshaun
    Try TheBestSpinner. Thats what I used to use as well as the guys I used to buy my manual spins off.
  • DeeeeeeeeDeeeeeeee the Americas
    Thanks, Shaun.  I will check it out. 

    I continued to re-edit this ridiculousness for another 40 minutes after your reply. lol
  • SvenSven
    @DeeeeeeeeIm not arrogant to think I can not learn from you or anyone else. I always listen to what you or others say. But creating a spin-db or tool is really hard. I made GSA Content Generator and dealt with this before.

    There is so much to take care of. Also see my English, it's something hard to understand as it's not my mother tong. So I better not try to make it for English without any help from you guys.

    I can surly do it in German but that would not help you at all and even that would be hard work.

    I am open for ideas though.
  • DeeeeeeeeDeeeeeeee the Americas
    Thanks, Sven.. :)

    I'm happy to hear you have an open mind. I didn't think you were arrogant at all, just extremely proficient and knowledgeable. :)

    I'm glad to know something I write may help others. You've done the same for so many users, bro. :)

    But I guess we all can learn from everyone and everything.

    I know Im going to be thinking about all this spinning stuff some more.  lol
  • SvenSven
    yea keep me updated. Maybe we can work on Content Generator and improve this.
  • DeeeeeeeeDeeeeeeee the Americas
    OK, cool. :) Sounds good, Sven.
  • andrzejekandrzejek Polska
    edited January 2018
    @Sven @Deeeeeeee spinners will be dead soon, future is: natural language processing:

    And its more interesting to build :D

  • shaunshaun
    @andrzejek That's some interesting tech, from my experience, it seems Google already have something like that implemented as my auto-generated content projects either held their initial ranks or fell where as human wrote content climbed.
  • @shaun it all depends on language i guess, beacuse while ranking ex. polish content i remember using random garbage words like "askx9ak skaklxo <link> akxja amskxi" and it worked:)
  • shaunshaun
    .com seems to always get the updates first then rollout to the other English speaking TDLs.

    I might try a non-English market if I end up getting some spare time.
  • DeeeeeeeeDeeeeeeee the Americas
    edited January 2018
    "spinners will be dead soon, future is: natural language processing:"

    andrzejek: I definitely agree! ! !

    I've been thinking about spinning all night, and I've found this is a deep topic. lol

    We're really talking about language comprehension, or at least something akin to this.  I also watched some videos on YouTube as well last night on AI and "emotional" robots. (Found a seeming fake video with "Sylvia" and her maker, where she says, "EXTERMINATE! EXTERMINATE!" at the end.)

    "And its more interesting to build"

    I agree. The current level of spinning (I've encountered) is just Mad-Libs®.  Now we're actually getting into AI and the future. Wow. I didn't realize the implications, just wanted a better spinner. lol

    When I began playing with computers, I remember how much fun it was to write a random sentence generator using the names of my friends and putting in humorous verbs, adverbs, prepositional phrases, and other parts of speech. :D

    But truthfully  I still crack up on the floor reading some spun output or even spintax choices I get.  Hilarious!!!!! :) But..back then it was deliberately silly. Now it's accidental.

    My issue with the software I've used is that it's not even a good sentence generator. At least the one from when I was in middle school worked based on parts of speech. But of course, that's pre-loaded word choices, and so not quite the same.

    "it seems Google already have something like that implemented as my auto-generated content projects either held their initial ranks or fell where as human wrote content climbed."

    This is very interesting. Auto-gen projects fell?! So we're in a new era, I guess.  Right now I'm thinking about various basic sentence forms and types, as well as the linguistics of HOW to recognize parts of speech, and how context reveals which pos is in use for words with multiple possibilities. I'm enthused!!

  • @Deeeeeeee take a look at github, there is a lot of libraries already done to handle a lot of content related tasks
  • DeeeeeeeeDeeeeeeee the Americas
    edited January 2018
    OK, thanks. I've learned in recent years not to waste time re-doing stuff already out there. I already have an account..or two..on there. I appreciate the help. :)
Sign In or Register to comment.