This is where JayBonci keeps all of his proposals for widespread sweeping changes across the system before he implements them. This is a general request for feedback on any one of these proposals. Click on the link to go to the appropriate section.


Database topicing

A new method of finding writeups



So it struck me today that we could organize the entries in the database into a topic tree. People who wish to have their items be more easily found can add their items to some place in the tree. (as a low level power). This will make content easier to find without searching. Many search engines (see Google and Yahoo) take advantage of this sort of organization to control their millions of entries. Why shouldn't we.

This could be yet another form of writeup integration and topical cohesion between writeups. It would also be an alternate search or browsing mechanism. We could give the ability to add to a topic (and we would only display it when viewing the particular writeup (not the entire node that it is in), thus it would be a nominal hit on the db.

So we invent these topics:

  • Everything.Arts (Music.whatever)
    • Movies
    • Music
      • Classical
        • Beethoven
    • Everything.Fiction
    • Everything.Reference
    • Everything.Original (.Poetry.etc)
    • Everything.Administrative
    We could organize content however which way we want (the rest of you are the smart and literate ones), and showcase what we have (and holes in our current stuff) very quickly if we got this categorization together. Like firmlinking, this will help only a little in the short term, but in the long run (over a few months), it may really give this place some cohesion and organization.

    And allow users to help their writeups get read by associating them with topics. The topic would only get displayed when viewing the particular writeup (as the rest wouldn't have much to do with it.

    This is just a rough idea, but I think this might be another powerful tool to allow editors and gods to do more than just kill. It will allow them to shape and mold a little more precisely. We can showcase the top level stuff on the front page, and maybe pick a random topic every day. Also topics could be searched for. Chess would show up in a search as a (topic) even though it's fulll topic chain would really be Everything.Entertainment.Games.Board Games.Chess ... This would be very easy to do, and at nil db lag cost. Thoughts on this would be great.

    For the techheads, a topic would inherit a type of nodegroup, except that it can only contain writeups and other groups. We can let the editors and gods nest them in there, and let users add something to a topic that's at least three levels down. (We can lock the top db stuff). You can probably see how this would play.


    Database compacting

    A potential speed increase by deleting broken records.



    I was tossing and turning in bed tonight, and it hit me how we can pick up some speed. I had this idea vaguely before, but now I actually know how we can implement it. Basically any index we can check for invalidity with any certainty. Others may be checkable as well. The speed boost from toasting beaten up records may be as much as 5-10%. I really don't know. We need to shave records out of the following dbtables:
    • users -> (items that don't correspond to a node_id)
    • links -> (bookmarks that aren't anything)
    • links -> (softlinks from dead nodes)
    • links -> (softlinks to dead nodes)
    • coolwriteups -> (cools on nodes that are deleted)
    • writeups ->(bad parents) *DB SCOUT
    • writeups ->(deleted users) *DB SCOUT
    • writeups ->(parent id of zero)
    • messages ->(deleted from_user)
    • messages ->(deleted to_user)
    • nodes -> invalid nodetypes
    More to be added later.


    Everything2.com Tutorial

    An everything2 tutorial should include the following (* means an interactive step)

    • Sending messages *
    • Deleting messages *
    • Submitting writeups *
    • Choosing titles
    • Finding prefs
    • Adding personal nodes
    • Homenodes (editing your homenode) *
    • Jargon, etc
    • Chatterbox
    • Bookmarking *
    • Personal nodes *
    • Levels
    • Other users: Gods and editors.



    The Best of Everything

    Problems

    Real world incentiveThere is not a lot of real world incentive for people to join Everything2.com. While people join it and gain prestige as a great writer on the system, the community is largely inbred in its ideas, talents, and personalities. There is no large draw for other people to join in.

    Bad business model Everything2.com has a lot of repeat traffic and thus is hard to market on a level that would make it profitable. E2, while really a great web experiment, does not provide a product in any way. This proposal is a method to identify the quality in workmanship that exists here, and to market it as a product.



    Implementation

    I would propose that we create a new system called Best of Everything (from here on out denoted as BoE). BoE would accomplish three things: Identify and heavily reward quality on the system, keep content moving and always fresh, and give an incentive in the real world for a contribution to the Everything2.com system.

    It is my proposal that we create an Everything2 "power" similar to C! that is Best! or B!. Bests would be considerably more rare in distribution, level seven and higher, one per month. Gods and editors would have unlimited B!s and every ed-cool would be considered towards the Best pool. Bests are a special notification that can only be used on timely content, IE within the last month or two months. The user receives a B! and some manner of arbitrary XP award.

    The recipients of a B! would then be thrown into a pool from which to publish a bi-monthly literary magazine for sale (or even general distribution initially). The publication (roughly titled in my mind "The Best of Everything") would be something that displays the best that our site has had to offer in the last few months. Excellent stories, biographies, etc. The gods (and possibly a select group of editors would decide which items made it into the final publication.

    This publication has its copyright implications. Thus, every recipient of a Best! from a high-level user or god will be brought (or directed) to an authorization screen, where a person can accept or decline the nomination for publication. The person will then give their consent to Everything2.com for the purpose of publishing the work in a newsletter. All of the copyright implications would be spelled out there, and users would be required to authorize each best they received. This allows us to maintain a record, should something arise. The point would be to acknowledge it as either yay or nay, and not to default to either (nay on the safe side). A user should also be given the choice to publish under their real name, or a pseudonym. The point here is to cater to the writer, and to do what they ask with their publication.

    The thought of real world publication, fame on the system, and possibly a resume builder through publication on the site and on this system would be a huge draw for many authors and fledgling writers who want to try their hand. We'd see more fiction, better writeups, and an increase of overall quality in which to receive that coveted B! from a user or god. Even if we cannot print a version of the newsletter or literary magazine for the first few months, it still gives us an incredible jumping block for the material used here.

    Technically speaking, this is a very simple extension of our current C! system is now. This will not replace the C! system, as it is for site recognition. In fact, I can see a C!ed writeup in consideration for a Best!, due to the increased exposure it has. I believe that this could in fact add a real element to the site, that we are lacking now.




    Level Powers

    Shielding: (proposed level 9 power)

    "Your integration with the database has increased with your ascension to Level 9. The database now surrounds you and protects you from harm." The basic gist behind this one is that it's a $USER var that protects a level 9 user against the loss of XP from two downvotes a day. They still lose the rep, but they don't lose the XP. It's a guaranteed way out of the XP loss. Not that the level 9 people need XP, per se, but still. A message will show up that a shield has been engaged (so that they still get the message that something is being downvoted). It's a cool power that would be easy to implement. Currently lv 9 is without a power.



    Writeup Re-casting (proposed level 7 power)

    One recast per day, regardless of level. It would allow you to stick a writeup yours or someone else's on the new writeups list again (presuming it has been changed / modified. There would be restrictions on what could be recast:

    • Nothing that has been C!ed
    • Nothing that was written within the last week
    • Nothing that has a reputation above 5 (10, maybe? Balance issue)
    • Optionally, recasting it may reset the reputation to zero.
    This would provide incentive for a person to rehash their old, dead writeups. Ones that just didn't do well a while ago, and need to be re-thought and re-worked on. This would also help level 7 Mentors with their Mentees to make changes to a writeup, and to give it another seed of exposure, without Cool!ing it. At one per day, people

    Technically speaking, this would be a simple hack of the newwriteups table, and there are no ill technical side-effects to it. It would be implemented as an opcode.



    dem bones says :

    • I don't see much point in "shielding" and would generally like to avoid going *towards* too much more new xp-related stuff.
    • The idea of an E2 paper publication has been kicked around much and what JayBonci describes above has generally been the idea ... but we're not a funded company and the legal / publication fees and investments are not available. Rancid_Pickle has looked into this as well and we've sent a few e-mails around. For now it's a no go unless there are some financial volunteers and some experienced people willing to put in the work of organizing a publication.
    • The 'recast' idea is part of what I was going for with the multiple C!s allowed on writeups now. People 'recast' by putting the writeup back on top of the Cool Archive. I'm not too keen on seeing ENN be mixed old and new.

    All that being said I'm certainly open to more comments here.

    Whatcha all thinking?

    Gritchka babbles:

    Tutorial

    Sounds a very good idea: and should be made compulsory for new users.

    The FAQ is not clear, for someone starting from nothing, especially if they're not computer types. Highly intelligent people have been baffled about what to do, where to begin. I've been writing my own solution to this; I was hoping to have more done before today, and might try to finish it over the weekend. This is how i mention it on my homenode:

    I am creating a larger website to act as a kind of FAQ and beginner's guide. It will not replace any detailed explanations, but will put simple explanations in one place, with pointers to where to go for more. Eventually I might bring it into E2. It's not finished yet, but you might be interested in what I'm trying to do. Comments are always welcome. It's at
    http://uk.geocities.com/zetete/e2whatis.html

    Best of Everything

    It doesn't excite me too much but it could well be of great interest to others: worth asking for expressions of interest.

    I don't think anyone should have unlimited B!s. I'm getting comfortable with my unlimited C!s but would be happy to have a constraint on a significantly stronger power.

    How can we ask for public expressions of interest? Could we perhaps designate a poll node where everyone is welcome to contribute a short write-up (perhaps even with a size limit), and where all contributions will be removed after a fixed time? Mentioned in front page news?

    Shielding

    No need; XP and write-up loss is a perennial fact of life for anyone who's got that far, and they should know what the quality of their material is and how much of it is likely to disappear. A high-level user who wants XP can spend a couple of hours voting; this totally dominates all other effects.

    Re-casting

    I don't like the idea of using New Writeups for this. Some official advertisement of a major change is a good idea. Everyone's got four-line write-ups when they were at level 1 or 2 which they could turn into a barnstorming two-page thing of their present high standard. How often do you see people advertising these on their homenodes, the only possibility at present? Very rare.

    How many level 7 and above users are there, who'd take regular advantage of it? Would it distort New Writeups? Would we need some evidence that there'd been a big change? (I don't think comparing before and after is possible now, is it?)

    What about a dynamic node Revised Writeups where people of the requisite level can add a write-up mentioning, say, two updated nodes and their dates. Editors remove them after a certain period. No fancy code: just a list like the title edit node.

    wharfinger mutters:

    This word "topicing" drives me batshit. "Topicking", maybe? There's got to be a better way, or maybe the Good Lord never meant for "topic" to be a verb.

    The feature itself might be groovy. The problem is that we'd be forever at loggerheads with idionice people who feel that their poem about how their mean mean ex-girlfriend stomped on their tender wiggly little heart with a great big frozen boot of cruel indifference and turpitude and stuff really belongs under the NASCAR topic, because like they were watching like Bobby Rahall on TV the day they bought her that kitty cat that she put in the blender and stuff which isn't mentioned in the poem but it's like subjective and personal, man!.

    I mean, it's one more thing for people to fuck up, which means one more thing for us to enforce, explain, and beat people up about in the chatterbox. I don't know if I'm for it or against it, but there is a definite downside.

    There's another way to put an old writeup back into New Writeups: Get it nuked and re-submit the new version. Of course that would be subject to abuse, but so's anything. The only real objection is wear and tear on the CE's.

    clampe chimes in, belatedly:

    I'm not sure about doing a periodical publication based on E2, for the cost reasons mentioned above, but a book of some of the best writings from the site might be a good idea. We could sell the book to a publisher, and use the proceeds to maintain the site.

    Obviously, this would not be a simple process. I'm drafting a query letter to a few publishing houses to see if I get any bites. Managing IP issues will be a pain, but not insurmountable. The real trick would be to find the right tone. We want people to hear our stories, and need to think up some way that the writing of the site could help accomplish that. Anyway, let me know what you think.