h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

OK - it's time we had some new ideas to discuss. There have been suggestions about category selection, and all that sort of thing. In the next few screens of text , I intend to explain my idea and dismantle the usual arguments against to try and get you to think of new reasons why what I've written is Belg*um.

There is currently, on the front page, a fat directory of Edited pages and Help pages on h2g2. No, I didn't think that you used it often . Why not create a volunteer group that categorises things from across the guide. Obviously, the Edited things would be categorised by default, but it would allow any Entries to be incorporated. The whole thing could work on the following basis:

* Edited Entries
* Entries
* Conversations of Note
* Other DNA Pages
* Useful Web Links

One page could be made to stand out in particular - a great Entry or hilarious Conversation, perhaps. The whole thing could sit in the same box, working with the search engine, and be edited by category editors, and a few others, with various privileges.

Pages could be added in two ways - by submission to the category and acceptance by a category editor, or by being found in the Guide.

The system would be more efficient than the current search, would be easy to use and comprehensive, and provide links with a short, informative description. Certain things could be appended automatically - author/editor, date etc.

It'll never work, I hear you say. Never is a dirty word. Look at this: http://www.dmoz.org/ - it does work!

Now you know where I got the idea from. Of course, not all of the pages on the net are covered by the ODP, but I've not found a page anywhere else that wasn't there, and I've seen pages there that weren't easy to find on, say, Google. With the web, it's a massive challenge. h2g2, by comparison, is just a big challenge.

Now discuss.

Whoami?

As they say, discuss... because the technology is already there, we just don't let non-staff see it. So this is less a feature request, and more a good discussion to be had about how on earth one would organise such a volunteer scheme, and whether *further* tools beyond basic categorisation would be a good idea.

Over to you, Community.

Excellent news! I intend to set up a page for this idea, so that it can be all discussed and the progress documented.

If you read the entry on The Open Directory Project, which is due on the Front Page soon and is to be found at A718553, you'll get some idea of how this kind of system works.

A few things to discuss:

1) The process of allocating a category to a volunteer need not be so difficult to pass - if the user is around often, and experienced enough to manage, then they can be accepted.

2) Category editors ought to be acknowledged, and would need a bookmarks subcategory - like mine at http://dmoz.org/Bookmarks/W/whoamiuk/ to place entries that they personally like, or are in the process of moving/listing, however they would't need a profile - just a PS link.

3) A submission system would need to be set up. A link to an entry, description and comments form could be made, but rather than the optional email address system used by the ODP, we could just list the user like we would a conversation.

That's all I can think of right now...

Whoami?

A743014 is the page that I have created for this idea, but its purpose is explanation and documentation - please keep dicsussing here and do as Mark says - discussion of more general points is also vital!

Thanks, Mark. Let's see how this one works out.

Whoami?

-Spike A.

(how's that for discussion)

I think the idea of getting the community more involved in the system of creating the category system is a good one, but I'm not sure it should be extended beyond what it already covers. Effort should just be spent on making the existing category system as good as it can be, and allowing it to be modified/updated by users.

Looking at your list:

* Edited Entries
- Yes, obviously...

* Entries
- I don't see the point in categorising non-edited entries beyond the scope of 'World of h2g2' pages (fictional, community pages). Non-edited entries generally aren't as high quality as Edited ones, can't be updated by other people, and are in control of the author, who could delete/modify the page at any time. It would be better just to turn the good non-edited pages into part of the Edited guide.

* Conversations of Note
- Interesting idea, but don't conversations quickly become stale (or dead), so the idea of re-visiting old ones would be a bit odd. Besides, the fun is usually in being part of conversations rather than reading a huge archive. Also, in a way, conversations _are_ categorised, as conversations are attributed to pages, of which Edited ones and the best community pages are categorised.

* Other DNA Pages
- Not sure how cross-DNA categorisation would work. Other DNA sites might want to control their own categorisation systems eventually.

* Useful Web Links
- Edited entries already have a list of useful web links. Would there be any point in categorising web links that aren't attributed to Edited entries? This would be a lot of work, and would be very wide, and seems to me to be unneccessary duplication of dmoz. Producing a categorisation of hyperlinks is a lot less useful than a categorisation of original, useful entries that won't be found elsewhere on the net.

--

That's not to say that I think it's not worth taking up some of these ideas. I DO think it would be worthwhile to bring in some mechanisms for the community to take control of maintaining and updating edited entries and the categorisation system. Casting aside the updating of edited entries (which has been discussed elsewhere), these are all the functions I could think of that would be useful to the community:

- ENTRIES -

* Adding entry to a category
Obvious really. Usually this happens every time new Edited Entries are released.

* Deletion of entries from a category
Shouldn't happen too often, but sometimes I've spotted entries that have been listed in a category twice by mistake. Might also be useful for cleaning up dead pages in 'World of h2g2'.

* Moving of entries into different categories
e.g the Card Games listed in category C798 (Games) should be moved into the category C906 (Card Games).

* Creating a duplicate of an entry in another category
Some entries might also belong in another category.

- CATEGORIES -

* Creation of a category
e.g You might want to create a sub-category of 'Games' called 'Board Games'. This function would probably be used in conjunction with moving entries, so that you could create a sub-category and then move entries into it. You might specify some kind of rule saying there should be atleast five entries before a sub-category is considered.

(Note: I'm not including 'delete a category', because I can't think of an instance where this might be used. Perhaps it would be worth including anyway just in case though.)

* Creation of a link to a category in a seperate category.
Some categories might be linked to in other categories that aren't their direct parent. DNA 1.0 is suppossed to have the ability to do this (I think it was specified as 'italic links'). Yahoo use the '@' symbol to show where this happens.

* Moving of a category
You might want to move a category into a different one. eg. 'Computer Games' might be a sub-category of 'Games'.

--

So those are the functions I can think of. Since you probably wouldn't want people to just be able to make these changes happen straight away (someone might delete all the entries), you'd need some kind of approval system. I'd suggest that rather than allocating each category to a volunteer, which might be a bit boring, as some categories might never see any changes, that a volunteer system is set up where a pool of researchers decide on whether to accept or reject suggested changes. These suggested changes might be allocated on a random basis (like Edited Entries are to the Sub-Editors) or on a pooled basis (like new users are to the Aces).

I'd also suggest that the complicated tools to make the changes are made available to the volunteers, and that general researchers make suggestions to the volunteers by writing a short description of the suggestion in plain text via either a pop-up box, e-mail, or simply a special forum.
(Another way might be some kind of form researchers fill in, with volunteers either pressing 'accept' or 'reject' to make it happen).

Comments anyone?
(phew, I'm exhausted)

Hi Frankie
----------

I think you have made some useful additions - but my points are outlined on the project page that I'm building - under 'Advantages' in the longer version.

I find it difficult to see what allowing the Community to only deal with the EG would achieve. The idea is to bring together the community. Your posting is very long, so I've picked out what really needs answering...

"I don't see the point in categorising non-edited entries beyond the scope of 'World of h2g2' pages"

The Edited Guide is important too! Just because a piece doesn't fit into the Edited Guide doesn't mean it isn't a vital part of a topic. By listing everything, it helps people to come together and makes finding opinion pieces and other non-edited Entries easier than with the Search. Besides which, the EG is already categorised, what I suggest is so much more...

"conversations quickly become stale (or dead), so the idea of re-visiting old ones would be a bit odd. Besides, the fun is usually in being part of conversations rather than reading a huge archive. Also, in a way, conversations _are_ categorised, as conversations are attributed to pages, of which Edited ones and the best community pages are categorised."

I have to take your point on this one. I wouldn't want to categorise *all* conversations, but there are some that contain discussion on a topic that makes great reading...

"Not sure how cross-DNA categorisation would work. Other DNA sites might want to control their own categorisation systems eventually."

Yes, they might, but would it not be foolish to ignore the other parts of DNA if they had useful information to contribute?

"Edited entries already have a list of useful web links. Would there be any point in categorising web links that aren't attributed to Edited entries? This ... seems to me to be unneccessary duplication of dmoz."

The idea is that a few of the very best links that are particularly good further reading - possibly pages to be viewed researching further coverage of a topic - could be added in order to get people started. We can't just shut out the rest of the web, but I wasn't suggesting we index the whole web - people are already doing that, as you so rightly pointed out!

"Producing a categorisation of hyperlinks is a lot less useful than a categorisation of original, useful entries that won't be found elsewhere on the net."

No, it's an added bonus. And h2g2 doesn't cover everything (yet).

The rest of what you have said, what functions are needed, etc. is all stuff that I'm going to add to the proposal - it isn't quite finished yet. Since you beat me to it, I'll be crediting you (not as co-auth, since you don't support the main idea, but in line with text)

There we go - I've answered your points - lets see what others think!
*goes off to publicise...*

Whoami?

Oh, and 'approval system'. Dmoz has the right answer to this one - changes are instant, but if you mess up, you're unlikely to stay long. Besides, the biggest evils that you could commit are already against the house rules and would be moderated...

Whoami?

Hi Whoami,

Didn't mean to rain on your parade with my post, just adding ideas and comments into the melting pot...

I'll answer some of your points:

"I find it difficult to see what allowing the Community to only deal with the EG would achieve. The idea is to bring together the community. ... The Edited Guide is important too! Just because a piece doesn't fit into the Edited Guide doesn't mean it isn't a vital part of a topic. By listing everything, it helps people to come together and makes finding opinion pieces and other non-edited Entries easier than with the Search. Besides which, the EG is already categorised, what I suggest is so much more..."

My main point was that it wasn't worth categorising entries which DO fit the remit of the Edited Guide but aren't edited (such as a non-edited factual piece on Carbon60 - it's best just to get these into the Edited Guide). I take your point about entries that will never be suitible for the edited guide though, perhaps AGG/GAG could be responsible for maintaining a particular sub-section of the categorisation system (like 'world of h2g2')?

"I have to take your point on this one. I wouldn't want to categorise *all* conversations, but there are some that contain discussion on a topic that makes great reading..."

Hmm, perhaps. Can you give a practical example. Most of the useful topic-related conversations I can think of take place on forums attachted to Edited entries.

"Yes, they might, but would it not be foolish to ignore the other parts of DNA if they had useful information to contribute?"

I think this depends on what the overall strategy is for DNA sites and how they will interact. Comes down to policy.

"The idea is that a few of the very best links that are particularly good further reading - possibly pages to be viewed researching further coverage of a topic - could be added in order to get people started. We can't just shut out the rest of the web, but I wasn't suggesting we index the whole web - people are already doing that, as you so rightly pointed out!"

Again, I'm unsure of the practical use for this. Why not just use dmoz if you're researching a topic? Can you give an example of what you're thinking of.

"The rest of what you have said, what functions are needed, etc. is all stuff that I'm going to add to the proposal - it isn't quite finished yet."

Okay, no problem.

"Oh, and 'approval system'. Dmoz has the right answer to this one - changes are instant, but if you mess up, you're unlikely to stay long. Besides, the biggest evils that you could commit are already against the house rules and would be moderated..."

As far as I can see, changes on dmoz aren't instant, they need to be approved by the Editor. I agree you don't really have to worry about the volunteer's messing up.

Frankie, your input is great - a stimulating debate that makes me think, and hopefully others will too. I've decided that conversation forums are not something I can give an on-the-spot example of, so I'll back down on that one and see what other people think, too.

Non-edited:

Non-edited would include Entries that added too small a point to a topic to make Edited status, but were none the less helpful for it, or those, such as the Web-Safe Colours chart, that are important but not an Edited style piece. OK, so the AGG/GAG could perhaps deal with fiction, but opinion pieces provoke factual thought and debate, include factual evidence and are therefore not the sort of thing I'd want to marginalise. The idea of this directory is to bring the Guide together and give it a more helpful structure - not to further marginalise some things because they get excluded from listings. In THHGTTG, Adams introduces us only to an alphabetical index. This is the next logical step...

DNA:

I don't see how policy should really affect this per se. Linking out to other DNA sites, however we do it, as well as to other parts of the BBC, can only be helpful. For example, topics covered on h2g2 overlap with the 360 Book of Solutions in that the studies given on 360 are a logical extension of the recognition of the problem on h2g2 - further reading. It's only fair that other DNA/BBC links are prominently displayed - isn't it?


Approval System:

Ah. I'll explain this in the proposal later - you see, I missed your point entirely. Edits can only be made by the volunteer Editors of that category, or those who have futher privileges. Editing would be a privilege, not a right, as with the other volgroups.

I think that's it, apart from an example page - A743357. Keep those ideas and opinions, suggestions and additions coming!

Whoami?

"Non-edited would include Entries that added too small a point to a topic to make Edited status, but were none the less helpful for it, or those, such as the Web-Safe Colours chart, that are important but not an Edited style piece."

I take your point here, as I can't see any really good argument for not containing this content.

"In THHGTTG, Adams introduces us only to an alphabetical index. This is the next logical step..."

We have had a category system for community pages and the edited guide for a while (though it's not particularly visible). The strength of this proposal I think is to involve the community further in this process (this has been on the feature suggestions list for ages).

"I don't see how policy should really affect this per se. Linking out to other DNA sites, however we do it, as well as to other parts of the BBC, can only be helpful."

I agree, but is this something that will be controlled by h2g2, or will the other sites do this themselves. Will, for example, 360 set up a category system using C-numbers, which will work alongside the h2g2 one?

"Ah. I'll explain this in the proposal later - you see, I missed your point entirely. Edits can only be made by the volunteer Editors of that category, or those who have futher privileges. Editing would be a privilege, not a right, as with the other volgroups."

Yeah, I think we're thinking along the same lines here.

What do you think of the pooling v. random ideas I outlined in my first post though? Or should responsibilities be subdivided by categories (like dmoz)? My feelings tend to go towards the pooling idea at the mo, as it'd be faster, but it might mean a few people doing all the work...

Oh another thing, I'm not sure it's worth using descriptions. It'd mean changing the way the existing categorisation system works and writing descriptions for over 4004 entries!

regarding the improvement of the system, zyra's catagories system for his website (see http://www.zyra.org.uk/category.htm ) shows how it can be when it is done well.

as for indexing web links, they have to be checked anyway by the moderators to see that they are suitable, so why not have them add a paragraph about them into an index.

if you then combine this with an index of what different (internal and external) links are linked to as, then you could relatively easily add some form of smart tag type system to the preview mode, making it easier to find stuff to link to when you are writting your entry.

while this does take a little bit of programming, it wouldn't be particularly hard to write, so why not consider it?

any comments mark?

"I take your point here, as I can't see any really good argument for not containing this content." We agree on something, then!

"We have had a category system for community pages ... The strength of this proposal I think is to involve the community further in this process..."

Yep. The thing is to do the whole of h2g2 - it would be an ongoing project and will attract a different kind of user in addition to authors and editors. The involvement of the community isn't that useful unless people can submit their pages - whatever they're about - to an appropriate category. I don't see the benefit in doing this in a half-hearted sort of way.

"...Is this something that will be controlled by h2g2, or will the other sites do this themselves? Will, for example, 360 set up a category system using C-numbers, which will work alongside the h2g2 one?"

They might, but my idea is that pages useful to the h2g2 audience that would fit in well to a topic should be included in the directory in order to allow the user to get an overview of what they can find on one page.

Descriptions:

Descriptions are a useful tool for keeping track of things as category editors, and are useful to A search engine displays context text from the page, a directory should go a step further. A description is about 20-30 words and takes under a minute to write - I know, I've done loads of them . It adds so much to the usefulness of a page of links...

"What do you think of the pooling v. random ideas I outlined in my first post though? Or should responsibilities be subdivided by categories (like dmoz)? My feelings tend to go towards the pooling idea at the mo, as it'd be faster, but it might mean a few people doing all the work..."

A pooling idea wouldn't give a visible credit to the person doing the work - there's no percieved reward for the work. Also, while some may have general privileges, like a dmoz editall or greenbuster, running a category is a matter of pride.

Another reason for doing this is it'll show people just how much Guide there is - if people have to sort through information, they pick some of it up as they pass, and so learn more! Users benefit from the increased knowledge of the volunteers, and the cycle repeats itself as more is written...

Whoami ?

As an added bonus, category editors usually prefer to edit something they know a little about - making the system more accurate.

Whoami?

If the directory is revamped does that mean that plant entries will be moved from the animal section to plants and vice versa? Cause that would be kinda useful.

Whew! That was close!

I saw our name <./>agggag</.> and thought someone was volunteering us to do more than we do. We just stir the pot. We don't actually serve or cater.

Call it minestrone, soup de jour, gallimauph... gollimof ..Mulligan Stew, ..or any of the other wonderful names for a pot of leftovers and otherwise inedibles, but <./>agggag</.> is by definition not organised or organisable. It is a blog. A personal blog with highly subjective criteria depending on the weather. It is best defined by what it ain't, and it ain't Edited or organised.

If we unearth a lost treasure, mix a particularly sweet metaphor or three, distort logic and syntax in painful new ways or just give a home to the homeless and down trodden, we have done our fair share.

We also recently fell into the dangerously complex idea of 'themes' and 'categories' and have suffered much from this intrusion of 'systems'. We began to argue more about 'how and why' than 'who, what, when and where' and almost lost our way. But we will soon be back on the anarchy track of totally random nonsense once I finish #24 the "All Weird Science" issue coming May 16th.

Ask no more of <./>agggag</.>

Mind you I don't speak for the 'others'.

~jwf~

Thanks for that, jwf. I wasn't sure myself. However, agggaggers would be ideal candidates for looking after fiction subcats!

Oh, and yes, the plants and animals thing would be sorted out, given time...

Whoami?

as for indexing web links, they have to be checked anyway by the moderators to see that they are suitable, so why not have them add a paragraph about them into an index.

Directory pages would have to be moderated.
However, I don't think that the mods would necessarily have time to do this. [not to moderator: feel free to remove the 'necessarily' ] Their job isn't to involve themselves in the community, just to make sure we don't do anything naughty, like say b**gium all the time...

Whoami?