Currently there are a variety of ways to communicate. Each one with its own unique flaws. (The list is at WhatCommunicationSoftwareToUse).

Perhaps it could be possible to design a system that blends mailing lists with a few other things, in order to create a system better in all ways than mailing lists alone.

So basically you have a mailing list, plus some or all of these features: … (fill in details here)

Many of these features have been implemented in one place or another. How do I combine them into one system?

Perhaps it would be better to connect various systems somehow – a SideSystemsFirst approach – rather than building one huge monolithic do-everything application?

Other pages that talk about creating a better system by blending 2 or more different systems:

Starting Points for New Mailing List Software

hard core:

See Also


(inspired by LimitationsOfMailingLists)

I’m imagining something for a software development list.

We should be able to ask the list a question, and just get responses to that question.

We should be able to consolidate questions. If someone asks something that’s already come up, we want to just say, “That was already answered over here.”

We should be able to see a list of questions that haven’t been replied to yet. Most recent to oldest, but also by what people thought was of special value. Maybe a lot of people want to hear the answer to some question, or a dev thinks that a certain question is of particular value.

You should be able to subscribe to read the responses of people who’ve asked questions, and for which answers are still coming in.

You should be able to delete old questions.

You should be able to think outside of the “question” plane. You should be able to just hold a casual conversation. “So I’ve been thinking about…” This isn’t part of the formal question sphere. It doesn’t have a “ticket” attached to it, or anything like that.

However, it may have a “room” assigned to it, like chat rooms in IRC. People may say, “get another room,” like we do in IRC. And we should be able to trust that “another room” can be just spontaneously created. Not like in mailing lists, where it takes like forever to make a new mailing list. Rather, just by changing some meta-data in the email being sent, you should be able to make a new “room,” just like in IRC.

You should, if you really really want, be able to see all traffic.

You should be able to take frequently asked questions (quite literally!), and start construction of a human edited FAQ based on it. The construction of the FAQ should be assisted by technology.

You should be able to associate questions and their answer-threads with other questions and their answer-threads. “Interested in this question? Probably interested in this other stuff as well…” This would be wiki-like, and wiki-nodes-like. Anyone could update the settings.

Of course, being wiki-like, you need a SharedAwarenessSystem to observe that transformations are good.

When people get their questions answered, or while they are in the process of being answered (perhaps there is a thread of it,) they should be able to receive notice by Jabber, or whatever other protocols they’d like to use.

People should be able to peruse the list of questions, and subscribe to hear what’s going on in the answering of a particular question.

You should be able to annotate old discussions, and organize them. Or, not organize them. You should be free to just let them remain anonymous in the gigantic heap of stuff called “the past.” But casual searches should be able to say, “I read this message, but this other thing is really much more current.” If I may say so, the ability to forget and re-retrieve the past and then forget it again is really useful. ;) Maybe we just want things to automatically slip into a “this is sooo old,…” mode, and give google some sort of “archaeology” hint.

There’s a lot of cool stuff you could do.

This is sort of a mixture of IRC, Mailing List, Wiki, and Blog technology.

IRC: Immediate notification, and “rooms.”

Wiki: “Rooms,” and being able to reorganize things, to a degree. Also, the ability to summarize, and in the tech-assisted construction of the FAQ.

Mailing List: Well, it’s a principle line of communication. People ask questions, and get answers. There are people just asking questions, people helping out, and people

Blog: You can shoot the shit, talk about whatever you want. You can say, “So I was thinking the other day,…” People interested in whatever’s in the room could stay in it. People not interested in it could just leave for another room, or tell the people having a particular conversation to take it to another room.

This is a lot of technology, that I’ve just described. The way to build it, I think, is to focus on protocols, and tie together a bunch of little systems. OneBigSoup.

If nobody’s interested, I’ll move this on over to IntComm?. We can cluster it, but I’m not sure what we’d cluster it with, and thus not sure what gains we’d really make by clustering it.

Wouldn’t it be nice if (WIBNI) when I had a question, I could simply find the relevant experts, ask them, and get a response back in less than an hour ? Then just walk away with no further effort on my part ?

I suspect that’s impossible in any medium.

(Mailing list archives,, and Google prove that I can get answers to (some previously-asked) questions in just a few seconds.)

Let’s say there were already a group of the relevant experts gathered and chatting about some of the finer details / latest news in the field.

There seems to be only 2 possibilities for any possible interaction:

  • 1. Any random stranger can walk up and blurt out whatever is on his mind, so that everyone in the group can hear it.
  • 2. There’s some sort of filter that stops (1) from happening.

Techologies that allow (1) allow

  • a student to ask about problem #7 on his homework. That’s not so bad; after the obligatory “we can’t do your homework for you – read the FAQ” remark, it might spark some discussion about related topics.
  • 500 students to all ask about their homework, drowning out the previous discussion.
  • spammers that try to sell something that isn’t even relevant to the topic under discussion.

After either of these last 2 problems occur, I expect the experts to give up on that communication technology and wander off.

So there must be some sort of filter. But we don’t want to shut everyone out. We still want to see new, interesting, challenging, educational questions every once in a while. And we want people to learn enough to able to ask those questions.

How can we avoid those time-wasters ?

Some filters:

  • typical mailing list subscription weeds out people that are too clueless / lazy to figure it out.
  • LimitGrowth
  • wiki allows anyone to say whatever is on his mind, but if the first person who reads it decides it’s irrelevant, that person deletes it and no one else in the group has to suffer through it.
  • moderated mailing lists / moderated newsgroups typically funnel all posts through a single moderator. It was a lot of work for a single moderator to delete all the spam, and approve and forward the “good” posts in a timely manner.
  • A few moderated mailing lists / moderated newsgroups, such as the Usenet Oracle, had an elaborate distributed moderation scheme. (Why doesn’t anyone else do this today?)
  • Some mailing lists / auto-moderated newsgroups / air traffic control towers / etc. delete all messages that don’t have the special 3-letter code of the day. (Sticking in that code proves you’ve at least skimmed the latest version of the FAQ or automated airport weather broadcast).
  • [Please add any filters I’ve missed.]

The root problem is that humans have a far-too-short lifespan. If we could fix that, then all these other problems fade away. :-).


That sounds like an argument against the invention of Instant Messaging.

“Gee, we’d really like to invent instant messaging…”

“Wait! Don’t do that- people already get enough messages!”

You might think that IM would make our info-glut worse. But it actually helps a lot.

It takes the load that was going into emails, and lets you move a bunch of it into IM.

IM is basically e-mail lite. You don’t expect people to archive it. You don’t expect replies, much of the time. It’s felt to be a “weak” link, for “small” messages.

Another way that this system I described above helps save time, is that if an e-mail thread isn’t interesting to you, you just unsubscribe to it. (And the unsubscribe doesn’t take 5 minutes.) You just leave that e-mail “room.”

Usually if there’s a thread going on within a mailing list that you have no interest in, you have to read enough to know that you don’t care for the thread, and read enough to know that there aren’t any other important ideas in there either.

By this system I described above, you can just completely ignore a thread you don’t care about.

Another nice thing about the system is that if a dev isn’t in the mood to answer questions, the dev can just ignore them. All of them. Until she’s in a mood to answer questions. At which point she can look at the list of questions, and see how the community outside has reordered them, in order of preference, care.

So, you see- there are plenty of ways that we can improve the technology.

The point isn’t to get perfect- that is, to invent little pocket universes that we can give to developers, for infinite time.

The point is to improve what we have.

You want a mixture of looking long-term, looking near-term. But don’t look rediculous-term: Don’t look out to forever. Unless you want to talk about meditation or something like that, in which case I’m game. Just not right now. :)

I agree that even minor improvements are worthwhile ( SmallChangesYieldGreatChanges ), even if the result is still very far from perfection.

“At which point she can look at the list of questions, and see how the community outside has reordered them, in order of preference, care.”

So a community using IM/IRC can re-order lists of questions? Really? I thought (currently) only wiki does that. I guess I need to look at IM/IRC much more closely.

Ah, I see now – this not-yet-implemented MailingListsReplacement, a mixture of IRC and wiki and a bunch of other things, can re-order lists of questions. OK, great.

It’s just that I’m annoyed by spammers discovering Usenet news, then email, then wiki. If this new mailing list replacement ever became popular, I expect spammers to eventually discover it. I was hoping this new mailing list replacement would somehow magically block all such annoyances, while still letting all the good stuff through :-).

To bootstrap this idea of a discussion board replacing mailing lists one could write a mail archive that would allow people to read the email list as easily as they do with their current email readers (if those readers are web mails this should not be too difficult) and also do all the other things that you can do with a centralized solution (like voting, doing statistics etc).

There was an interesting discussion in FoRK on mailing list replacements, started by LuisVilla.

Some of the ideas that came out were:

  • Affix URLs – Attach the URL of each post’s web archive position, TO the post. (So you can post the URL to a social bookmarking site, or otherwise publicly link to the post or thread.) (cnntp may do this already)
  • Announce & Hide – Send only the first post in a thread by default. If the user wants to see more in the thread, they can click a web link that came with the post.
  • Collective Rating – Allow for individuals to vote, “uninteresting.” Have the software automatically cloak conversations to non-participants, upon sufficient votes reached. It could be implemented by clicking links at the bottom of posts. [1]

I’d add [2]:

  • Sub-Lists – Honestly, a lot of times, we need to create new lists straight out of conversations in existing lists. It’s really hard to get people signed up and so on- if there were an easy and transparently social way of creating new lists in view of others, and subscribing people, within the context of a super-list, that’d be really awesome. It could be as simple as saying, “if you put [blah] in the subject line, it forms a new list.” How do people join the list? Perhaps just sending a post with [blah] in the subject line is sufficient to subscribe you. Not sure how unsubscribing would work. Some of these rooms could have moderators, perhaps the first person to post to the room sets up how it works.
  • OverHear – make it possible to send a directed message to another person, through the list, that others don’t see, but can see via subscription or via website archives and so on.
  • web/list crossover – This is fairly common these days – you can post from either the web, or your mailing list software. It’s nice to participate in mailing lists “casually,” (by visitation,) rather than getting everything in your in-box, some times.
  • wiki-like linking – link to other threads, or pages in a wiki (LocalNames)

I think you can get a lot out of the concept: “Let the e-mail point back to the website used to maintain the list.”

Wouldn’t it be nice if (WIBNI) when I had a question, I could simply find the relevant experts, ask them, and get a response back in less than an hour? Then just walk away with no further effort on my part?

That’s exactly how freenode (IRC network) works, at least the sane channels. Of course, some channels are filled with unfriendly people, weird rules, etc. and many channels are simply missing, lacking enough experts to make them go (or overcrowded). But practically any active, helathy opensource project has at least one channel on freenode, where the developers and “core users” lurk.

I don’t have a cute picture, but this is a new addition anyway.

Postulate: The problem with existing mailing lists is they combine usage with content. So, if we were to separate the two, we’d have some clarity to the problem. One way to start to do that is to ensure that however a comment / post got authored, whether reply, or new conversation, or comment etc, it would have the same underlying technical structure (read: schema). That way each object can be structured any way without having to revise the content itself. Permalinks are essential, not just from a URL standpoint but also from a primary key / linked list type of standpoint.

In terms of usage, there are readers, contributors, and editors. Right now there are only readers and contributors. There is not the concept of editors.

The reader thumbs through the threads in the structure in which they were created. Then there is a branch, the previous thread is forgotten, and the new branch has all the attention but none of the history. This is what the reader gets now. What the reader would like is either a) a highly edited narrative of the topic as it evolves, a la my dinner with andre, or b) a specialized narrowcast which filters by topic, person, attitude, insert attribute here.

The contributor, right now, jumps on and can contribute at the item level but cannot arrange the way that the thread unfolded. The contributor can only throw the post in the mix and wait. NO WONDER we see trolling and such, it’s hard to make any other kind of impact without succumbing to collectivism. The contributor wants editing capabilities, to say “this is the real thread I’m responding to” so she can stand just aside from the troll, drawing the conversation away from the noise, and adding clarity for the individuals sharing the same filter attention.

The editor would operate at the item level, assigning attributes as the schema evolves. The editor could also create a new thread with the items themselves. Say you had a lively debate on soup going, and a particular contributor wanted to comment on mushrooms. With careful selection / tagging and filtered viewing, the thread that the contributor responded to would be the exact one s/he intended to respond to. Minus the demi-glace detour. Another example is to branch a thread and start a new topic. If we get the item level schema correct, there would not be a need for either of these terms. To bring bouillabaise into its own “thread” really all you’d do is create a new tag for it, filter on that tag, and the resulting items the reader would see would be the “thread.” Another example is the person with the hopeless task of bringing a group to consensus. Perhaps after careful analysis of the posts, it seems the group actually agrees on 1 or 2 things. This visionary individual creates tags for the side topics so they are filtered out of the main thread, and then we’re able to notice that agreement does exist, where before there was noise. Hey, it could happen?

Right now we have the software equivalent of the irritating meta-conversationalist, who always has his own thing to say, and is constantly monitoring the conversation itself. “Are we done talking about soup yet?” “OK, now to change the subject…” There is something kind, familial, and historical about letting the threads of live conversations take their branch. I think it’s because historically people have been able to walk away, which is something our online threads are unable to do. Walking away is not explicit, it’s an innate property of people. Unfortunately in mailing lists we must explicitly create new threads, a task which is completely unnecessary and as many people here have realized, intellectually lazy.

Wow I didn’t even know I could get all worked up about that! :)

So, every important idea in the conversation would be some sort of an object, somehow grouping and/or referencing to all relevant information. Then the editors would be able to rearrange these objects, link them, add references to other objects and even merge and split them.

I think that this is how wikis work, although there is no hard division of roles. There is also no “overview” of the structure (although there are some experiments with that) and no specialized tools for the “arrangers” – they just edit text like everyone else.

You can even say “I don’t think it really belongs here” (a little stronger than just walking away) by marking a page for deletion.

I wonder if a public “kilfile” list of ignored pages for every user would actually add to the community. Smells of inflated ego somehow.

Well, yes; … …except that wiki doesn’t appear in your inbox. (…and some other properties and consequences.)

I’m envisioning something similar to ConnectionPoint– people use e-mail as the primary vehicle of communication, but that e-mail process is closely tied to the web. Each post is distributed with a URL to click, at the bottom, that take you to the web-world’s archive / presentation of that email, and it’s context.

You should be able to rework the email, tag it, connect it with other emails, and so on.

I am envisioning functional emails; emails that include commands. For example:

 Subject: BBQ on Saturday
   Hey guys!  How are you doing?
   I'm having a BBQ on Saturday;  Anyone want to come?
 > CALENDAR 2007-10-06
 > REMINDUS 2007-10-05
   It's going to be great;  Sean and Bayle are going to be there.  :)

This message is “functional,” in that the “> CALENDAR” line is interpreted as a command to the server, and it puts this email on the calendar, using the subject line as the link text, and the target being the very post.

Everybody sees the commands in their email, so that way everybody learns the commands (by watching others use them.)

So, every important idea in the conversation would be some sort of an object, somehow grouping and/or referencing to all relevant information. Then the editors would be able to rearrange these objects, link them, add references to other objects and even merge and split them.

If you could say, “Make an object, call it Foo, connect it to X, Y, Z, …” …that’d be really freakin’ neat.

I once made something like that, scraping off CW RecentChanges, and then building a navigable data structure from the activity. (Or was it building off unalog? I forget specifically, …)

The thing you care about is the UserInterface for traveling over the structure, and issuing queries over it. (Or am I just thinking of my SemanticNetworksForSearch program and getting confused?)

This is only my personal experience, and it’s very limited. I also have a bd habit of making up arbitrary rules and assumptions that fit to my experiences and then following them – so I probably fail to notice anything that doesn’t fit. Anyways, it’s probably just me but:

I find e-mail interfaces that manipulate any kind of internal state extremely awkward and unfriendly. Not only they are slow (because my mail is forwarded several times and some of the servers on the way use graylisting), they also fail to communicate the status of the system and the outcome of my commands within my attention span (roughly 10 seconds to 20 minutes, depending on what I do at the moment). I suppose this can be remedied partially by sending some representation of the system state with every e-mail (no, just messages about the results of my actions are not enough – there are also other users and the interaction is rare enough to prohibit keeping a mental model of the system in my head all the time). Of course, by necessity, such a representation is simplified. Like matching a puzzle piece through a keyhole.

Second problem is the learning curve – can be probably remedied by proper help system, hints sent in the e-mails, and of course the “how did you do that” you mentioned.

Third problem I have with e-mail interfaces is that e-mails are not guaranteed to come in the order you sent them. On the contrary, they often come in random order or in batches. This forces you to put commands that rely on each other in a single mail, in a sort of a script. In a programming language. Se where I go?

Lastly, designing command languages (and writing parsers for them) is extremely fun and neat – and you can use the languages you designed as if… as if you designed them. But try to explain them to people who are not really interested, just want to have the job done. It’s fun to play with, sure.

I will try to finish my comment with something positive, because I agree that the current technologies (and the way we use them, that’s not technology, more like culture of tool) are not good for most people, and I’d like to find a way to improve them. I really like the “mail notification” of MoinMoin wiki engine – at least I loved it until I learned to use a feed reader. The notifications contain a link to the page that was changed, and the diff – although reading diffs requires some training, it’s doable. You see what is happening and you can quickly and effectively go to that page and do whatever you want to do using the same interface you use normally – no need to learn a command language. Sometimes I think it would be nice to be able to add a comment to a page by replying to such an e-mail, and to create a new page by sending an e-mail with the page’s name and text. But I don’t really know how that would actually work – it would have to be tried. I think that MoinMoin is the closest thing to what is being discussed here.

I like this conversation, I am actually setting up a MediaWiki instance to allow for EmailToWiki? right now. I think TheSheep makes some important points, and I will actually encourage users not to use EmailToWiki? as a primary interface for wiki. Instead, I see specific role for EmailToWiki?, which is to push email threads into wiki from email discussions, email lists, etc which would then be reworekd into pages on the wiki. The EmailToWiki? just makes this more convenient

I see another role in related function, which is a “send to email” function that allows MediaWiki to send an article to email (both of these functions are also possible in MoinMoin engine, I beleive) The emailed wiki page contains “edit” links that take you directly to the article itself.

But, these are more like a “gateway” for people who are not used to wiki. It can help them mentally process how people use wikis effectively, by seeing their email discussion reworked this way. Although, it really takes as much guidance as pretty much any other way of introducing wiki.


My primary theory, driving the command language, is exactly this:

What people can observe others do in totality, and see the results, they can repeat.

This came from my observation of SecondLife: Because I could watch the person make their thing, I could replicate it. And because they could watch me make mistakes, they could correct me. (The example is imperfect, because I can’t watch the mouse and menu selections from the outside. But I can see quite a bit more into construction process, than in most environments.)

It also comes from observing the efficacy of over-the-shoulder sessions, and screen-casts.

If you are in a social situation, and someone issues a command incorrectly, you can say, “Oh, wait- I see what you are trying to do; Here is how to do it correctly.”

Here’s an example exchange:

 From: LionKimbro
 Subject: PermaCultureList!
   Hi, my name is LionKimbro, and I'm really interested in talking
   about PermaCulture.  If you'd like to discuss Permaculture with
   me, please join:
 ! CREATE PermaCultureList
   ** LIST:  I have created mailing list PermaCultureList.
             Visit: to join. **

(For those who worry: the URL does not appear in the archive version, and expires automatically after a few days. It’s customized to the e-mail target.)

 From: AlexSchroeder
 Subject: Wiki!
   Hi!  My name is Jean, and I'm really interested in talking about
   wiki!  Please join me!
 CREATE  WikiTechnologyList

 From: MattisManzel
 Subject: Re: Wiki!
   Hey Alex!  I'm interested in that, too!  Just a quick hint--
   Whenever you issue a command, you preface it with an
   exclamation mark (!).
   Here, I'll make the list for you:
 ! CREATE WikiTechnologyList
   ** LIST:  I have created mailing list WikiTechnologyList.
             Visit: to join. **
   There you go!  It takes just a little getting used to.
   See CommandLanguageHelp for more details.

Doing everything before others, and making everything a literal mimic (“I just type what you typed,”) my theory is that essentials can be learned, fairly easily.

My theory is that people learn far more by watching others, than by reading help documentation, and such, which is only really used by experts, or people who want to become experts.

Anything that you do with buttons, or special links from a webpage, or pretty much anything else– and it’s all invisible. I’ve met so many newbies to wiki, who do not understand that the action’s at RecentChanges.

(In fact, I think wiki should perhaps just make RecentChanges their FrontPage, if they’re not using it…)

I agree with Lion about the benefits of OverHear and how they make learning and teaching much easier. I’m more concerned with inefficiency and inadequacy of e-mail for the purpose. I will try to describe this in depth tomorrow, but let me do a quick sketch here:

  • E-mail is slow and does not help to keep your focus (which is sometimes good, as you can reply with cold head, but may not be necessarily good if you require to connect several things).
  • Mass-mailing messages are expensive – most users don’t sort their mail automatically, so large traffic is tiring for them. If messages are tiring, people will refrain from trivial actions and small corrections. I think the vital prerequisite for the “here I’ll help you” is that it’s easy to help and doesn’t involve making everyone around angry at you or the newbie.
  • Command languages usually require people to remember the state. If the state is not clearly visible all the time, people who just joined or passers by can’t help without first interrogating the newbie about what he’s done already.
  • Wiki pages have the “overhear” property to a large degree – once people figure out how to see and edit the page’s text, they can just look at other pages to see how things are done, and correct small mistakes. The challenge here is removing as much magic as possible: making things explicit, making links that do more than just link to pages (like category links) actually look like they do something more, possibly splitting the incredibly unibversal and powerful ‘backlink search’ into smaller, easier to understand, more specialized functions, making things that act differently actually look different, etc.
  • the uses of e-mail I can imagine on a wiki:
    • notifications about changed pages (done in MoinMoin, can be global or selective) – not only containing the diff, but also the url to the page (or to it’s editor) and the e-mail of the author
    • commenting on a page: just send a reply to a change notification or click a ‘mailto:’ link on the page, the page name is in the topic with ‘Re:’ (or somrthing else with a colon), and the text is appended to the end of the page (headers and signature stripped).
    • creating a new page (thread): just send an e-mail with the page’s title in the subject (without any Re:) and the page’s text in the body. Not sure what should happen when the page exists (replace? append?)
    • attachments are uploaded as binaries, at least if the message’s body references them
    • redirections will redirect the edits too
    • mail a page to someone (Sam mentioned there is something like this for Mediawiki) – the message would contain a link to the page and replying to it would work as replying to a change notification. I don’t know if the others were ever tried. I imagine some anti-spam and anti-abuse measures would be needed – but then again maybe not, I guess it requires an experiment. Sorry for being chaotic, I will try to clean it up tomorrow.

Nrr, …

I agree with you, Radomir.

Here’s the thing: The people I work with, will not use anything other than e-mail.

They won’t open a browser, and click on a link. That’s “entertainment” for them, or something. I don’t know, exactly.

I just know that the only technology they know how to use and will use is e-mail. Anything else, and you don’t have their attention, and they’re not talking.

E-mail is the center of their lives.

They know how to hit “reply,” they do not know how to add a comment in a blog. They see it, and they get confused.

So, I’m just working in a different project space.

The scenario is this: You go to a conference. There are, let us say, 100 people at the conference. They all agree that they want to maintain the connection, after the conference.

You can’t possibly teach these people all how to use wiki, or PHP BB forums, or something there. They’re just not there yet. You might be able to get 5-15 people who would be comfortable with one of these, or the other of these, but otherwise, you’re not talking with them.

The only technology that everybody (every last one) understands, is email. (You’re not going to leave anyone behind.)

The catch here is that (with the e-mail notifications and comments) you don’t actually need to use anything else than e-mail to participate in the discussion – you just won’t be able to make the corrections or refactoring, at least not until you realize that others are doing it and how (the links are supposed to help you in this).

I know that many people will open pages from links in their e-mails (our anti-virsu software is always busy), so if these links point exactly to the place they need, they will use it. At least that’s what I hope. If not, they will remain a “second-category wikizen” with just e-mail – but they won’t be completely cut off.

You see, I really believe in “There should be one – and preferably only one – obvious way to do it. Although that way may not be obvious at first unless you’re Dutch.” (from PEP20)

Discussion moved to StaffExpected.

Heh; ;)

We used Google Groups, and we sent invitations to ~100 people.

Of the ~100 people, about 20 were able to click on the link, in order to join the forum.

Of the remainder, we got a bunch of mail saying, “Hey! We don’t know how to do this! I want to subscribe! What do I do?”

(“Well, you just click on the link, exactly like it said...”)

In the end, what we ended up having to do, was just force-subscribe everybody. Since there are anti-spam throttling limits, it took several days to do.

So, in my command language, I’ve included a bunch of commands for manipulating others’ subscription status to lists:

  CREATE HolyGrailList
  PUSH Lancelot HolyGrailList
  PULL LancelotsBestSquire
  KICK ConfusedAndLostGuy
  • CREATE – make a new list
  • PUSH – put someone into a list
  • PULL – pull someone into the present list
  • KICK – remove someone from the present list
  • EXIT – leave the present list

I have a theory that people do not read machine text.

That is, if there’s some text that looked automatically generated, that people just mentally do not look at it, even if it’s exactly the text they need: “CLICK THIS TO SUBSCRIBE:” …

I think some sort of “Oh, but I’m not technical” circuitry shoots off, or something.

Bizarre, but I think it’s true.

I could tell you other, very similar stories, but they’re about the company I’m working at right now, so I won’t… But you’d be surprised about what people do not see.

You just made me have a wacky idea: WikiOverMail

Heh, … Well, so, …

One of the commands I want to make is… “NAMEPOST.”

KingArthur? sends:

 From:  KingArthur
   Hey guys!
 ! NAMEPOST RoundTable
   We moved the round table from CamelotCastle to StoneHenge!
   That's where we'll be having our meetings in the future.
   Take care!

When people receive this mail, they get, …

 From:  KingArthur
   Hey guys!
 ! NAMEPOST RoundTable
   We moved the round table from CamelotCastle [1] to StoneHenge! [2]
   That's where we'll be having our meetings in the future.
   Take care!
 #XX# Present: KingArthur, QueenGuinevere, SirGalahad, SirKay, SirTristan, SirLancelot, SirPercival, ...
 #XX# List:
 #XX# Post:
 #XX# 1. CamelotCastle:
 #XX# 2. StoneHenge:

Today I see yet another person on a mailing list wish for a better communication technology than email. "The need for better communication tools & a semantic web"

This one suggests that “tags” will be helpful, specifically semantic tagging. And suggests that “weighted tags” (?) will be helpful.

Define external redirect: KingArthur IntComm EmailToWiki

EditNearLinks: MoinMoin UserInterface MediaWiki