# \[ANN\] Ruby Forum

**URL:** <https://www.ruby-forum.com/t/ann-ruby-forum/47958>\
**Category:** Ruby\
**Created:** [November 14, 2005, 12:02am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958 "2005-11-14T00:02:16Z")\
**Posts on this page:** 20\
**Page:** 2

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 4:11pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/21 "2005-11-14T16:11:07Z")

</div>

James Edward G. II wrote:

> On Nov 14, 2005, at 7:23 AM, Nikolai W. wrote:

> > What about google-groups?

> The Rails mailing list is on Google G.??? Excellent! Can I have  
> a link please?

Ah.

Still, I don’t see the point of having ruby-talk in forum form?

```
    nikolai
```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 7:06pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/22 "2005-11-14T19:06:35Z")

</div>

Kero wrote:

> > In threaded mode, posts are grouped together and put in pretty  
> > threads. However, a thread that was created at the dawn of time  
> > stays at the very bottom my list, even if it was just replied to a  
> > minute ago. Since I only read about 30% of ruby-talk, that means I’d  
> > never notice it, 'cause it’s just another old unread message.  
> > Rather, if threads were sorted by _most_ recent posting, I’d see it  
> > as I scroll by the new posts, and get to decide what I want to do  
> > with it. (You can argue about the utility of this feature with  
> > yourself, but it is [AFAIK] different from what Thunderbird  
> > provides.)

> That’s what you would use scoring for, or delete a thread that is not  
> interesting, or other ways of archiving. mutt provides me with TAB to  
> go to New mails (or old&unread if there’s no New mail).

I agree.

> I read ruby-lang on usenet, because I prefer pulling for such a  
> (reasonably) high volume list. gmail is out because it represents  
> threads as lists, not as trees.

Yes, I’ve long been thinking of reinstalling slrn and going the usenet  
way for this list instead. But as more and more usenet servers are  
closing, it really hasn’t been an alternative. It’d be nice if Google’d  
provide an NNTP service.

The really perverted way of reading ruby-lang is reading it through  
GMANE’s NNTP server :-).

> IMHO, it’s in the tools, not in another archive that holds the same  
> content.

Yes, so far that’s what I’ve been thinking for every suggestion for  
improvement (to the forum) that has come up in this thread. Still,  
perhaps not everyone is comfortable in using mutt. I wouldn’t know why,  
but that may be so. I’m uncomfortable in using anything _but_ mutt, but  
that’s me.

```
    nikolai
```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 7:33pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/23 "2005-11-14T19:33:36Z")

</div>

On 11/13/05, Nathaniel S. H. Brown [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> - It needs a RegEX to remove all email addresses from the message body  
> (removed\_email\_address@domain.invalid → john.doe@…)

This is ultimately nonsensical, as the public archives have the email  
addresses in plaintext.

-austin

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 5:14pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/24 "2005-11-14T17:14:17Z")

</div>

> lie (or are confused) about the current datetime.  
> Or the machines they run on. As I recall from the days I used elm to  
> read my  
> mail and most ppl replied from PCs using a puny form of DOS or win31  
> that  
> hadn’t even heard of NTP. But an occasional high rate of writing emails.

Or clients that ignore the IDs that are put in various headers to make  
threading easier (that’s where google mail gets bonus points).

> In threaded mode, posts are grouped together and put in pretty threads.  
> However, a thread that was created at the dawn of time stays at the very  
> bottom my list, even if it was just replied to a minute ago. Since I  
> only read about 30% of ruby-talk, that means I’d never notice it, 'cause  
> it’s just another old unread message. Rather, if threads were sorted by  
> _most_ recent posting, I’d see it as I scroll by the new posts, and get  
> to decide what I want to do with it. (You can argue about the utility of  
> this feature with yourself, but it is [AFAIK] different from what  
> Thunderbird provides.)

That’s what you would use scoring for, or delete a thread that is not  
interesting, or other ways of archiving. mutt provides me with TAB to go  
to  
New mails (or old&unread if there’s no New mail).

I read ruby-lang on usenet, because I prefer pulling for such a  
(reasonably)  
high volume list. gmail is out because it represents threads as lists,  
not  
as trees.

IMHO, it’s in the tools, not in another archive that holds the same  
content.

Bye,  
Kero.

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 10:17pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/25 "2005-11-14T22:17:23Z")

</div>

On 11/14/05, Nathaniel S. H. Brown [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> Having one less place where spam filters can grab my email address, I would  
> indeed like to see added 🙂
> 
> The public archives _hopefully_ have filters for headers that prevent this  
> sort of thing, but in any case, this is something that should be done by  
> default. Google has implemented this as well.

Hm? Filters for headers? As I said, the public archives pretty much  
have the email addresses. The only distinction is that they have  
replaced ‘@’ with ’ '. It’s, quite honestly, something that probably  
isn’t worth the coding time to change at this point.

-austin

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 1:14am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/26 "2005-11-15T01:14:51Z")

</div>

Kero wrote:

> That’s what you would use scoring for, or delete a thread that is not  
> interesting, or other ways of archiving. mutt provides me with TAB to go to  
> New mails (or old&unread if there’s no New mail).

No, it’s what _you_ would use those things for. None of them (I Googled  
to find out what scoring is) suit my needs/taste adequately.  
Nonetheless, why are we arguing about an off-hand Thunderbird feature  
request I made on this list?

> IMHO, it’s in the tools, not in another archive that holds the same content.

What is mutt, but another (local) archive that holds the same content?  
What makes the web forum view not just another tool?

Devin

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 14, 2005, 8:17pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/27 "2005-11-14T20:17:16Z")

</div>

Having one less place where spam filters can grab my email address, I  
would  
indeed like to see added 🙂

The public archives _hopefully_ have filters for headers that prevent  
this  
sort of thing, but in any case, this is something that should be done by  
default. Google has implemented this as well.

Warmest regards,  
Nathan.

* * *

Nathaniel S. H. Brown Toll Free 1.877.4.INIMIT  
Inimit Innovations Phone 604.724.6624  
[www.inimit.com](http://www.inimit.com) Fax 604.444.9942

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 2:20am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/28 "2005-11-15T02:20:59Z")

</div>

Devin M. wrote:

> Kero wrote:

> > That’s what you would use scoring for, or delete a thread that is  
> > not interesting, or other ways of archiving. mutt provides me with  
> > TAB to go to New mails (or old&unread if there’s no New mail).

> No, it’s what _you_ would use those things for. None of them (I  
> Googled to find out what scoring is) suit my needs/taste adequately.  
> Nonetheless, why are we arguing about an off-hand Thunderbird feature  
> request I made on this list?

Scoring is precisely what you described yourself. You can give  
messages/threads positive and negative scores depending on various  
attributes of said entity.

```
    nikolai
```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 1:28am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/29 "2005-11-15T01:28:57Z")

</div>

Andreas;

This looks splendid. Well done.

> If you notice problems or have any suggestions, please send me a mail.

My only suggestion is kinda knit-pickedy, but it’s hard to tell which  
threads have already been read in the thread list. Maybe a bit more  
contrast for the followed links?

Again, great job.

–  
Lou

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 12:30pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/30 "2005-11-15T12:30:01Z")

</div>

ljscoras wrote:

> Andreas;
> 
> This looks splendid. Well done.

Thanks.

> > If you notice problems or have any suggestions, please send me a mail.
> 
> My only suggestion is kinda knit-pickedy, but it’s hard to tell which  
> threads have already been read in the thread list. Maybe a bit more  
> contrast for the followed links?

If you are logged in threads with posts that you have not yet read are  
displayed in bold face.

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 5:27am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/31 "2005-11-15T05:27:24Z")

</div>

You guys all need imap + a mail client that threads well. I’m on  
dovecot+Mail.app(+postfix+procmail), and it’s just fine. Use  
Thunderbird sometimes too. Whatever…

For those of you that can’t get such a setup, use wtf you want,  
webforum, nntp, imap, pop3, straight up mail, whatever who cares.

–Steve

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 2:33am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/32 "2005-11-15T02:33:01Z")

</div>

Nikolai W. wrote:

> Scoring is precisely what you described yourself. You can give  
> messages/threads positive and negative scores depending on various  
> attributes of said entity.

Err? Honest confusion. How is that the same as (or isomorphic to)  
sorting threads by date of most recent post?

Devin  
oh yeah… [OT]

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 10:38pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/33 "2005-11-15T22:38:15Z")

</div>

Devin M. wrote:

> Nikolai W. wrote:

> > Scoring is precisely what you described yourself. You can give  
> > messages/threads positive and negative scores depending on various  
> > attributes of said entity.

> Err? Honest confusion. How is that the same as (or isomorphic to)  
> sorting threads by date of most recent post?

Ah, I misread. Scoring allows you to give messages a positive or  
negative score depending on attributes of a message. The attributes  
that may be scored upon are numerous, so you best consult mutt’s manual  
(which has scoring), or if you use NNTP, slrn’s manual. There’s also a  
separate document about it: [http://slrn.sourceforge.net/docs/score.txt](http://slrn.sourceforge.net/docs/score.txt).

Thus, instead of sorting your threads in “having the most recent  
messages first” mode, you could see to it that uninteresting threads are  
removed/ignored, that certain threads are given a higher score than  
others and sort based on score (set sort=score in mutt). (You could  
also do ‘set sort=threads’ and ‘set sort\_aux=score’. I don’t know for  
sure, but hopefully it’ll sort on date if two threads have the same  
score.)

```
    nikolai

```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 10:52pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/34 "2005-11-15T22:52:29Z")

</div>

This is really cool, I’m not a big fan of high-traffic mailinglists (I  
tend to not read them for a couple days then find I’m swamped). This  
should fit my reading habits better (I can swing by and see what’s  
interesting without pulling 500 emails).

I have a couple ideas; would it be possible to have a generic “reply”  
option in addition to the “reply with quote”(so I don’t have to delete  
text)? Would it be possible to try to join broken threads with the  
original parent using some heuristic? What about expanding the  
line-length so it actually uses more then 1/4 of my screen width? How  
about displaying quotes better? Previewing might be handy too…

I actually tried something like this myself a couple years back, except  
I did it in PHP so I was doomed to frustration and failure. I’m annoyed  
that I don’t have enough free time so I could offer to help with this,  
but maybe that will change.

Andrew

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 11:11pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/35 "2005-11-15T23:11:23Z")

</div>

Andrew T. wrote:

> This is really cool, I’m not a big fan of high-traffic mailinglists (I  
> tend to not read them for a couple days then find I’m swamped). This  
> should fit my reading habits better (I can swing by and see what’s  
> interesting without pulling 500 emails).

How is it easier to read 500 emails in forum form?

> I have a couple ideas; would it be possible to have a generic “reply”  
> option in addition to the “reply with quote”(so I don’t have to delete  
> text)?

Or better yet, reply to the pertinent message and remove any text that  
isn’t relevant to your reply, instead of replying to some random message  
in the thread.

> Would it be possible to try to join broken threads with the original  
> parent using some heuristic?

In a perfect world, all MUAs would produce In-Reply-To and References  
headers.

> What about expanding the line-length so it actually uses more then 1/4  
> of my screen width?

The line-length is probably set at 72 or some such, as most people send  
emails with 72-characters-per-line-long lines

```
    nikolai
```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 11:54pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/36 "2005-11-15T23:54:05Z")

</div>

> How is it easier to read 500 emails in forum form?

Well, I can easily see what threads have had updates since I’ve read  
them, and I can easily ignore entire threads. Also, I spend a lot more  
time in a browser then a mail client

> Or better yet, reply to the pertinent message and remove any text that  
> isn’t relevant to your reply, instead of replying to some random message  
> in the thread.

And when you don’t have a particular reply to reply to? (I usually do  
the above when I’m replying to something specific, like right now).

> In a perfect world, all MUAs would produce In-Reply-To and References  
> headers.

Yeah, but we self evidently do not, which is why I proposed something to  
augment those.

> The line-length is probably set at 72 or some such, as most people send  
> emails with 72-characters-per-line-long lines

I’m in a browser, I don’t need my lines wrapped at 72 characters, I’d  
prefer to have them wrapped depending on the size of my browser window.

Andrew

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 15, 2005, 11:20pm UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/37 "2005-11-15T23:20:26Z")

</div>

Nikolai W. wrote:

> Devin M. wrote:

> > Nikolai W. wrote:

> > > Scoring is precisely what you described yourself. You can give  
> > > messages/threads positive and negative scores depending on various  
> > > attributes of said entity.

> > Err? Honest confusion. How is that the same as (or isomorphic to)  
> > sorting threads by date of most recent post?

> also do ‘set sort=threads’ and ‘set sort\_aux=score’. I don’t know for  
> sure, but hopefully it’ll sort on date if two threads have the same  
> score.)

Actually, the best way to do this exactly what you wanted, i.e., sorting  
threads by most recent mail is (in mutt):

set sort=threads  
set sort\_aux=last-date-received

```
    nikolai
```

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 16, 2005, 12:34am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/38 "2005-11-16T00:34:38Z")

</div>

> Likewise. Thank you mutt(1).

Yeah, I saw this one coming 🙂 I’m just stating my personal preference,  
I prefer a forum to a mailinglist any day.

> And why was my message the one you replied to? (I’m not looking for an  
> argument, honest.)

Last one in the thread at the time, what else should I have done?

> Both you and I have MUAs that add those headers (well ruby-forum does  
> it, even though it’s strictly not an MUA).

Indeed, but there’s _always_ broken threads on just about any  
mailinglist I’ve ever been on and it annoys me, especially when its an  
interesting topic and I’d like to see the whole thing in one piece, not  
all scattered across the archives

> Yes, but how would you determine what lines to join?

Well, one possibility might be to run all lines together that did not  
have an empty line between them and let the browser wrap them. I haven’t  
really thought about _how_ this would be done though. It might be more  
problematic then I thought.

> Furthermore, the human brain, together with the human eye, prefers to  
> process lines that contain about 72 (note the about) characters.

If you say so, I find the wrapping on most emails kind of annoying, it  
also makes things a lot longer vertically then they would be if they  
wrapped in a more efficent fashion.

Andrew

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 16, 2005, 1:05am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/39 "2005-11-16T01:05:50Z")

</div>

> > IMHO, it’s in the tools, not in another archive that holds the same content.
> 
> What is mutt, but another (local) archive that holds the same content?  
> What makes the web forum view not just another tool?

mutt is a local tool for a local copy. It does not need to scale,  
because it  
is an endpoint. my endpoint.

a webbased forum is a public proxy for an already complicated  
newsgroup-mailinglist combo. The fact that it is possible to make this  
does  
not make it a good idea by itself.

The specific argument I responded to is like “it provides me the view on  
the  
list I never had” and that’s IMHO an _extremely_ thin argument.

The reason behind all this is that I fear needless fragmentation. The  
Ruby  
community grows, it will have to fragment to some extent, we can’t  
populate  
a mailing list with ten times more people. Rails split off the main  
lists,  
main IRC, even gets its own conference. Fine, it seems a clear cut  
distinction (but given all other useful web frameworks in Ruby, I doubt  
it  
deserves to be this clear cut).

We have to be critical of new development.

Kero.

---

<div class="post-metadata">

**Author:** ![andreas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/ee7513/32.png) [@andreas](https://www.ruby-forum.com/u/andreas)\
**Post date:** [November 16, 2005, 1:44am UTC](https://www.ruby-forum.com/t/ann-ruby-forum/47958/40 "2005-11-16T01:44:57Z")

</div>

On 11/15/05, Kero [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> The reason behind all this is that I fear needless fragmentation. The Ruby  
> community grows, it will have to fragment to some extent, we can’t populate  
> a mailing list with ten times more people. Rails split off the main lists,  
> main IRC, even gets its own conference. Fine, it seems a clear cut  
> distinction (but given all other useful web frameworks in Ruby, I doubt it  
> deserves to be this clear cut).
> 
> We have to be critical of new development.

It will only fragment if the shards are appealing enough to attract  
people.

I personally use a combination of mailing list and reading  
comp.lang.ruby online via google groups, and I don’t feel like it  
takes me away from any of that.

Instead, I think it’s encouraging us to get new traffic from people  
who have not yet found the list or usenet group. Your personal choice  
has not been effected in anyway.

If the new forum results in an influx of spam, poorly formatted  
emails, etc, then yeah, we can complain then. But right now, I see  
it as a unifier and a new point of entry for people into the ruby  
community, rather than a fragmentation.

I only recently signed up on the list, but I’ve been monitoring  
comp.lang.ruby for about a year. I think the trend goes towards the  
center, not away from it. I doubt this forum is going to steal people  
away from ruby-talk :-/

As far as the Rails conference goes… I think that’s great too. It  
lightens the pressure on the RubyConf to be rails oriented, and if you  
enjoy both, you have two conferences to look forward two now! 🙂 (i’m  
planning on attending both Rails and RubyConf 2006)

[Previous page](https://www.ruby-forum.com/t/ann-ruby-forum/47958.md?page=1)

[Next page](https://www.ruby-forum.com/t/ann-ruby-forum/47958.md?page=3)
