> The documents are part of the group's history, and are often the most
> important products of the group and the things a new person to the list
> needs to know about first. The trouble is, if the group has been
> talkative, it takes quite a while to weed through all the messages using
> the existing indexes to find the ones with attachments.
I agree. I've often felt that pain when using a list to which source patches are posted every once in a while, intermixed with a lively discussion. It is annoyingly hard to find the patches...
> So I think it might be useful to have a page that has links only to
> messages with attachments. It's important to link the messages, and not
> just the attachments, because the metadata for the document often exists
> nowhere but in the message that accompanies the document.
Good thinking. I had envisioned a list with attachments only, but you're right. Without the meta data in the messages, many attachements will become useless or at least senseless.
> I'd thought about using the existing indexes and marking the subject
> lines of messages that have attachments, but in the end I decided an
> attachments index would be easier to use.
Yes, I'd prefer a separate index.
> I'm at work on a change to print.c (and a small change to hypermail.c)
> that will add an index page for messages that have attachments.
Sounds reasonable.
> I'm also going to do it the right way and put any strings in lang.h.
> But I'll tell you what -- that's nasty.
The "i18n" approach of hypermail is messy and generally not good. Someone should take a grab at it and make it use a proper system. Like GNU gettext.
> I started with the hypermail-2b30.tar.gz distribution.
Sounds like a start.
> First off, am I using the right codebase, or should I start with
> something else? Second, do I have permission to check stuff in once I
> confirm it works, or should I send it to someone?
No, you won't have the permission to check in code. Let's do like this: you send me the patches and I apply and commit them to CVS. If the general quality is fine (no, I don't know how to measure that! ;-) I'm all for giving away write access to you for your next improvement!
I hope that's fine with you. It is nothing personal, I just think it is good to make sure all developers that commit things have a sense for programming and not only talking, and the best way is to let everyone prove that before given access. This is more or less how it has been dealt with before and it has turned out working pretty good.
> Also, if you could let me know what you have in mind for release
> milestones, I'll do my best to get it in, or else put it off until after
> the release if one's imminent.
Now, I am not an official spokes-person for hypermail and this is my personal view.
The b30 was supposed to be the last beta release before the true hypermail 2.0 was about to get released. The lack of enthusiasm and project management, however, has made this not happen. There is no one that takes a hold of this project and performs actions like that. If we were to release hypermail 2.0 non-beta, I'd say that would be the b30 renamed to 2.0.
To not let that be an obstacle in development and adding new ideas, I say we carry on with the CVS version towards a 2.1 release some time in the future.
Unfortunately, there are no "milestones" or planned features for any special releases. There's simply the TODO file that mentions a set of various ideas that may or may not get included in a future relelase.
I think we should prioritize those and try to sort them into milestones and try to work against those goals.
This said, I am interested to participate in this, not lead it.
> But first things first. If I'm the only one who thinks an index of
> messages with attachments might be useful, I'll just keep the changes
> here and save everybody the trouble.
I've missed such an index since the day we could extract the attachments! ;-)
--
Daniel Stenberg - http://daniel.haxx.se - +46-705-44 31 77
ech`echo xiun|tr nu oc|sed 'sx\([sx]\)\([xoi]\)xo un\2\1 is xg'`ol
Received on Wed 13 Dec 2000 10:21:44 AM GMT
This archive was generated by hypermail 2.3.0 : Sat 13 Mar 2010 03:46:12 AM GMT GMT