Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > alt.comp.software.thunderbird > #13418 > unrolled thread

crossposting

Back to article view | Back to alt.comp.software.thunderbird

Contents

  crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-04 14:03 -0400
    Re: crossposting Andy Burns <usenet@andyburns.uk> - 2024-07-04 19:10 +0100
      Re: crossposting Big Al <alan@invalid.com> - 2024-07-04 14:19 -0400
      Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-04 14:45 -0400
    Re: crossposting Jörg Lorenz <hugybear@gmx.net> - 2024-07-04 21:25 +0200
      Re: crossposting "J.B. Nicholson" <jbn@forestfield.org> - 2024-07-06 18:07 +0000
        Re: crossposting sticks <wolverine01@charter.net> - 2024-07-06 13:41 -0500
    Re: crossposting "David E. Ross" <nobody@nowhere.invalid> - 2024-07-04 12:34 -0700
      Re: crossposting ant@zimage.comANT (Ant) - 2024-07-04 20:19 +0000
      Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-04 17:28 -0400
        Re: crossposting Paul <nospam@needed.invalid> - 2024-07-05 00:04 -0400
      Re: crossposting "David E. Ross" <nobody@nowhere.invalid> - 2024-07-04 16:01 -0700
        Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-04 22:34 -0400
          Re: crossposting sticks <wolverine01@charter.net> - 2024-07-04 22:11 -0500
            Re: crossposting Andy Burns <usenet@andyburns.uk> - 2024-07-05 08:10 +0100
            Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-05 08:06 -0400
              Re: crossposting Big Al <alan@invalid.com> - 2024-07-05 08:36 -0400
                Re: crossposting "Carlos E. R." <robin_listas@es.invalid> - 2024-07-05 15:28 +0200
                  Re: crossposting Big Al <alan@invalid.com> - 2024-07-05 10:12 -0400
                  Re: crossposting Stan Brown <the_stan_brown@fastmail.fm> - 2024-07-07 09:15 -0700
                    Re: crossposting "Carlos E. R." <robin_listas@es.invalid> - 2024-07-07 20:13 +0200
              Re: crossposting "Gary R. Schmidt" <grschmidt@acm.org> - 2024-07-06 00:41 +1000
              Re: crossposting Paul <nospam@needed.invalid> - 2024-07-06 08:52 -0400
                Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-06 09:17 -0400
                  Re: crossposting Stan Brown <the_stan_brown@fastmail.fm> - 2024-07-07 09:17 -0700
          Re: crossposting Jörg Lorenz <hugybear@gmx.net> - 2024-07-07 07:42 +0200
            Re: crossposting "David E. Ross" <nobody@nowhere.invalid> - 2024-07-06 22:55 -0700
              Re: crossposting Newyana2 <newyana@invalid.nospam> - 2024-07-07 07:26 -0400
                Re: crossposting Jörg Lorenz <hugybear@gmx.net> - 2024-07-07 22:45 +0200
        Re: crossposting Stan Brown <the_stan_brown@fastmail.fm> - 2024-07-07 09:11 -0700
          Re: crossposting Big Al <alan@invalid.com> - 2024-07-07 12:34 -0400
            Re: crossposting "Carlos E. R." <robin_listas@es.invalid> - 2024-07-07 20:15 +0200

Page 2 of 2 — ← Prev page 1 [2]

#13468

On 2024-07-07 18:15, Stan Brown wrote:
> On Fri, 5 Jul 2024 15:28:37 +0200, Carlos E. R. wrote:
>>
>> On 2024-07-05 14:36, Big Al wrote:
>>>>
>>> If the message ID were the same in every group, seems like a table of
>>> message IDs and the read status would be a simple lookup.   Of course
>>> what do I know.
>>
>> That needs a largish database on disk with some kind of ram cache, with
>> very fast read operation, and nearly as fast write operation.
> 
> Why would a database be needed? The Newsgroups line in the headers
> shows the groups the message was posted to.  It should be simple to
> mark the message as read in the other groups where it is posted.

And how do you find it? Becase it takes seconds to scan another group.


-- 
Cheers,
        Carlos E.R.

[toc] | [prev] | [next] | [standalone]

#13442

On 05/07/2024 22:06, Newyana2 wrote:
> On 7/4/2024 11:11 PM, sticks wrote:
> 
>> It's not a minor bug.  Jörg commented on this in article 
>> <soqt4u$f9b$2@dont-email.me> and this is the relevant factor causing 
>> the difficulty:  "So you want to mark something read in a different news
>> group? Well, the folder per news groups doesn't lend itself to this."
>>
> 
>    I don't buy it. That's just making excuses. The software should
> do the job well. If they'd stop rushing new versions out the door
> for no good reason then maybe they'd have time to actually clean
> up the code. (Culling and documenting prefs wouldn't hurt, either.)
> And maybe someone in charge to actually make those things happen.
> 
>     OE had this in 1998. It didn't take 100 versions to get there.
>   And it worked fine on old computers running 300 MHz
> Celerons with 64 MB RAM. Whenever a newsgroup is loaded,
> it has to sort which messages are marked as read. The only
> additional action needed is to cross-reference a master list,
> which would presumably be generated as groups are loaded.
> 

Since rn, rrn, trn, nn, and late incarnations of readnews can do it, why 
can't Thunderbird do it?

For those who came in late, those were the programs we ran - and some 
still do - on 80x24 terminals at 9600baud, on multi-user computers with 
what seems like bugger-all memory, and sometimes seem to have been so 
much better than what is available today.

Occasionally I contemplate getting the latest patches to elm working, 
and using that, but there's apparently a bit of a problem getting it to 
work with Panda-IMAP.  :-)

	Cheers,
		Gary	B-)

[toc] | [prev] | [next] | [standalone]

#13447

On 7/5/2024 8:06 AM, Newyana2 wrote:
> On 7/4/2024 11:11 PM, sticks wrote:
> 
>> It's not a minor bug.  Jörg commented on this in article <soqt4u$f9b$2@dont-email.me> and this is the relevant factor causing the difficulty:  "So you want to mark something read in a different news
>> group? Well, the folder per news groups doesn't lend itself to this."
>>
> 
>   I don't buy it. That's just making excuses. The software should
> do the job well. If they'd stop rushing new versions out the door
> for no good reason then maybe they'd have time to actually clean
> up the code. (Culling and documenting prefs wouldn't hurt, either.)
> And maybe someone in charge to actually make those things happen.
> 
>    OE had this in 1998. It didn't take 100 versions to get there.
>  And it worked fine on old computers running 300 MHz
> Celerons with 64 MB RAM. Whenever a newsgroup is loaded,
> it has to sort which messages are marked as read. The only
> additional action needed is to cross-reference a master list,
> which would presumably be generated as groups are loaded.

The Mail and USENET News sides of the program, share some
properties in terms of the basic protocols with the two
server types they contact.

But other aspects of the two functions, are not the same.

Some individuals have 50GB of email collected over 40 years.
Guess what. Their client is slow. Talking to them does not help.
I used to deal with people like this at work :-) "Paul,
come quick, my email is slow" :-) Then you'd look, vomit
in an adjacent bucket and try to help them without
insulting their sense of taste. It does not matter what email
tool you use, those 50GB people are "out there".

USENET News users, based on "various calamities" that befall
the metadata storage, the dimension tends to not be as big.
Because of Compaction, the metadata need be no bigger than
the event horizon of the server (two years retention equals
not a lot of articles in the Mork file). A user on Eternal-Seotember
does not necessarily have large Mork files. Compaction, whether
intentional or not, sees to it. A person on a binary server,
will have larger files. Much larger.

Consequently, I might agree, that the two sides of the
program, could use different mechanisms for the metadata.

The developers do not see it this way. They seek to have
one subsystem for both environments. They want to reuse
components, without writing (and more importantly, supporting)
two metadata trees or tree types.

A textual format, would be too slow for someone with 50GB
of emails. Mork files (a weak as piss form of database),
that is not a particularly scalable mechanism. And the
comment I saw before in Bugzilla about this, is they
*don't* ever want all the Mork files opening at the
same time. Does that happen when you scan for new
articles on a server ? That, I do not know. Thunderbird
is limited by the number of connections it can open
to the server, and more Mork files open than the connection
limit does not necessarily have to be the case. They could
choose to close and put away the Mork files quickly, during
a server scan for new mail.

As far as I know, the server takes care of the cross post.
If you cross to A,B,C then the server puts an entry
for the same <MID> into A, B, and C queue. This way,
when a client fetches a group, only the metadata for
that group need be consulted.

But you also have the opportunity to document each MID
when it comes in, and store the MID in a master list for
that server. Later, if someone reads the article, by
consulting the master list, you could have materials
to deal with the side effects.

One problem is, you don't know how I use Thunderbird.
I *never* Compact. *Never*.  This means I carry more
metadata than you do, and my storage requirements
for that metadata are different than yours. There is
about eight years of history in here, which is
not on the server. Clicking any of these links, displays
the <MID> but says the article is not available. The
<MID> is valuable information.

    [Picture]

     https://i.postimg.cc/8Cz8HhDd/w7-history-thunderbird-nocompact.gif

Everyone is different. There could be a Thunderbird user with
512MB of free RAM available who expects performant Thunderbird service.
We can't go around assuming everyone has 16GB like we do.
This means the developers have to suit a wide audience.
And do the best they can. Not just ship the same "fast prototype"
for the next twenty years.

Nobody in the Thunderbird project wants to open the database
part of the program. We saw this, when the guy who added Filters
to Thunderbird, did his work. He had to learn how Mork worked,
basically by himself. There is an internal program of some sort,
for viewing Mork files, at Mozilla. But it was not released to the
public. There are also two dumpers written outside Mozilla. One
of them works. The last time I wanted a copy, the Python version
thing prevented me from succeeding.

That guy did not stick around. He did his Filters prototype,
listened to the "style feedback" about his source code, and
when all that was tidy, he seemed to disappear. He had no plan
to be a object to be abused over the years, by maintaining or
extending that Filters prototype.

This should tell you that the developers know their limits.
They're good at some of the subsystems. They really did not
want to touch the USENET code AT ALL, but the opportunity
to hire someone new and make them rewrite part of that in
HTML/js, is how that code got opened up.

Does anyone want to touch Mork. Fuck, no :-)

Project objectives:

1) Think up a good technical solution.
2) Do it, without pissing off the staff.

There is no intersection in the Venn Diagram for (1) and (2).
Only by parachuting someone in, can you make progress on some topics.

https://en.wikipedia.org/wiki/Mozilla_Thunderbird

   "Thunderbird also uses Mork and (since version 3) MozStorage
    (which is based on SQLite) for its internal database. Mork
    was due to be replaced with MozStorage in Thunderbird 3.0,[20]
    but the 8.0 release still uses the Mork file format.
   "

https://bugzilla.mozilla.org/show_bug.cgi?id=11050

   "Summary: [HELP WANTED] Implement MDB in terms of an open source database which is more memory-efficient than Mork
    ...
   "

   "davidmc   Comment 2 -  25 years ago

    I thought this was a request for outside help; why is it a bug assigned
    to me?  I think the milestone should be a much bigger number than m15,
    since I don't plan to ever fix this bug in the forseeable future.
   "

:-)

   Paul

[toc] | [prev] | [next] | [standalone]

#13448

On 7/6/2024 8:52 AM, Paul wrote:

> 
>     "Summary: [HELP WANTED] Implement MDB in terms of an open source database which is more memory-efficient than Mork
>      ...
>     "
> 
>     "davidmc   Comment 2 -  25 years ago
> 
>      I thought this was a request for outside help; why is it a bug assigned
>      to me?  I think the milestone should be a much bigger number than m15,
>      since I don't plan to ever fix this bug in the forseeable future.
>     "

   Interesting insider tidbits. I remember seeing a discussion
somewhere recently that seemed to imply that Mozilla are
coming out with their insanely frequent updates because
they believe that once a month is what people want.

   Maybe someday someone will actually open up all that code
and find an unpublished anti-Vietnam war song.

   So I guess we count our blessings. TBird is free and there
never was much other option. Young people now don't even
know what usenet is. Maybe betterbird really is better?
When I looked at their feature list there was nothing I was
interested in. But to be honest, Betterbird, for me, falls
into the same category as MyPal. I get stopped by the
cutesiness of the name. It would have to be amazingly
good to get past that.

[toc] | [prev] | [next] | [standalone]

#13466

On Sat, 6 Jul 2024 09:17:33 -0400, Newyana2 wrote:
> On 7/6/2024 8:52 AM, Paul wrote:
> 
> > [quoted text muted]
> >      I thought this was a request for outside help; why is it a bug assigned
> >      to me?  I think the milestone should be a much bigger number than m15,
> >      since I don't plan to ever fix this bug in the forseeable future.
> >     "
> 
>    Interesting insider tidbits. I remember seeing a discussion
> somewhere recently that seemed to imply that Mozilla are
> coming out with their insanely frequent updates because
> they believe that once a month is what people want.

That falls to the ground when we reflect that iupdates come closer to 
once a week than once a month.

-- 
Stan Brown, Tehachapi, California, USA         https://BrownMath.com/
Shikata ga nai...

[toc] | [prev] | [next] | [standalone]

#13457

Am 05.07.24 um 04:34 schrieb Newyana2:
> On 7/4/2024 7:01 PM, David E. Ross wrote:
> 
>> I found four Bugzilla reports about this, all closed as duplicates of
>> another report.  That report was #43278 at
>> <https://bugzilla.mozilla.org/show_bug.cgi?id=43278>, which is now 24
>> years old.  The last comment to that report was added two years ago, but
>> it did not indicate any progress in fixing.  A suggested fix was posted
>> six years ago but did not result in any action.
>>
> 
>     Crazy, isn't it? They've probably released 100 versions since then.
> Maybe 500 subversions. Yet no one could get their act together
> to fix one minor bug.

If it is a bug at all it is absolutely minor. As I said: IMHO it is not 
a bug.

The way I use the Usenet it "fixing" could have very negative effects. 
And btw X-posting is impolite to the highest degree.

-- 
"Gutta cavat lapidem." (Ovid)

[toc] | [prev] | [next] | [standalone]

#13458

On 7/6/2024 10:42 PM, Jörg Lorenz wrote:
> Am 05.07.24 um 04:34 schrieb Newyana2:
>> On 7/4/2024 7:01 PM, David E. Ross wrote:
>>
>>> I found four Bugzilla reports about this, all closed as duplicates of
>>> another report.  That report was #43278 at
>>> <https://bugzilla.mozilla.org/show_bug.cgi?id=43278>, which is now 24
>>> years old.  The last comment to that report was added two years ago, but
>>> it did not indicate any progress in fixing.  A suggested fix was posted
>>> six years ago but did not result in any action.
>>>
>>
>>     Crazy, isn't it? They've probably released 100 versions since then.
>> Maybe 500 subversions. Yet no one could get their act together
>> to fix one minor bug.
> 
> If it is a bug at all it is absolutely minor. As I said: IMHO it is not 
> a bug.
> 
> The way I use the Usenet it "fixing" could have very negative effects. 
> And btw X-posting is impolite to the highest degree.
> 

Before Thunderbird was spun off, the Mozillia organization acknowledged
that this is a regression bug since it worked in Netscape Communicator.

The Usenet Board cross-posted notices of changes to any "Big 8"
newsgroup or requests to create a new newsgroup.  The cross-posts went
to news.groups.proposals, to news.admin.announce, to the affected
newsgroup if it already existed, and to any newsgroups whose topics were
similar to the subject newsgroup.  This was to ensure that any conflicts
could be reported before action was taken.

-- 
David E. Ross
<http://www.rossde.com/>

Demonstrators worldwide are demanding that Israel stop
fighting in Gaza.  Why does it seem that no one is demanding
that Hamas stop fighting?  And where are the demonstrations
against Russia fighting in the Ukraine.

[toc] | [prev] | [next] | [standalone]

#13463

On 7/7/2024 1:55 AM, David E. Ross wrote:

>> If it is a bug at all it is absolutely minor. As I said: IMHO it is not
>> a bug.
>>
>> The way I use the Usenet it "fixing" could have very negative effects.
>> And btw X-posting is impolite to the highest degree.
>>
> 
> Before Thunderbird was spun off, the Mozillia organization acknowledged
> that this is a regression bug since it worked in Netscape Communicator.
> 
> The Usenet Board cross-posted notices of changes to any "Big 8"
> newsgroup or requests to create a new newsgroup.  The cross-posts went
> to news.groups.proposals, to news.admin.announce, to the affected
> newsgroup if it already existed, and to any newsgroups whose topics were
> similar to the subject newsgroup.  This was to ensure that any conflicts
> could be reported before action was taken.
> 

    I've yet to see a post from this Jorg that isn't just bitter
derision. There's no sense trying to discuss anything with him.
(I blocked him the first time he hurled nonsensical insults
just to be mean.)

[toc] | [prev] | [next] | [standalone]

#13474

Am 07.07.24 um 13:26 schrieb Newyana2:
>     I've yet to see a post from this Jorg that isn't just bitter
> derision. There's no sense trying to discuss anything with him.
> (I blocked him the first time he hurled nonsensical insults
> just to be mean.)

THX! :-D

-- 
"Manus manum lavat."

[toc] | [prev] | [next] | [standalone]

#13464

On Thu, 4 Jul 2024 16:01:08 -0700, David E. Ross wrote:
> On 7/4/2024 12:34 PM, David E. Ross wrote:
> > [quoted text muted]
> > message in a crossposted thread as read marked all instances as read.
> > That never worked in Thunderbird.  There is a bug report for Thunderbird
> > about this.  That report might be more than a decade old.
> > 
> 
> I found four Bugzilla reports about this, all closed as duplicates of
> another report.  That report was #43278 at
> <https://bugzilla.mozilla.org/show_bug.cgi?id=43278>, which is now 24
> years old.  The last comment to that report was added two years ago, but
> it did not indicate any progress in fixing.  A suggested fix was posted
> six years ago but did not result in any action.
 

It might be worth checking Betterbird to see if this is one of the 
Thunderbird bugs that are fixed in Betterbird.

-- 
Stan Brown, Tehachapi, California, USA         https://BrownMath.com/
Shikata ga nai...

[toc] | [prev] | [next] | [standalone]

#13467

On 7/7/24 12:11 PM, Stan Brown wrote:
> On Thu, 4 Jul 2024 16:01:08 -0700, David E. Ross wrote:
>> On 7/4/2024 12:34 PM, David E. Ross wrote:
>>> [quoted text muted]
>>> message in a crossposted thread as read marked all instances as read.
>>> That never worked in Thunderbird.  There is a bug report for Thunderbird
>>> about this.  That report might be more than a decade old.
>>>
>>
>> I found four Bugzilla reports about this, all closed as duplicates of
>> another report.  That report was #43278 at
>> <https://bugzilla.mozilla.org/show_bug.cgi?id=43278>, which is now 24
>> years old.  The last comment to that report was added two years ago, but
>> it did not indicate any progress in fixing.  A suggested fix was posted
>> six years ago but did not result in any action.
>   
> 
> It might be worth checking Betterbird to see if this is one of the
> Thunderbird bugs that are fixed in Betterbird.
> 
If Betterbird has so many fixes, why can't the TB crew just apply them to TB?  If code is open it 
should be a no-brainer.
-- 
Linux Mint 21.3, Cinnamon 6.0.4,  Kernel 5.15.0-113-generic
Al

[toc] | [prev] | [next] | [standalone]

#13469

On 2024-07-07 18:34, Big Al wrote:
> On 7/7/24 12:11 PM, Stan Brown wrote:
>> On Thu, 4 Jul 2024 16:01:08 -0700, David E. Ross wrote:
>>> On 7/4/2024 12:34 PM, David E. Ross wrote:
>>>> [quoted text muted]
>>>> message in a crossposted thread as read marked all instances as read.
>>>> That never worked in Thunderbird.  There is a bug report for 
>>>> Thunderbird
>>>> about this.  That report might be more than a decade old.
>>>>
>>>
>>> I found four Bugzilla reports about this, all closed as duplicates of
>>> another report.  That report was #43278 at
>>> <https://bugzilla.mozilla.org/show_bug.cgi?id=43278>, which is now 24
>>> years old.  The last comment to that report was added two years ago, but
>>> it did not indicate any progress in fixing.  A suggested fix was posted
>>> six years ago but did not result in any action.
>>
>> It might be worth checking Betterbird to see if this is one of the
>> Thunderbird bugs that are fixed in Betterbird.
>>
> If Betterbird has so many fixes, why can't the TB crew just apply them 
> to TB?  If code is open it should be a no-brainer.

Some BB user has said already that BB has the same problem.

-- 
Cheers,
        Carlos E.R.

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | alt.comp.software.thunderbird

csiph-web