Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:

> On Tue, 04 May 2021 17:54:46 -0300, Boris Dorestand wrote:
>
>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>
>>> On Fri, 30 Apr 2021 16:29:59 -0700, Nicolas Vazquez wrote:
>>>
>>>> Yall still here
>>>
>>> You /do/ realize that you replied to a thread last posted to over 24
>>> years ago, right?
>>
>> How did you know? Did you look the original message in the references
>> header? How do you do that so easily? I'd love to the same. Thanks!
>
> I looked it up in google's C.U.P mirror
>
> https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ

But how did you know that this was the URL? Did you have to search? I
wish such mirror would index it by message-id --- or perhaps (group,
message-id). It doesn't seem to be like that.

The message included

In-Reply-To: <32DA98BA.4457@softart.com>#1/1>

The references were

<5amorl$d2c@rtpnews.raleigh.ibm.com>
<32DA98BA.4457@softart.com>#1/1>

My desire is to look it up from my news reader.

On Fri, 07 May 2021 17:02:52 -0300, Boris Dorestand wrote:

> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>
>> On Tue, 04 May 2021 17:54:46 -0300, Boris Dorestand wrote:
>>
>>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>>
>>>> On Fri, 30 Apr 2021 16:29:59 -0700, Nicolas Vazquez wrote:
>>>>
>>>>> Yall still here
>>>>
>>>> You /do/ realize that you replied to a thread last posted to over 24
>>>> years ago, right?
>>>
>>> How did you know? Did you look the original message in the references
>>> header? How do you do that so easily? I'd love to the same. Thanks!
>>
>> I looked it up in google's C.U.P mirror
>>
>> https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ
>
> But how did you know that this was the URL? Did you have to search? I
> wish such mirror would index it by message-id --- or perhaps (group,
> message-id). It doesn't seem to be like that.
>
> The message included
>
> In-Reply-To: <32DA98BA.4457@softart.com>#1/1>
>
> The references were
>
> <5amorl$d2c@rtpnews.raleigh.ibm.com>
> <32DA98BA.4457@softart.com>#1/1>
>
> My desire is to look it up from my news reader.

Depending on your news provider, that may not be possible at all.

Best case, you can use the
Message-ID:
of the message you are looking for, or the
In-Reply-To:
or
References:
that lead to it.

If your news provider keeps an archive, then your news reader /might/
be able search that archive. But, FWIW, most news providers don't keep
archives of past articles, preferring to drop articles from their store
after a reasonable (and often short) period of time.

HTH
--
Lew Pitcher
"In Skills, We Trust"

On 5/7/21 4:02 PM, Boris Dorestand wrote:
> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>
>> On Tue, 04 May 2021 17:54:46 -0300, Boris Dorestand wrote:
>>
>>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>>
>>>> On Fri, 30 Apr 2021 16:29:59 -0700, Nicolas Vazquez wrote:
>>>>
>>>>> Yall still here
>>>>
>>>> You /do/ realize that you replied to a thread last posted to over 24
>>>> years ago, right?
>>>
>>> How did you know? Did you look the original message in the references

The first message I saw in my newsreader recently had a title of "Re:
How to check (in C) if a process (with". The "Re:" indicates that it was
a response to some earlier message, so I knew there should be earlier
messages. I could have also figured it out by looking at the message
headers, but just noticing the "Re:" is much simpler.

>>> header? How do you do that so easily? I'd love to the same. Thanks!
>>
>> I looked it up in google's C.U.P mirror
>>
>> https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ
>
> But how did you know that this was the URL? Did you have to search? I

Since I knew it was posted to comp.unix.programmer, I went to

https://groups.google.com/g/comp.unix.programmer

I then searched manually through the threads to find the title of this
thread. Since the most recent message on that thread was posted quite
recently, it was near the top. However, if you're interested in an older
message, there's a box near the top of the screen which says "Search
conversations within comp.unix.prog...". You can just type a simple
search string in that box, or you can fill out a slightly more
sophisticated search form by clicking the downward-pointing triangle on
the right hand side of that search box.
If you want to search all groups, click on the downward-pointing
triangle to the left of the search box.

They used to have an option of searching for a particular message using
its usenet message ID, but that's one of the many useful features that
they have slowly been removing over the years. Even as recently as a
year ago, Google Groups' search capabilities were significantly more
sophisticated than they are today (while being positively primitive
compared to what you could do with Deja News).

Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:

> On Fri, 07 May 2021 17:02:52 -0300, Boris Dorestand wrote:
>
>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>
>>> On Tue, 04 May 2021 17:54:46 -0300, Boris Dorestand wrote:
>>>
>>>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>>>
>>>>> On Fri, 30 Apr 2021 16:29:59 -0700, Nicolas Vazquez wrote:
>>>>>
>>>>>> Yall still here
>>>>>
>>>>> You /do/ realize that you replied to a thread last posted to over 24
>>>>> years ago, right?
>>>>
>>>> How did you know? Did you look the original message in the references
>>>> header? How do you do that so easily? I'd love to the same. Thanks!
>>>
>>> I looked it up in google's C.U.P mirror
>>>
>>> https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ
>>
>> But how did you know that this was the URL? Did you have to search? I
>> wish such mirror would index it by message-id --- or perhaps (group,
>> message-id). It doesn't seem to be like that.
>>
>> The message included
>>
>> In-Reply-To: <32DA98BA.4457@softart.com>#1/1>
>>
>> The references were
>>
>> <5amorl$d2c@rtpnews.raleigh.ibm.com>
>> <32DA98BA.4457@softart.com>#1/1>
>>
>> My desire is to look it up from my news reader.
>
> Depending on your news provider, that may not be possible at all.
>
> Best case, you can use the
> Message-ID:
> of the message you are looking for, or the
> In-Reply-To:
> or
> References:
> that lead to it.
>
> If your news provider keeps an archive, then your news reader /might/
> be able search that archive. But, FWIW, most news providers don't keep
> archives of past articles, preferring to drop articles from their store
> after a reasonable (and often short) period of time.

I meant to look it up on an HTTP server, say, not from an NNTP server or
my NNTP server. For instance, why doesn't Google Groups lets us do that
right now since it probably has it all archived and available on the
web?

James Kuyper <jameskuyper@alumni.caltech.edu> writes:

> On 5/7/21 4:02 PM, Boris Dorestand wrote:
>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>
>>> On Tue, 04 May 2021 17:54:46 -0300, Boris Dorestand wrote:
>>>
>>>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> writes:
>>>>
>>>>> On Fri, 30 Apr 2021 16:29:59 -0700, Nicolas Vazquez wrote:
>>>>>
>>>>>> Yall still here
>>>>>
>>>>> You /do/ realize that you replied to a thread last posted to over 24
>>>>> years ago, right?
>>>>
>>>> How did you know? Did you look the original message in the references
>
> The first message I saw in my newsreader recently had a title of "Re:
> How to check (in C) if a process (with". The "Re:" indicates that it was
> a response to some earlier message, so I knew there should be earlier
> messages. I could have also figured it out by looking at the message
> headers, but just noticing the "Re:" is much simpler.
>
>>>> header? How do you do that so easily? I'd love to the same. Thanks!
>>>
>>> I looked it up in google's C.U.P mirror
>>>
>>> https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ" rel="nofollow" target="_blank">https://groups.google.com/g/comp.unix.programmer/c/xN3-YrzTeLk/m/e-wNQjBOzsEJ
>>
>> But how did you know that this was the URL? Did you have to search? I
>
> Since I knew it was posted to comp.unix.programmer, I went to
>
> https://groups.google.com/g/comp.unix.programmer
>
> I then searched manually through the threads to find the title of this
> thread. Since the most recent message on that thread was posted quite
> recently, it was near the top. However, if you're interested in an older
> message, there's a box near the top of the screen which says "Search
> conversations within comp.unix.prog...". You can just type a simple
> search string in that box, or you can fill out a slightly more
> sophisticated search form by clicking the downward-pointing triangle on
> the right hand side of that search box.
> If you want to search all groups, click on the downward-pointing
> triangle to the left of the search box.

Thank you. But I was hoping we could just look it up as in

curl http://somewhere.example.com/comp.unix.programmer/message-id-xyz

or even better

curl http://somewhere.example.com/message-id-xyz.

It's a pity no archive lets us do that.

> They used to have an option of searching for a particular message using
> its usenet message ID, but that's one of the many useful features that
> they have slowly been removing over the years. Even as recently as a
> year ago, Google Groups' search capabilities were significantly more
> sophisticated than they are today (while being positively primitive
> compared to what you could do with Deja News).

It's a shame.

I think it's pretty simple. Index it properly and let it be fetched
through an HTTP API protected by keys that limit usage. Then let people
build their software on it. If it gets expensive, handle it to some
organization such as archive.org or create a non-profit for it. A lot
of people would devote themselves to keep this up.

On 5/8/21 8:36 AM, Boris Dorestand wrote:
....
> I meant to look it up on an HTTP server, say, not from an NNTP server or
> my NNTP server. For instance, why doesn't Google Groups lets us do that
> right now since it probably has it all archived and available on the
> web?

They used to offer such a feature, and as far as I can tell, have
dropped it. Why? I have no idea - nothing seems particularly sane about
the way Google Groups has been managed. The simplest explanation is that
Google has no particular interest in providing useful access to their
usenet archives, but if that's the case, why do they bother providing
any access to them at all? Why do they regularly modify their interface
to make it less useful? Surely just maintaining their existing software
would have been easier?

James Kuyper <jameskuyper@alumni.caltech.edu> writes:

> On 5/8/21 8:36 AM, Boris Dorestand wrote:
> ...
>> I meant to look it up on an HTTP server, say, not from an NNTP server or
>> my NNTP server. For instance, why doesn't Google Groups lets us do that
>> right now since it probably has it all archived and available on the
>> web?
>
> They used to offer such a feature, and as far as I can tell, have
> dropped it. Why? I have no idea - nothing seems particularly sane about
> the way Google Groups has been managed. The simplest explanation is that
> Google has no particular interest in providing useful access to their
> usenet archives, but if that's the case, why do they bother providing
> any access to them at all? Why do they regularly modify their interface
> to make it less useful? Surely just maintaining their existing software
> would have been easier?

All good questions. They should just provide us with a series of
packages of all the archive and let us handle it some other way. I
wonder what is the size it of all.

On Monday, May 10, 2021 at 2:38:13 PM UTC-4, Boris Dorestand wrote:
> James Kuyper <james...@alumni.caltech.edu> writes:
....
> > the way Google Groups has been managed. The simplest explanation is that
> > Google has no particular interest in providing useful access to their
> > usenet archives, but if that's the case, why do they bother providing
> > any access to them at all? Why do they regularly modify their interface
> > to make it less useful? Surely just maintaining their existing software
> > would have been easier?
> All good questions. They should just provide us with a series of
> packages of all the archive and let us handle it some other way. I
> wonder what is the size it of all.

See <https://en.wikipedia.org/wiki/Usenet#Usenet_traffic_changes>. Unfortunately, those numbers include the binaries newsgroups, which would require enormously more space to archive than the purely text messages, which is why most places don't even consider archiving the binaries. Still, I suspect that even the text archives are too big for most users to even consider downloading them.

James Kuyper <jameskuyper@alumni.caltech.edu> writes:

> On Monday, May 10, 2021 at 2:38:13 PM UTC-4, Boris Dorestand wrote:
>> James Kuyper <james...@alumni.caltech.edu> writes:
> ...
>> > the way Google Groups has been managed. The simplest explanation is that
>> > Google has no particular interest in providing useful access to their
>> > usenet archives, but if that's the case, why do they bother providing
>> > any access to them at all? Why do they regularly modify their interface
>> > to make it less useful? Surely just maintaining their existing software
>> > would have been easier?
>> All good questions. They should just provide us with a series of
>> packages of all the archive and let us handle it some other way. I
>> wonder what is the size it of all.
>
> See
> <https://en.wikipedia.org/wiki/Usenet#Usenet_traffic_changes>. Unfortunately,
> those numbers include the binaries newsgroups, which would require
> enormously more space to archive than the purely text messages, which
> is why most places don't even consider archiving the binaries. Still,
> I suspect that even the text archives are too big for most users to
> even consider downloading them.

Very interesting. Thank you.