Tuesday, January 24, 2023

Digest for comp.lang.c++@googlegroups.com - 25 updates in 8 topics

David Brown <david.brown@hesbynett.no>: Jan 24 02:20PM +0100

On 23/01/2023 21:46, David Ritz wrote:
>> saw it with his own eyes.
 
> I'm of the "Show Me" school. I do not care about spurious,
> unsupported claims. The only glitch was human error.
 
I'm sorry, but you are wrong.
 
If you think servers - hardware and/or software - can never fail, you
are naïve beyond comprehension.
 
If you think it is appropriate to accuse one person you have never met
of lying, and another of trolling, based solely on your belief in the
perfection of all Usenet servers, then you should not be involving
yourself in any kind of abuse resolution or advice. You should be
asking questions first - not passing arbitrary judgement.
 
Someone of the "Show me" school would ask for more information. You
appear to be in the "Jump to conclusions" school.
 
 
 
There are limits to how much I can "show you" what happened. But I can
describe things in as much detail as practical.
 
I use news.eternal-september.org as a Usenet server, with Thunderbird as
the client. I am currently looking at the thread with subject "Re:
Compute Unique Numbers in a Set" started (via a cross-post from
comp.lang.c) on 08.01.2023. On one branch, posters "Bonita Montero" and
"Muttley@dastardlyhq.com" had been arguing back and forth, here in
comp.lang.c++. The penultimate post in the branch was made by Bonita
18.01.2023, 18:23, message id "tq99tp$vtqp$1@dont-email.me". Everything
looks normal.
 
When downloading message headers, the next message on the thread appears
to be the expected reply from Muttley. It has the same subject, and the
timestamp 19.01.2023, 10:31.
 
The message body here is, however, completely unexpected. It's headers
do not match the previously downloaded headers. In the message body,
the subject is "Change and Choice", the poster is "Ilya Shambat", the
newsgroup is "rec.arts.books", the id is
"36403165-3cf1-4b73-8ad1-da339b960339n@googlegroups.com", there are no
reference links to the other messages in the thread. The message
contains a rambling essay.
 
 
If I attempt to follow-up the message in Thunderbird, the Thunderbird
client sets the newsgroup to "rec.arts.books", the subject to "Re:
Compute Unique Numbers in a Set" (i.e., from the pre-downloaded headers,
not the body), and the quotation is given as "Muttley@dastardlyhq.com
wrote:". The body of the quotation is the rambling essay. This is
entirely consistent with Bonita's post - she replied to what she
believed to be a message from Muttley.
 
 
 
It is quite clear that there has been a glitch on the
news.eternal-september.org server. Somehow, a message body from
rec.arts.books has been crosslinked, was delivered as though it matched
the normal-looking pre-downloaded header in the thread. (I'm guessing
Muttley actually made a post in the original thread, but I am unable to
see the real message body.) The glitch appears to be limited to
news.eternal-september.org - other users of that server have said they
saw it, while it did not appear on other servers. People have reported
the existence of the rambling essay post in rec.arts.books, made by Ilya
Shambat - it was a real post, but on news.eternal-september.org the
message body was accidentally attached to the wrong header summaries for
comp.lang.c++.
 
 
Usenet server glitches are rare - client glitches are more likely. The
same glitch occurring for at least three different people is, however,
highly unlikely. And the idea that Bonita would jumped through hoops to
repost some random post from another group but made to look like it came
from Muttley so that she could comment on it - it's ridiculous.
 
 
 
> I recommend a visit to an optician.
 
I recommend you either listen to what happened, or bow out of this
conversation. It is possible that you have experiences or knowledge
that could help explain things - but if you think you know all the
answers before you've heard the question, there's no point in commenting.
scott@slp53.sl.home (Scott Lurndal): Jan 24 03:36PM


>I'm sorry, but you are wrong.
 
>If you think servers - hardware and/or software - can never fail, you
>are naïve beyond comprehension.
 
I'm not sure that is really relevent.
 
In this case, an unsupported assumption has been made that news
server software (e.g. Cnews, INN and successors) "glitched"
by changing the headers of a post and reposting it.
 
My exposure to both Cnews and INN would lead me to assert
that such a glitch is extremely unlikely. It's far more
likely that someone, perhaps Christoph/Bonita, was playing
games.
Malcolm McLean <malcolm.arthur.mclean@gmail.com>: Jan 24 08:06AM -0800

On Tuesday, 24 January 2023 at 15:36:23 UTC, Scott Lurndal wrote:
> that such a glitch is extremely unlikely. It's far more
> likely that someone, perhaps Christoph/Bonita, was playing
> games.
 
As a general rule you don't accuse someone of lying unless you're
absolutely certain that this is the case, and it's about an important
matter. Maybe Bonita pressed a few buttons on her web client and
somehow inadvertently generated a spurious crosspost. But it seems
to me neither here nor there.
David Brown <david.brown@hesbynett.no>: Jan 24 06:30PM +0100

On 24/01/2023 16:36, Scott Lurndal wrote:
> that such a glitch is extremely unlikely. It's far more
> likely that someone, perhaps Christoph/Bonita, was playing
> games.
 
The assertion of a server glitch is /not/ unsupported. I have
explained, in detail, exactly how things appeared. Both I and another
news.eternal-september.org user (red floyd) have said they saw the
original spurious post - the one Bonita replied to. I am not talking
about Bonita's reply that everyone saw, but the original spurious post
that she replied to.
 
If the only evidence was Bonita's post, then I could understand if
people thought she was playing games - truly weird, and out of character
games though they may be. But do you really believe that I, and red
floyd, have been conspiring with Bonita and are making up everything?
 
Unless you think I am deliberately lying, you will have to accept that
the spurious post existed. We all know that such a newsgroup-jumping
post should not have occurred in normal functioning behaviour. I assert
that this could only have come from one of two causes - a server glitch,
or the result (or side-effect) of a deliberate hack. I certainly do not
believe that Bonita would be behind such a hack, though it is vaguely
conceivable that there has been a completely different hack into the
news.eternal-september.org machine and this jumbled post was an
unexpected side-effect.
 
I can find no explanation that is more feasible than a server glitch.
Note that this does not necessarily mean a flaw in the software.
news.eternal-september.org is a low budget affair - in the past, they
have been pulled completely offline by hardware faults, dead disks, and
the like. We don't need to look any further than a single-event upset -
a bad memory cell, a bad disk read, a power fluctuation, a cosmic ray in
the wrong place - to see a single mistake. Or maybe it /is/ software -
an obscure race condition, a bug in the kernel.
 
I am quite happy to accept that such server glitches are extremely
unlikely. But extremely unlikely events happen sometimes.
Spiros Bousbouras <spibou@gmail.com>: Jan 24 05:47PM

On Tue, 24 Jan 2023 14:20:19 +0100
 
> Usenet server glitches are rare - client glitches are more likely. The
> same glitch occurring for at least three different people is, however,
> highly unlikely.
 
It could be a Thunderbird issue. Note the following :
 
From: Bonita Montero <Bonita.Montero@gmail.com>
Newsgroups: rec.arts.books,comp.lang.c++
Subject: Re: Compute Unique Numbers in a Set
Date: Thu, 19 Jan 2023 15:42:54 +0100
Message-ID: <tqbkrr$1ja1c$1@dont-email.me>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
Thunderbird/102.6.1
 
From: red floyd <no.spam.here@its.invalid>
Newsgroups: comp.lang.c++
Subject: Re: Compute Unique Numbers in a Set
Date: Thu, 19 Jan 2023 08:42:51 -0800
Message-ID: <tqbrub$1kco3$1@redfloyd.dont-email.me>
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101
Thunderbird/102.6.1
 
From: David Brown <david.brown@hesbynett.no>
Newsgroups: comp.lang.c++,news.admin.net-abuse.usenet
Subject: Re: Server glitch vs other possibilities (Was : Compute Unique
Numbers in a Set)
Date: Tue, 24 Jan 2023 18:30:21 +0100
Message-ID: <tqp4jd$7ik0$1@dont-email.me>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
Thunderbird/102.4.2
 
All 3 of you use Thunderbird. Not precisely the same version but close. I assume
you have tried to reproduce the steps which originally led you to see the defective
post and you can no longer reproduce them.
 
--
vlaho.ninja/prog
James Kuyper <jameskuyper@alumni.caltech.edu>: Jan 24 01:30PM -0500

On 1/24/23 12:47, Spiros Bousbouras wrote:
> you have tried to reproduce the steps which originally led you to see
> the defective
> post and you can no longer reproduce them.
 
I use Thunderbird 102.4.2, the same as David, but never saw the spurious
post that he saw.
 
However, with both him and Red Floyd claiming to have seen it, I see no
plausible reason to disbelieve them. I would not be surprised if the
cleanup for such a glitch would have erased the evidence that it had
occurred (I am not suggesting any attempt to hide the evidence, just a
normal clean-up after a mess occurs).
My opinion of Bonita's competence is low, and my opinion of her social
skills is even lower. However, I don't see any plausible reason for her
to fake this. My own first assumption, had I seen it, would have been
that he was posting to two different newsgroups using two different
identities, and slipped up somehow. Therefore, I won't fault her for
failing to consider the possibility of a server glitch instead.
scott@slp53.sl.home (Scott Lurndal): Jan 24 06:50PM

>people thought she was playing games - truly weird, and out of character
>games though they may be. But do you really believe that I, and red
>floyd, have been conspiring with Bonita and are making up everything?
 
It was not my intention to cast doubt on observations, but rather
to cast doubt that it was due to a "server glitch".
 
 
>I can find no explanation that is more feasible than a server glitch.
 
Does eternel-september.org honor cancels?
David Brown <david.brown@hesbynett.no>: Jan 24 08:36PM +0100

On 24/01/2023 19:50, Scott Lurndal wrote:
>> floyd, have been conspiring with Bonita and are making up everything?
 
> It was not my intention to cast doubt on observations, but rather
> to cast doubt that it was due to a "server glitch".
 
Fair enough. But how would a joke post by Bonita be visible in one
server only? How would it break Usenet server setups by having a
mismatch between the header and the message body?
 
 
>> I can find no explanation that is more feasible than a server glitch.
 
> Does eternel-september.org honor cancels?
 
I don't know.
 
 
I have just tried taking a new newsreader (pan - since I am somewhat
familiar with it, and it's a quick apt-get away) and connected to
news.eternal-september.org. I downloaded the past 20 days of headers
for comp.lang.c++. Sure enough, the bad post is there. In the list of
headers, it appears to be a post from Muttley on 19.01.2023 at 10:31, in
reply to Bonita. Click on the header and look at the post, and the
subject, from field, data, newsgroup, etc., are all wrong.
 
If I look in Google Groups, I see the same header for a post from
Muttley at that time, but this time the body is correct - it is clearly
a reply from Muttley to Bonita.
 
news.eternal-september.org is giving out the wrong message body for that
header.
 
I encourage you and anyone else who doubts this to connect to
news.eternal-september.org and try it yourself. You need an account,
but it is free and quickly made, and the experiment will only take a few
minutes.
 
 
I give no credit to the idea that Bonita is capable of making a hack on
that server to produce this effect. I give no credit to the idea that
she would do so, in this way, if she were able to. If there is an
alternative explanation to a server glitch (hardware and/or software), I
would be happy to hear it.
David Brown <david.brown@hesbynett.no>: Jan 24 08:42PM +0100

On 24/01/2023 19:30, James Kuyper wrote:
> skills is even lower. However, I don't see any plausible reason for her
> to fake this. My own first assumption, had I seen it, would have been
> that he was posting to two different newsgroups using two different
("he" being Muttley in this context, I presume?)
> identities, and slipped up somehow. Therefore, I won't fault her for
> failing to consider the possibility of a server glitch instead.
 
That is exactly how I see it.
David Brown <david.brown@hesbynett.no>: Jan 24 08:46PM +0100

On 24/01/2023 18:47, Spiros Bousbouras wrote:
 
> All 3 of you use Thunderbird. Not precisely the same version but close. I assume
> you have tried to reproduce the steps which originally led you to see the defective
> post and you can no longer reproduce them.
 
You are correct that we all use Thunderbird, and it is definitely
something I considered. More relevantly, however, we all use
news.eternal-september.org. And I have just tested using a clean
installation of pan as an alternative newsreader, and seen exactly the
same effect.
 
I don't know why you think I can no longer reproduce the effect of the
broken post - I have made no such suggestion, precisely because I /can/
replicate it. I still see it on two different computers with
Thunderbird, and now also with Pan.
 
For convenience, I have taken a screenshot of the effect. I have not
used this site "paste.pics" before, and I apologise if there are
unwanted adverts or other effects. If you are sceptical of clicking the
link, then I can happily email the screenshot. But it seems a quick and
easy way to make a link to the screenshot.
 
<https://paste.pics/b4149f38abb4e210da0a71886714d014>
"Adam H. Kerman" <ahk@chinet.com>: Jan 24 07:50PM


>>. . .
 
>Does eternel-september.org honor cancels?
 
It does not honor third-party cancels. It will honor first-party cancels
using the Cancel-Lock protocol but not a cancel control message without
using Cancel-Lock.
Muttley@dastardlyhq.com: Jan 24 05:05PM

On Mon, 23 Jan 2023 16:44:01 -0300
>> }
 
>OMG! that's why we still have plenty of CVEs in C/C++ applications and
>also new programming languages like Rust... :-O
 
It was example code. It would have far more checks in a real program.
Juha Nieminen <nospam@thanks.invalid>: Jan 24 07:17AM


>> You read: "Nothing of what I wrote is controversial, how could anyone
>> disagree with me?"
 
> Perhaps you don't know what the word "controversial" actually means?
 
I am using the word "controversial" to mean something outrageous and
provocative, something that's likely to cause outrage or offense,
something that's disrespectful, insulting, offensive, derogatory, smug,
something that's likely to rile people up, something that stirs heavy
dispute, something very fringe, extremist and outrageous, or something
that goes so blatantly against established fact that it's outright
arrogant to state it.
 
In other words, something that stirs controversy and dispute.
 
If you say something like "I think apple pie is better than cherry pie",
that's not very controversial. You may disagree with it, but it's not
something you would get highly aggressive or outraged about. You could
have a reasonable conversation about the merits of that opinion, but
certainly you wouldn't go on a warpath trying to convince that person
otherwise and make absolutely certain he understands how utterly wrong
and arrogant he is for stating such an outrageous obvious falsehood.
Certainly you wouldn't start reading between the lines and putting
words in his mouth in order to try to discredit his opinion. In
other words, it's not a controversial statement.
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 11:23PM -0800

On 1/11/2023 4:20 AM, David Brown wrote:
> personally.  But even then, it is your posts that appear arrogant - no
> one suggests that you are an arrogant person.  There's a vital
> difference there.)
 
What is the most verbose code? An entire program within a variable name?
 
int a = return_int_main_arg_0_void_EXIT_SUCCESS = EXIT_SUCCESS?
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 11:26PM -0800

On 1/23/2023 11:23 PM, Chris M. Thomasson wrote:
 
> What is the most verbose code? An entire program within a variable name?
 
> int a = return_int_main_arg_0_void_EXIT_SUCCESS = EXIT_SUCCESS?
 
Sorry for being an ahole here Juha.
 
int a = just_taking_verbosity_to_an_extreme_ocd = 42;
David Brown <david.brown@hesbynett.no>: Jan 24 09:35AM +0100

On 24/01/2023 08:17, Juha Nieminen wrote:
> Certainly you wouldn't start reading between the lines and putting
> words in his mouth in order to try to discredit his opinion. In
> other words, it's not a controversial statement.
 
Thanks for writing that explanation.
 
 
It can be difficult to pin down exact meanings of words, and certainly
different people use them in different ways, and their strength varies
by context.
 
Matters of taste and opinion are not controversial - they are just
personal preferences.
 
I would say "controversy" requires one of two things. You can have a
passionate disagreement (not necessarily involving outrage or offence,
but at least involving a number of people or parties over time). Or you
can have at least one side of a disagreement considering their side to
be "fact", despite there being clearly significant arguments to the
contrary.
 
By definition, if one side of a disagreement says they think it is
"uncontroversial", and the other side disagrees with the opinion, then
the issue is "controversial".
 
 
So I think at least some of this discussion actually does come down
using the term "controversial" (and "uncontroversial") in a different
manner. (There have been other miscommunications, and there's at least
one poster who appeared simply to enjoy rocking the boat and making
others loose their cool.)
 
I believe, and you'll correct me if I interpret you badly, that what you
were intending to say can be paraphrased as saying that you are
surprised people get so angry, offensive, insulted and worked-up about a
relatively small matter such as identifier naming habits, and that it is
ridiculous that people are making such a fuss about it.
 
Unfortunately, that is not what you /actually/ said. Paraphrasing
again, what you actually said is that you believe it is a /fact/ that
the only way to write good code is to use long, full-word identifiers on
every occasion, and anyone who believes anything else is being ridiculous.
 
 
I hope you can appreciate that no one here is agreeing to that second
interpretation. We see it as extremist, and counter-productive -
descriptive identifiers are good, but /too/ much of a good thing is
usually bad. And while people are welcome to prefer descriptive
identifiers in all cases in their own code, they cannot claim that it is
always "better" or "clearer".
 
With the first interpretation, you'll see a lot more agreement. It is
surprising that a thread on this topic has evoked such passion. It is
worth noting, however, that the person that stands out as being a great
deal more worked-up is /you/.
 
 
Perhaps we can put this all to rest by agreeing on a few points? Or if
we can't agree on them, at least we can agree on what we disagree on?
 
1. We all approve of writing code to make it clear, easy to read, and
easy to understand.
 
2. Some code can be written in a way that experienced programmers can
jump into the code and quickly understand what is happening. If it is
possible to write the code in such a way, then that is a good thing.
Other types of code requires significant knowledge of the application or
field as a whole, and code should assume that knowledge rather than
duplicate it. (For example, it could use niche abbreviations rather
than cluttering the code with excessively long identifiers or comments.)
 
3. Idioms and conventions make code faster and easier to read and
understand, when used appropriately.
 
4. Descriptive names with full words are generally helpful for clarity
and understanding what code does. But excessive length makes code
harder to read and obscures useful information.
 
5. Google's style guide section on naming conventions - the whole
section - is pretty good for a lot of code. But some code may have
different requirements.
 
6. There is no one "perfect" style that applies everywhere. Clarity
should always be a goal, but there can be variations as to what makes
code clear.
Malcolm McLean <malcolm.arthur.mclean@gmail.com>: Jan 24 03:09AM -0800

On Tuesday, 24 January 2023 at 07:17:54 UTC, Juha Nieminen wrote:
 
> If you say something like "I think apple pie is better than cherry pie",
> that's not very controversial. You may disagree with it, but it's not
> something you would get highly aggressive or outraged about. x
 
Normally if we say that something is "controversial" we mean that it
supports the weaker or more stigmatised side in an established public
dispute. Sometimes we mean that it is likely to spark such a public dispute.
 
The word is used in a light-hearted way about trivial things, such as putting milk
in tea before or after the tea. So apple pie and cherry pie would be another
example.
 
It's uncontroversial that clear variable naming should be used. No-one would
dispute that. However it is not established that full English language words
should always be used. That mathematical convention is to use symbols. In
C++, you are restricted to Latin. Though it's not uncommon to see Greek
symbols like "theta" or "epsilon" spelt out. And some programming identifiers
have a lot in common with mathematical symbols, whilst others do not.
 
Most real world code uses a mixture of single letters, full English words,
and abbreviations.
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 07:59PM -0800

On 1/20/2023 5:56 PM, Sebastian Nibisz wrote:
>> comp.programming.threads. I did one a while back:
 
>> https://groups.google.com/g/comp.lang.c++/c/FBqOMvqWpR4/m/gfFd4J2GBAAJ
> I don't know this thread. I will read.
 
Thank you. Any input is welcome, indeed.
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 11:15PM -0800

On 1/20/2023 6:01 PM, Chris M. Thomasson wrote:
>>> without having to worry about the ABA problem. Is that correct?
>> Yes.
 
> Great: I need to find some time to study up on your work! Well done.
 
Fwiw, when you get some really free time to burn, bored, try to run your
code through Relacy:
 
https://github.com/dvyukov/relacy
 
https://www.1024cores.net/home/relacy-race-detector
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 10:46PM -0800

On 5/12/2022 12:16 PM, Chris M. Thomasson wrote:
> Using my experimental vector field to generate a fractal formation. Here
> is generation two:
 
> https://fractalforums.org/gallery/1612-120522191048.png
 
Well, I was only able to generate a finite view, but, it actually is
infinite:
 
https://youtu.be/bpBvK-VhSjA
Jack L <invalid@invalid.net>: Jan 24 03:15AM

Some of you might be interested in this document:
 
<https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/n4928.pdf>
 
Document is dated: 2022-12-18
 
I am not cross posting this to C newsgroup as most of them visit here
during the day.
red floyd <no.spam.here@its.invalid>: Jan 23 10:19PM -0800

On 1/23/2023 7:15 PM, Jack L wrote:
 
> Document is dated: 2022-12-18
 
> I am not cross posting this to C newsgroup as most of them visit here
> during the day.
 
Why would you even WANT to cross post this? It's about C++,
not about C.
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 08:11PM -0800

On 1/18/2023 11:55 PM, David Brown wrote:
 
>> Yes. I have been told this, damn near, exact same warning before David
>> on a couple of jobs.
 
> That should give you a clue that it is a good idea!
 
Nodding. So far, I have found a bug in my code yet. During a refactor.
"Chris M. Thomasson" <chris.m.thomasson.1@gmail.com>: Jan 23 08:12PM -0800

On 1/23/2023 8:11 PM, Chris M. Thomasson wrote:
>>> David on a couple of jobs.
 
>> That should give you a clue that it is a good idea!
 
> Nodding. So far, I have found a bug in my code yet. During a refactor.
 
God damn it! I have NOT found a bug yet. f'ing typos!
 
Shit.
Lynn McGuire <lynnmcguire5@gmail.com>: Jan 23 08:21PM -0600

"C++ programming language and safety: Here's where it goes next"

https://www.zdnet.com/article/c-programming-language-and-safety-heres-where-it-goes-next/
 
"There's been a shift towards 'memory safe' languages. So, can updates
to C++ help it catch up in the eyes of developers?"
 
Lynn
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)