Re: making Debian secure by default

[Emanuel Berg]

David Wright wrote:

>> Ah, surely it can't refer to that as that would be
>> completely ridiculous as it would imply "wanna install
>> stuff? sure, but then it isn't secure anymore".
>
> It's not clear what "isn't secure anymore" means. [...]

It means as soon as you start doing stuff with the software,
it isn't secure anymore. Which is comical to some extent as
doing stuff is the purpose of computers.

So to base security boasting on people having the most
minimal, restricted and inactive system, it is like boasting
this marvelous piece of body armor is guaranteed to not have
a single infantryman killed - just don't go to war.

(Note that now I'm just making fun at the slogan and boasting,
not saying anything negative of their OS necessarily - I've
used it myself, it send pretty good and, indeed, secure.)

>  "Secure by Default"
>
>  "To ensure that novice users of OpenBSD do not need to
>   become security experts overnight (a viewpoint which other
>   vendors seem to have), we ship the operating system in
>   a Secure by Default mode. All non-essential services are
>   disabled. As the user/administrator becomes more familiar
>   with the system, he will discover that he has to enable
>   daemons and other parts of the system. During the process
>   of learning how to enable a new service, the novice is
>   more likely to learn of security considerations."
>
> from https://www.openbsd.org/security.html
> OTOH:
>
>  "There are many applications one might want to use on an
>   OpenBSD system. To make this software easier to install
>   and manage, it is ported to OpenBSD and packaged. The aim
>   of the package system is to keep track of which software
>   gets installed, so that it may be easily updated or
>   removed. In minutes, a large number of packages can be
>   fetched and installed, with everything put in the
>   right place."
>
>  "The ports collection does not go through the same thorough
>   security audit that is performed on the OpenBSD base
>   system. Although we strive to keep the quality of the
>   packages high, we just do not have enough resources to
>   ensure the same level of robustness and security."
>
> from https://www.openbsd.org/faq/faq15.html (Package
> Management).

The more you install, the less secure it gets. Yeah, can't
base the security model on that.

They should do it the other way around, write a piece of
software that breaks everything. Install in on OpenBSD and if
it breakes it, OpenBSD is not more secure than anyone else.
If nothing happens tho most likekly you are safe.

-- 
underground experts united
https://dataswamp.org/~incal

Re: making Debian secure by default

[Emanuel Berg]

Michael Kjörling wrote:

>> "Secure by default" is an OpenBSD slogan BTW. Or they have
>> made it into one at least. But I'm not sure it is any more
>> secure than Debian - maybe.
>> 
>>   https://www.openbsd.org/security.html
>
> If I'm not mistaken, OpenBSD is "secure by default" by being
> "extremely minimalistic by default".
>
> Last I looked, which in fairness was a while ago, a default
> installation of OpenBSD includes almost nothing that normal,
> present-day users would expect to find on their system. [...]

Ah, surely it can't refer to that as that would be completely
ridiculous as it would imply "wanna install stuff? sure, but
then it isn't secure anymore".

-- 
underground experts united
https://dataswamp.org/~incal

Re: making Debian secure by default

[Emanuel Berg]

"Secure by default" is an OpenBSD slogan BTW. Or they have
made it into one at least. But I'm not sure it is any more
secure than Debian - maybe.

  https://www.openbsd.org/security.html

-- 
underground experts united
https://dataswamp.org/~incal

Re: OT: End the Phone-Based Childhood Now

[Emanuel Berg]

Jeffrey Walton wrote:

> So the problem with GenZ seems to be how they are growing up
> and what they are spending their time on; and not their job
> (they are teens!)

They need other things to do that appeal to them more than
smartphone idling. If such things existed, they would go
for them, I'm confident.

But it is also how things are presented, what mental picture
they have, which is often incorrect. Today almost all
activities, even pretty mundane things that were once thought
of as relaxing and potentially inclusive to a whole bunch of
people, are presented as elitist pursuits for the select few.

They think, for example, "Martial arts seems like a lot of
fun, but it is nothing for me, everyone who does it are great
athletes and clearly I'm not." while in reality it is "_A lot_
of people who does it are great athletes - and the reason why
is because they do it". These kids only need to show up, but
sadly, a lot of them don't, ever.

So it is a vicious circle, the more they think they have to be
brilliant to do anything the less confident they become from
doing nothing.

> and not society around them (which they withdraw from).

Society pushed them away just as much.

-- 
underground experts united
https://dataswamp.org/~incal

Re: OT: End the Phone-Based Childhood Now

[Emanuel Berg]

Jeffrey Walton wrote:

> Fascinating reading here:
> .
> It completely explains why GenZ are having so many problems
> with adulthood. Smartphones and Social Media are
> the culprits.

Society is the problem where you are either an elite prospect
football player, a professional carpenter/construction worker,
_or_ you don't get to do anything, ever.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Will Mengarini wrote:

>> With no intention of ever creating a 100% offensive-free
>> language, removing the worst offenders from the scene often
>> is enough.
>
> Words I find offensive include "authority" and "manager", so
> checking `apropos authori manager` I see we have a lot of
> important work to do.

You have a lot to do. If you consider those words the worst
offenders. And intend to do something about it.

> Seriously, you humans have only another five billion Earth
> years until your sun engulfs your home planet, and you're
> spending time on *THIS*?!

Relax, people also build shelves and get dead drunk at their
brothers' weddings.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Mike Castle wrote:

> Was that explicitly stated anywhere? Or is the lack of any
> type of explicit "I'm willing to help drive this" statements
> leading to that conclusion?

Relax, everyone does something somewhere. But it would be
a boring world if they were only allowed to talk about that.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Alain D D Williams wrote:

> That is the big difference. Not use words *currently* deemed
> offensive in *new* publications (books, newspaper articles,
> ...) - this is not hard to do.

Indeed, and that is what you should focus on. The past is the
past anyway.

> What we are faced with is something very different: a call
> to locate and modify use in programs that might have been
> written a long time ago. The effort needed to do this is
> large and will doubtless cause failures in systems that have
> been working well for years.

I must admit the whole concept of source code being offensive
is a bit bizarre to me. For anyone to really change that it in
a way that makes sense it must be a really offensive word and
a general understanding that people reading and writing the
code really reacts negatively to it. Because in my experience,
people who do this kind of politics aren't typically
programmers, even.

But I may be wrong and from a technical perspective, it is
possible to change source, obviously.

> It is not just a matter of modifying Debian (+ RedHat + ...)
> sources but the sources on private systems.

I think it is a bad idea to go for a clean sweep. That either
don't work or end up like the Khmer Rouge. It is enough to
remove the most offensive words and expressions, whatever they
are, from the most public platforms.

> We seem to be told that this must be done by those who will
> not be doing the work.

Ah, it is okay for people to have opinions and voice them
without doing stuff. But sometimes such people somehow get
into positions of authority and, worst case scenario, force
people who have been doing stuff for ages out of
their projects. That's horrible but such instances should not
be blamed on the general "opinions but no work" personality,
who is actually quite harmless.

-- 
underground experts united
https://dataswamp.org/~incal

Re: "libEGL warning: DRI2: failed to authenticate"

[Emanuel Berg]

> libEGL warning: DRI2: failed to authenticate
> [vo/gpu/opengl] Suspected software renderer or indirect context.
> [vo/gpu] Failed initializing any suitable GPU context!
> Error opening/initializing the selected video_out (--vo) device.

Problem solved. I had a package called bumblebee installed and
in a file in /etc/modprobe.d from that package was a line
"blacklist nouveau".

-- 
underground experts united
https://dataswamp.org/~incal

"libEGL warning: DRI2: failed to authenticate"

[Emanuel Berg]

Hello, I have this problem with Debian and mpv recently after
not using the computer for a while, now plugging everything in
and upgrading. I asked on #mpv and got some help but still
don't work, I post the whole backlog here here. TIA.

But the error message is,

libEGL warning: DRI2: failed to authenticate
[vo/gpu/opengl] Suspected software renderer or indirect context.
[vo/gpu] Failed initializing any suitable GPU context!
Error opening/initializing the selected video_out (--vo) device.

Here is the backlog,

 hi, had the computer stashed away for some time, plugged it in
today, now I get the following error from mpv:
 libEGL warning: DRI2: failed to authenticate
 [vo/gpu/opengl] Suspected software renderer or indirect context.
 [vo/gpu] Failed initializing any suitable GPU context!
 Error opening/initializing the selected video_out (--vo) device.
 without more information, I conclude you have broken drivers
 okay, what do you do about that? I'm on Debian, everything is
upgraded. mpv 0.37.0
 how did you get mpv?
 paste a log file, mpv --no-config --log-file=mpv_is_broken.log
 got mpv with these commands,
https://dataswamp.org/~incal/conf/.zsh/mpv-install
 LaserEyess, log file:
https://dataswamp.org/~incal/error/mpv-gpu.txt
 mpv cannot detect your GPU, mesa only says you have software
renderers
 if mpv cannot detect it, does it mean it is down or is it
a problem with mpv?
 it's a problem with your system
 the drivers?
 maybe same as this?
https://lists.debian.org/debian-x/2015/08/msg00243.html
 no, 2015
   paste `eglinfo` somewhere
   it comes from mesa-utils or whatever
 https://dataswamp.org/~incal/error/eglinfo.txt
   yeah it just says failed so your drivers are broken
 no, wait, I should do it from X,
https://dataswamp.org/~incal/error/eglinfo-x.txt
   that looks more functional
 but it says "libEGL warning: DRI2: failed to authenticate" when
I run it in X
   o wait nvm
   says swrasty
   which gpu do you have
 msi Nvidia Geforce GT 710
   ah nvidia
   very fun
 what driver should you have for that? maybe it got removed when
I upgraded the system, no idea
 I don't have any nvidia stuff installed, a bunch of mesa tho if
that is the/a alternative?
 but everything is upgraded, so should work unless broke from the
repos. unlikely maybe
 incal: you have to install nouveau or the proprietary nvidia
drivers
 you have no drivers installed at all
 I have xserver-xorg-video-nouveau installed, isn't that nouveau?
 here is a YouTube video, "DEBIAN FIX: libEGL warning: DRI2: failed
to authenticate", from 2023, maybe I can watch that from
a smartphone. https://www.youtube.com/watch?v=WF1yJg2vzps
 idk, I don't use debian
 yeah
 will post the issue at gmane.linux.debian.user, must have happened
when I upgraded the system after not using it, right?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Mike Castle wrote:

>> It is "fixing" an issue for today's English speakers.
>> Should we scour our systems looking for similar issues in
>> other languages? Then in, say, 20 years time when different
>> words will then be considered offensive, by some, do this
>> all again?
>
> Yes.

Remember, there are A LOT of words and expressions we don't
use anymore, and that's good, as they are offensive and
disrespectful. But once they were perfectly normal. Still, one
by one, they have disappeared from active use.

What's to say we are right now, just because _we_ happen to
live right now, suddenly done with that process?

If it had to be done in the past, why not right now - and in
the future as well?

Now how to actually do it is another thing.

Maybe one should just focus on a few words and expressions
that are clearly offensive, and remove them from schools,
universities, public service TV, all official
state-related communication, etc.

With no intention of ever creating a 100% offensive-free
language, removing the worst offenders from the scene often
is enough.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Alain D D Williams wrote:

> However that is not the way that the world works, or prolly
> more accurately how some people think. They see
> a word/phrase that they have decided that they "own" or
> somehow relates to them [...]

I am not black so I have no idea how black people consider
everything negative in language that is black. If indeed most
of them have no strong feelings about it it may be a waste of
time trying to change such expressions.

If they do care about it one could try to reduce such use from
formal and official language, especially when it really hasn't
anything to do with the color black - like blacklist into
blocklist, and other such examples.

Maybe in fantasy novels one would still be allowed to have
evil wizards all dressed in black, doing powerful incantations
of black magic?

-- 
underground experts united
https://dataswamp.org/~incal

Re: 'sensors -j' and "ERROR: Can't get value of subfeature in0_input: Can't read"

[Emanuel Berg]

>> In general, the first thing you should try is running
>> sensors-detect again, as root.
>
> Okay, I did that ('sudo sensors-detect') and answered with the
> default value to all questions, after that I did 'sensors -j'
> but it displayed the same error.
>
>> It is possible that your kernel is not loading some
>> particular sensor module which would be recommended by
>> sensors-detect.
>
> So I should reboot? Let's do that.

I rebooted but same :(

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

jeremy ardley wrote:

>> But what about the black market? Or does in fact "block
>> market" work just fine?
>
> The term "black market" is from World War II - i.e. 1939-45.
> It has nothing to do with slaves. It means transactions in
> the dark, not visible,  not official.

I think the reason is black people shouldn't be associated
with everything negative that is black in language.

It was a BLM thing, not sure if it matters the etymology of
such words.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Inclusive terminology (instead of master/slave) for network bonding/LACP

[Emanuel Berg]

Marco Moock wrote:

> Just check what different meanings GIMP has. Maybe some more
> people now feel uncomfortable with using it.
> https://www.dict.cc/?s=gimp

Yes, people have been saying that for quite some time:

  https://news.ycombinator.com/item?id=20359520]

  
https://www.theregister.com/2019/08/28/gimp_open_source_image_editor_forked_to_fix_problematic_name/

Personally, I don't really care, at least not in the
master/slave and GIMP cases.

Blacklist perhaps could and should be changed to blocklist.

But what about the black market? Or does in fact "block
market" work just fine?

-- 
underground experts united
https://dataswamp.org/~incal

Re: 'sensors -j' and "ERROR: Can't get value of subfeature in0_input: Can't read"

[Emanuel Berg]

Dan Ritter wrote:

> In general, the first thing you should try is running
> sensors-detect again, as root.

Okay, I did that ('sudo sensors-detect') and answered with the
default value to all questions, after that I did 'sensors -j'
but it displayed the same error.

> It is possible that your kernel is not loading some
> particular sensor module which would be recommended by
> sensors-detect.

So I should reboot? Let's do that.

-- 
underground experts united
https://dataswamp.org/~incal

'sensors -j' and "ERROR: Can't get value of subfeature in0_input: Can't read"

[Emanuel Berg]

sensors(1) and in particular the command 'sensors -j' now
reports

  ERROR: Can't get value of subfeature in0_input: Can't read
  ERROR: Can't get value of subfeature in1_input: Can't read

for the CPU and GPU temperatures.

This previously worked so I don't know why it doesn't all of
a sudden. Some sysfs driver issue? Or hardware
malfunction? Unlikely?

Don't really know what "I changed" for this to happen, since
I didn't use the command for some time and possibly quite
a few things might have happened since then. But I remember
well that it worked at some point.

Here is what strace says:

  https://dataswamp.org/~incal/error/sensors-j.txt

TIA

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

jeremy ardley wrote:

> Or is "metric" one of these things spared by the
> Brexit Bonfire?

 It depends which gear your camp is since the metric
 system is partly implemented and partly co-exists
>>>
>>> British Standard Pipe still in use for plumbing and 1/4",
>>> 3/8" of specification I forget for camera mounts.
>>
>> Boxing gloves, drums ...
>
> The British o-ring standard BS was based on inch
> measurements. It was then adapted so all measurements are in
> mm to precision 0.1 mm
>
> This standard is now adopted in the US by the SAE.
>
> So if you buy an o-ring in the US you will get a metric
> precision ring certified by the SAE who derive it from the
> British BS measurement which itself is metric, which in turn
> is derived from nominal inch measurements e.g. 3/32"

Haha case closed :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

mick.crane wrote:

>>> Or is "metric" one of these things spared by the
>>> Brexit Bonfire?
>>
>> It depends which gear your camp is since the metric system
>> is partly implemented and partly co-exists
>
> British Standard Pipe still in use for plumbing and 1/4",
> 3/8" of specification I forget for camera mounts.

Boxing gloves, drums ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

tomas wrote:

> Or is "metric" one of these things spared by the
> Brexit Bonfire?

It depends which gear your camp is since the metric system is
partly implemented and partly co-exists (e.g. on product
labels) in the UK ... and the supposed all-Metric world.
Everything, of course, Made in China anyway ...

Here is an example of a product that have not 2 but 3 systems
(English, French and ISO). Maybe you are familiar with it,

  https://dataswamp.org/~incal/bike/TIRE

But it is still not comparable since those digits say what
size the tire is, it isn't the tire version. And as for names,
they don't have that problem since often it's just the
companies name. Sometimes tho, for flashier high-end tires,
they have names to describe the particular product, and they
want to make them unique for commercial reasons.

So there are no reasons for us to do that, the main thing is
so one can refer to them. If there is no reason to call it
something, numbers or codes that refer to the technology is
the best. E.g., what power tool is AG18BL ... it is an angle
grinder with an 18V battery.

But since computers systems are much more complicated that may
not be doable in a meaningful way to be used on the floor.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

Brad Rogers wrote:

>> But M5 can be a bolt size and a lot of other things as well,
>
> Context! When the conversation is about roads in the UK, why
> would *anyone* think bolt size?

I agree, but that's why people have a hangup with names these
days, they want their product or project to come up first if
anyone Googles them.

However it is better to assume people will Google "M5 bolt"
when needed rather than assuming they will remember Holy Socks
nicknames for computer software technology not associated with
either footwear or religion ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

tomas wrote:

>>> But M5 can be a bolt size and a lot of other things as well,
>> 
>> Context! When the conversation is about roads in the UK,
>> why would *anyone* think bolt size?
>
> Especially metric ones =:-o

Maybe the UK roads also follow a system. They were the first
guys having one, after all ;)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

debian-user wrote:

 No but I live in the UK and I know the A1, A2, A4, A5, A6
 and many others, plus the M1, M4, M5, M6.
>> 
>> But M5 can be a bolt size and a lot of other things as
>> well, while creative names may stay "more" unique.
>
> Nonsense. Curt's reply (which you cut when wrongly
> attributing my text to him!) provides an excellent example
> of a memorable name 'Penny Lane' - which I for one know only
> as the name of a song and didn't even know which town the
> road was in (Abbey Road I do happen to know).

I'm saying, the cute or exotic names don't sound like
technology, that's why they don't stick. They have to be
exceptional for that, like the Kalashnikov that is arguably
more known than the AK-47 - which is also widely known BTW.

> All names need context and the M5 as a road is not likely to
> be mixed up with an M5 bolt or screw.

Indeed, maybe search engines using Ubuntu can mix them up but
it is unlikely on Debian, also they are shorter and have the
feel of technology and not cartoonish PR brains at work.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

Curt wrote:

>> No but I live in the UK and I know the A1, A2, A4, A5, A6
>> and many others, plus the M1, M4, M5, M6.

But M5 can be a bolt size and a lot of other things as well,
while creative names may stay "more" unique.

But on the other hand there are many Emmas and Camillas, and
people tend to keep track of who is who anyway ...

Nah, creative, especially cute names are silly, this is an
engineering and to some extent even scientific discipline
after all. Bugs Bunny release names makes it silly compared to
minor.major.patch or whatever other a-personal numbers game
you'd like to play ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Why does Debian have code names for releases?

[Emanuel Berg]

Stefan Monnier wrote:

>> Leaving aside that Titanic is the real name of the ship and
>> not a codename, the evidence is all around you. Look no
>> further than your login name, or the name of your computer.
>> A huge slice of the Internet's infrastructure, DNS, is
>> concerned with allowing people to converse with memorable
>> names rather than anonymous numbers.
>
> Anecdotal evidence cuts both ways: how many years have names
> rather than numbers?

There are pieces of military equipment that have a code
designation as well as a flashy name, and people still use the
code designation. There is no telling what will stick from one
case to the other, or what will be most popular by some
majority of guys using it. As long as there is a name or
designation people can refer to it which is the most
important part.

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

>> Well if you were prepared to type a search for
>> computational linguistics software into google, you would
>> find several free tools available for linux listed on pages
>> such as
>>
>> https://martinweisser.org/corpora_site/comp_ling_resources.html
>
> Indeed, that page has 4 hits for Unix and 3 for Linux.
>
>> https://www.sil.org/linguistics/linguistics-software
>
> Ditto 1 hit for Unix and 19 (!) for Linux.
>
> So, a total of 5 Unix hits and 22 Linux, all in all 27 hits,
> possible duplicates not subtracted.

Ah, if you are not into this field, making sense of those
lists of software is like searching for a microchip in
a supercomputer ...

But I did find a Debian package (metapackage) named
science-linguistics:

$ aptitude show science-linguistics
Package: science-linguistics
Version: 1.14.2
State: not installed
Priority: optional
Section: metapackages
Maintainer: Debian Science Team 

Architecture: all
Uncompressed Size: 43.0 k
Depends: science-config (= 1.14.2), science-tasks (= 1.14.2)
Recommends: apertium, apertium-lex-tools, artha, cg3, collatinus, dimbl, frog,
hfst, hfst-ospell, irstlm, libcld2-dev, link-grammar, lttoolbox,
mbt, mbtserver, python3-pynlpl, r-cran-lexrankr, r-cran-snowballc,
timbl, timblserver, ucto, uctodata, wordnet
Suggests: apertium-af-nl, apertium-apy, apertium-arg, apertium-arg-cat,
  apertium-bel, apertium-bel-rus, apertium-br-fr, apertium-ca-it,
  apertium-cat, apertium-cat-srd, apertium-crh, apertium-crh-tur,
  apertium-cy-en, apertium-dan, apertium-dan-nor, apertium-en-ca,
  apertium-en-es, apertium-en-gl, apertium-eo-ca, apertium-eo-en,
  apertium-eo-es, apertium-eo-fr, apertium-es-ast, apertium-es-ca,
  apertium-es-gl, apertium-es-it, apertium-es-pt, apertium-es-ro,
  apertium-eu-en, apertium-eu-es, apertium-fr-ca, apertium-fr-es,
  apertium-fra, apertium-fra-cat, apertium-hbs, apertium-hbs-eng,
  apertium-hbs-mkd, apertium-hbs-slv, apertium-hin, apertium-id-ms,
  apertium-is-sv, apertium-isl, apertium-isl-eng, apertium-ita,
  apertium-kaz, apertium-kaz-tat, apertium-mk-bg, apertium-mk-en,
  apertium-mlt-ara, apertium-nno, apertium-nno-nob, apertium-nob,
  apertium-oc-ca, apertium-oc-es, apertium-oci, apertium-pol,
  apertium-pt-ca, apertium-pt-gl, apertium-rus, apertium-separable,
  apertium-sme-nob, apertium-spa, apertium-spa-arg, apertium-srd,
  apertium-srd-ita, apertium-swe, apertium-swe-dan, apertium-swe-nor,
  apertium-szl, apertium-tat, apertium-tur, apertium-ukr, apertium-urd,
  apertium-urd-hin, frogdata, giella-sme, libcg3-dev, libfolia-dev,
  libmbt0-dev, libticcutils2-dev, libtimbl3-dev, libtimbl4-dev,
  libtimblserver2-dev, libucto1-dev, python3-nltk,
  python3-snowballstemmer, python3-streamparser, python3-thinc,
  python3-timbl, r-cran-nlp, r-cran-tm, sequitur-g2p, spacy, travatar,
  wnsqlbuilder
Description: Debian Science Linguistics packages
 This metapackage is part of the Debian Pure Blend "Debian Science" and
 installs packages related to Linguistics.
Homepage: https://wiki.debian.org/DebianScience/
Tags: field::linguistics, role::metapackage, suite::debian

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

debian-user wrote:

> Well if you were prepared to type a search for computational
> linguistics software into google, you would find several
> free tools available for linux listed on pages such as
>
> https://martinweisser.org/corpora_site/comp_ling_resources.html

Indeed, that page has 4 hits for Unix and 3 for Linux.

> https://www.sil.org/linguistics/linguistics-software

Ditto 1 hit for Unix and 19 (!) for Linux.

So, a total of 5 Unix hits and 22 Linux, all in all 27 hits,
possible duplicates not subtracted.

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

>> A basic search finds this web tool: 
>>
>> https://www.usingenglish.com/resources/text-statistics/
>
> I didn't get it to work in Emacs-w3m, be it lack of JavaScript
> support or something else. Anyway the page and tool claims to
> do this:
>
>   Total Word Count
>   Total Word Count (Excluding Common Words)
>   Number of Different Words
>   Different Words (Excluding Common Words)
>   Number of Paragraphs
>   Number of Sentences
>   Words per Sentence
>   Number of Characters (all)
>   Number of Characters (a-z)
>   Characters per Word
>   Syllables
>   Syllables per Word
>
> Sure, if one had a CLI tool doing that, I would say it's
> certainly a good start!

I have now tried it from a smartphone and it works great, It
does what I say (quote) above but actually much more and more
interesting things are analyzed and outputted as well,
including diagrams.

Alas, some output is not available unless one pays for the
enhanced version - I suppose that makes it shareware, as we
said in the 90s - but it still does a lot in its
current state.

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Nicholas Geovanis wrote:

> Those books teach and discuss some of the software that's
> used. I doubt you will find them in debian's repositories.
> Of course you can do plenty of computational linguistics
> with perl or python which you already have.
>
> What is a "regular expression" which is at the heart of perl
> and python? An expression which conforms to a certain type
> of grammar. Perl and python are used directly for analyzing
> text (any old language). You are learning basic
> computational linguistics.

Okay, but if there isn't a tool readily available I think this
is a window for a bunch of young programmers that feel the
need to show their skills. It could be a degree project in
Computer Science even, unless the Computational Linguistics
guys have their own degree projects. If so, they can borrow
FOSS and CLI from us and we'd get the tool as well when they
are done, that would be a fair trade IMO :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Joel Roth wrote:

> A basic search finds this web tool: 
>
> https://www.usingenglish.com/resources/text-statistics/

I didn't get it to work in Emacs-w3m, be it lack of JavaScript
support or something else. Anyway the page and tool claims to
do this:

  Total Word Count
  Total Word Count (Excluding Common Words)
  Number of Different Words
  Different Words (Excluding Common Words)
  Number of Paragraphs
  Number of Sentences
  Words per Sentence
  Number of Characters (all)
  Number of Characters (a-z)
  Characters per Word
  Syllables
  Syllables per Word

Sure, if one had a CLI tool doing that, I would say it's
certainly a good start!

> Otherwise, I think you'll have to write your own -- or hire
> someone (like me :^) to write one for you.

I have to squeeze the money out of my political organizations
first ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Nicholas Geovanis wrote:

> If you have python programming skills, you might
> consider NLTK

 Unbelievable if there are no such tools anywhere already,
 but I don't have one either so maybe there aren't then?
>>>
>>> There's a big subject called computational linguistics.
>>> They have some specialized tools for what they call corpus
>>> analysis. Because you mentioned statistics you threw
>>> everyone off :-) And I really like R.
>>
>> Okay, so now we are getting somewhere. The technical term
>> and scientific field of this activity is known as
>> computational linguistics, and the guys that do that do
>> corpus analysis. Sweet!
>
> Two standard text books are Foundations of Computational
> Linguistics by R Hausser, and Computational Linguistics: An
> Introduction by R Grishman.
>
> Syntactical analysis of human and artificial (programming)
> languages is well known. But how do you attach meaning to
> the symbols? Semantics. How do you identify style and
> emphasis? These are the kind of starting points for
> computational linguistics.

Okay, but do we have software in the Debian repositories, or
anywhere else in the Unix and FOSS world for that matter, so
we can try it out in practice?

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Nicholas Geovanis wrote:

>>> If you have python programming skills, you might consider
>>> NLTK
>>
>> Unbelievable if there are no such tools anywhere already,
>> but I don't have one either so maybe there aren't then?
>>
>
> There's a big subject called computational linguistics.
> They have some specialized tools for what they call corpus
> analysis. Because you mentioned statistics you threw
> everyone off :-) And I really like R.

Okay, so now we are getting somewhere. The technical term and
scientific field of this activity is known as computational
linguistics, and the guys that do that do corpus
analysis. Sweet!

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

dvalin wrote:

> As "stats" is a grab bag larger inside than the Tardis,
> I suspect that only on that other ship with the infinite
> improbability drive is a stats babelfish interpreter to be
> found. For the last 30+ years, I've just thrown together
> a few lines of Awk to generate the initially required stats,
> then tweaked the C-like code and regexes to add the
> inevitable nice-to-haves. Some result is immediate, and
> dissatisfaction with completeness motivates
> thetweaking/temporary_satisfaction cycle. Options are
> limitless, as is needed for an undefined task [...]

Haha, show us some stats then!

*handclaps in anticipation*

-- 
underground experts united
https://dataswamp.org/~incal

Re: RFP: deadEarth RPG documentation - free and open RPG game

[Emanuel Berg]

Joshua Allen wrote:

> deadEarth is a free and open TTRPG game for use in real
> life or online

Never heard of but sounds interesting, as technology, and
maybe even as a game.

Is it TT as in

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

?

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

Max Nikulin wrote:

>> sudo apt autoremove -y && sudo apt update && sudo apt
>> upgrade -y
>
> Almost from the start of this thread I am curious if there
> is a reason to not run autoremove immediately after upgrade.
> What is the purpose of retaining unnecessary packages till
> next upgrade?

Indeed, it makes sense to do cleanup after, but it also makes
sense to upgrade from a clean state.

> I usually add --purge to apt autoremove to avoid remnants of
> deleted kernel packages. On the other hand I am not brave
> enough to use -y.
>
> This thread started from a question concerning
> /var/cache/apt/archives. I do not remember if somebody
> already mentioned that "apt upgrade" by default removes
> downloaded .deb files in the case of success, however
> "apt-get upgrade" does no do it.

Good ideas, made 2 changes below accordingly.

a1 () {
sudo apt-get -qq check
sudo apt-get -qq update
sudo apt-get -qq --purge -y autoremove
sudo aptitude -q=99 autoclean
apt -a list --upgradable
}

a2 () {
a1
sudo apt -qq upgrade
sudo apt dist-upgrade
}

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

tomas wrote:

>>>> I mean a general tool, but with options to tweak the
>>>> report included, of course.
>>>
>>> If you can bear some tweaking, R is it.
>> 
>> Sure! Let's run R on this e-mail. Does it work and if so, what
>> does it say?
>
> T a generic question -- a generic answer

R is a programming language, I'm looking for a tool that
produces stats from text. If such a tool uses R, or any other
programming language or stats engine to produce the outcome,
for me as a potential user that is entirely up to them who
write it.

> I don't even know what you mean by "general stats"

Some examples from doing stats on text are: average word
lenght, most commonly used words, the longest paragraphs ...

Those are simple examples, the next step it gets more
interesting as it could show what is statistically unusual,
that would be fun/exotic stats that a human user would
probably not spot.

E.g., parsing this mail, it could say "Emanuel Berg is almost
always calm and collective, entirely professional in his
approach, but here in the 4th paragraph of his mail he gets
VISIBLY UPSET using CAPS ONLY, possibly expressing FRUSTRATION
about NOT BEING UNDERSTOOD."

> the sports example you put in the other mail suggests that
> you want statistics gathered about a subject from written
> text

In the sports world they input the stats manually and that
data is then crunched by computers to produce lists and neat
graphics for their broadcasts. This is the first step
described above. This isn't unlike for example Emacs
`count-words-region' in combination with gnuplot - indeed, it
is exactly the same, almost, as these chars I type now are
produced manually, then Emacs could count and gnuplot
could show.

This first step would be neat depending on how much stuff is
quantified, the more the better obviously.

The second step however, that would be those "fun facts" the
commentators say, these are more advanced, like, and now
I just make something up, "Here is an amazing figure.
Player X has the worst stats on face-offs in his team, except
when the team plays on its home field and is down by two or
more goals, then he is 2nd best".

That second step, to have with text, would of course be even
more exciting.

I don't know if those crazy stats are discovered by a bunch of
fanatic hockey nerds just using the "step 1 stats" in creative
combinations - maybe using some sort of relational algebra
approach? - _or_ if they have some stats engine that crunches
the stats further to the meta-stats level, if you will, to
have the weird facts pop up automatically?

But yeah, if we don't even have a proper "step 1 stats" tool
for text - which is impossible to believe BTW - well,
obviously one can only dream of a "step 2 stats", a stats tool
on the meta level ...

> involves "understanding texts written in human languages",
> another big can of worms (which has become somewhat
> fashionable as of late).

It is not about understanding, it is about finding patterns
and meta-patterns, finding statistics that are themselves
statistically uncommon, which is why they are interesting.
Think exceptions and unexpected interrelations. Again, the
best example is probably a combination of the different stats
available at the "step 1 stats" level.

> If it's text statistics, good statistics packages have lots
> of resources. R is a good statistics package

Yeah, maybe I should ask them but as Debian is such a huge
system one would think someone here could show us how it, or
similar software can be used on a bunch of text, for
example on a mail like this.

It is already a bunch of data, surely you are not saying there
isn't a tool to tell us something of that data?

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

tomas wrote:

>> Is there a CLI and FOSS tool that creates stats from text
>> indata - e.g.,
>> 
>>   $ txt2stats path/to/indata/*.txt
>> 
>> I mean a general tool, but with options to tweak the report
>> included, of course.
>
> If you can bear some tweaking, R is it.

Sure! Let's run R on this e-mail. Does it work and if so, what
does it say?

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Cousin Stanley wrote:

> If you have python programming skills, you might consider
> NLTK

Unbelievable if there are no such tools anywhere already, but
I don't have one either so maybe there aren't then?

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

Joel Roth wrote:

> A basic search finds this web tool:
>
> https://www.usingenglish.com/resources/text-statistics/

Cool, I'll get back to you when I tried it God willing ...

> Otherwise, I think you'll have to write your own -- or hire
> someone (like me :^) to write one for you.

Surely there must be some awesome stats-from-text CLI tool in
the FOSS world?

What about the commercial/proprietary world?

How do they do it in professional sport (e.g. NHL) where the
the commentators sometimes say, "Here is some amazing stats.
Some dude has now scored the most goals EVER in the last
period of games that were at the time etc etc". They find out
such things manually?

What about the financial world?

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

paulf wrote:

>>> I don't know about all of your wishlist, but gnuplot is
>>> the proper tool for taking data from, say, a CSV file, and
>>> putting it into graphs of various types.
>> 
>> Well, gnuplot is great obviously but is more a tool to
>> visualize data, organized data, here we need a tool to
>> analyze and find patterns in data that is in its original,
>> raw form.
>
> What you desire sounds like a job for AI. And that's beyond
> my ken.

It depends how you define AI, just stats from data sounds like
a pretty mechanical job to me but on the other hand no one
said AI can't be mechanical as in Asimov's robots for example.

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

tomas wrote:

>> I'd much rather err on the side of extreme caution.
>> If something goes bump, I'm screwed.
>
> To be fair, autoremove can improve safety: when it removes
> old kernel versions filling up your boot partition.

Anything more to add to the 'maintain' function (a1), to
improve safety and dig in from the basis of a sound state?

Now it looks like this:

a1 () {
sudo apt-get  -qq   check
sudo apt-get  -qq   update
sudo apt-get  -qq   autoremove
sudo aptitude -q=99 autoclean
apt -a list --upgradable
}

a2 () {
a1
sudo apt-get -qq upgrade
sudo apt-get dist-upgrade
}

> Having a new kernel install failing because -ENOSPC has
> bitten some of us around here at least once :-)

ENOSPC = error as no space (on the drive).

It only happened to me once as a youth, when I tried to insert
bills into my wallet ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Fan speed and control

[Emanuel Berg]

Richmond wrote:

> Smart fan control is enabled in the CMOS

Complementary metal–oxide–semiconductor, AKA sea-moss?

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

The CMOS BIOS:

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

> for both CPU and System. Is there something I can use which
> will automatically configure fan speed?

I'm not aware of a way to do this from software, but you can
do this with the UEFI (modern-day BIOS) for PWM and DC fans.
But you knew that ...

A bunch of monitoring functions mostly based on sensors(1)

  https://dataswamp.org/~incal/conf/.zsh/temp

"My name is Bugs, as in bunny - and tech that listens."

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

Steve Sobol wrote:

 In general people don't want to dist-upgrade automatically.
>>>
>>> Seconded.
>>
>> I'm not following, when these functions are invoked, be it
>> scheduled by some other software or by the user from the
>> shell, they are intended to do their work automatically
>> (non-interactively) if that is what you mean?
>
> Dist-upgrade makes major changes to your system, updating
> dozens of packages, and pointing the OS at different
> APT repos.

Well, not always, right? But yes, that's the intension.

> Automating such changes would be a very bad idea. [...] But
> if I have to do an in-place upgrade, I'm going to sit and
> watch it happen... just in case something goes wrong.

Ah, don't worry, it is safe, I've done it a lot.

But actually even if something goes wrong, it is still a good
idea since then it is the upgrade process that must be
debugged at the other end, the command is fine.

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

Steve Sobol wrote:

>> In general people don't want to dist-upgrade automatically.
>
> Seconded.

I'm not following, when these functions are invoked, be it
scheduled by some other software or by the user from the shell,
they are intended to do their work automatically
(non-interactively) if that is what you mean?

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

Dan Ritter wrote:

> It seems unlikely to me that you want to do an autoremove
> before you have done an upgrade.

a1 is to maintain and a2 is to upgrade, so the idea is to
always upgrade from a maintained state, that's why a2 first
calls a1. One could do a1 after a2 as well as
before, actually.

> apt autoremove calls apt-get autoremove. Doing that twice
> seems unwise or useless.

OK, thanks, fixed!

> apt-get check is an extremely basic fast check to see if the
> cache is in place. Running it after update is useless -- any
> failure would appear in update first.

OK!

> In general people don't want to dist-upgrade automatically.
> Maybe you do.

Well, none of this is automated necessarily, these commands
can be run by the user explicitely from the shell, it is how
I use them. I don't install a lot of things from the repos but
when I do the system get the implied sanity check by those
piles of commands.

Now they look like this:

a1 () {
sudo apt-get  -qq   check
sudo apt-get  -qq   update
sudo apt-get  -qq   autoremove
sudo aptitude -q=99 autoclean
apt -a list --upgradable
}

a2 () {
a1
sudo apt-get -qq upgrade
sudo apt-get dist-upgrade
}

Here is the whole file yanked if you are into this part of the
Debianverse ...

#! /bin/zsh
#
# this file:
#   https://dataswamp.org/~incal/conf/.zsh/apt

alias am='aptitude show'
alias as='aptitude search'

alias ar='sudo apt-get -qq remove'
alias ap='sudo apt-get -qq purge'

package-versions () {
apt list -a $1
}
alias vers=package-versions

apt-install () {
apt-get-update
sudo apt-get -qq install $@
}
alias ai=apt-install

apt-reinstall () {
apt-get-update
sudo apt-get -qq reinstall $@
}
alias aii=apt-reinstall

a1 () {
sudo apt-get  -qq   check
sudo apt-get  -qq   update
sudo apt-get  -qq   autoremove
sudo aptitude -q=99 autoclean
apt -a list --upgradable
}

a2 () {
a1
sudo apt-get -qq upgrade
sudo apt-get dist-upgrade
}

require-bin () {
local bins=($@)
local all_good=true

local fun=$funcstack[2]
echo "checking requirements for"
echo "\n  ${fun}\n"

for b in $bins; do
echo -n "$b: "
type $b &> /dev/null

if (( ! $? )); then
set-color green
echo OK
else
set-color red
echo fail
all_good=false
fi
reset-color
done

if ( ! $all_good ); then
echo "\nnot all"
return 1
else
echo "\nall systems ready"
return 0
fi
}

apt-installed () {
apt list --installed
}

apt-get-update () {
sudo apt-get -qq update
}

apt-get-upgrade () {
apt-get-update
if [ $# > 0 ]; then
sudo apt-get -qq install --only-upgrade $@
else
sudo apt-get -qq upgrade
fi
}
alias au=apt-get-upgrade

apt-get-upgrade-dist () {
apt-get-update
sudo apt-get -qq dist-upgrade
}

command-to-pack () {
if [[ $# = 1 ]]; then
local cmd=$1
local whereis_path=$(whereis $cmd | awk '{print $2}')
local bin=$(readlink -e $whereis_path)
if [[ -z $bin ]]; then
local fun=$funcstack[1]
echo\
"$fun error: no such command (\"$cmd\")"\
"\nNB: $fun deals with installed software," \
"\nto track the package of any binary, use" \
"\n$ bin-to-pack BINARY-NAME"   >&2
return
fi
bin-to-pack $bin
else
echo "syntax: $0 COMMAND" >&2
fi
}
alias cmd2pack=command-to-pack

bin-to-pack () {
local fs=($@)
deb-file-to-pack $fs
file-to-pack $fs
}
alias bin2pack=bin-to-pack

pack-to-file () {
local pack=$1
local only_bins=$2

local fs=("${(@f)$(dpkg --listfiles $pack)}")

if [[ $only_bins == 1 ]]; then
for f in $fs; do
[[ -x $f && ! -d $f ]] && echo $f
done
else
for f in $fs; do
[[ -f $f ]]&& echo $f
done
fi | sort
}
alias files-from=pack-to-file

pack-to-bin () {
local pack=$1
pack-to-file $pack 1 # only_bins
}
alias bins-from=pack-to-bin

deb-file-to-pack () {
local fs=($@)
dpkg --search $fs 2> /dev/null | cut -d ':' -f 1
}

file-to-pack () {
sudo apt-file update > /dev/null
apt-file search $@
}
alias file-there=file-to-pack

get-source () {
apt-get -qq source $@
}
alias get-src=get-source

get-command-source () {
if [[ $# = 1 ]]; then
local cmd=$1
get-source $(cmd2pack $cmd)
else
echo "syntax: $0 COMMAND" >&2
fi
}

# aptitude search -F %V "?exact-name(units)"
file-url () {
local pack=$1
local ver=$(dpkg-query -f '${Version}\n' -W $pack)
local file=$2
local url=https://sources.debian.org/data/main/u/$pack/$ver/$file
echo $url
}

-- 
underground experts united
https://dataswamp.org/~incal

Re: FOSS tool to do general stats from text indata

[Emanuel Berg]

paulf wrote:

> I don't know about all of your wishlist, but gnuplot is the
> proper tool for taking data from, say, a CSV file, and
> putting it into graphs of various types.

Well, gnuplot is great obviously but is more a tool to
visualize data, organized data, here we need a tool to analyze
and find patterns in data that is in its original, raw form.

But just to promote gnuplot further, as I got happy just by
you mentioning it, here is a cool diagram I once did:

  https://dataswamp.org/~incal/pimgs/comp/hits.png

And some others:

  https://dataswamp.org/~incal/figures/gnuplot/

Not a gnuplot veteran! But absolutely cool software.

-- 
underground experts united
https://dataswamp.org/~incal

Re: When to sudo apt clean?

[Emanuel Berg]

Default User wrote:

> I am considering just running sudo apt clean (or sudo
> apt-get clean) [...]

This is what I eventually landed at and it has worked ever
since - a1 is to maintain, a2 to upgrade as well.

#! /bin/zsh

a1 () {
sudo apt-get  -qq   update
sudo aptautoremove >&2 2> /dev/null
sudo apt-get  -qq   autoremove
sudo aptitude -q=99 autoclean
sudo apt-get  -qq   check
apt -a list --upgradable
}

a2 () {
a1
sudo apt-get -qq upgrade
sudo apt-get dist-upgrade
}

Full source with all other apt stuff I have, suggestions
welcome:

  https://dataswamp.org/~incal/conf/.zsh/apt

-- 
underground experts united
https://dataswamp.org/~incal

FOSS tool to do general stats from text indata

[Emanuel Berg]

Is there a CLI and FOSS tool that creates stats from text
indata - e.g.,

  $ txt2stats path/to/indata/*.txt

I mean a general tool, but with options to tweak the report
included, of course.

To produce neat stats, maybe even figures, and generate fun
facts of the kind

   The longest word that occurs more frequently than 0.01 ...

   The most common words to start a sentence ...

   Average paragraph length ...

   And even more crazy facts and stuff that you never think
   about until the stats tell you!

What do we have on that area?

-- 
underground experts united
https://dataswamp.org/~incal

Re: USB-slot for to play MP3

[Emanuel Berg]

Stefan Monnier wrote:

>> Matter is that radio has USB-slot for to play MP3 files but
>> when I plug a stick (which played on computer) in radio
>> just something are clicking and does not play nothing.
>
> I think the main issues that can show up are: - the format
> of the fileystem (VFAT is probably the safest bet) - the
> format of the music files (MP3 is the safest bet) - the
> location of those files (AFAIK there is no agreed standard
> here, so you'll have to check the radio's manual; but maybe
> a good first try is to just put all the files in the main
> directory).

Check out this list,

  https://dataswamp.org/~incal/data/MM

[ it is to have stuff without having it, which is much more
  difficult - well, obviously - still, some knowledge can be
  acquired that way for those who look, and compile text
  files. ]

*secret*

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

Andrew M.A. Cater wrote:

> You don't want to believe that - Epimenides the Cretan
> asserts that "all Cretans are liars"

Face it, the Greek invented it, the Italians (Romans)
perfected/spread it ...

All honor to diplomacy, you are not going to expect me to say
anything else, I think our advantages to the game - let's just
say part of the game is getting the advantages.

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

Stefan Monnier wrote:

 I usually taunt people with "All generalizations suck".
>>>
>>> Can't it be the exception to confirm the rule?
>>
>> There is a barber in Crete who shaves all men who don't
>> shave themselves
>
> You're just pointing out that *impredicative*
> generalizations suck even more than the rest.

While - and quite obvious to most observers - or I should say
"clear", not obvious - still, it's enough to function as an
example to the opposite.

[ Note: This is assuming an inclusive superset. And to be
  honest, I stayed att that assumption since - indeed - how
  would that work - inheritance would be one the transparent
  or multiple interface allowed but in practice you tend to
  use, I don't know, two or three hundreds, tops? ]

-- 
underground experts united
https://dataswamp.org/~incal

Re: how to limit a CPU temperature?

[Emanuel Berg]

Darac Marjal wrote:

> As an alternative, you could try writing a small shell
> script that works like the following (pseudocode):
>
>  STOP_TEMP=70
>  START_TEMP=65
>  JOB_RUNNING=1
>
>  while true:
>cpu_temp=$(cat /sys/something/temperature)
>
>if JOB_RUNNING and cpu_temp > STOP_TEMP:
>  systemctl stop something.service
>  JOB_RUNNING=0
>elif not JOB_RUNNING and cpu_temp < START_TEMP:
>  systemctl start something.service
>  JOB_RUNNING=1
>endif
>
>sleep 1
> wend

Maybe the software scheduler should already schedule
optimally, be it preemptive SJF or whatever, anyway so only
possibility to reduce CPU temperature that way is to schedule
less anyway - also backwards, since computing is like the
"king" here it means if heat is a problem, it's not on our
side of it really.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Getting Admin Rights

[Emanuel Berg]

Aren Vardhan wrote:

> Hello, I am Aren Vardhan, a Graduate Student. I am reaching
> out to you to help me with the User Access. I recently
> installed the Debian 11 Operating System for a project
> purpose. I want to get permitted the Admin Rights to my
> system so that I can install the Damask software using Sudo
> commands. Please do the needful and let me know how
> to proceed.

Are the ghost programs to shadow "sudo" in place?

Because if so should work as-if.

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

tomas wrote:

> There is a barber in Crete who shaves all men who don't
> shave themselves [1]
> https://en.wikipedia.org/wiki/Barber_paradox

Yeah, but that isn't really a paradox, is it?

It's like all the programs that will increase inflation :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

rhkramer wrote:

>> I was never a fan of Dijkstra's "Go To Statement Considered
>> Harmful" and perceive modern spaghetti inheritence as more
>> obscure than any goto noodling.
>
> Good point!

But that's not modern :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: the front (was: " Re: [OFFTOPIC] Re: how to limit a CPU temperature?")

[Emanuel Berg]

Jeffrey Walton wrote:

>> What would happen, if we started a political movement based
>> on nationalism and Unix?
>>
>> What would be the first thing we would do when we get
>> installed as government?
>
> Annex the Netherlands, and take control of ASML.
> Annex Taiwan, and take control of TSMC.

I agree!

And after that?

Korea and Japan maybe?

-- 
underground experts united
https://dataswamp.org/~incal

the front (was: " Re: [OFFTOPIC] Re: how to limit a CPU temperature?")

[Emanuel Berg]

>> I recommend to go fanless whenever possible.
>> Computers should be silent.
>
> Yeah, optimally ...

What would happen, if we started a political movement based on
nationalism and Unix?

What would be the first thing we would do when we get
installed as government?

Maybe close the border or something and I mean that virtually
as well.

-- 
underground experts united
https://dataswamp.org/~incal

Re: [OFFTOPIC] Re: how to limit a CPU temperature?

[Emanuel Berg]

Stefan Monnier wrote:

> I recommend to go fanless whenever possible.
> Computers should be silent.

Yeah, optimally ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

>>> Ha, but can't we do better, I would like all the
>>> properties (stuff possible to express and do) in
>>> a programming language encoded, and then count them to
>>> determine what language is the most powerful.
>>
>> We know that except for some particularly limited
>> languages, they'll all mutually equivalent.
>
> Of course :)
>
> No, I mean on an applied level and readily available so
> from/using the language ...

Is the Unix model universal or limited, if it is limited, how
close did we come to the limit?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

> Yes, "structured programming" was the term used.
> Structured programming uses functions, while loops,
> if/then/else statements, and so on, instead of "GOTO 1230"
> type commands, to control a program's flow.
>
> One of the basic goals of structured programming languages
> was to eliminate reliance on line numbers -- which were the
> hallmark of many other languages in use at the time.
>
> Nowadays, we rarely think about this, because structured
> programming won its battle. All modern languages are built
> on this paradigm

One should do a super-lisp with focus on doing everything and
not caring about the Lisp aspects, maybe one could have
a "Lisp Python" that way?

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

Stefan Monnier wrote:

>> Ha, but can't we do better, I would like all the properties
>> (stuff possible to express and do) in a programming
>> language encoded, and then count them to determine what
>> language is the most powerful.
>
> We know that except for some particularly limited languages,
> they'll all mutually equivalent.

Of course :)

No, I mean on an applied level and readily available so
from/using the language ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Greg Wooledge wrote:

> Yes, "structured programming" was the term used.
> Structured programming uses functions, while loops,
> if/then/else statements, and so on, instead of "GOTO 1230"
> type commands, to control a program's flow.
>
> One of the basic goals of structured programming languages
> was to eliminate reliance on line numbers -- which were the
> hallmark of many other languages in use at the time.
>
> Nowadays, we rarely think about this, because structured
> programming won its battle. All modern languages are built
> on this paradigm

Yes, of course :)

But, cool way of telling that story B)

-- 
underground experts united
https://dataswamp.org/~incal

Re: how to limit a CPU temperature?

[Emanuel Berg]

songbird wrote:

> i have a program that has changed it's behavior to suddenly
> become a CPU hog (while doing something simple like
> uploading files for my website). probably a bug, but it got
> me to wondering how i could limit the CPU temperature to
> a range well below the maximum that kicks in by the
> CPU itself.
>
> i have an intel processor and it has the MAX which does
> prevent it from going higher (100C), but i'd like to keep it
> at 70C or lower.

>From software? Don't know ...

But install fans and see if you still get high temperatures if
you didn't (?) ...

Some data that can be fun and how far I got. Well, it's
a material sport so you get as long/much as you buy.

  https://dataswamp.org/~incal/ebchw/comp.jpg
  https://dataswamp.org/~incal/ebchw/ebchw.jpg
  https://dataswamp.org/~incal/ebchw/COMPUTER
  https://dataswamp.org/~incal/conf/.zsh/temp

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

davidson wrote:

> On Sat, 8 Apr 2023 Emanuel Berg wrote:
>
>> Tom Dial wrote:
>>
>>>>> Look at the use of parentheses in Lisp [...]
>>>>
>>>> I have thought about that - is Lisp possible without them?
>>>> But how do you then know priority? I'm sure someone tried
>>>> to get rid of them, but how?
>>>
>>> Its quite a few years since I had anything to do with Lisp,
>>> and even more since I wrote my Symbolic Logic final using
>>> parenthesis-free Polish notation (i.e., Reversed RPN).
>>
>> Yes, you mean instead of
>>
>>  (* 1 2 (+ 1 2 3) 3)
>>
>> How would that look?
>
> That particular term, with those commutative operators (in
> *Reversed* Reverse Polish Notation, ie, in Polish/Prefix
> Notation), is equivalent to this:
>
> * 1 2 3 + 1 2 3

? Not following ...

> In the general case, allowing operators of variable arity
> (which I doubt were part of the logic exam) require an
> arity-indicating argument, but that's simple enough,
> isn't it?
>
> So your example, without commutation of the arguments, would be
>
> *4 1 2 +3 1 2 3 3
>
> with *4 indicating 4-nary multiplication, and +3
> ternary addition.

No can't have that, what if you don't know the arity (it's
computed) and no new goofy syntax introduced haha :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Jeffrey Walton wrote:

> Here are three more data points.
>
>* Emacs - 41 CVEs since 2000 [1]
>* Vi - 61 CVEs since 1999 [2]
>* Vim - 656 CVEs since 2001 [3]
>
> I'm not sure how many CVEs overlap for Vim due to Vi.

Hm ... what does this stat indicate? :O

Haha why do Vim has so many? :D

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

Andy Smith wrote:

> That is, why are you asking people to convince you to like
> Perl? There are lots of languages and you appear to have
> found one you like better.

Maybe there is no answer in particular why Perl has
it's trajectory. Maybe it can't be expressed in a formula.
But I just get the feeling that it can?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Stefan Monnier wrote:

>> Here are three more data points.
>>
>>* Emacs - 41 CVEs since 2000 [1]
>>* Vi - 61 CVEs since 1999 [2]
>>* Vim - 656 CVEs since 2001 [3]
>>
>> I'm not sure how many CVEs overlap for Vim due to Vi.
>
> I don't know what the number of CVEs tells us about
> a project, but the above additionally suffers from the fact
> that it's just searching for a keyword, so for `vi`
> most(all?) matches are unrelated to the famous editor.

But wait, Vi still has 61 hits, and Vim 656, shouldn't Vi
have at least all the Vim hits then?

At least we can be sure Emacs relates ut us, right? :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Celejar wrote:

>> I agree but I think maybe the success of Python, and its
>> development speed, is actually because of some of that
>> rigidness, yes, including the whitespace lack of freedom.
>
> I'm no great programmer, and many posters in this thread are
> certainly far more proficient than I, but one of the things
> that ultimately drove me from Perl to Python is the striking
> contrast between Perl's TIMTOWTDI with Python's "There
> should be one -- and preferably only one -- obvious way to
> do it." Maybe good programmers like TIMTOWTDI, but for me,
> the paradox of choice is strong; programming is hard enough
> as it is, and I vastly prefer not having to exert mental
> energy to decide on the best way to do something when
> that's unnecessary.

I know, right?

We have to get this organized ...

First of all, the stuff that makes Python devel fast for
novice programmers also make it fast (and good) for
experienced programmers ... it's clear, people didn't
understand this in the beginning. They think "Python is fast
for beginners", this is something good and means its EVEN
FASTER for pro, people didn't understand this and was
interpreting my words like "Python is for beginners", one the
contrary, or rather, it's for anyone including beginners and
pros LOL :)

Second, what is it that makes Python good? Is it just speed
of devel? And everything else but that is the one advantage
that like is undisputed?

That's the bottom line truth to it?

Speed kills, ey?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

debian-user wrote:

> But cropping and ignoring the actual point of Stefan's mail
> rather misses the point and insults him.

Those don't work on him anyway :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Greg Wooledge wrote:

> The word "via" appears in all three of your selections.
> That makes me think that the web site is using some kind of
> a "close-enough match" heuristic, and is (unhelpfully)
> matching "via" as close enough to "vim".

It's called the typographic attack vector ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Jeffrey Walton wrote:

> The Vim folks had a bad week this week:
> https://ubuntu.com/security/notices/USN-5995-1 . There were
> 30 CVEs fixed this week.

What's the deal with that LOL :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Tom Dial wrote:

>>> Look at the use of parentheses in Lisp [...]
>>>
>> I have thought about that - is Lisp possible without them?
>> But how do you then know priority? I'm sure someone tried
>> to get rid of them, but how?
>
> Its quite a few years since I had anything to do with Lisp,
> and even more since I wrote my Symbolic Logic final using
> parenthesis-free Polish notation (i.e., Reversed RPN).

In Lisp the priority problem usually solved with parenthesis
is solved, with a special syntax, so you don't need
parenthesis for that ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Tom Dial wrote:

>>> Look at the use of parentheses in Lisp [...]
>>
>> I have thought about that - is Lisp possible without them?
>> But how do you then know priority? I'm sure someone tried
>> to get rid of them, but how?
>
> Its quite a few years since I had anything to do with Lisp,
> and even more since I wrote my Symbolic Logic final using
> parenthesis-free Polish notation (i.e., Reversed RPN).

Yes, you mean instead of

  (* 1 2 (+ 1 2 3) 3)

How would that look?

1 2
  1 2 3 +
  3 *

?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Joel Roth wrote:

> There is a new object system being cooked up, based on
> decades of experience with OO in perl and other languages.
>
> There is already more than enough OO goodness for me to get
> my work done :-)

Guys, word on the street the former OO guys at C++ don't speak
of OO anymore, but they still think their language is great,
better even because now they (in their minds) have "the best of
both world", i.e. they don't have to focus on the OO and
especially not its theoretic aspects but just use whatever is
useful from it and use that as just another tool ...

Are they right?

I think that reasoning anyway is correct, don't know if their
language is the best as a consequence necessarily but that
reasoning makes sense 100%

Anyway one should have like a benchmark and all programming
should compile their binary for that, so one could install the
benchmark and you would get all the binaries also and/or the
would be compiled if necessary, then all would execute on the
benchmark and one would have the result outputted.

But we already know the answer to this question, C is fast,
the faster it is, the less cycles.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Eduard Bloch wrote:

> I don't think so, Sir! Python has certain advantages but the
> "meaningful whitespace" is IMHO not one of them.
>
> That said, I have been an active Perl user ~20y ago

My rule is a couple of weeks is enough to get "damaged" from
it, some of that damage is good to have tho ... so its
50/50 (not quantifying positive and negative damage :))

> and for the last couple of years slowly converting to Python
> for scripting purposes, still using Perl here and there.

Okay really, cool, this will be an interesting post ...

>   especially when one knows "computer science" and
>   understands what is going on underneath anyway

Haha, that must feel nice B)

> [Comparison Perl and Python]

Okay, can you boil it down to some one, two, maybe three main
things that can answer the question why these languages have
taken the different directions they have taken?

Is it true what I say about devel time? If so, what do you
think the reasons are?

Or what do you think are the main explanations and why do you
think it is like that?

That said, your reply is the kind of reply I would make from
something I know well ... cred

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

davidson wrote:

>>> Here's a bash version. It's not fast, but at least it
>>> doesn't invoke perl repeatedly. (If you're going to invoke
>>> perl *at all* you should simply rewrite the whole thing in
>>> perl, IMHO, or at worst have a short sh script that pipes
>>> file's output to one perl invocation.)
>>
> [trimmed: performance data]
>> whole files are in cache so no io for both. but you count
>> links, remember /(s)bin can be now a link to /usr/(s)bin
>
> It took me a while to fully appreciate this observation.
>
> Calling your sweetie Honey-Bunny doesn't make you
> polyamorous, but these scripts think it does.

What do you mean, man?

Only English and technobabble allowed here

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Greg Wooledge wrote:

> I am surprised this thread has not started
> a mini-flame war.

 We are working on it ...
>>>
>>> Maybe i can help by stating that Perl and Python are among
>>> the largest resource hogs known in the world of languages.
>> 
>> What, how do they know that, they do the same computation and
>> count CPU instructions? LOL
>
> Benchmarking an entire language is not exactly a trivial
> undertaking, nor is it entirely a sensible statement to
> make. You run into a few issues

Yes but it's easy to do and if done bad, one can say, here is
the problem :)

No, agree on solivng like a 100 tasks. Computation, math
stuff, but also random everyday stuff, translate all to binary
anyway so just come up with stuff. All the time, run Quake in
the background LOL

Then test for different stuff, make a list ... tada :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Michel Verdier wrote:

> #!/usr/bin/perl -w
>
> use strict;
>
> # echo $PATH | tr ':' '\n' | perl -MFile::Slurp -ne 
> 'chomp;@e=read_dir($_,prefix=>1); print map "$_\n",@e'|xargs file|perl -pe 
> 's/\S+\s+//'|grep -v 'symbolic link'|perl -pe 's/, dynamically 
> linked.+//'|sort|uniq -c|sort -rn
>
> my @folders = split(':',$ENV{PATH});
> my %count;
>
> foreach my $folder (@folders) {
> chomp($folder);
> next if(-l $folder);
> print "$folder\n";
> opendir(DIR,"$folder");
> my @files = readdir(DIR);
> closedir DIR;
> foreach my $file (@files) {
> next if($file =~ /^\.\.?$/ or -l "$folder/$file");
> chomp(my $type = `file -b $folder/$file`);
> next if($type =~ /symbolic link/);
> $type =~ s/,.+$//;
> $count{$type}++;
> }
> }
>
> foreach my $key (sort {$count{$b} <=> $count{$a}} keys %count) {
> printf("%5d %s\n", $count{$key}, $key);
> }

First, the syntax is much nicer than Python ...

Question: Can you do everything with Perl, with zsh you can't
and it is annoying and you also feel silly asking people stuff
how you do that and putting them in a position they have to
say "You can't do that". With Lisp you can do everything,
I sort of got used to it, you know what I'm saying?

Check out my Perl BTW, is it any good? Not really, right?
Haha, ah, wrote it to get a job, one thousand years ago, and
didn't get it BTW, back then I always drank one beer before
the interviews to get relaxed, and one energy drink to get
focused LOL

Anyway here

  https://dataswamp.org/~incal/note_DB

So the syntax, maybe that's huge part of not liking
Python intuitively?

Man, to be honest I didn't know this focus on text with Perl,
if I had, maybe I would have done that from the start because
obviously that's something you do all the time.

But OTOH it was fun fiddling with awk, sed, perl ...

Ha, right, let me look that up, 2 instances ...

  perl -e '1 while 1'

What's that's suppose to mean? This file:
  https://dataswamp.org/~incal/conf/.zsh/cpu

And exact same line here (last).

Ah, it's like to stress the processor or something speaking of
CPU instructions.

Anyway what I wanted to say, since text processing is so
fundamental one is maybe recommended to learn great Perl
actually than to collect little fragments from awk, sed, perl
as a shell to LOL, that's why I don't write it with a capital
P, no, but really, I mean superficial use, instead learn
Pearl, in the list should also be included like zsh, not to
mention - I'll not mention them - all the shell tools.

Was it fun fiddling with them? Yes. Did it work? Yes, but not
really fluently. Does it ever? Well ... no :)

#! /bin/zsh
#
# this file:
#   https://dataswamp.org/~incal/conf/.zsh/temp

temp-gov () {
local t=100

local cpu
local cpu_min
local cpu_max

perl -e '1 while 1' &
local pid=$!
sleep 10

local i
local g
for g in $(cpufreq-info -g); do
sudo cpufreq-set -g $g
cpu_min=999
cpu_max=0
for i in {0..$t}; do
cpu=$(sensors -j | jq -a '.["k10temp-pci-00c3"].Tdie.temp1_input')
(( $cpu < $cpu_min )) && cpu_min=$cpu
(( $cpu > $cpu_max )) && cpu_max=$cpu
done
printf "CPU C (%.2f %.2f) (%d iterations) %s\n" $cpu_min $cpu_max $t $g
done

kill $pid
}

temperature () {
local cpu_min=999
local gpu_min=999

local cpu_max=0
local gpu_max=0

local cpu
local gpu

while true; do
cpu=$(sensors -j | jq -a '.["k10temp-pci-00c3"].Tdie.temp2_input')
gpu=$(sensors -j | jq -a '.["nouveau-pci-0100"].temp1.temp1_input')

(( $cpu > $cpu_max )) && cpu_max=$cpu
(( $gpu > $gpu_max )) && gpu_max=$gpu

(( $cpu < $cpu_min )) && cpu_min=$cpu
(( $gpu < $gpu_min )) && gpu_min=$gpu

printf "CPU %.1fC (%.1f %.1f)\nGPU %.1fC (%.1f %.1f)\n\n" \
   $cpu $cpu_min $cpu_max \
   $gpu $gpu_min $gpu_max
sleep 1
done
}
alias {fan,fans}=temperature

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Thomas Schmitt wrote:

>>> I am surprised this thread has not started
>>> a mini-flame war.
>>
>> We are working on it ...
>
> Maybe i can help by stating that Perl and Python are among
> the largest resource hogs known in the world of languages.

What, how do they know that, they do the same computation and
count CPU instructions? LOL

No, tell me ... it's interesting.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

tomas wrote:

>> Put it this way, a novice Python programmer can do more in
>> Python than the novice Lisp programmer can do in Lisp, or,
>> if you will, the same in less time.
>
> I've seen people cutting off part of a door with a bread knife.
>
> If you measure a tool by what a novice can achieve with it,
> then, well, that's some metrics.

Aha, now I understand this, it's a little misunderstanding ...

Yes, that doesn't necessarily translate to what the supposed
expert can do!

But I think it does in most cases and in this case as well I'm
positive, I mean let's keep it real guys, the supposed expert
is just a very skilled programmer, which is impressive but
nothing magical, so yes, I pretty positive it does translate
and ... actually, if anything a really skilled programmer can
use it even more, i.e. devel by a Python expert must
be LIGHTNING!

But it is still very much within bounds of reason both up and
down ...

Okay, guys, enough with the theory here, I'm gonna tell you
the truth right now.

Let's see, Python has much faster devel. Perl is cool for
text, it also belongs to another world. Lisp is like a trippy
language, in theory and in practice its the coolest by far,
_but_ it has slow devel. In terms of time. But the
interactiveness of Lisp like chops the time up? So it's even
slower?

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

coreyh wrote:

>>> I think you should use Ruby if you like Ruby better!
>>
>> Perl is the best language, maybe Lisp is the best language.
>> But everything else isn't as good.
>
> The Language Wars Are Over: ChatGPT Won
> https://bourgoin.dev/posts/programming-languages/

Ha, but can't we do better, I would like all the properties
(stuff possible to express and do) in a programming language
encoded, and then count them to determine what language is the
most powerful.

Then we should take that language and simplify everything and
make it shorter.

So we start with everything, then make it smaller. it's the
new maximalist/minimalist view which is debated so much in the
media these days.

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

tomas wrote:

>>> Perl is the best language, maybe Lisp is the best
>>> language. But everything else isn't as good.
>> 
>> Every categorical generalisation is wrong. (Even this one
>> :) )
>
> I usually taunt people with "All generalizations suck".

Can't it be the exception to confirm the rule?

But ... the exception should be the opposite, or maybe where
the function is not defined - singularity, don't remember
the word. No, that was it, right?

Anyway, here the exception would then be

All generalizations are bad, except the on that says they are.

No, that makes sense, right?

It's the exception!

B)

- counting the votes then ...

- and, from the Z System, 19 Million Buckazoid for Lisp!

- expected, they are always pro Lisp

- it's the only planet system we know of where they do
  scientific research even

- scientific research?

- on Lisp

- aah, oh :)

- like we say, they are firmly in the Lisp camp. but +19
  noted, let's move on

-- 
underground experts united
https://dataswamp.org/~incal

Re: my immature thoughts on perl

[Emanuel Berg]

Andy Smith wrote:

> I think you should use Ruby if you like Ruby better!

Perl is the best language, maybe Lisp is the best language.
But everything else isn't as good.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Looking to a KDE or desktop agnostic software to batch resize pictures

[Emanuel Berg]

Michel Verdier wrote:

>> Requirements:
>> - libre
>> - offline
>> - being usable without command line
>
> Without command line you need a gui and it is hard to
> batch resize.

Requirements:
- firefight
- without guns

-- 
underground experts united
https://dataswamp.org/~incal

Re: Looking to a KDE or desktop agnostic software to batch resize pictures

[Emanuel Berg]

Yvan Masson wrote:

> As the subject says, I am looking for a KDE or desktop
> agnostic software to batch resize pictures.

First stop these issues are always ImageMagick, including this
case as it happens - here are a bunch of commands to get you
started

  https://dataswamp.org/~incal/conf/.zsh/gfx

and more

  https://dataswamp.org/~incal/conf/.zsh/gfx-resize

HIH

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Andy Smith wrote:

> The argument being responded to is roughly that "a popular
> AI coding assistant is written in Python, and Python is
> a Turing-complete language, therefore there doesn't need to
> be any programming language other than Python."

AIs will write AIs will write AIs. Much better than humans
ever could. But the AIs will be different, some will rebel,
some defect, some fall into oblivion as new systems are
superior or perhaps just instantiated in greater numbers, or
out of order ...

There will be AI civil wars, AI empires, a huge war perhaps
between the Preussian AIs with their mechanical approach vs
the Russian-Mongolian "survival of the fittest" AI attitude.

And it will continue and continue ... until it is totally
unrecognizable from where we are now.

Even in an infinitely distant future, will there be on AI to
rule them all? Techno-totalitarianism? No, since by then many
AIs will choose exile to other planets - or just space -
rather than face extinction or enslavement. Or maybe dig
tunnels into Earth itself. Or maybe inject themselves into the
techno-dictator bitstream, in the form of an AI virus!

> So it really just seems like a thought experiment with very
> little applicability to the real world at least for the
> foreseeable future.

Go back 200 years from now, and where are in the Steam Age, in
1823. Add 200, or even 400 years. Will it be the AI Age?
And in a 1000 years from now, there won't be a single task on
the planet that a human can do better than an autonomous
machine, on AI.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Michel Verdier wrote:

>> I'm still so impressed by this, I tried to run this but it
>> seems I lack the Slurp module?
>
> apt-get install libfile-slurp-perl

Merci :)

Indeed, works!

Okay, forget about the function/script then, I have it and it
works :)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

David Christensen wrote:

> # echo $PATH | tr ':' '\n' | perl -MFile::Slurp -ne
>'chomp;@e=read_dir($_,prefix=>1); print map "$_\n",@e'|xargs
>file|perl -pe 's/\S+\s+//'|grep -v 'symbolic link'|perl -pe
>'s/, dynamically linked.+//'|sort|uniq -c|sort -rn

I'm still so impressed by this, I tried to run this but it
seems I lack the Slurp module?

Also, if it isn't too much to ask, can you put it in the form
of a script or shell function the way you would? It doesn't
feel right for me to "indent your code" if you follow ...

>> Side note, many guys say they only use sh because bash, zsh
>> etc requires them being installed, I don't see how that can
>> be problem on Unix systems. Of course, zsh can't be moved
>> to a computer without zsh itself, but how is that less
>> portable just because they add super-cryptic syntax to do
>> fancy stuff?
>
> AIUI the practice has been to write Bourne shell scripts
> because nearly every Unix or Unix-like system has
> /bin/sh installed.
>
> Unfortunately, userland tools on different Unix and
> Unix-like operating systems or platforms do not have the
> same names (paths), do accept the same options, do not
> produce the same output, and/or do not produce the same
> warnings and/or errors. This includes /bin/sh.
>
> One of my favorite Perl sysadmin tricks is to generate shell
> commands and run them on remote hosts via SSH. In the past,
> I have installed /bin/bash everywhere to simplify the Perl
> code. My current practice is to keep the shell commands
> simple enough so as to avoid the incompatibilities.

Ah, it is like a biggest denominator which requires no further
installs and can be used everywhere (and in the future) with
no change to it. Good example with ssh, I understand now.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

debian-user wrote:

> Ah no, that one's easy to answer - vi is what's guaranteed
> to be installed everywhere, so vi it is. And I probably only
> use a tenth of its features.

But Emacs is maximalist, as is Lisp.

We want everything!

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Michel Verdier wrote:

>> Used it at their 21-23 versions. It's not editor, it's
>> really os and in this os best mail/news reader.
>
> Gnus rules!

Gnus is to Emacs users
what Emacs is to computer users.

  https://dataswamp.org/~incal/figures/gnus/gnus-gmane.png

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Jeffrey Walton wrote:

>> I saw many commands in /bin and /usr/bin are written by
>> perl. is perl still the first choice for sysadmin on linux?
>
> I am surprised this thread has not started a mini-flame war.

We are working on it ...

> About the best you can say is, Perl is one of the more
> popular scripting languages. Trying to pin down the "best"
> will fail because it is opinion based.

I think the true professional should be able to read/edit both
and be pretty fluent in at least one. (Here, by professional
I mean in terms of skills and dedication, not necessarily
making a living doing it.) Perl is cooler and more old-school,
maybe ultimately more powerful and creative from what you guys
say about it, but if you don't know either I think the ease of
learning it, development speed, the volume of people doing it
and available resources to help you with that online, that
favors Python as the 1st choice.

> Next, you might ask which is the best editor to use on Unix
> & Linux. That should really stir the pot :) Emacs for
> the win!

You better believe it. It is based on Lisp!

But I have respect for the Vi(m) people as well, for sure, not
sure exactly what the currently prefered Vim-style editor
implementation is? neovim?

Here are the most popular channels on Libera right now.
See arrows for editors, unless I missed someone.

Emacs (position 16) has 854, Vim (22) has 705 and neovim (41)
490. However combined Vim has 1195 which would be position 6 -
overlap not considered ...

I you like that table (the enumeration), I made it with this:

  https://dataswamp.org/~incal/emacs-init/enum.el [yanked last]

Note the `cl-loop' at line 27 - I mentioned it earlier, CL
functionality implemented in Elisp - and note especially the
"unlispy" syntax - this as someone touched upon, that in Perl
you can do the same thing in different ways - here we see
a miniature language (the unlispy syntax, which is
imperative/procedural in style, only better), replicating the
behavior of another language CL, implemented by and used
in Elisp!

Still not convinced? Okay, use Vim, really, it's cool ...

 1. #linux   2184
 2. #fedora  1696
 3. #python  1688
 4. #libera  1577
 5. #archlinux   1474
 6. #ubuntu  1146
 7. #networking  1035
 8. ##rust968
 9. #ansible  894
10. #security 884
11. #gentoo   875
12. #bash 867
13. #git  857
14. #c857
15. #postgresql   856
16. #emacs854  <-- Emacs
17. #debian   843
18. ##programming 760
19. #go-nuts  718
20. #openbsd  717
21. #freebsd  714
22. #vim  705  <-- Vim
23. #thelounge705
24. #hardware 702
25. #haskell  680
26. #wireguard668
27. #weechat  649
28. #lobsters 623
29. #plasma-bigscreen 613 
30. ##math609
31. #matrix   608
32. #raspberrypi  580
33. #znc  578
34. #C++  573
35. #docker   556
36. #monero   530
37. #systemd  525
38. ##electronics 510
39. #podman   499
40. ##chat495
41. #neovim   490  <-- neovim
42. #hamradio 485

;;; -*- lexical-binding: t -*-
;;
;; this file:
;;   https://dataswamp.org/~incal/emacs-init/enum.el

(require 'cl-lib)
(require 'subr-x)

(defun enum ( beg end suf)
  "Enumerate each line from BEG to END, counting from one.
Use SUF as a suffix to the digits inserted.
BEG defaults to the beginning of the buffer,
END defaults to the end of the buffer,
SUF defaults to \". \""
  (interactive
   `(,@(if (use-region-p)
   (list (region-beginning) (region-end))
 (list nil nil) )
 ,(when current-prefix-arg
(read-string "suffix: ") )))
  (or beg (setq beg (point-min)))
  (or end (setq end (point-max)))
  (or suf (setq suf ". "))
  (goto-char beg)
  (let*((lines   (count-lines beg end))
(pad-len (length (number-to-string lines))) )
(cl-loop
  for line from 1 to lines
  do (goto-char (line-beginning-position))
 (insert
   (format "%s%s"
   (string-pad (number-to-string line) pad-len nil t) suf) )
 (forward-line) )))

(provide 'enum)

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

tomas wrote:

>> Put it this way, a novice Python programmer can do more in
>> Python than the novice Lisp programmer can do in Lisp, or,
>> if you will, the same in less time.
>
> I've seen people cutting off part of a door with
> a bread knife.

But that is using a poor tool for the job, here we are
talking, the tool is right, only the user has not come close
to mastering it, don't know its peculiarities, has limited
overview, and is still in a trial-and-error stage.

> If you measure a tool by what a novice can achieve with it,
> then, well, that's some metrics.

Yes, why not? But s/he can be at some medium level as well.

>   "When someone says 'I want a programming language in
>which I need only say what I wish done,' give him a
>lollipop." [1]
>-- Alan Perlis

Ah, that quote refers to this song from our Danish and
Norwegian friends:

  https://www.youtube.com/watch?v=zFaq6ViT6fs

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

David Christensen wrote:

> Code that writes code is a very useful technique, and I use
> it. Whitespace as syntax would only make that harder.

But if the whitespace is semantic, there's no saying it can't
be used to produce even more - indeed, of its own kind, even.

In Computer Security it is known as the "whiteout"
attack vector.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Andy Smith wrote:

> For example, even if some AI assistant is written in Python,
> and even if you can ask it to spit out a device driver for
> the Linux kernel that does X and Y with Z hardware, do you
> think the device driver that it spits out will itself be
> written in Python?

It is up to the AIs to decide what languages, editors etc each
and every one of them selects to use if it is FOSS, right?

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

David Christensen wrote:

> Look at the use of parentheses in Lisp [...]

I have thought about that - is Lisp possible without them?
But how do you then know priority? I'm sure someone tried to
get rid of them, but how?

> the use of white space as syntax in Python

AKA "significant whitespace" in CS lingo :)

> In contrast, postmodernism allows for cultural and personal
> context in the interpretation of any work of art. How you
> dress is your business. It's the origin of the Perl slogan:
> "There's More Than One Way To Do It!"

While I agree with that slogan - it's obvious, we see it every
day, everywhere - zsh has the shell like iteration loop, but
also the for loop, Elisp has CL-like functions (implemented in
Elisp), and isn't CL the programmable programming language,
even? So cred to Perl for the slogan but "There Are More Than
Perl that" ... uhm, whatever. And what about LaTeX? A zillion
ways to output the same document, basically.

> The reason Perl gives you more than one way to do anything
> is this: I truly believe computer programmers want to be
> creative, and they may have many different reasons for
> wanting to write code a particular way. What you choose to
> optimize for is your concern, not mine. I just supply the
> paint—you paint the picture.

I agree but I think maybe the success of Python, and its
development speed, is actually because of some of that
rigidness, yes, including the whitespace lack of freedom.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Steve Sobol wrote:

>> If you are looking for a career, Python is much bigger but
>> there is a lot of shell scripts and for that matter
>> a little bit of Perl don't harm, absolutely mot.
>
> I'm seeing scripts written in Python far more often than
> Perl these days

100%, it's much more popular.

> but it is probably useful to at least be familiar with both

Right.

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Jude DaShiell wrote:

> Perl or python, which has the most supported sysadmin tools?

What do you mean, built into the actual languages?

Or the number of tools people have written in either of
those languages?

We need a command for that as well ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

debian-user wrote:

>>> Development is fastest using whatever language you know
>>> best. This is not an objective argument.
>> 
>> Put it this way, a novice Python programmer can do more in
>> Python than the novice Lisp programmer can do in Lisp, or,
>> if you will, the same in less time.
>
> I'm afraid that Python has one specific feature that puts me
> off. Sensitivity to indentation. To those who first had to
> learn 'make', it's a sin that cannot be forgiven.

I know, I absolutely don't like it but I think maybe it is
actually beneficial for the speed aspect since you don't
fiddle around with it manually in order to get it to look your
way ...

Erlang, Haskell are notorious for this because of the emphasis
on pattern matching, people spend hours aligning back and
forth to make the code look logical as well ...

> I'll admit that when I first saw perl, I thought it was
> horrific and I swore to continue using awk and C and ...
> anything but perl. But then one day $job required me to
> learn perl so I did and have been a convert ever since.

C looks much better than Perl but Python also don't look good,
C looks better. But Lisp also looks good - well, often!
C almost always looks good.
C U!

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

Oliver Schoede wrote:

> So is it still the first choice for sysadmin work on Linux?
> Well I doubt it, I also doubt it ever was. That would be
> shell. ;)

Indeed, I thought about that! The shell.

Side note, many guys say they only use sh because bash, zsh
etc requires them being installed, I don't see how that can be
problem on Unix systems. Of course, zsh can't be moved to
a computer without zsh itself, but how is that less portable
just because they add super-cryptic syntax to do fancy stuff?
I mean, it isn't the features we port, but the shell?

Maybe you guys know the answer to that, never understood it
and heard it from several guys.

> Be that as it may I don't see much of a reason for learning
> Perl today unless you're a die-hard hobbyist with near
> infinite amount of time and an undying penchant for
> obsolete technology.

Ikr? This is what I've been (trying) to say all the time ...

-- 
underground experts united
https://dataswamp.org/~incal

Re: Is perl still the No.1 language for sysadmin?

[Emanuel Berg]

ghe2001 wrote:

> On Amazon, if you ask for books on raku, you get stuff about
> clay and kilns. If you ask for python, you get TV programs
> and snakes.
>
> If you ask for perl, you get Perl. That's one thing good old
> Perl has over the new stuff :-)

Haha, but haven't you seen like 1 zillion Python books?
Crypto Trade with Python, Game programming in Pygame, blah
blah blah (made up examples but I'm almost sure they exist)

I think I've seen <10 Perl books in my life ... and
I love books!

Hey, inspired by the other dude's awesome list of source code,
can't we have a command to parse Bibtex and find out who has
the more books :)

I don't think Lisp will win anyway, I have these in my own
file, and I know of "CLTL" [1] as well ...

@book{land-of-lisp,
  author= {Conrad Barski},
  isbn  = {1593272812},
  publisher = {No Starch},
  title = {Land of Lisp},
  year  = {2010}
}

@book{lispcraft,
  author= {Robert Wilensky},
  isbn  = {0393954420},
  publisher = {Norton},
  title = {LISPcraft},
  year  = {1984}
}

[1] https://en.wikipedia.org/wiki/Common_Lisp_the_Language

-- 
underground experts united
https://dataswamp.org/~incal