Re: [DNG] Please keep 32-bits alive

[Enrico Weigelt, metux IT consult]

On 26.07.2017 04:47, Christopher Clements wrote:


My high-school programming class was advertised as teaching people how to
program in C and do all sorts of low-level stuff.  I signed up thinking
I might finally meet a "computer expert" that actually knew what they
were talking about...

The teacher began by forcing us all to make "hello world" applications
IN JAVA!


For a beginner, java might be a good choice (plain old of course).

At our local university, the students (lower semesters) should just
learn simple things like sorting algorithms (wikipedia level), but
then are asked to repair the teacher's broken/incomplete C++ code,
with lots of template magic (hmm, do we have some hall of shame, where
I can upload his bloat ?).

As a friendly repititor I showed students how to do that in plain C.
Of course, they didn't pass the exam, as they didn't actually repair
his shitty code, but thrown it away.

From my perspective as seasoned consultant, this teacher failed his job
miserably (assuming, teaching software engineering really was his job
in the first place). And his code doesn't even look like he's got much
idea of C/C++ programming. I'm thinking about subscribing in the
university again and join that coures, just to have some fun ...


When the assignments got more complex (still in Java), I decided to run
vim off a flash drive to edit the sources because I was self-taught and
doing something (somewhat) low-level. The teacher threatened to have me
expelled for "hacking" because he didn't like the white text on a black
background; we were only allowed to use the Eclipse IDE.


*rofl* that guy seemed to be totally incompetent. yes, I also had such
folks back in school. I still remember a situation (early 90th) when the
teacher tried to explain how hard drivers work. After his longer
monologue I had to step in and explain that ATA block addressing doesn't
have much to do w/ physical locations and asked him to explain why the
machine in front of me claims to have an hdd w/ 32 heads.

After some more similar actions I had my const 1+ for the rest of the
year and didn't need to be present anymore ...


Please, KEEP DOING WHAT YOU'RE DOING!
I have seen our "successors", and they are all brain-dead MCSEs!
As members of the steadily-declining demographic of "old-fashioned"
hackers, it falls to us to keep things in somewhat-usable states!


True. OTOH, just let folks make their horrible experiences (just from
time to time give some little notes about how easy, robust and cheap
things can be) ... sooner or later they come back to us, and we can
charge any rate :p


I think that everyone should have to learn how to write a simple shell
for an 8-bit architecture before they can be considered a real
programmer. ;P


ACK.


--mtx

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[Christopher Clements]

On Tue, Jul 25, 2017 at 12:49:17PM +0200, Dr. Nikolaus Klepp wrote:

Am Dienstag, 25. Juli 2017 schrieb Didier Kryn:

> https://around.com/ariane.html

 Even with a high level language carefully designed to discourage
loose programming and dirty low-level tricks, it is still possible,
although with strong efforts, to fuck the compiler and keep programming
at a very low level. And, unfortunately, there are programmers who like
doing such things.


And there is a whole generation of programmers comming fresh from university 
that do not know the difference between bits and bytes. For me that's even 
scarier.


In 2014, I was still in high school.

My high-school programming class was advertised as teaching people how to
program in C and do all sorts of low-level stuff.  I signed up thinking
I might finally meet a "computer expert" that actually knew what they
were talking about...

The teacher began by forcing us all to make "hello world" applications
IN JAVA!

When the assignments got more complex (still in Java), I decided to run
vim off a flash drive to edit the sources because I was self-taught and
doing something (somewhat) low-level. The teacher threatened to have me
expelled for "hacking" because he didn't like the white text on a black
background; we were only allowed to use the Eclipse IDE.

The whole experience would be laughable if it weren't affecting my GPA.
Once I set the vim colorscheme to "black on white" and set :syn off, I
recieved the "most improved" award at the end of the year. (I _was_ a
self-taught C programmer since I was 12, after all...)

If it weren't for prior experience, the internet, and people like you guys
(and girls, etc.), I would have ended up as some idiot who does all their
important work in Java and thinks that writing an OS in Visual Basic is
a good idea. (That was actually an extra credit assignment... the "OS"
was actually just a script that ran under Windows.)

At 16, I found myself mentioning to a classmate "Back in MY day, we
could process comma-seperated-values without loading up a graphics API!"

Please, KEEP DOING WHAT YOU'RE DOING!
I have seen our "successors", and they are all brain-dead MCSEs!
As members of the steadily-declining demographic of "old-fashioned"
hackers, it falls to us to keep things in somewhat-usable states!

(I am ~19 years old now, and I am considered to be a "Linux-loving old
fart". Go figure...)

I think that everyone should have to learn how to write a simple shell
for an 8-bit architecture before they can be considered a real programmer. ;P

--
This email has NOT been checked for viruses by Avast antivirus software.
http://0xc9.net

Of course, there is _absolutely nothing_ stopping me from saying it _has_.

Use common sense and most of your "security" becomes redundant.

* * *

GPG Key: 0769 AFCF 681E F61E 2137  F4CB 5044 1726 610D 5AE0


signature.asc
Description: Digital signature
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[Christopher Clements]

On Sun, Jul 23, 2017 at 04:56:56AM -0400, Hendrik Boom wrote:

On Sat, Jul 22, 2017 at 11:26:56PM +0200, Adam Borowski wrote:

On Sat, Jul 22, 2017 at 06:50:19AM +0100, KatolaZ wrote:
> you  might probably want to have a look at:
>
>   http://popcon.devuan.org/
>
> Whatever the statistical significance of those data, it seems that
> between 15% and 20% of Devuan installations are on i386. So apparently
> there is no reason at all to drop it, rather the opposite.

Then this looks like a problem that needs to be looked at.  There's no way
that many people use gear from ≤ 2004 (or a brief throwback of early Atoms
from 2008).


THere's at least one.  My laptop runs an Atom processor.  It was the
first EEEPC that was completely Linux-compatible without requiring any
nonfree drivers, and came out anout a year after the first EEEPC.

Until it came on the market, I  was looking with dismay at all the
hugely expensive, overpowered, battery draining machines on the
market.  Years later, I can still go to a coffee chop for an
afternoon's writing without needing to bring along a power supply.

It is still performing well.  In fact, for some things (videoo codecs
coe to mind) it runs a lot better now than it did when I bought it)
Google has abandoned it, but I haven't, and Linux hasn't either.

I have upgraded the RAM from 1 to 2 G, and replaced the hard drive
with a much bigger one.  It's running fine, and hasn't needed repairs.

By contrast, my 64-bit server is on its last legs.

All of these machines are old enough not to have malware built into
the hardware, as far as I can tell.  I'm not looking forward to having
to upgrade to hardwar containing malware.

Please help keep 32-bit architecture alive.  I've been running Devuan
since the alpha-2 release.

-- hendrik



--
⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ A dumb species has no way to open a tuna can.
⢿⡄⠘⠷⠚⠋⠀ A smart species invents a can opener.
⠈⠳⣄ A master species delegates.


I'm probably a little late to comment, but here's my two cents' worth.

My laptop, which runs Devuan, is a heavily-used Samsung N150p which was
purchased from a friend for $40 USD, it also runs an Intel Atom processor
(although I prefer ARM), and it is predominantly 32-bit:

$ uname -a
Linux MB 3.16.0-4-686-pae #1 SMP Debian 3.16.39-1 (2016-12-30) i686 GNU/Linux

$ lshw
...
product: Intel(R) Atom(TM) CPU N450   @ 1.66GHz
vendor: Intel Corp.
physical id: 1
bus info: cpu@0
version: 6.12.10
serial: 0001-06CA----
size: 1667MHz
capacity: 1667MHz
width: 64 bits
...

There is 2GB of RAM, and anything that uses "hardware acceleration"
grinds to a halt and can take minutes between screen refreshes.

$ free -h
total   used   free sharedbuffers cached
Mem:  2.0G   186M   1.8G   460K28M90M
-/+ buffers/cache:67M   1.9G
Swap: 9.3G 0B   9.3G

I love my laptop, however, because the whole thing is ~8x11 inches, and
can be brought everywhere. I use it for programming from the termninal;
there is no "greeter" installed and it boots straight to getty. I use
a text-mode editor (which will not be mentioned to avoid flames (it's
not emacs ;P)) and that's it.

While the Atom processor DOES support 64-bit computing, it seems that
every instruction takes twice as long -- it takes 30 minutes to boot a
Tails live image. (Yes, I know about heads; it didn't boot.)

For people like me, the "latest and greatest" features are simply useless
overhead that makes things bigger, slower, heavier, hotter, and/or use
more electricity.

With all due respect, I would appreciate it if 32-bit processors would
continue to be supported. (ARM too, I'll finish my Raspberry PI 3B-based
laptop Real Soon Now! (Maybe next weekend... next month...))

--
This email has NOT been checked for viruses by Avast antivirus software.
http://0xc9.net

Of course, there is absolutely nothing stopping me from saying it has.

Use common sense and most of your "security" becomes redundant.

* * *

GPG Key: 0769 AFCF 681E F61E 2137  F4CB 5044 1726 610D 5AE0


signature.asc
Description: Digital signature
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[Didier Kryn]

Le 25/07/2017 à 12:49, Dr. Nikolaus Klepp a écrit :

Am Dienstag, 25. Juli 2017 schrieb Didier Kryn:

https://around.com/ariane.html

  Even with a high level language carefully designed to discourage
loose programming and dirty low-level tricks, it is still possible,
although with strong efforts, to fuck the compiler and keep programming
at a very low level. And, unfortunately, there are programmers who like
doing such things.

And there is a whole generation of programmers comming fresh from university 
that do not know the difference between bits and bytes. For me that's even 
scarier.
When you program at low level, you must know bytes and bits; when 
you program at high level you should better forget everything about such 
things. I mean if your program is critical in any sense.


Didier

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[Dr. Nikolaus Klepp]

Am Dienstag, 25. Juli 2017 schrieb Didier Kryn:
> > https://around.com/ariane.html
> 
>  Even with a high level language carefully designed to discourage 
> loose programming and dirty low-level tricks, it is still possible, 
> although with strong efforts, to fuck the compiler and keep programming 
> at a very low level. And, unfortunately, there are programmers who like 
> doing such things.

And there is a whole generation of programmers comming fresh from university 
that do not know the difference between bits and bytes. For me that's even 
scarier.

Nik



-- 
Please do not email me anything that you are not comfortable also sharing with 
the NSA, CIA ...
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[m712]

Sometimes, though, you *need* to keep doing low-level things. It's not much of 
a necessity in the current state of technology anymore, but if you recall back 
in the 90s people used to hand-tune assembly code because the C code generated 
by the compiler was not efficient enough. The hardware was also the cause of 
such trickery, like Quake's fast inverse square root, which was created because 
integer aritmetic was much faster than floating point arithmetic back then. 
Most of these problems are solved by compiler optimizations and hardware 
improvements but people working on embedded systems sometimes still need to do 
low-level magic.

On July 25, 2017 12:04:15 PM GMT+03:00, Didier Kryn  wrote:
>Le 25/07/2017 à 00:34, Alessandro Selli a écrit :
>> On Mon, 24 Jul 2017 at 21:52:23 +0200
>> Ruediger Meier  wrote:
>>
>>> On Monday 24 July 2017, Joachim Fahrner wrote:
 Am 2017-07-24 20:34, schrieb Hendrik Boom:
> How much source code actually cares whether pointers are 32 or 64
> bits?
 Clean written code should not care about pointers or integers are
>32
 or 64 bit or byte order. Code written in a higher language should
>run
 on any hardware, otherwise I call it "defect".
>>> In theory you may have right but in practice I guess most currently
>used
>>> code would not run correctly on 16bit machines.
>>>
>>> In practice you have to run and test the code on all target
>>> architectures to keep it portable.
>>Right.  Remember the Ariane!
>>
>> https://around.com/ariane.html
>
> Even with a high level language carefully designed to discourage 
>loose programming and dirty low-level tricks, it is still possible, 
>although with strong efforts, to fuck the compiler and keep programming
>
>at a very low level. And, unfortunately, there are programmers who like
>
>doing such things.
>
> Didier
>
>___
>Dng mailing list
>Dng@lists.dyne.org
>https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

--- :^) --- :^) --- :^) --- :^) --- :^) --- :^) --- :^) --- :^) ---
https://blaze.nextchan.org - https://gitgud.io/m712/blazechan
https://nextchan.org - https://gitgud.io/nextchan/infinity-next
I am awake between 7AM-12AM UTC, hit me up if something's wrong
garbage of the day: hola

signature.asc
Description: PGP signature
___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Please keep 32-bits alive

[Didier Kryn]

Le 25/07/2017 à 00:34, Alessandro Selli a écrit :

On Mon, 24 Jul 2017 at 21:52:23 +0200
Ruediger Meier  wrote:


On Monday 24 July 2017, Joachim Fahrner wrote:

Am 2017-07-24 20:34, schrieb Hendrik Boom:

How much source code actually cares whether pointers are 32 or 64
bits?

Clean written code should not care about pointers or integers are 32
or 64 bit or byte order. Code written in a higher language should run
on any hardware, otherwise I call it "defect".

In theory you may have right but in practice I guess most currently used
code would not run correctly on 16bit machines.

In practice you have to run and test the code on all target
architectures to keep it portable.

   Right.  Remember the Ariane!

https://around.com/ariane.html


Even with a high level language carefully designed to discourage 
loose programming and dirty low-level tricks, it is still possible, 
although with strong efforts, to fuck the compiler and keep programming 
at a very low level. And, unfortunately, there are programmers who like 
doing such things.


Didier

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] Why am I being stonewalled (GRSecurity discussion)?

[aconcernedfossdev]

I was responding to this by Bradley M. Kuhn:
https://lists.debian.org/debian-user/2017/07/msg00811.html

Rather, the giant worldwide queue of known GPL violations
should be prioritized by figuring out which ones, if solved, will do 
the

most to maximize software freedom for all users.


Explaining why resting on ones rights is a very inadvisable move.
Laches will bar you from equitable relief after a time, the statute of 
limitations for the claim will bar you from relief under law (money 
damages).


Bruce then made a claim of "well a programmer could claim they had no 
money to hire a lawyer". I said that a court would not be impressed by a 
plaintiff making such an argument.

For that I get stonewalled and libeled.

I also started to brainstorm various defenses the opposition might 
bring. This offended Bruce (who, BTW, lets everyone know that he is not 
a lawyer.)


We are not supposed to anticipate an opponents moves apparently...



As for competence.
I am competent at making 3d video game maps.
I am competent at 3d modeling.
I am competent at game code programming.
I am competent at argument to help induce you to fork Debian.
I am competent at inducing Bruce Perens to publish an advisory regarding 
GRSecurity and it's copyright violation, which was my goal.


What do you find me incompetent in?


On 2017-07-25 03:39, Se7en wrote:
On Tue, Jul 25, 2017 at 03:18:25AM +, aconcernedfoss...@airmail.cc 
wrote:

Why am I being stonewalled from the discussion now?


So you were kicked out of a discussion thread for being weird, you go
to 8chan to complain, they figure out you're a sockpuppet of
MikeeeUSA, and then you leave there and come to the devuan mailing
list and don't even put OT in the subject header.

Mikeee, you're a weirdo and outside of maybe 10 people no one thinks
you are even competent.

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] systemd'oh! DNS lib underscore bug bites everyone's favorite init tool, blanks Netflix

[Jaromil]

On Tue, 25 Jul 2017, Dragan FOSS wrote:

> On 07/24/2017 11:37 PM, Ozi Traveller wrote:
> > https://www.theregister.co.uk/2017/07/24/underscore_domain_name_bug/
> > 
> > Oops!
> Actually (and TBH..), this is a bug in libidn2, which isn't in systemd
> source tree...at least for now ;>

well, one would expect a software as important as systemd to have
integration tests covering these sort of changes. sure is libidn2
fault, but then this episode shows systemd developers are updating a
core library without even testing the consequences.

ciao

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng

Re: [DNG] apt-file doesn't work (ascii/ceres)

[Jaromil]

On Tue, 25 Jul 2017, Yevgeny Kosarzhevsky wrote:
>Is there anything I am missing at my end or is it a known issue?

this is a known issue, it is fixed already by the amprolla3 rewrite
which will be in production soon.

ciao

___
Dng mailing list
Dng@lists.dyne.org
https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/dng