Register
It is currently Sat Apr 21, 2018 6:57 pm

Unix philosophy


All times are UTC


Post new topic Reply to topic  [ 5 posts ] 
Author Message
 PostPosted: Wed Oct 18, 2017 1:08 pm   
Linux Guru
User avatar

Joined: Sat Apr 03, 2004 12:39 am
Posts: 12378
Location: Clinton Township, Michigan
From the Wikipedia at https://en.wikipedia.org/wiki/Unix_philosophy

There is a very interesting article about the Unix philosophy. Included are a number of citations, including references from the original yellow "The Bell System Technical Journal". (At one time, I actually owned my own copy of this journal and a couple from that series. I doubt that I still have them unless they are in the garage; they may be a casualty of my many moves over the past 5-10 years). Nevertheless, I am quite familiar with several of these classic articles and they are worth reading if you have the time and the interest.

_________________
Brian Masinick
Distros: MX-17, antiX, Debian


Top
 Profile WWW YIM  
 PostPosted: Wed Oct 18, 2017 1:19 pm   
Linux Guru
User avatar

Joined: Sat Apr 03, 2004 12:39 am
Posts: 12378
Location: Clinton Township, Michigan
This reference is also one of my old favorites. In 1989 when I took my first graduate course in Operating Systems Theory (the one course I took at the University of New Hampshire), I went to MIT in Cambridge, MA to find a copy of: "Final Report of the Multics Kernel Design Project by M. D. Schroeder, D. D. Clark, J. H. Saltzer, and D. H. Wells, 1977."

I first encountered MULTICS in 1982 when I was working on Hourly Personnel Systems @ the GM Tech Center and I actually had an opportunity to use a Honeywell MULTICS system; later that same year, and going into 1983 I got my hands on some Linux systems. I left the Hourly Personnel Systems group and joined an Advanced Development team that was researching alternatives to using only mainframe computing for development. We were looking into a distributed computing approach with personal computers and UNIX departmental servers - an idea that was early because networks were nowhere near as powerful or prevalent as they are now, but nevertheless, we were making substantial investments in networking technology, including fiber, microwave, satellite, and Ethernet local area networks, none of which were very widespread at the time.

The first time I got my hands on a Unix system in 1982, I had the same feeling as the first time I got my hands on a microcomputer - a Heathkit microprocessor that didn't have much power, but was a lot of fun to experiment with. I first experimented with microprocessors at Michigan Tech in the late seventies, went to the Wayne State University of Lifelong Learning (hosted at the GM Tech Center) to pursue this further, and these things led to my work with Unix in 1982 and 1983 - which eventually led to finding a telecommunications engineering position at Digital Equipment Corporation (DEC) in early 1985.

So this stuff has historical and personal significance and interest to me, because it's what shaped my entire career.

_________________
Brian Masinick
Distros: MX-17, antiX, Debian


Top
 Profile WWW YIM  
 PostPosted: Wed Oct 18, 2017 1:28 pm   
Linux Guru
User avatar

Joined: Sat Apr 03, 2004 12:39 am
Posts: 12378
Location: Clinton Township, Michigan
Isn't it interesting to see how far these "research projects" have advanced to the current state of the art?

I believe that one of the next major advancements - if you consider it an "advancement", is the widespread use of Artificial Intelligence (AI) for solving business problems in every day business. It's underway now, with major accomplishments expected in the next 1-2 years.

As engineers, business people, and responsible members of society as a whole, our jobs are to balance the development of these things with reason and positive results, and protect the work from malicious, negative and damaging consequences, which are certainly possible, and quite likely, given malice and greed that we see - along with the good that the majority of us earnestly seek.

Therefore I implore each of you that are able to do constructive work in your respective areas, always be vigilant to make sure that the technology we develop also has safeguards as much as possible to deter those who seek to harm others. Integrity, security, and safety are every bit as important as technology for the sake of profit or convenience. The failure to protect, secure, and balance the information we use is the greatest risk to the cultures we live in.

_________________
Brian Masinick
Distros: MX-17, antiX, Debian


Top
 Profile WWW YIM  
 PostPosted: Wed Oct 25, 2017 11:37 am   
. . . .
User avatar

Joined: Mon Apr 21, 2003 3:56 pm
Posts: 8302
masinick wrote:
Therefore I implore each of you that are able to do constructive work in your respective areas, always be vigilant to make sure that the technology we develop also has safeguards as much as possible to deter those who seek to harm others. Integrity, security, and safety are every bit as important as technology for the sake of profit or convenience. The failure to protect, secure, and balance the information we use is the greatest risk to the cultures we live in.
Myself included, there are some here who feel that we are giving away our personal data to the likes of Google, Akonadi and Nepomuk.

The problem resolves to several issues for the sum of all of our concerns.

One glaring issue for me is how certain developments present very little of use to me. VERY little. Yet, when programs such as these are entrenched within the core of our desktop software (true blurb: tried once, long ago, to remove Nepomuk from a *buntu, using Synaptic; I ended up removing 90% of the KDE installation, rendering KDE useless), users must suffer incompatibilities with, and maintenance of, that package (feature pushes disguised as, or forcibly incorporated with, "security updates"). Who gets the greater benefit of these? This user doesn't feel it.

More, I say we need a push towards deobfuscation, creating a process where developers expect user input to be posted in areas where the public expects to report their matters. With each passing year, newly created feedback paths are increasingly scattered, and the sheer quantity of feedback paths increases in step with the quantity of packages to report to. Remember: sometimes, a package has a team which develops it, and we might see 3 or more feedback paths for each package.

I'd say users are in the middle of losing control over software development. No, I cannot read lines of code and conclude what the developer is doing by these stages, so I can only report disagreeable end results. But where? How do we users, as a result of us users being vigilant users, call out an unwanted coding decision? I'm not talking about broken code, I'm talking about core concepts, whole paradigms.

Well, we have to follow tedious processes:
1] hope that the feedback information in the package is accurate (Hint: man pages still get displayed stating being written over 10 years ago, with scattered absences of accurate maintainer addresses),
2] follow that (low overall efficacy) path and chase through spotty forwarding,
3] join a mailing list or some obscure forum for EACH package we are concerned about (github, sourceforge, savannah, KDE developers forum are just a few of the grouped fora),
4] try to not irritate the developers by being uncertain of what has happened to cause the end result that we are reporting,
and such like.
I've even read at one developer site where it demanded that our feedback to them must follow a format they specify or they will discard the feedback. In the end, it sounded like this: If you do not tell us which line of code has the typo, we aren't going to listen to you. User feedback must never be constricted to only a checklist or predefined outcomes, since we may be trying to report whole paradigms as being a problem, yet that feedback site wanted their own decisions to continue.

Yes, somehow, we need to address the manifold users needs along with the coders needs.

Then vigilance can resume.

_________________
eMachines T5246 AMD 64 X2 w/Ubuntu Mate 16.04.4
EeePC 900A w/Antix 16.2 32 bit
Dell Inspiron 1545 w/Neptune 5.1


Top
 Profile  
 PostPosted: Wed Oct 25, 2017 12:25 pm   
Linux Guru
User avatar

Joined: Sat Apr 03, 2004 12:39 am
Posts: 12378
Location: Clinton Township, Michigan
I do not have any good answer to those who do not listen or accept feedback concerning core infrastructure concerns. Such refusal to accept solid feedback are without question - at least to me - and it appears to you as well - are primary reasons why the integrity and security of things that matter greatly to all of us (that care) continue to be a significant issue.

When such things are ignored, we don't have too many other options, other than the freedom to refuse to use such software and infrastructure. Sometimes that's not an easy decision to make because for other reasons we may need to use the same software in spite of our concerns because the alternatives are even more concerning and the inability to use them results in other difficulties, such as lacking the information we need for other important life decisions. That of course is a major frustration - having to use technology that we'd prefer to leave out of our lives entirely. So in essence, something that's called a freedom really isn't providing the real freedom we're seeking; instead, it's constraining our ability to enjoy privacy. The "information" becomes a risk and a great intrusion to that which we may be seeking.

That's the way that I see quite a bit of this type of "choice": anything BUT true freedom of choice!

_________________
Brian Masinick
Distros: MX-17, antiX, Debian


Top
 Profile WWW YIM  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 5 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
cron


Powered by phpBB © 2012 phpBB Group
© 2003 - 2012 USA LINUX USERS GROUP