Tiny Core Linux
Off-Topic => Off-Topic - Tiny Core Lounge => Topic started by: PDP-8 on July 18, 2020, 04:31:55 AM
-
I won't ever touch the nano editor again purely for political / personal reasons.
*** Warning - dumb political opinion ahead *** skip now!
I never thought I'd feel this way, but maybe important world events have my dander up. Surely, this subject is pithy compared to the real world, however it is strictly computer related and an OT thread ...
Two things: Gary Kildall (R.I.P) being erased from the University of Washington as a student and teacher, and memories coming back about the PINE email program.
Yeah, so I won't repeat the whole CP/M vs msDOS saga. You can research that yourself. I remembered PINE and it's Pico editor though, and marvelled at how the licensing seemed to look for loopholes in the nascent free-software movement. You can look that up too. It bugged me back then, so I used the older ELM from Dave Taylor. I quickly moved to MUTT but that's another story.
So I look through Wikipedia, and was reminded about PINE. Sometimes referred to as "Pine Is Not Elm", or "Pine Is Near Elm".
Then I snapped thinking, oh just like "DOS Is Not CP/M", or "DOS Is Near CP/M" (my words) when thinking about Qdos.
I will say this - Paul Allen (R.I.P) had the decency to establish a Gary Kildall Endowed Fellowship, but you have to look for it. Gary - a former student and teacher - is nowhere to be found on the Wikipedia page for the UofWash.
You'd think that "one of their own", who went on to establish the boom of the CP/M era, (despite the following dos drama), basically pushing the hardest to get microprocessor use beyond the calculator to act more like a mini for consumers, would be acknowledged.
Or not be slapped in the face by being invited to the 25th anniversary of the UofWashington's computer department to be a mere spectator, and reward the keynote speaker to his johnny-come-lately competitor. (again, don't want to rehash that all here - look it up)
Seems like a competetive computing atmosphere at the Uni.
So competetive, that in my mind, thinks of their very own Gary Kildall as a total Loser with a big -L because he wasn't a cut-throat computer business man, despite being successfull and world-changing in the early PC era. Sadly, it took Paul Allen to even recognize him.
I don't know - somehow the erasure of Gary's existence, and the Nano editor reminding me of Pine, sticks in my craw.
And that's why I'll never use Nano, or anything from the Uni, or any mirrors from them. Maybe not a world-changing act, it just makes me feel better in my own way.
-
Addendum:
Before anyone gets bent out of shape over the QDOS thing, the situation is eerily similar to the ATT / Coherent Unix thing. That is, ATT dragged Ken Thompson to Coherent to look for tell-tale things only he would know. He couldn't find any of his secrets in the corners, and could not testify in court that outright code copying, even though it looked and acted similarly, occurred.
So this isn't about that old CP/M / Dos issue which one can easily look up for themselves. One can easily jump to the wrong conclusion.
Same with the PINE mailer. It's not about the code, nor even the programmer.
It's only about the nano editor reminding me of things I don't like in my own research, and only *my* reaction to that research on an overarching level. At least Paul Allen recognized Gary.
Gary was there in 1970, when the University had computers I'm familiar with on a historical level, like a CDC, Xerox/SDS Sigma 5, and Burroughs B5500 with a six-mb disk! (look up the Arpanet's "imp" computers for that one - the equivalent of your home router today) and published some awesome work for APL and PL/I subset-G, but I digress ...
Maybe there's even a Gary Kildall "hall", room, or even broom-closet with his name on it somewhere there. I don't know.
-
Sadly, I think I know what is prompting this - the recent loss of key people during the pandemic who should not be forgotten in the real world.
Seems like teachers always get the shaft. So if you go into your bios to change anything, thank Gary.
Fire up FreeDos - and thank Gary first.
Two other teachers long forgotten about their skills and vision, before computing became productized:
Prof John Kemeny and Thomas Kurtz. Think BASIC. At Dartmouth University. On a timesharing system no less! Designed to bring computers to the average John and Jane Doe, not just the comp-sci students who make computing their careers.
Prof Kemeny had the knack for knowing his audience, and as such could demonstrate how E=mc 2 to non-physics students, and bring it to their level of understanding.
Such was the use of the Dartmouth TSS and Basic. But unlike the silly street-basics which came later, this was compiled, and had matrix operations and more. CP/M machines of the day simply didn't have the room for advanced functions, so compiling and matrix ops were usually cut out. And then of course the non-portable extensions by the rest of the micro makers. Ugh.
Rather than this be a rehash of the BASIC history, you can look it up.
What makes me sad is that like Gary Kildall, Profs Kemeny & Kurtz rarely get mention in some of the early CP/M basics - or perhaps a throwaway line without even mentioning their names like
"This is a subset of Dartmouth Basic". Maybe not even that. Erase the teachers and even the University that brought it to you to make it appear like it was yours.
Just saying - if you fire up Basic itself, or heck any other simple to use high-level language, tip your hat to Professors Kemeny and Kurtz, who had the vision to bring simplified computing power to the people - rather than being solely the realm of comp-sci students learning lisp.
I guess I got this out of my system. Look beyond the surface ...
-
Hi, PDP-8!
Thanks again for Your retrospectives. Those bright-minded men put the seeds into the ground, but I think not all plants thrived are what they were dreaming about. And the only chance to repair this world is not to loose the chance for ability to build up the house from ground zero, and not their final results, but the ways of their minds is what we need to learn.
You have observed and participated in the computer transformation of our world close enough, can You please say a few words about Your attitude to plan9?
Regards!
-
Plan 9 - that's hard to answer. One could write a book. Hopefully I won't do that here.
As a pure o/s reserch project, I think it was a natural follow-on to Unix, written by a team fully aware of not making the mistakes of "second system" syndrome such as was witnessed in the not too distant past of the transition from CTSS to Multics.
Whether Plan 9 became marketable or popular to any large degree is immaterial from a pure research standpoint. All that was *supposed* to matter was that it bent your mind - which could produce fruit in other areas if Plan9 itself wasn't a commercial success - even if it proved superior to Unix.
We actually might be at some of the ideals initially proposed by Multics and Plan9 today - albeit in a roundabout way. Remote filesystems, AWS instances (or whatever instance floats your boat) - and on and on.
There's always room for improvement around the corner - the question is what do we do with it to improve ourselves singularly and as a society?
That's what I dig about *nix. You don't have to be an AWS expert to bring value to your life - even if it means nothing to others. Even with the most simplistic of environment, like TinyCore, you can almost feel like you were part of the userbase of that original crew (Ken, Dennis, Lorinda, Doug, Kernighan, Aho, Roberts, Conlon etc etc) simply doing it for each other ... and oh, now you.
Welcome to the team of mind-benders.
-
The spirit of research -
I should mention Wollongong University and their outstanding demonstration of *nix spirit.
Their feat of bringing V6 to an alien architecture like the Interdata 7/32 machine was nothing short of outstanding.
Even though ATT was doing similarly a little bit later to the 8/32 box, (with no communication between them) what those Wollongong guys had to go through to achieve the V6 port was figuratively (and literally) driven by spirit, not commercial success.
Although yes, they did become commercially successful later as the Wollongong Group, the original intent was not to be an OEM for Interdata.
More of a spiritual journey / drive to make that port happen - which coincidentally proved the long-promised portabilty of *nix, even though by circumstance was never actually proven for a long time simply because everyone was buying DEC PDP's.
"Good on ya' mate" indeed!
-
Hi, PDP-8!
I've re-read "The Art Of Unix Programming" part about Plan9, and now I have the strong feeling, that GNU Hurd is the second coming of those ideas in our world. And UTF-8 is already here with us :-)
Cheers!
-
Now you got me wondering in hindsight of today's world re microkernels ...
Makes me wonder if System-D isn't kind of a microkernel kinda' band-aid placed upon a monolithic kernel?
In that light, maybe the HURD kernel is the answer to system-d! :)
.. talking out my you-know-what here ...
-
Hi, PDP-8!
Following Your retrospections' mood I am reading some papers. The next was "The Text Editor sam" by Rob Pike, thanks to cat-v.org. Besides sam description (that I'm very interested in), I've learned from this paper that non-deterministic finite state machines were invented by Ken Thompson too! Does this man invented everything? It was year of 1968!
Regards!
-
To show I'm not biased technically, only superficially for my own reasons, pico, er GNU NANO 5.0 is out!
https://www.nano-editor.org/
Re looking back to the early days: What fascinates me is not so much the editor-wars stuff like this evokes, but when using muscle-memory in BusyBox vi on TC, you are touching something that goes back a looong way and can appreciate the work put into it.
Ultimately, ed represents a cut-down QED editor which was run on much bigger timesharing machines of the day. Both Ken and Dennis implemented QED (introducing regular expressions into an actual tool!) on various systems long before Unix. CTSS, TSS etc. The other popular option was TECO, but it had it's own editing macros and such. Which lead to Emacs and that whole history back to Stallman.
But it's not about the editor themselves that fascinates me - it is about discovering the people involved and what they were working with at the time (starting from CTSS right after Sputnik totally freaked everyone out).
The thing to do is temporarily sweep away our commercial-mindset about it all, and see what it took just to get bits on a filesystem, get 'em out, and share them. :)
There are a LOT of people we owe a debt of gratitude to, long before *nix was on the scene. Using DD is probably the closest we can get using living history on a modern machine.
-
Of course, I am not interested in some editor "itself" :-) I'm trying to understand the way the concept of editing had come, and observing this way to understand, what concepts and relationships stands behind, and were the driving forces of this mind movement. Sorry for my English. I think that recent for example around-systemd holywars are manifestation of the same forces struggle, taking place and making their work under the surface of the sea of minds, involved in the ongoing transformation of information dissemination process. And if we want to guess the forecast, we can use the passed, as You reminded, editor-wars, as some model, which evolution and output variables we already have stored on long-term storing devices.
Of course, some fundamental obstacles must be taken into consideration. The main is, that in the times of CTSS financial world was intersted in computer progress, and now, I think, they say "enough". And they are not asking for our opinion. Even the enormously sophisticated software can not overcome hardware limitations, You know and feel this very well, UEFI-wars is the closest example.
The cage is closed, and the keys are in the others pocket. Now the way out I see is RISC-V on FPGAs, but this way is not cheap, powersaving and lightweight one. And FPGA producers have a lot of time to install the mouse-traps.
Sorry hundred times for bad English.
Contradictions are the driving force of our universe. As I can understand, one of the reasons for Ken Thomson's team to cut-down QED into ed was PDP-11 small amount of memory insufficient to hold the full QED. Nowadays You need 4G of RAM to read the html page with Firefox, and everybody consider this normal.
First, software designers want more memory. Next they got more of it, and they are happy to implement their ideas. And further their ideas are sunk in the ocean of memory, filled with the junk.
-
Your english is just fine! I wish a knew a second language as well as you do english!
It is all relative to managing resources. At first, they were only limited to the computing elite simply due to cost. Millions of dollars. Memory - especially core - was unbelievably costly.
The funny thing now is that I run TinyCore / dCore on relatively huge machines, which are actually "low end" toys. 4gb ram, 64gb sticks, etc. Then firing up seeing tc run with only about 92mb used. Even after running sudo cache-clear ; free -m
Almost a waste - unless I want to fill it up emulating a full-blown desktop distro most are used to.
I think we're on the same page. Just do whatever gives you the most satisfaction.
The core values (pun intended) seem to stay the same...
-
academies; rather than seeking universal truths;
academies; they try to exist with dissenting thought and criticism
-
Well - at MIT, where it exposed two diametrically opposing choices to users...
1) The darpa-funded CTSS timesharing system user, progressing into Multics
or
2) The counter-revolutionary AI department users running the ITS timesharing system.
Underneath the fingertips on your keyboard running TC are mixtures of both driving forces.
-
@PDP-8
I don't know why and what for, but I am spending certain amount of time trying to understand the influence of Plan9 on our present environment. There exists plan9port and 9base ports available for Linux to be run in the userspace. If I am not mistaken, in TinyCore early versions there was an extension of the later. Information from suckless.org and about werc too. I am not IT-guy and am uncertain understanding the full meaning, but werc is declared to be the minimalistic anti-framework for sites hosts. And - surprise! - it is running completely under 9base under rc shell! List of sites, powered by werc is quite small (not to speak tiny ) but among them here it is cat-v.org :-)
Guys from suckless.org, while describing 9base port, point that its size is 66k SLOCs, which is less than bash size, besides this port includes all base utilities, including rc shell. And the latest werc version is - surprise again! - 2020/06/16. I've built 9base and plan9port for TC, it is interesting for me to try rc shell.
Please correct me, if I am going wrong, but procfs and sysfs we have in Linux are the Plan9 heritage?
Regards!
@xor
Hi, xor! You're quite right - the only universal truth is, that there are no universal truth :-)
-
Well, we could get into a long discussion of how the dream of Multics / Plan 9 is already here for most of the population. Or other minutae of software and hardware issues.
Look at where Rob Pike works. And many other early *nix luminaries.
The original dream of the public having access to "data as a service", much like water, gas, electrical utilities up 24/7 in the 60's and late 70's has been achieved in a slightly different form.
Instead of using a teletype terminal attached to the wall jack, most of us see the browser as a terminal replacement, and ethernet / wireless have replaced the rs-232 data jack.
Instead of having the end-user store his files locally on paper-tape, it would have been stored in the GE-645 / Honeywell mainframes, all backboned together across the nation. The user would have remote filesystems looking like it was local to him or her, when in fact it might be stored on a GE-645 mainframe across the country.
Security on multics truly needed specialized hardware to achieve it's goals of ring-security, but today one might think of a Yubi-Key or some other hardware dongle in addition to normal o/s security methods.
Today - can you say "cloud" - simply a reference to your files being remote on some other machine where you don't have to know or care. Can you say "Google Drive" today? In the Multics day, the plan was not to have users needing local paper-tape storage either.
So much so, that today the only thing a user deals with is a browser, and no need to know the innards of Multics / Plan 9 / or *nix whatsoever.
For most, that means the worry of "cat -v is harmful" is a quaint concern of history.
Heh, but not for us - the people who like to mess around with computing, the old-school way with minimal applications and tools to build tools. :) I totally get what that organization's goals are, and if it matters to you (like it does me too) great. Go for it. But change the world? Hardly.
But sure - go ahead and study Plan 9 for your own edification. Great stuff.
When your bank / school / office or other organization demands you use a specific set of browsers, and one wants to argue about the Plan9 rc-shell being adequate for your needs, see how far you'll get. :)
I'm just saying it's all great stuff, but make no mistake - TinyCore isn't going to change the world. However, it may bring great joy and enrichment to our lives as individuals.
-
Oh, just a sad conclusion. You were reading this book for so long time, and You have a few pages unread, but You are certain, what will be the tale ending? Let's see some time later ;-) I hope that we will return to this thread in about a year. Running TinyCore 12!
Yes, I know, were Rob Pike works now, I've read his recent interview. This huge companies are exchanging people's confidence to dependencies (in a bad meaning) and further to manipilation, which is the form of violence. One moment people may encounter exchage rate unfair. In my opinion, corps are in a hurry, and i don't think, that the game score is predefined. Of course it's just an opinion :-)
And thanks for Your warning :-)
-
... I hope that we will return to this thread in about a year. ...
well it has been more than a year but I wanted to put some links into an existing thread that already referenced suckless.org
was here:
[rabbithole]
https://sysdfree.wordpress.com/downloads/
checked out:
https://fleshless.org/pages/spark.html
led to:
https://fleshless.org/pages/links.html
and so:
https://suckless.org
then:
https://suckless.org/philosophy/
related link titled:
Use of command-line tools for effective data processing
https://adamdrake.com/command-line-tools-can-be-235x-faster-than-your-hadoop-cluster.html
[/rabbithole]