Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Gentoo leaves democracy
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... 6, 7, 8, 9, 10, 11  Next  
Reply to topic    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
1clue
Advocate
Advocate


Joined: 05 Feb 2006
Posts: 2569

PostPosted: Thu Apr 12, 2018 8:01 pm    Post subject: Reply with quote

It's curious that we're talking about sock puppets with someone who joined the forum today, and who has only posted in this thread. And who is suggesting we split the tree and maintain a legacy branch. On THIS thread, where the OP wants Gentoo to change dramatically and/or fail entirely by his own statement.
Back to top
View user's profile Send private message
Zucca
Moderator
Moderator


Joined: 14 Jun 2007
Posts: 3339
Location: Rasi, Finland

PostPosted: Thu Apr 12, 2018 8:20 pm    Post subject: Reply with quote

This whole thread detached itself from reality long time ago. ;)
_________________
..: Zucca :..
Gentoo IRC channels reside on Libera.Chat.
--
Quote:
I am NaN! I am a man!
Back to top
View user's profile Send private message
bobbymcgee
n00b
n00b


Joined: 12 Apr 2018
Posts: 55

PostPosted: Thu Apr 12, 2018 8:38 pm    Post subject: Reply with quote

I was trying to dig down to the root cause of all this. It's just an idea, I imagine it's been discussed before? I do not mean all clusterfckd like funtoo is. What are the drawbacks? The code is there. Seems like many people are unhappy because opposing ideologies are all locked in the same cage. It would solve all the problems I have seen lately and there would not be people harassing devs in the first place. It would spawn more active participation and get rid of most of the drama. Again, what are the drawbacks? What I meant by "legacy" branch is it would maintain the namespace as well.

As far as hierarchy, neddy should be the king of the devs :)
Back to top
View user's profile Send private message
asturm
Developer
Developer


Joined: 05 Apr 2007
Posts: 8935

PostPosted: Thu Apr 12, 2018 9:15 pm    Post subject: Reply with quote

We have tons of participation through proxied maintainers.
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6227
Location: Room 101

PostPosted: Fri Apr 13, 2018 1:56 am    Post subject: Reply with quote

1clue wrote:
It's curious that we're talking about sock puppets with someone who joined the forum today, and who has only posted in this thread. And who is suggesting we split the tree and maintain a legacy branch. On THIS thread, where the OP wants Gentoo to change dramatically and/or fail entirely by his own statement.

1clue ... you do realise that you're suggesting mv (the OP) is using a sockpuppet account (in violation of the forum guidelines)? ... and that without evidence of this being the case your engaging in a slur?
Back to top
View user's profile Send private message
pjp
Administrator
Administrator


Joined: 16 Apr 2002
Posts: 20067

PostPosted: Fri Apr 13, 2018 3:06 am    Post subject: Reply with quote

It could just as easily be a sockpuppet account created by someone else. It could even be 1clue. But there is little to be gained from making assumptions.
_________________
Quis separabit? Quo animo?
Back to top
View user's profile Send private message
1clue
Advocate
Advocate


Joined: 05 Feb 2006
Posts: 2569

PostPosted: Fri Apr 13, 2018 3:11 am    Post subject: Reply with quote

Note: This post has been edited multiple times.

I have no idea who the sockpuppet is. Only pointing out that it's highly suspicious that a new user was created today and posted with rather specific information about events and a proposed course of action.

I'm not pointing fingers at anyone in particular. For all I know it could be a longtime lurker who actually reads documentation rather than posting. Weirder things have happened. But why create an account now, when there were so many fun opportunities with the systemd excitement?

It was khayyam who first seemed suspicious IMO. Which strongly implies that he's not the sockpuppet.

FWIW I see no sense in a person creating a duplicate account and then pointing it out right away.

It occurs to me that someone with ssh access to the forum server could find out what IP address the sockpuppet account is coming from and see who it might match to.
Back to top
View user's profile Send private message
pjp
Administrator
Administrator


Joined: 16 Apr 2002
Posts: 20067

PostPosted: Fri Apr 13, 2018 3:34 am    Post subject: Reply with quote

I agree that it is highly suspicious, but with the exception of the funtoo comment, the posts aren't "unreasonable."

As for the IP address, there are many common use domains, so it doesn't provide much proof of anything.
_________________
Quis separabit? Quo animo?
Back to top
View user's profile Send private message
1clue
Advocate
Advocate


Joined: 05 Feb 2006
Posts: 2569

PostPosted: Fri Apr 13, 2018 4:24 am    Post subject: Reply with quote

khayyam wrote:
1clue wrote:
It's curious that we're talking about sock puppets with someone who joined the forum today, and who has only posted in this thread. And who is suggesting we split the tree and maintain a legacy branch. On THIS thread, where the OP wants Gentoo to change dramatically and/or fail entirely by his own statement.

1clue ... you do realise that you're suggesting mv (the OP) is using a sockpuppet account (in violation of the forum guidelines)? ... and that without evidence of this being the case your engaging in a slur?


Only now, reading back on this did I see an implied connection pointing at mv. Not my intent. I was pointing out the intent of the OP and the intended nature of the thread, and the presence of a likely sock puppet. The former is potential cause for the latter, but who the actor is I have no idea. Frankly I would suspect mv to bash heads directly under his own name rather than play the sock puppet game.
Back to top
View user's profile Send private message
Ant P.
Watchman
Watchman


Joined: 18 Apr 2009
Posts: 6920

PostPosted: Fri Apr 13, 2018 11:57 am    Post subject: Reply with quote

The only thing a new account is a likely indicator of is a spambot, and that's not happening here. I used Gentoo for three or four years before getting a forums account.

You're ascribing far too much status to arbitrary labels, the last six pages have been you attacking developers and contributors in defense of people whose only differentiator is a red warning flag following them everywhere signalling that they're more than willing to go through a hazing ritual to engage in office politics.

Maybe you should get one, because that's the only thing I see you bringing to Gentoo.
Back to top
View user's profile Send private message
1clue
Advocate
Advocate


Joined: 05 Feb 2006
Posts: 2569

PostPosted: Fri Apr 13, 2018 1:18 pm    Post subject: Reply with quote

Ant P. wrote:
The only thing a new account is a likely indicator of is a spambot, and that's not happening here. I used Gentoo for three or four years before getting a forums account.

You're ascribing far too much status to arbitrary labels, the last six pages have been you attacking developers and contributors in defense of people whose only differentiator is a red warning flag following them everywhere signalling that they're more than willing to go through a hazing ritual to engage in office politics.

Maybe you should get one, because that's the only thing I see you bringing to Gentoo.


Ant P., why don't you go back and look at who said "sockpuppet account" first? And if you'll look I also mentioned it could be a long-time user who just recently got an account. Others attributed an accusation to me, when there was none.
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 7470

PostPosted: Fri Apr 13, 2018 1:32 pm    Post subject: Reply with quote

1clue wrote:
It was khayyam who first seemed suspicious IMO. Which strongly implies that he's not the sockpuppet.

That's the first thing i would do myself. So what, should we take blood of everyone, burn it, and see whom blood jump out?*
What do you think inspector 1clue ? :D

* i keep failing miserably with my reference, so my new modus operandi is to actually point the reference, maybe my joke and reference would be better. So it's just a reference to "the thing" were the scientists have hard time finding who has the thing in them, and get paranoid about each other as result. Yeah, great movie, and old movie, must be because i'm old
Back to top
View user's profile Send private message
1clue
Advocate
Advocate


Joined: 05 Feb 2006
Posts: 2569

PostPosted: Fri Apr 13, 2018 1:37 pm    Post subject: Reply with quote

@Ant P.

Up until your post this thread has been heated but mostly free of personal slurs deliberately aimed at specific users and their character. Thank you for your new additions.

@krinn,

I pointed out that there was a suspicious account, and that this thread being what it is, somebody may have a reason to use a sock puppet. I speculated on how its owner might be determined if it is a sock puppet. At least one moderator is following so I'm done with that.
Back to top
View user's profile Send private message
khayyam
Watchman
Watchman


Joined: 07 Jun 2012
Posts: 6227
Location: Room 101

PostPosted: Fri Apr 13, 2018 3:51 pm    Post subject: Reply with quote

1clue wrote:
Ant P., why don't you go back and look at who said "sockpuppet account" first?

1clue ... that would be me, but context is everything, so let me remind you of that context:

khayyam wrote:
bobbymcgee ... is that what we're discussing? Oh, and when I read "extremely vocal minority" I think sockpuppet account.

... that is a far cry from an attribution to sockpuppetry to someone ... anyone. The "thought" comes as the result of 1). they didn't seem to have read the thread, 2). they used the commonly voiced (and loaded) description of dissenting individuals as an "extremely vocal minority", 3). first post.

You immediately jumped in to think it "curious", and proceeded to fill in the blanks with "[o]n THIS thread, where the OP wants Gentoo to change dramatically and/or fail entirely by his own statement". You've since claimed that it was "not the intent" to suggest that mv was behind it, but a reasonable person might ask, if that wasn't the intent then what is it that connects what comes before it with "THIS thread [...] where the OP" in that entire statement? ... because it seems clear to me the "curious" does exactly that.

1clue wrote:
It was khayyam who first seemed suspicious IMO. Which strongly implies that he's not the sockpuppet. FWIW I see no sense in a person creating a duplicate account and then pointing it out right away.

You're not counting on me being an evil mastermind ... double-bind, triple-bind, double-triple-bind, take your pick ;). Oh, and BTW, what sockpuppet? ... you do understand what happens when you use the definite article "the"?

best ... khay
Back to top
View user's profile Send private message
Marcih
Apprentice
Apprentice


Joined: 19 Feb 2018
Posts: 213

PostPosted: Fri Apr 13, 2018 7:33 pm    Post subject: Reply with quote

Last time I checked this thread it was a fairly on-topic and reasonable one. I come back and I see a (sock) puppet theater!
_________________
Bones McCracker wrote:
It wouldn't be so bad, if it didn't suck.

NeddySeagoon wrote:
The problem with leaving is that you can only do it once and it reduces your influence.
Back to top
View user's profile Send private message
Ant P.
Watchman
Watchman


Joined: 18 Apr 2009
Posts: 6920

PostPosted: Sat Apr 14, 2018 8:15 am    Post subject: Reply with quote

1clue wrote:
@Ant P.

Up until your post this thread has been heated but mostly free of personal slurs deliberately aimed at specific users and their character. Thank you for your new additions.

Keep the weak passive-aggressive snipes to yourself unless you want to get laughed out of the room. Your office politics hold no power here.

You were very insistent — before derailing the thread into 120 posts of arguing over the definition of "is" — that this is no democracy, that our input is worthless to Gentoo, and that anyone who doesn't like it can gtfo. The fact that you did or didn't single people out is irrelevant.

Okay, so that means you want no more decent search tools for portage, say bye-bye to a bunch of the front-line support on the forums, and there'll be nobody working any more on init or plumbing-layer software that isn't dependent on Lennartware. And that's just the first-order consequences of you insulting people in *this thread* until you get your way. What is your end goal here?

Here's a better idea: if you're going to reject democracy (who gave you the authority?) then we'll play by meritocracy rules, because this is a community-run FOSS project, not whatever cesspit you acquired that nasty-ass attitude from. Are you going to start writing code, reporting and fixing bugs, fielding support questions? What do *you* bring to this table? Step up or go off. We aren't paid enough to put up with the shit you're flinging at us.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Sat Apr 14, 2018 3:25 pm    Post subject: Re: Gentoo leaves democracy Reply with quote

mv wrote:
Perhaps not everybody here is following dev-ml regularly.

So I feel the need to make public here that it was practically decided to destroy democracy in gentoo:
Allow write access to dev-ml only to developers and perhaps to a few explicitly allowed by developers.

This is not about moderation or filtering spam (which so far never was a serious problem on that list) but about completely taking away non-developers the voice to be able to be involved with developers decisions.
Background is clearly that some developers do not want to be criticized anymore. (The "attacker" being the cause of this has left dev-ml, but this seems to play no role for the decision).

The decision has already been made by the council and is already going into technical implementation.
The non-democratic tendencies which came with certain developers in the previous years have won.
Ouch, that really is an awful idea.

The whole point of the dev m-l (vs the IRC channel) is to run implementation ideas by the wider user-base, since they have a great deal of collective knowledge. After all the user-base is where the entire developer-pool comes from.
This is well-known, and has been since the beginning; drobbins was quite insistent on it, based on his experience at Enoch, and the echo-chamber it became.
Quote:
For me, this is enough: I will probably switch distribution (hence stopping any explicit or implicit support of gentoo) when the policy really gets in force, and I hope that I will not be the only one.
Hmm, I don't think I can stop using the portage ebuild tree, which is the real output. I need to consider it, though.

Seems to me that you're right about the overall; the RedHat brigade, along with all their bulshytt rationalisations for cargo-cult elitism, have finally got Gentoo.

Wonder how long it'll be before we see the exodus of the essential developers: the quiet ones who get on with all the really interesting work in the background, without ever posturing anywhere at all.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Sat Apr 14, 2018 3:34 pm    Post subject: Reply with quote

CasperVector wrote:
See also rc(1) from Plan 9 and its Unix port which fixes some of its unsatisfactory aspects.
Dr.Willy wrote:

There is also fish and ion - not that I would recommend using them in a ebuild-esque setting.
If you really wanted to provide some sort of alternative, the best bet would be mksh.
With my scripting-hat on, I can envisage ebuilds that work in both bash and mksh, with the admin deciding which to use at the specific installation.

You'd start with ebuild.sh, which I rewrote to clean bash a few years ago (so I do know it fairly well.) There's nothing especially clever about any of it, and there is an awful lot that isn't (or wasn't, last time I checked the code.)

Then a few of the eclasses, autotools first.

They're both korn-shell derivatives, and mksh supports quite a few bash builtins and idioms, as well as being much lighter, and having other capability that would be useful in the backend, like hash-vals and unicode display lengths (where some uchrs take x columns, and some take none.)
Back to top
View user's profile Send private message
CasperVector
Apprentice
Apprentice


Joined: 03 Apr 2012
Posts: 156

PostPosted: Sat Apr 14, 2018 4:45 pm    Post subject: Reply with quote

steveL wrote:
CasperVector wrote:
See also rc(1) from Plan 9 and its Unix port which fixes some of its unsatisfactory aspects.
Dr.Willy wrote:
There is also fish and ion - not that I would recommend using them in a ebuild-esque setting.
If you really wanted to provide some sort of alternative, the best bet would be mksh.

If we are to find the best (TM) alternative, wouldn't it be some kind of Scheme? (Cf. GNU Guix ;)
_________________
My current OpenPGP key:
RSA4096/0x227E8CAAB7AA186C (expires: 2020.10.19)
7077 7781 B859 5166 AE07 0286 227E 8CAA B7AA 186C
Back to top
View user's profile Send private message
Dr.Willy
Guru
Guru


Joined: 15 Jul 2007
Posts: 547
Location: NRW, Germany

PostPosted: Mon Apr 16, 2018 11:55 am    Post subject: Reply with quote

steveL wrote:
CasperVector wrote:
See also rc(1) from Plan 9 and its Unix port which fixes some of its unsatisfactory aspects.
Dr.Willy wrote:
There is also fish and ion - not that I would recommend using them in a ebuild-esque setting.
If you really wanted to provide some sort of alternative, the best bet would be mksh.
With my scripting-hat on, I can envisage ebuilds that work in both bash and mksh, with the admin deciding which to use at the specific installation.

That would be an improvement from the technical side of things, but I'm still skeptical.
Behind the scenes more and more code is moved to the python side of things (like the doins, dodoc, … functions) and bash only serves as a frontend.
I remember talking to radhermit about pkgcore and he voiced the idea to drop the bash-executable entirely and parse the ebuilds from python instead. (wasn't that what libbash was about?)
Anyway I thought that idea was crazy. mksh might offer a better implementation, but if the implementation were to be dropped anyway, it is only a marginal improvement in terms of the language itself.
On the other hand ebuilds make only limited use of shell features anyway. POSIX-sh probably covers 95% of what is done in an ebuild.
… and yeah, there's always someone to suggest using a lisp. Only in this case it's not even that far fetched.

Another interesting approach might be to move the ebuild interface from function-level to program-level. So instead of a src_install function you'd have a src_install script. That would give you complete freedom wrt to the implementation.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Apr 16, 2018 3:22 pm    Post subject: Reply with quote

steveL wrote:
If you really wanted to provide some sort of alternative, the best bet would be mksh.
CasperVector wrote:
If we are to find the best (TM) alternative, wouldn't it be some kind of Scheme? (Cf. GNU Guix ;)
Lul, no, though I do understand where you're coming from (I still have a soft spot for Standard ML..)
The reason a shell front-end is the best approach, is because of the immediacy with which anyone can start working on an ebuild, which is after all simply automating the configure, make && make install process, perhaps with patches before configure (anywhen else and YDIW, ime.)
Note that all of those use shell, too; awfully in the case of autoconf, as the author himself stated a year or so after first implementing it ("maybe m4 wasn't the best choice, but it was just an undergrad project.")

The shell is central to UNIX; it always has been. Standard C is an abstraction of the high-level parts of UNIX, and the low-level facilities needed to implement the middle-layer, and the reason the argc/argv interface is the way it is, is because UNIX programs are started by shell, which handles argument expansion and splitting.
This was radically different to every other system, as still evinced in the awful way MS-doze handles arguments.
OFC it is really about splitting out, so the process code can be minimal, and use the same argv interface to start other programs (eg: make for most commands, if you're doing it right.)
The point being that this design was thought through *before* it was implemented (again, in stark contrast to practically every other approach.)

Good luck taking build-system authors away from shell.. (however much Guile you stick in gmake ;-)

A large aspect of shell you only really grok, when you see how it is used in make; there are a few things POSIX standardises, which really help makefile authors. (consider the exit status of an if with no else, when no branch is taken, for example; the same thing applies to case. Or how case does not word-split its switch value.)
So no, I don't see shell going away, and the same things that make it so convenient for make (which is all about double macro expansion, given that every shell is a macro expansion processor at heart) also make it incredibly convenient for wrapping the build. Especially when you add proper arrays at the language-level, as bash does, and eg: printf -v for backend work.
None of this is surprising when you consider that shells (when scripted) are designed for automating canned command-sequences, with dynamically-evaluated predicates and parameters (to get flowery.)

Best job-control language going.
Back to top
View user's profile Send private message
CasperVector
Apprentice
Apprentice


Joined: 03 Apr 2012
Posts: 156

PostPosted: Mon Apr 16, 2018 3:59 pm    Post subject: Reply with quote

steveL wrote:
The reason a shell front-end is the best approach, is because of the immediacy with which anyone can start working on an ebuild, which is after all simply automating the configure, make && make install process, perhaps with patches before configure (anywhen else and YDIW, ime.)
Note that all of those use shell, too; awfully in the case of autoconf, as the author himself stated a year or so after first implementing it ("maybe m4 wasn't the best choice, but it was just an undergrad project.")

Good luck then when you have time to flirt with scsh.

steveL wrote:
A large aspect of shell you only really grok, when you see how it is used in make; there are a few things POSIX standardises, which really help makefile authors. (consider the exit status of an if with no else, when no branch is taken, for example; the same thing applies to case. Or how case does not word-split its switch value.)
So no, I don't see shell going away, and the same things that make it so convenient for make (which is all about double macro expansion, given that every shell is a macro expansion processor at heart) also make it incredibly convenient for wrapping the build. Especially when you add proper arrays at the language-level, as bash does, and eg: printf -v for backend work.
None of this is surprising when you consider that shells (when scripted) are designed for automating canned command-sequences, with dynamically-evaluated predicates and parameters (to get flowery.)

Note how easy it is to emulate Makefile behaviours with Lisp, and how scsh does the shell right. See also how rc(1) managed to get rid of most historical baggages (btw, I do *not* like execline) from the Bourne shell while keeping roughly as powerful as bash for scripts.
_________________
My current OpenPGP key:
RSA4096/0x227E8CAAB7AA186C (expires: 2020.10.19)
7077 7781 B859 5166 AE07 0286 227E 8CAA B7AA 186C


Last edited by CasperVector on Mon Apr 16, 2018 4:17 pm; edited 2 times in total
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Apr 16, 2018 4:10 pm    Post subject: Reply with quote

Dr.Willy wrote:
mksh might offer a better implementation, but if the implementation were to be dropped anyway, it is only a marginal improvement in terms of the language itself.
On the other hand ebuilds make only limited use of shell features anyway. POSIX-sh probably covers 95% of what is done in an ebuild.
Heh, I refer you back to the long-running discussion I had with Roy Marples about moving ebuilds to POSIX sh.

I really would not want to maintain ebuild.sh and eclasses without bash, or an equivalent ksh derivative.
Recipe for complete head-fsck, if you ask me. (and I say this is as someone who really loves POSIX sh.)
Quote:
Another interesting approach might be to move the ebuild interface from function-level to program-level. So instead of a src_install function you'd have a src_install script. That would give you complete freedom wrt to the implementation.
Sure, it sounds plausible.
What you are saying is already what happens in implementation terms; that is why there is complete freedom of implementation-language at package-manager level.

I think the format is already proven, however. Consider that it is so nice, Roy went and used the exact same metaphor for openrc initscripts, which we know and love (or used to, when he was still lead.)

I don't think throwing away the baby will help deal with the dirty bathwater.

mksh at least addresses the concerns from the minimalist crowd, that bash is too heavy (and too large to be reliable.)
mksh is in fact an excellent first-bootstrap program, once you have a toolchain going. IF people do jump ship and start another distro, I'd recommend it be considered for /bin/sh symlink, in fact.
Back to top
View user's profile Send private message
steveL
Watchman
Watchman


Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery

PostPosted: Mon Apr 16, 2018 4:23 pm    Post subject: Reply with quote

CasperVector wrote:
Good luck then when you have time to flirt with scsh.
A quick glance at the docs page tells me it's functional, which means most admins are going to have switch mental gears when they work in it.

By all means show us some examples which prove me wrong, or I'll check for some next time I come online. I know functional languages are good for writing parsers for other languages.
Quote:
Consider how easy it is to emulate Makefile behaviours with Lisp, and how scsh does shell right.
Sure you can provide the same format.
However if I have to switch to (functional (programming) (in (order to) implement)) anything, then it's not going to fly, afaic.
Quote:
See also how rc(1) managed to get rid of most historical baggages from the Bourne shell (btw, I do *not* like execline).
Yes, I like rc; it's aimed explicitly at the scripting side of it, which is ofc much less to worry about.
Most of the "baggage" in shell comes from it having to cope with random user input at the terminal. (It helps to bear that in mind when scripting it, ime; it makes it much easier to relax into quoting, for a start.)

I like execline too; it has a different niche. Though personally I'd roll it into standard arg processing, on a "--!" parameter.

For the reasons stated above, you will never supplant the shell in UNIX. As a direct consequence, every admin worth the name, and every user who is not a total end-user, knows it, and uses it regularly; daily and routinely on the part of admins.

It makes much more sense to let that knowledge work for you, rather than force busy admins and other people who might be able to patch and will definitely be able to QA, to switch headspace.
At least from where I'm sitting, typing on a machine built with the results of that proven approach.
Back to top
View user's profile Send private message
Dr.Willy
Guru
Guru


Joined: 15 Jul 2007
Posts: 547
Location: NRW, Germany

PostPosted: Mon Apr 16, 2018 6:05 pm    Post subject: Reply with quote

steveL wrote:
The reason a shell front-end is the best approach, is because of the immediacy with which anyone can start working on an ebuild, which is after all simply automating the configure, make && make install process, perhaps with patches before configure (anywhen else and YDIW, ime.)

That argument sounds compelling, but in reality you mostly call obscure functions and rely on variables assigned who-knows-where.
If you know the "API" you don't really have to know a lot of shell at all.

steveL wrote:
The shell is central to UNIX; it always has been. Standard C is an abstraction of the high-level parts of UNIX, and the low-level facilities needed to implement the middle-layer, and the reason the argc/argv interface is the way it is, is because UNIX programs are started by shell, which handles argument expansion and splitting.

…and it works so well, that correctly quoting variables in a shellscript is an art by itself? ;)
Also, what you discribe there is "a" shell, not necessarily a bourne-derivate.

steveL wrote:
Sure, it sounds plausible.
What you are saying is already what happens in implementation terms; that is why there is complete freedom of implementation-language at package-manager level.

What do you mean by package-manager level?

steveL wrote:
mksh is in fact an excellent first-bootstrap program, once you have a toolchain going. IF people do jump ship and start another distro, I'd recommend it be considered for /bin/sh symlink, in fact.

Well, what we can agree on is that it is a better basis than bash.

steveL wrote:
CasperVector wrote:
Good luck then when you have time to flirt with scsh.
A quick glance at the docs page tells me it's functional, which means most admins are going to have switch mental gears when they work in it.

Well, at least a very commonly used shell feature, pipelines, is very much functional. ;)
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3 ... 6, 7, 8, 9, 10, 11  Next
Page 7 of 11

 
Jump to:  
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 vote in polls in this forum