Re: [9fans] sad commentary - Plan9

This is a discussion on Re: [9fans] sad commentary - Plan9 ; > is not available under Plan 9. (Or is it?) As there is no simple > introduction to Plan 9 new users will just go the easy way and get > Windows or Linux. lack of an introduction is not ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 21

Thread: Re: [9fans] sad commentary

  1. Re: [9fans] sad commentary

    > is not available under Plan 9. (Or is it?) As there is no simple
    > introduction to Plan 9 new users will just go the easy way and get
    > Windows or Linux.


    lack of an introduction is not the problem. not being unix
    is the problem.

    > For example the
    > role of make as an equivalent for cc is not self-evident for a
    > traditional normal OS-user.


    come again?

    - erik


  2. Re: [9fans] sad commentary

    >> For example the
    >> role of make as an equivalent for cc is not self-evident for a
    >> traditional normal OS-user.


    >come again?


    i thought it meant that he always types in cc commands on unix.
    of course you could do that too with 8c/8l but normally on plan 9 i
    create a mkfile except for the tiniest one-off things.


  3. Re: [9fans] sad commentary

    erik quanstrom wrote:
    >> is not available under Plan 9. (Or is it?) As there is no simple
    >> introduction to Plan 9 new users will just go the easy way and get
    >> Windows or Linux.

    >
    > lack of an introduction is not the problem. not being unix
    > is the problem.
    >


    Looking too much like UNIX while acting differently is part of the
    problem. However, the bigger part is that the existing documentation
    can be a bit daunting for someone who is new to Plan 9, and still has
    only a vague notion of how the system works. Like the UNIX man pages,
    the documentation is very detailed, and great for a reference. But many
    new users need a bit of hand-holding, of the "Trust me, you want to run
    this command. You'll learn why/how later, but for now, just RUN THIS
    COMMAND." sort. At least until the 'new user' anxiety dies down a bit,
    and the return of rational thought allows one to digest the more
    extensive documentation.

    Besides, isn't not being UNIX one of the prominent features of Plan 9?

    Steven Vormwald

    PS: John, thanks for the link to the RIT Intro paper (in another message
    in this thread). It helped a lot.


  4. Re: [9fans] sad commentary

    I do this:

    fn build {
    8c $1^.c && 8l -o $1 $1^.8 && rm $1^.8
    }
    acmef sieve.c # acmef opens acme with fixed-width font in one column
    build sieve
    sieve 2053

    when I need to build a one-file program.

    That's another good argument against Plan 9 that Unix users can make:
    why do I have to run two programs to compile a C source code file when
    Unix needed only one? I've grown used to it - build and mk help.

    On Jun 30, 2008, at 11:36 AM, Charles Forsyth wrote:

    >>> For example the
    >>> role of make as an equivalent for cc is not self-evident for a
    >>> traditional normal OS-user.

    >
    >> come again?

    >
    > i thought it meant that he always types in cc commands on unix.
    > of course you could do that too with 8c/8l but normally on plan 9 i
    > create a mkfile except for the tiniest one-off things.
    >




  5. Re: [9fans] sad commentary

    > erik quanstrom wrote:
    >>> is not available under Plan 9. (Or is it?) As there is no simple
    >>> introduction to Plan 9 new users will just go the easy way and get
    >>> Windows or Linux.

    >>
    >> lack of an introduction is not the problem. not being unix
    >> is the problem.
    >>

    >
    > Looking too much like UNIX while acting differently is part of the
    > problem. However, the bigger part is that the existing documentation
    > can be a bit daunting for someone who is new to Plan 9, and still has
    > only a vague notion of how the system works. Like the UNIX man pages,
    > the documentation is very detailed, and great for a reference. But many
    > new users need a bit of hand-holding, of the "Trust me, you want to run
    > this command. You'll learn why/how later, but for now, just RUN THIS
    > COMMAND." sort. At least until the 'new user' anxiety dies down a bit,
    > and the return of rational thought allows one to digest the more
    > extensive documentation.
    >
    > Besides, isn't not being UNIX one of the prominent features of Plan 9?
    >
    > Steven Vormwald
    >
    > PS: John, thanks for the link to the RIT Intro paper (in another message
    > in this thread). It helped a lot.


    I've been thinking of writing a "Plan 9 for Dummies" style thing;
    Nemo's book is excellent but definitely aimed at someone most
    interested in writing code immediately. Basically stealing the format
    from all UNIX beginner's books ever written, it would have a chapters
    about logging on, basic rio usage, basic commands, the file system
    layout, acme and sam (to match the standard vi and emacs sections!),
    rc programming, and C under Plan 9. Imagine chapter one of Nemo's
    book except greatly expanded.

    Now, before I set quill to parchment (or fingers to keyboard as may
    be), has anyone else started something like this?


    John



  6. Re: [9fans] sad commentary

    > Looking too much like UNIX while acting differently is part of the
    > problem. However, the bigger part is that the existing documentation
    > can be a bit daunting for someone who is new to Plan 9, and still has
    > only a vague notion of how the system works. Like the UNIX man pages,
    > the documentation is very detailed, and great for a reference. But many
    > new users need a bit of hand-holding, of the "Trust me, you want to run
    > this command. You'll learn why/how later, but for now, just RUN THIS
    > COMMAND." sort. At least until the 'new user' anxiety dies down a bit,
    > and the return of rational thought allows one to digest the more
    > extensive documentation.


    I think there is another, somewhat related, problem. Like most of my
    generation, I learned UNIX at university, on machines administered by
    other people. By the time I had to install and administer systems
    myself I already knew a lot. With plan9 you have to learn to be a
    user and administrator at the same time. That's one reason I would be
    very reluctant to recommend trying plan9 to most people I know. I'm
    afraid there's not much we can do about this.
    --
    John Stalker
    School of Mathematics
    Trinity College Dublin
    tel +353 1 896 1983
    fax +353 1 896 2282


  7. Re: [9fans] sad commentary

    This is a very good point. I mostly learned Unix in a corporate
    environment, but the same logic holds: somebody else had set
    up and maintained the systems.

    // I'm afraid there's not much we can do about this.

    Other, obviously, than getting uni types to use it there. Plan 9
    (like Inferno) has quite a bit to offer from pedagogical view.

    // Trinity College Dublin

    Pretty campus, warm sweatshirts. Convince your IT folks. ;-)

    anthony



  8. Re: [9fans] sad commentary

    I have not even started such thing, but, if you go for it and want help,
    count me in

    > I've been thinking of writing a "Plan 9 for Dummies" style thing;
    > Nemo's book is excellent but definitely aimed at someone most
    > interested in writing code immediately. Basically stealing the format
    > from all UNIX beginner's books ever written, it would have a chapters
    > about logging on, basic rio usage, basic commands, the file system
    > layout, acme and sam (to match the standard vi and emacs sections!),
    > rc programming, and C under Plan 9. Imagine chapter one of Nemo's
    > book except greatly expanded.
    >
    > Now, before I set quill to parchment (or fingers to keyboard as may
    > be), has anyone else started something like this?
    >



  9. Re: [9fans] sad commentary

    On Mon, Jun 30, 2008 at 2:33 PM, Francisco J Ballesteros wrote:
    > I have not even started such thing, but, if you go for it and want help,
    > count me in


    And I would read it!

    --
    Tom Lieber
    http://AllTom.com/


  10. Re: [9fans] sad commentary

    > Besides, isn't not being UNIX one of the prominent features of Plan 9?

    tautology, no? to be plan 9 it must be different. if it were not, it would be unix.

    - erik



  11. Re: [9fans] sad commentary

    On Jun 30, 2008, at 1:34 PM, john@csplan9.rit.edu wrote:
    > Now, before I set quill to parchment (or fingers to keyboard as may
    > be), has anyone else started something like this?


    I was planning on doing something of the sort...

    On Jun 30, 2008, at 5:46 PM, erik quanstrom wrote:
    > this guide was writen at coraid by michael covington.
    > the document proclaims itself to be:


    ....until I saw this. Should this go into /sys/doc unmodified? Good
    luck, Mr. Covington.



  12. Re: [9fans] sad commentary

    On Tue, Jul 1, 2008 at 2:27 AM, wrote:
    > This is a very good point. I mostly learned Unix in a corporate
    > environment, but the same logic holds: somebody else had set
    > up and maintained the systems.
    >
    > // I'm afraid there's not much we can do about this.
    >
    > Other, obviously, than getting uni types to use it there. Plan 9
    > (like Inferno) has quite a bit to offer from pedagogical view.


    Also, public 9grids. Though judging by gdiaz's experiences with
    sirviente, there's a bit of work to be done in that area - I get the
    impression things are fairly unstable once the machine gets under
    memory pressure.
    -sqweek


  13. Re: [9fans] sad commentary

    hello

    256 MB of ram fills quite easily when using a fossil+venti and when trying new incarnations of upas/fs , i can't even compile some ports of gnu things ☺. Fortunately this will change in august, as 9grid.es will have 1Gb of memory.

    about the unstability, i should disable swap partition to see if that fix something ☺

    greetings,

    gabi
    PS: sorry for the off-topic non-sad comentary :P


  14. Re: [9fans] sad commentary

    > 256 MB of ram fills quite easily when using a fossil+venti
    > and when trying new incarnations of upas/fs ,


    the manual pages for venti and fossil do spell out
    a number of parameters to control memory usage.
    but you probablly knew that.

    running the fs on the same machine as the cpu server
    can be a problem when used heavily. plan 9 overcommits
    physical memory. and fossil or venti may be left holding the bag.

    ndb/dns also has an ever-growing footprint.

    > i can't even compile some ports of gnu things ☺.


    glad to hear that there is a silver lining.

    > about the unstability, i should disable swap partition to see if that fix something ☺


    i find important to comb the kernel panic messages.
    running out of kernel memory is fatal. however,
    it could be that you are using very little and can
    decrease kernelpercent.

    - erik



  15. Re: [9fans] sad commentary

    > Also, public 9grids. Though judging by gdiaz's experiences with
    > sirviente, there's a bit of work to be done in that area - I get the
    > impression things are fairly unstable once the machine gets under
    > memory pressure.
    > -sqweek


    i think this is an artifact of setting up heavily-used systems
    combining venti, fossil, auth and cpu server.

    i think you may be blaming
    plan 9 when in fact there's just not enough hardware
    to go around.

    sure crashing is antisocial. the alternative is to add very
    large amounts of code to the kernel. but even linux doesn't
    solve this problem. my 256mb linux machine with only me
    on it, locks solid due to oom conditions more often than the
    entire coraid plan 9 system with 20 users.

    - erik



  16. Re: [9fans] sad commentary

    On Wed, Jul 2, 2008 at 8:35 PM, erik quanstrom wrote:
    >> Also, public 9grids. Though judging by gdiaz's experiences with
    >> sirviente, there's a bit of work to be done in that area - I get the
    >> impression things are fairly unstable once the machine gets under
    >> memory pressure.
    >> -sqweek

    >
    > i think this is an artifact of setting up heavily-used systems
    > combining venti, fossil, auth and cpu server.

    ....
    > sure crashing is antisocial. the alternative is to add very
    > large amounts of code to the kernel.


    Back when this was first posted I wanted to protest the point that a
    large kernel modification is necessary, since I figured you can do a
    "good enough" job with just an interface to tell the kernel not to
    kill the important server processes.
    Obviously I decided to let it lie, but I just discovered this can be
    done without modifying the kernel at all when I happened across an
    interesting line in termrc:

    /rc/bin/termrc:dontkill
    '^(ipconfig|factotum|mntgen|fossil|cs|dns|listen|r eboot)$'

    The default cpurc doesn't use dontkill, but I suspect it could be a
    big help for all-in-one servers. Figured I'd point it out as it seems
    easy to miss.
    ... plus everyone can use a good scare every now and then, and what
    better way than to resurrect sad commentry?
    -sqweek


  17. Re: [9fans] sad commentary

    sqweek wrote:
    > On Wed, Jul 2, 2008 at 8:35 PM, erik quanstrom wrote:
    >
    >>> Also, public 9grids. Though judging by gdiaz's experiences with
    >>> sirviente, there's a bit of work to be done in that area - I get the
    >>> impression things are fairly unstable once the machine gets under
    >>> memory pressure.
    >>> -sqweek
    >>>

    >> i think this is an artifact of setting up heavily-used systems
    >> combining venti, fossil, auth and cpu server.
    >>

    > ...
    >
    >> sure crashing is antisocial. the alternative is to add very
    >> large amounts of code to the kernel.
    >>

    >
    >

    Swap doesnt work reliable here. :-(
    I have disabled swaping and let the kernel kill the biggest process
    skipping any critical server processes and it works well.
    got ~100 days uptime and i use this machine for linuxemu
    development/testing.

    no adding very large amounts of code... maybe fix the swap... or even
    remove it alltogether.
    > Back when this was first posted I wanted to protest the point that a
    > large kernel modification is necessary, since I figured you can do a
    > "good enough" job with just an interface to tell the kernel not to
    > kill the important server processes.
    > Obviously I decided to let it lie, but I just discovered this can be
    > done without modifying the kernel at all when I happened across an
    > interesting line in termrc:
    >
    > /rc/bin/termrc:dontkill
    > '^(ipconfig|factotum|mntgen|fossil|cs|dns|listen|r eboot)$'
    >
    > The default cpurc doesn't use dontkill, but I suspect it could be a
    > big help for all-in-one servers. Figured I'd point it out as it seems
    > easy to miss.
    > ... plus everyone can use a good scare every now and then, and what
    > better way than to resurrect sad commentry?
    > -sqweek
    >

    good to know :-)

    cinap



  18. Re: [9fans] sad commentary

    > Swap doesnt work reliable here. :-(
    > I have disabled swaping and let the kernel kill the biggest process
    > skipping any critical server processes and it works well.
    > got ~100 days uptime and i use this machine for linuxemu
    > development/testing.


    i'm curious as to what is taking so much memory.

    - erik



  19. Re: [9fans] sad commentary

    > the alternative is to add very
    > large amounts of code to the kernel.


    not really. the alternative is to add some code to the
    kernel, and varying amounts of code to quite a few applications.


  20. Re: [9fans] sad commentary

    > no adding very large amounts of code... maybe fix the swap... or even
    > remove it alltogether.


    assuming it doesn't work now, the paging code used to work, at least
    in the sense of survive --i used 8l to link kernels on a 4mbyte 386sx16 --
    so i imagine it's just a matter of repairing it, if it indeed is responsible.
    unfortunately it's hard to tell because "doesn't work" is a little vague.



+ Reply to Thread
Page 1 of 2 1 2 LastLast