Linux-Advocacy Digest #542, Volume #33 Thu, 12 Apr 01 12:13:03 EDT
Contents:
Re: New virus attacks Linux and MS OS ("Mikkel Elmholdt")
Re: Need your recommendation for a full-featured text editor (Thore B. Karlsen)
Re: Need your recommendation for a full-featured text editor (Phillip Lord)
Re: Need your recommendation for a full-featured text editor (Phillip Lord)
Re: Need your recommendation for a full-featured text editor (Ken Tough)
Re: Need your recommendation for a full-featured text editor (Phillip Lord)
Re: Blame it all on Microsoft (Craig Kelley)
Re: So much for modules in Linux! (Grant Fischer)
Re: Communism, Communist propagandists in the US...still..to this day. (Rob
Robertson)
Re: So much for modules in Linux! (Chad Everett)
Re: Blame it all on Microsoft (hack)
Re: Blame it all on Microsoft ([EMAIL PROTECTED])
hmm getting tired of this! ("tony roth")
----------------------------------------------------------------------------
From: "Mikkel Elmholdt" <[EMAIL PROTECTED]>
Subject: Re: New virus attacks Linux and MS OS
Date: Thu, 12 Apr 2001 16:56:15 +0200
"Aaron R. Kulkis" <[EMAIL PROTECTED]> skrev i en meddelelse
news:[EMAIL PROTECTED]...
<snip>
> > Most Linux users that I know boot into root and stay there. The easiest
of
>
> You must hang with some pretty stupid people.
>
> by the way, how do you become non-root in the typical Winblows desktop
machine
That's easy if you are running WinNT/2000 in a Windows domain. An ordinary
user is not automatically granted Administrator rights on their own machine,
unless the IT guys enable it. That means that you cannot install drivers,
services, and other stuff.
With Win9x it is a different matter, but if you're using that for serious
stuff, you had it coming anyway.
<snipped ridiculous log sig>
Mikkel Elmholdt
------------------------------
From: Thore B. Karlsen <[EMAIL PROTECTED]>
Crossposted-To:
24hoursupport.helpdesk,alt.comp.shareware.programmer,comp.editors,comp.lang.java.help,comp.lang.java.programmer,comp.lang.java.softwaretools,comp.os.linux.development.system
Subject: Re: Need your recommendation for a full-featured text editor
Date: Thu, 12 Apr 2001 10:03:43 -0500
On 12 Apr 2001 15:48:49 +0100, Phillip Lord <[EMAIL PROTECTED]>
wrote:
> >> You want the additional functionality you have to pay something
> >> for it.
> Thore> I _don't_ want the additional functionality! It's nice to
> Thore> have it, but as options, not as part of the standard
> Thore> package. I don't want 47MB worth of crud that I'm never going
> Thore> to use.
> Thore> Emacs should stick to being a text editor, not an application
> Thore> environment.
> I like it as an application environment.
>
> The idea of shipping a minimal emacs has been suggested for a
>while, on top of which you could add just the packages you use. I
>think that this would be a good idea, but I don't think that the woman
>power is there to do it at the moment. To be honest I doubt that it
>ever will be. 47MB of disk space is just not a lot these days.
Sigh..
So it's not a lot. It's still more than even Microsoft Office. It's more
than DJGPP.
That you in most cases have lots of space doesn't justify wasting it. I
just don't buy into that philosophy.
--
"By the time we've finished with him, he won't know whether
he's Number Six or the cube root of infinity!"
------------------------------
From: Phillip Lord <[EMAIL PROTECTED]>
Crossposted-To:
24hoursupport.helpdesk,alt.comp.shareware.programmer,comp.editors,comp.lang.java.help,comp.lang.java.programmer,comp.lang.java.softwaretools,comp.os.linux.development.system
Subject: Re: Need your recommendation for a full-featured text editor
Date: 12 Apr 2001 15:56:46 +0100
>>>>> "Aaron" == Aaron R Kulkis <[EMAIL PROTECTED]> writes:
Aaron> Dragan Cvetkovic wrote:
>> "Aaron R. Kulkis" <[EMAIL PROTECTED]> writes: > > If I moved
>> around from machine to machine a lot (which > > I don't) then I
>> would stick emacs with all my configuration and extra > >
>> packages and anything else that I wanted onto a CD-ROM, and run
>> it off > > there. > > In that case, you would have a lot of
>> clients pissed at you.
>>
>> Why? he is saying he is going to run it from CD-ROM. Nothing gets
>> onto client's hard disk.
Aaron> run emacs get software upgrade distribution CD swap CD's want
Aaron> to edit config files swap CD's want to do upgrade swap CD's
Aaron> want to clean up config files swap CD's.
Aaron> FUCK THAT!
I don't understand I am afraid. I upgrade the packages for
emacs probably a couple of times a month, as new versions come
out. But on my laptop I do so much more rarely, say once every 6
months, and would happily last for much longer.
You could do it every year at Christmas.
Config files would go onto the CD-ROM as I've said, so
the clean up consists of removing the CD-ROM when you leave. Not
hard.
Phil
------------------------------
From: Phillip Lord <[EMAIL PROTECTED]>
Crossposted-To:
24hoursupport.helpdesk,alt.comp.shareware.programmer,comp.editors,comp.lang.java.help,comp.lang.java.programmer,comp.lang.java.softwaretools,comp.os.linux.development.system
Subject: Re: Need your recommendation for a full-featured text editor
Date: 12 Apr 2001 15:57:52 +0100
>>>>> "Aaron" == Aaron R Kulkis <[EMAIL PROTECTED]> writes:
Aaron> Phillip Lord wrote:
>> >>>>> "Aaron" == Aaron R Kulkis <[EMAIL PROTECTED]> writes:
>>
Aaron> Phillip Lord wrote:
>> >> >>>>> "Aaron" == Aaron R Kulkis <[EMAIL PROTECTED]> writes: >>
Aaron> Which means that as soon as your on a new machine, your stuck
Aaron> editing WITHOUT your config file....
>> >>
Aaron> UGH.
>> >> This is why God invented NFS mounted home spaces. >> >> Phil
>>
Aaron> And if your behind a corporate firewall which doesn't permit
Aaron> NFS connections through it....
>> Fortunately god also invented floppy disks.
>>
Aaron> emacs that fits on a floppy. you're kidding, right?
We were talking about config files. 1.4M is a hell of a
lot of lisp.
Phil
------------------------------
From: Ken Tough <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.shareware.programmer,comp.editors,comp.lang.java.help,comp.lang.java.programmer,comp.lang.java.softwaretools,comp.os.linux.development.system
Subject: Re: Need your recommendation for a full-featured text editor
Date: Thu, 12 Apr 2001 10:53:23 -0400
Paul Shirley <[EMAIL PROTECTED]> wrote:
><[EMAIL PROTECTED]> writes
>>Syntax highlighting is useful for NOVICE programmers.
>>Most experienced programmers have used one-color text
>>for program code for years...
>...although the ones that earn a living at it mostly side with the
>novices.
I think there's probably a UNIX/realtime - "enterprise" divide
here. Quick straw poll -- how many UNIX programmers use 1 colour?
--
Ken Tough
------------------------------
From: Phillip Lord <[EMAIL PROTECTED]>
Crossposted-To:
24hoursupport.helpdesk,alt.comp.shareware.programmer,comp.editors,comp.lang.java.help,comp.lang.java.programmer,comp.lang.java.softwaretools,comp.os.linux.development.system
Subject: Re: Need your recommendation for a full-featured text editor
Date: 12 Apr 2001 16:06:17 +0100
>>>>> "Thore" == Thore B Karlsen <[EMAIL PROTECTED]> writes:
Thore> I _don't_ want the additional functionality! It's nice to
Thore> have it, but as options, not as part of the standard
Thore> package. I don't want 47MB worth of crud that I'm never going
Thore> to use.
Thore> Emacs should stick to being a text editor, not an application
Thore> environment.
>> I like it as an application environment.
>>
>> The idea of shipping a minimal emacs has been suggested for a
>> while, on top of which you could add just the packages you use. I
>> think that this would be a good idea, but I don't think that the
>> woman power is there to do it at the moment. To be honest I doubt
>> that it ever will be. 47MB of disk space is just not a lot these
>> days.
Thore> Sigh..
Thore> So it's not a lot. It's still more than even Microsoft
Thore> Office. It's more than DJGPP.
Thore> That you in most cases have lots of space doesn't justify
Thore> wasting it. I just don't buy into that philosophy.
Neither do I.
As I said I agree with you that reducing the amount of space
that emacs takes would be a nice idea. The core would probably fit
into 10M or so I think.
But at the moment emacs does not have an advanced package
manager (XEmacs has something of the sort, but I don't know how good
it is), which is what it would really need, as peoples favourite
packages vary pretty widely.
Saving space is generally a good idea, but there is a cost
benefit pay off here. Writing a package manager, producing a minimal
emacs, and packing up many of the current standard packages as
optional ones, would be a lot of effort. Reducing the disk space
requirement is I am afraid not as much of a pressing priority as it
once was. It is a good idea, and would be a good thing, but at the
moment its not that likely.
Phil
------------------------------
From: Craig Kelley <[EMAIL PROTECTED]>
Crossposted-To: comp.theory,comp.arch,comp.object
Subject: Re: Blame it all on Microsoft
Date: 12 Apr 2001 09:10:08 -0600
Jerry Coffin <[EMAIL PROTECTED]> writes:
> In article <LX7B6.2025$[EMAIL PROTECTED]>, amolitor-at@visi-
> dot-com.com says...
>
> [ ... ]
>
> > As I assume lots of people will tell you, there is no technical
> > reason why garbage collected systems cannot outperform explicitly
> > allocated systems.
>
> Actually, there's quite a bit of reason. At the very least every GC
> has to carry out essentially the same operation for marking objects
> that are still in use. This is linear on the number of objects
> currently visible in the system. Since you have to visit each object
> that's visible at least once to mark it as in use, there's
> essentially no getting away from it remaining linear.
You needn't walk over all objects; only those that are candidates for
losing scope; as soon as scope changes you only need to check those
that are visible in the current scope; the same applies with
intentional reference death, you only need to check the current
reference and it's children. Yes, this does mean you could have to
walk over all objects -- but more often it doesn't mean anything of
the sort.
[i slipped up and used the term 'object' when i wanted to keep it
generic; but it seems appropriate]
> Releasing memory manually can be done in constant time, regardless of
> the number of objects active in the system. In most cases you want
> to coalesce free space with adjacent free blocks, but this can quite
> easily be done in logarithmic time, and is often lazily, so it need
> only be done under roughly the same circumstances as GC would happen,
> when an attempt at allocating fails. In this case, the coalescing
> becomes approximately linear on the number of free blocks.
>
> In the end, comparing the two is extremely difficult even at best,
> but one thing becomes quickly apparent: with most GC systems, speed
> tends to depend primarily on the number of objects currently in use,
> while in explicit memory management, speed tends to depend on the
> number of objects that have been freed. Therefore in some
> situations, there's VERY good reason that GC can end up a lot slower.
Probably true, but we heard the same things about high-level (and 4GL)
languages back in the 80s and early 90s. People avoided C like the
plauge because it was so slow at certain things; it could never be as
fast has hand-crafted assembly. Today, it's difficult to beat the
compiler with your own assembly code -- and now people ususally use it
in cases where they don't want the compiler to optimize away some
special hardware property that they are accessing.
Memory managment is a mundane task that will eventually be handled
automatically every time, in every commonly-used language.
> > If the Java run-time people didn't bother to read the
> > extremely extensive literature on GC, well, that's another issue. That's
> > an implementation issue, to be exact.
>
> From what I've seen, it IS true that quite a few Java VMs _seem_ to
> have been written in complete ignorance of the 30 years or so of work
> that's been done on garbage collectors for Lisp, Smalltalk, Self,
> etc. At the same time, the speed of GC tends to depend heavily upon
> object usage. Oddly enough, the fact that Java looks like C and is
> probably attracting lots of programmers who aren't accustomed to
> having a GC around is likely to work against the speed of GC.
Bingo.
> When you're using GC, creating a temporary object to do something
> like returning a value tends to be _cheap_. When you're using
> something like C or Pascal, it tends to be expensive, so quite a few
> former C and Pascal programmers tend to shy away from it even when
> it would work perfectly well. In doing so, they often create
> situations that make even a really well written GC substantially
> less efficient than it could otherwise have been.
I think we're violently agreeing on this issue; I don't see GC as a
bad thing in Java 2.0 -- it had some problems with Sun's Java 1.0 JVM
to be sure, but those were implementation issues coupled with too much
beautiful architecture in the wrong places of the language (when was
the last time you saw a garbage man wearing a dinner suit?).
--
It won't be long before the CPU is a card in a slot on your ATX videoboard
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: [EMAIL PROTECTED] (Grant Fischer)
Subject: Re: So much for modules in Linux!
Date: Thu, 12 Apr 2001 15:20:37 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Thu, 12 Apr 2001 15:39:52 +0100, Pete Goodwin
<[EMAIL PROTECTED]> wrote:
>> Did you notice that the dhclient isn't named S05* to start
>> with? The intent is to move the dhclient after the network
>> startup, which I thought was exectly what you're talking
>> about.
>
> I moved dhcp to boot.local which I thought would run after everything
> else. It did but it still ran before the network modules were loaded.
>From boot.local:
#
# Here you should add things, that should happen directly after booting
# before we're going to the first run level.
#
Networking isn't started until run level 2 or 3. Try doing what
the article suggests; move the S02dhclient to say S06dhclient
(putting it after S05network in the startup.) You'll need to
do it in both the rc2.d and rc3.d directories to cover your
contingencies.
This assumes that their old advice is still helpful; they may
have changed the mechanism between 7.0 and 7.1. Their advice
may also depend on the static IP being on eth0.
The network modules should be loaded on demand when the scripts
configure them using ifconfig.
--
Grant Fischer (gfischer at the domain hub.org)
------------------------------
From: Rob Robertson <[EMAIL PROTECTED]>
Crossposted-To:
misc.survivalism,alt.fan.rush-limbaugh,soc.singles,alt.society.liberalism,talk.politics.guns
Subject: Re: Communism, Communist propagandists in the US...still..to this day.
Date: Thu, 12 Apr 2001 11:22:18 -0400
Reply-To: [EMAIL PROTECTED]
silverback wrote:
>
> Rob Robertson wrote:
>
> >silverback wrote:
> >>
> >> Rob Robertson wrote:
> >
> ><etc,... snip>
> >
> >> >> >> >> >What part of Islamic theology specifies the vertical integration of
> >> >> >> >> >industry within the country.
> >> >> >> >>
> >> >> >> >> fascism is corporate rule dummy. Maybe you should educate yerself and
> >> >> >> >> find out how many of those Iranian corporations are controlled by the
> >> >> >> >> religious allyotahs.
> >> >> >> >
> >> >> >> > So if the Ayatollahs who constitute the ruling *state* power control
> >> >> >> >those Iranian corporations, how does that support your "fascism is
> >> >> >> >corporate rule" thesis, Glen?
> >> >> >>
> >> >> >> [Scott Erb = disingenuous fraud] is playing dum.... err forgot who I was
> >> >> >> replying to there folks excuse me. [Scott Erb = disingenuous fraud] doesn't
> >> >> >> have to play dumb he is dumb. Gee [Scott Erb = disingenuous fraud] you
> >> >> >> might do a little checking and when you do you will find that the
> >> >> >> Ayatollahs own and control those corporations.
> >> >> >
> >> >> > Right. Fascism is characterized by the *state-directed* control of
> >> >> >the economy, not societal control by corporations. You and I have had
> >> >> >this conversation several times in the recent past, but only now do
> >> >> >you recognize that it is the state rulers controlling businesses that
> >> >> >constitutes fascism, not rule by corporations.
> >> >>
> >> >> wrong again [Scott Erb = disingenuous fraud] under fascism the corporations
> >> >> are the government.
> >> >
> >> > Oh dear, this really is getting 'complex' now. Up above you say that "those
> >> >Iranian corporations are controlled by the religious allyotahs [sic]", which
> >> >to me means that the *political* leaders (the allosaurus's, or whatever) are
> >> >in control of the *business* leaders, not the other way around. That's what
> >> >I'd said to you many times, as noted in the replays which you deleted (leftist
> >> >censorship in action again, eh Glen?).
> >>
> >> it figures that the simple minded lying [Scott Erb = disingenuous fraud] would
> >> be unable to understand.
> >
> > I've explained it to you many, many times over the past few months, Glen, and
> >I've explained and paraphrased it several times in this thread (and the proof
> >of that is in this very post). I think that you're just upset because I found
> >you eventually agreeing with me that fascism is the state-directed control of
>
> wrong again [Scott Erb = disingenuous fraud]. The Nazis allowed the corporations
> to write the laws.
Do the Ayatollahs allow Iranian companies to write the laws, Glen?
> >the economy as you note with your 'Ayatollah Fascism' comment above, and that
> >I would connect that back to your fascist call for the government take-over of
> >the electricity industry in California.
> >
> > Not very good advertisement for your book _The Nazi Hydra_, Glen!
> >
> >> > You repeat the claim that "the Ayatollahs own and control those corporations",
> >> >and at the risk of being redundant and repetitious, I'd note that Ayatollahs
> >> >are the *political* leaders. Now you contradict *yourself* (while leaving in
> >> >the evidence of the contradiction in your response, oddly enough) when you
> >> >claim that "the corporations are the government".
> >> >
> >> >> > Do you remember this exchange, Glen?;
> >> >>
> >> >> i remember yer a lying [Scott Erb = disingenuous fraud].
> >> >>
> >> >> remainder of this lying [Scott Erb = disingenuous fraud]'s tripe snipped,.
> >> >> He doesn;t have a point and is prone to lying.
> >> >
> >> > Au contraire, mon frere. The point I've been trying to make all along has
> >> >been that fascism is the state-directed *control* of the economy as opposed
> >> >to communism's state *ownership* of industry. You echoed my sentiment above
> >> >with your 'fascist Iran' claim, and I'd also note the irony of your calls
> >> >for the government takeover of the California electricity industry, which
> >> >is much more like fascism than my calls for laissez-faire capitalism.
> >> >
> >> > Thank you for the pleasant discussion.
_
Rob Robertson
Got freedom?
http://www.lewRockwell.com
http://www.free-market.net
http://www.mises.org
http://www.hazlitt.org
http://www.antiwar.com
http://www.unionsquarejournal.com
http://www.libertocracy.com
http://www.unknownideal.com
http://www.self-gov.org
http://www.lp.org
------------------------------
From: [EMAIL PROTECTED] (Chad Everett)
Subject: Re: So much for modules in Linux!
Reply-To: [EMAIL PROTECTED]
Date: 12 Apr 2001 10:12:26 -0500
On Thu, 12 Apr 2001 06:06:25 GMT, Pete Goodwin <[EMAIL PROTECTED]>
wrote:
>Chad Everett wrote:
>
>> I don't get it. The kernel and modules are distro independent. If there
>> is some ordering of modules issue, why don't you just modify your system's
>> startup scripts to load things in the order that works for you.?
>
>I tried moving DHCP to boot.local as suggested by SuSE support.
>Unfortunately, it is still too soon.
>
As boot.local says:
#
# Here you should add things, that should happen directly after booting
# before we're going to the first run level.
and as /etc/inittab says:
# runlevel 0 is System halt (Do never use this for initdefault)
# runlevel 1 is Single user mode
# runlevel 2 is Local multiuser without remote network (e.g. NFS)
# runlevel 3 is Full multiuser with network
# runlevel 4 is Not used
# runlevel 5 is Full multiuser with network and xdm
# runlevel 6 is System reboot (Do never use this for initdefault)
So you don't want to use boot.local. You want to put your DHCP startup
in the correct order with your other network startup stuff in the run
level 3 startup scripts.
------------------------------
From: [EMAIL PROTECTED] (hack)
Crossposted-To: comp.theory,comp.arch,comp.object
Subject: Re: Blame it all on Microsoft
Date: 12 Apr 2001 15:24:53 GMT
In article <[EMAIL PROTECTED]>,
Jerry Coffin <[EMAIL PROTECTED]> wrote:
>In article <LX7B6.2025$[EMAIL PROTECTED]>, amolitor-at@visi-
>dot-com.com says...
>
>[ ... ]
>
>> As I assume lots of people will tell you, there is no technical
>> reason why garbage collected systems cannot outperform explicitly
>> allocated systems.
>
>Actually, there's quite a bit of reason. At the very least every GC
>has to carry out essentially the same operation for marking objects
>that are still in use. This is linear on the number of objects
>currently visible in the system. Since you have to visit each object
>that's visible at least once to mark it as in use, there's
>essentially no getting away from it remaining linear.
Doesn't generational GC avoid some of this, in that only "old" objects
need to be scanned? With proper cooperation between VMM and GC (on systems
where that is possible) GC can be performed largely in the background.
Also, the fact that Java does not expose the representation of pointers
permits greater flexibility in GC mechanisms than might otherwise be the
case (e.g. pointer/object segregation, lookaside structures etc.).
Funnily enough, this comes from a person who actually believes in explicit
storage management (via address/length, not via address alone as for malloc),
relying on very fast storage allocators of the type you had in mind (e.g.
based on cartesian trees, exploiting double-ended allocation for storage of
different lifetime characteristics, and using extremely fast private sub-
allocation for small fixed-size pieces with known lifetime properties).
This tends to exploit the maximum knowlegde available for a particular
program or application. It is however more difficult to write.
If GC could easily be combined with explicit allocation/release (or, at
least, useful-hint-based release), we might have the best of both worlds.
That seems to be difficult to achieve, however, because of the need to
distinguish hint from guarantee.
Michel.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.theory,comp.arch,comp.object
Subject: Re: Blame it all on Microsoft
Date: 12 Apr 2001 15:27:09 GMT
In comp.theory Erik Funkenbusch <[EMAIL PROTECTED]> wrote:
> "Chad Everett" <[EMAIL PROTECTED]> wrote in message
>> Borland makes grep IDEs...much better than MS Visual Studio that's for
>> darn sure. Borland compilers are much superior to Microsoft's.
> Actually, Borland's IDE is quite buggy and error prone. It lacks key
> useability features. For instance, there is no way to easily manage all the
> windows opened up by BCB, and there isn't any way to split code windows to
> see two parts of the code at the same time. The only way to do it is to
> open two seperate edit windows.
I've used Borland before, and I'm just now starting to explore MS
Visual Studio. I agree about Borland -- I didn't like the user
interface at all... MS Visual Studio is marginally better, but I
can't see what all the fuss is about. The editor is pretty crappy,
and the debugger is almost useless except for very simple "watch the
variables" type debugging. About the only positive point was the
built-in help in the editor -- popping up a standard function
prototype when you start typing it is pretty nice...
Honestly, I'm not sure why people like an integrated environment. My
Unix/Linux environment is MUCH more powerful, even though the pieces
are separate. Emacs makes the MS editor look like junk, and DDD is
actually a useful debugger, unlike the MS version. I know there are
better editors you can add on -- for example, I've seen ads for one
(called something like "CodeWright"??) that looks pretty nice, but who
wants to pay an extra $300-something for an editor to add on to your
development environment (already costing big bucks). Of course, you
could get emacs for Windows, and I know a lot of people that do that,
but then you're really not talking about the MS environment any
more...
--
Steve Tate --- srt[At]cs.unt.edu | Gratuitously stolen quote:
Dept. of Computer Sciences | "The box said 'Requires Windows 95, NT,
University of North Texas | or better,' so I installed Linux."
Denton, TX 76201 |
------------------------------
From: "tony roth" <[EMAIL PROTECTED]>
Subject: hmm getting tired of this!
Date: Tue, 10 Apr 2001 13:42:48 -0700
hate to be a killjoy but another one bites the dust!
http://www.microsoft.com/technet/migration/hotmail/default.asp
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.advocacy.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Advocacy Digest
******************************
