Linux-Misc Digest #191, Volume #19               Fri, 26 Feb 99 15:13:10 EST

Contents:
  Re: Linux is not even in Windows 9X's class. (Matthias Warkus)
  Re: Any JDK for Linux? (Jet)
  Re: Best Free Unix? (why FreeBSD?) (Stephen J. Tremblett)
  Re: More bad news for NT (Larry)
  Re: Porting svgalib (Linux) to FreeBSD (Alexander Viro)
  Re: enlightenment themes? (Matthias Warkus)
  Re: Pentium III Boycott and survey info ("~The Seventh Sign~")
  Re: PPP for network connection?? ([EMAIL PROTECTED])
  Re: Web access to email needed, Re: HTML and/or Java email client for LInux??? 
("Mark Stoneman")
  Re: More bad news for NT ("Simon")
  Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?) ("David A. Frantz")
  Re: Red Hat's sick sense of humor (support) (Kyler B. Laird)
  Diald on Sparc Redhat 5.2 (Jeff Dearmin)
  Re: Is Linux better than DOS/Win3.1 on 486? (Greg H.)
  Re: Linux/FreeBSD compatability (Was Re: Best Free Unix? (why FreeBSD?)) (Robert 
Sexton)
  Re: Any JDK for Linux? (Alex Holloway)

----------------------------------------------------------------------------

From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To: alt.linux,alt.os.linux,comp.os.linux.advocacy
Subject: Re: Linux is not even in Windows 9X's class.
Date: Fri, 26 Feb 1999 17:50:29 +0000
Reply-To: [EMAIL PROTECTED]

It was the Thu, 25 Feb 1999 18:44:23 -0800...
..and Ryan Cumming <[EMAIL PROTECTED]> wrote:
> Matthias Warkus wrote:
> 
> > > 1. Better GUI
> >
> > Hmph. Depends on how you define a GUI and what kind of GUI you are
> > using on Linux.

> I dont't think any UI in Linux even touches Windows ease of use.
> Some of the new fangled desktop enviroments (Gnome, KDE, etc.) come
> close, but MS has done a good job of allowing the newbie be able to
> use the OS right away,

Hmph. Have you really actually used KDE or Gnome?

> > > 2. More software
> >
> > Certainly.
> >
> > > 3. More Hardware support
> >
> > Yes, but oh so crappy drivers...
> 
> I rather have a Winmodem with crappy drivers than a useless card in Linux

We all agree that Winmodems are evil and a moronic idea. They just
don't count.

[schnibble]
> > > 6. One word: Microkernel
> >
> > Can mean a performance penalty.

> Waiting 15 minute for my kernel to compile for a new driver is a
> performace penilty

You don't happen to have heard of loadable device modules, have you?

> > > 7. No mounting
> >
> > This is a disadvantage as it easily can mess up removable media and it
> > means a huge file system throughput penalty.
> 
> > > 8. Better file locking
> >
> > I don't know jack about this one.

> For the ill-educated, Unix's file locking has being crappily added
> on since its original design (like multithreading). Therefor,
> instead of getting an app to lock its files using one mechinism, it
> may break portablity to half of the Unixes. So many apps don't
> bother. I've upgraded X while it was running (stupid dea, BTW) and
> do to crappy file locking it went down in flaming glory instead of
> warning me.

Automatically locking a file when a process writes to it is even
crappier in my book.

mawa
-- 
When I went to the theatre, recently, I saw a woman who wore nothing
but a transparent catsuit over a black body that was cut out down to
the navel. I don't know whether she felt naked. She sure looked naked.
                                                               -- mawa

------------------------------

From: Jet <[EMAIL PROTECTED]>
Crossposted-To: comp.lang.java.setup,comp.lang.java.api
Subject: Re: Any JDK for Linux?
Date: Fri, 26 Feb 1999 11:07:55 -0800

You might want to write to solaris_jdk12_feedback at Sun dot COM.

J

James Tam wrote:
> 
> Hi,
> 
> I am looking for a JDK for Linux.
> Anyone know anything?
> 
> Thanks
>                 James

-- 
The human brain is the most incredible structure in the universe.
Yeah, but look what came to that conclusion.
email me at jetgal at earthlink dot net

------------------------------

From: [EMAIL PROTECTED] (Stephen J. Tremblett)
Crossposted-To: 
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Best Free Unix? (why FreeBSD?)
Date: 26 Feb 1999 18:39:31 GMT

John Hasler <[EMAIL PROTECTED]> writes:

>Stephen J. Tremblett writes:
>> I always assumed that was because the rpm/deb/etc packages are actually
>> were contributed by the maintainer to the FTP site.  There are hundreds
>> of ftp sites that have Linux packages, and no real control over them, so
>> maintainers often have more current versions on their personal web pages.

>This is not true of Debian.

>> The FreeBSD ports is managed a little differently in that there is ONE
>> central 'contrib' place to get the official ports, thus the maintainers
>> know that is the place to put the package.

>And there is ONE official central place to get official Debian packages.
>All 2200+ of them.

I must admit I was mistaken there.  I was thinking more of rpms and tgz
packages where sunsite is good, but by no means a central site.  The
'deb' popped out by mistake.


--
-=> Steve Tremblett       
-=> Memorial University Computer Science
-=> [EMAIL PROTECTED]; www.cs.mun.ca/~sjt 

------------------------------

From: [EMAIL PROTECTED] (Larry)
Crossposted-To:  alt.destroy.microsoft,comp.os.linux.advocacy,alt.linux
Subject: Re: More bad news for NT
Reply-To: [EMAIL PROTECTED]
Date: 25 Feb 1999 18:16:15 GMT

On Tue, 23 Feb 1999 18:55:07 -0600, Jon Wiest <[EMAIL PROTECTED]> wrote:
>Jason Clifford wrote in message ...
>>On Tue, 23 Feb 1999, Jon Wiest wrote:



>I'd say the whole process of getting it going is pretty stupid.  Heck
>setting up PPP requires me to read and absorb a 50 page HOWTO.  That's okay,
>I like learning, but gawd, doesn't everybody want PPP?

Gee, never having used PPP before on any machines that I run, it took me a
grand total of 2 hours to rebuild the kernels for ppp operation, install
pppd, and edit the dip configuration files for 4 machines. I didn't read
anything on the installation or operation of ppp but I did have to read the
man page on dip.

SO here's the 50 page HOWTO in a nut shell.

1) Build the kernel for ppp  operation.  
2) install pppd, if not already there.
3) Read the dip man page.
4) Build your dip script file to do the dialing and logging in.

And VIOLA, you're on the net.


A tip for those trying to install dip...
Login manually from a comm program first. Minicom works great.
then record on a piece of paper every key press and action you have to
make to get logged in correctly, then write your script according to your
actions used to log in manually. 
The example in the man page is ok but not great. A number of changes usually
need to be made to that script according to what your ISP activity is when
you dial up. My script file was much shorter and less complicated than the
example in the man page.

I would be glad to write a dip-script-HOWTO that is in fairly plain english.
Does anyone know where I could put it for ftp download though? My machine
isn't connected at all times right now or I would just put it here.






------------------------------

From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To: comp.unix.bsd.freebsd.misc
Subject: Re: Porting svgalib (Linux) to FreeBSD
Date: 26 Feb 1999 14:09:29 -0500

In article <[EMAIL PROTECTED]>,
Donn Miller  <[EMAIL PROTECTED]> wrote:
>If nobody objects, I think I'll start working on porting svgalib over to
>FreeBSD. I've seen posts where people have said no, svgalib is too low
>level to port over to FreeBSD.  It should actually be pretty
>straightforward.  
        SVGAlib was bad mistake on Linux and propagating it even further
instead of proper deep-sixing... Shrug.

>The one speed bump on the porting highway would be where svgalib interacts
>with the Linux console, which differs from the FreeBSD console code.  So
>at the very least, the code where svgalib interacts with the Linux console
>will have to be modified in order to support FreeBSD's console,
>which at the present is syscons. Unless svgalib is doing some low-level
>Linux kernel interaction, porting it to FreeBSD should be pretty 
>straightforward.
        SVGAlib basicly asks the console to f*ck off and let it manage the
whole thing. From that point it acts in raw mode and talks with the VGA
directly. I.e. does what the X does.

>Pros:
>
>* There's a lot of existing software that works with svgalib
>* Won't have to wait for FreeBSD's libvgl to be finished
>
>Cons:
>
>* We don't want to be copying a lot of Linux stuff for use on FreeBSD.
>The two OSes are unique, and to start borrowing ideas from one OS to use
>on the other would destroy that uniqueness.

        Huh??? "borrowing ideas" is perfectly fine. Now, borrowing mistakes...

>I saw where FreeBSD might have kernel threads, which would be "borrowed"
>from Linux threads.  Does anyone know if Linux has had any ideas borrowed
>spefically from FreeBSD, like drivers, etc.?  I don't know of any...

        D'oh. Dunno about FreeBSD per se, but lots of ideas are borrowed
from 4.{4,3}BSD (or happen to be common for all Unices). Damnit, old VFS comes
back to Sun works. Quota is derived from Melbourne one, just as in BSD.
Ext2fs shares a lot with UFS and has the common ancestor with it - FFS.

-- 
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid.  Get yourself a better computer" - Dilbert.

------------------------------

From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To: comp.windows.x,comp.linux.x
Subject: Re: enlightenment themes?
Date: Fri, 26 Feb 1999 17:45:47 +0000
Reply-To: [EMAIL PROTECTED]

It was the Thu, 25 Feb 1999 17:31:45 -0500...
..and Ross Vandegrift <[EMAIL PROTECTED]> wrote:
> "David A. Frantz" wrote:
> > 
> > Well I'm using E on and off, often switching back to FVWM and find it very
> > useful.    The thing is you have to be willing to give it a break after all
> > this is developers release 15.    Its not like its the 20th version of
> > EMACS, you can't even call it a new Born yet.   IT certainly is the coolest
> > windows manager going, I would love to see a stable release of E.    Just a
> > matter of time now.
> 
> Oh, I totally know, and wasn't downing it at all.  I was just
> commecnting on its *current* state.  As a matter of fact, I CVSed the
> new version last night and compiled it, and it is quite a bit better
> than the older release I had running.  The only problem:  Where has the
> E control tollbar on the lefthand side of the screen gone?

It has been discontinued, since now, there is the ultra-nifty e-conf
capplet for Gnome to control it. I don't know whether it really needs
the Gnome Control Center; perhaps it can run standalone.

mawa
-- 
As soon as a drunk driver sees a roadblock, what does he do?  He makes
a U-turn and *bang*, another roadblock related traffic accident!
                                       -- Jimmy Tingle, on:  Heat, NPR

------------------------------

From: "~The Seventh Sign~" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy,comp.os.linux.hardware
Subject: Re: Pentium III Boycott and survey info
Date: Fri, 26 Feb 1999 14:07:31 -0500
Reply-To: "~The Seventh Sign~" <[EMAIL PROTECTED]>


You might also try my sig!
--
~The �eventh �ign ~

Protect privacy, boycott Intel: http://www.bigbrotherinside.org

Boycott Swintel <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
: Pentium III chip with the individual serial number that can
:track your web surfing and buying habits can now have the ID number
:turned on and off by software.  Following some links I found the
:www.fightdivx.com website and noticed that they have a Intel Boycott
:page with links, quotes and info on why you should boycott the
:invasion of privacy Pentium III chips. Just like everyone suspected,
:the ID number can be taken without a customers knowledge. Just like
:cellular phone fraud, once someone has your unique ID number, they
:could pose as you on the internet.  Do not be fooled by reports that
:this problem is fixed because Intel disabled this feature by software
:on their up coming chips.  Information is power. They want to know
:your surfing and buying habits. That is what this is all about. Here
:is the link to the page with the boycott info and links.
:
:http://www.fightdivx.com/intelboycott.htm
:
:Also you will find a Boycott Intel screen saver and banner on their
:page above. Spread it around.
:
:
:
:
:Take the Pentium III Boycott Survey
:http://mail.infotrieve.com/isurvey/index.cfm?vendorid=6045&formid=F0006045
:
:
:


------------------------------

From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.setup,linux.redhat.misc
Subject: Re: PPP for network connection??
Date: Fri, 26 Feb 1999 18:55:36 GMT

On Fri, 26 Feb 1999 13:08:02 -0500, "Craig Shields"
<[EMAIL PROTECTED]> wrote:

>Hello,
>I am trying to get connected to the internet and do email through my network
>connection at work and I can't seem to "see" anything beyond my LAN.  I can
>ping any machine on the LAN, but can't get mail or PPP internet
>access to work.  I am yet another immigrant to Linux from Win95/98/NT and
>must be
>missing something.  Could someone PLEASE help me translate my TCP/IP
>settings in Win98 to Linux??   (i.e.,  IP, DNS, Gateway, WINS)  I've read
>many FAQ's and just seem to be missing something.
>thanks,
>Craig Shields
>
put the dns numbers in /etc/resolve.conf
ignore wins (wins is a DNS type of server on NT for WINDOWS Only
clients)

are you dialing in to work for internet??? or are you actually on the
lan that has internet access? or do you dial an ISP from work?
your setup is unknown.
if you are on your lan and dial an ISP your need to change your
default route.
route del default
route add default ppp0

this should work IF the rest of your routeing table is setup corectly.
mine looks like
# route
Kernel IP routing table
Destination   Gateway      Genmask       Flags Metric Ref    Use Iface
208.249.107.7   *         255.255.255.255 UH    0      0        4 ppp0
localnet        *         255.255.255.0   U     0      0       52 eth0
loopback        *           255.0.0.0       U     0      0       12 lo
default   208.249.107.7   0.0.0.0         UG    0      0       96 ppp0

where 208.249.107.7 is my ISP


------------------------------

From: "Mark Stoneman" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: Web access to email needed, Re: HTML and/or Java email client for LInux???
Date: Fri, 26 Feb 1999 21:13:14 +0200

I don't know much, but I *do* know this - check out
http://www.obsidian.co.za 's intranet demo.

William Burrow wrote in message ...
>On 18 Feb 1999 18:25:08 GMT,
>Cameron Spitzer <[EMAIL PROTECTED]> wrote:
>>>>Hi, is there some sort of email client that I can use in a web browser?
>>>>I know its not ideal, but I am looking for a VERY lightweight client.
>>>>
>>>>Basically I want to see something  kinda like HotMail.




------------------------------

From: "Simon" <[EMAIL PROTECTED]>
Crossposted-To: alt.destroy.microsoft,comp.os.linux.advocacy,alt.linux
Subject: Re: More bad news for NT
Date: Thu, 25 Feb 1999 19:23:53 -0000

Oh dear I seem to have started some sort of tit-for-tat debate over the
merits of Linux over NT and vice-versa. I don't really understand the fuss.
I mean if it works for you then use it.

It just seems a bit odd that people/companies will pay for a product (NT) in
the traditional sense of the term when they could get (judging by the
reaction) a similar result (maybe a bit worse, maybe a bit better) from a
low cost/no cost option.

Why pay hundreds/thousands of Sterling/Dollars for an OS and all associated
products when you could have an easy to install simple to operate OS
(Linux). Its not even as if you get any real support from Microsoft on NT
Server unless you pay more and sign a support contract.

I just do not understand the logic. Still maybe the NT people are working on
the theory that "No one ever got fired for buying IBM". I just think its
ironic that the modern day IBM is Microsoft - interesting that !

I do agree that MS will add more features to NT and associated products
making them more and more attractive. It just seems sad to spend good money
on software that is put together in a shoddy manner. If you read a book
called "Inside Intel" by Jackson you will find reference to Intel's distress
that Microsoft never made good use of new CPUs that they had to make break
throughs in technology to create. One section of the book sees an Intel
engineer dis-assemble a section of Excel code only to discover a small
coding mistake. He then informs Microsoft of this issue and says that a
change to one of the loops will see an eight fold increase in the apps
performance. Microsoft reply that they have no interest in this finding and
are only interested in adding more features and selling more software.

If this approach appeals to you then my advice is buy Microsoft after all
"whoever got fired for buying Microsoft NT".

Simon



Craig Kelley wrote in message ...
>[EMAIL PROTECTED] (Eric Lee Green) writes:
>
>> Frankly, I thought that was a very flattering article. Think about it.
>> Linux with one processor disabled is faster than NT with both
>> processors going on dynamic page generation. And frankly, that's the
>> benchmark that matters most to me. I can just about saturate a T1 with
>> our current web server on static page generation (our current web
>> server is a Pentium 166 with 64mb of RAM and a 1gb IDE hard drive),
>> but dynamic web generation is a different story.
>
>Funny, that's *our* main server as well.  A Pentium5/166 with 64MB of
>RAM.  It serves out web pages and is the logon/filesharing server for
>about 80 machines with 300+ users.  It also handles all of our email
>(both SMTP *and* POP/IMAP).  It runs RedHat 5.2 with a 2.0 kernel.
>
>--
>The wheel is turning but the hamster is dead.
>Craig Kelley  -- [EMAIL PROTECTED]
>http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block



------------------------------

From: "David A. Frantz" <[EMAIL PROTECTED]>
Crossposted-To: 
alt.os.linux.slackware,comp.os.linux.development.system,comp.os.linux.hardware,comp.os.linux.setup
Subject: Re: Overclocking (was: Re: K6-2 and Linux, Are there any Bug?)
Date: Thu, 25 Feb 1999 18:42:00 -0500


Bugs Bunny wrote in message <[EMAIL PROTECTED]>...
>
>Yair Paz wrote in message <[EMAIL PROTECTED]>...
>>Bugs Bunny wrote:
>>> you can overclock it it will run hotter and finally burn out like all
>chips
>>> that are over clocked it is a gamble
>>>
>>> do it aown risk
>>Bulls. some chips are meant to beoverclocked - the P2-300 for example,
>>in the last 6 months of it's manufacture, it was made with a P2-350 core
>>hard wired to be 300. you can easily un-wire it (OK , not so easily). or
>>the Celeron 300A which overclocks to 450 w/o a glitch just by changing
>>the FSB speed, or the K6-2 300 which can overclock to 375 w/o additional
>>cooling, or any P5 MMX chips which can usually be overclocked to about
>>20% more, if you have the right board, w/o needing for extra hardware
>>(coolants) or voltage tweeking. nowdays, you can even overclock
>>HardDisks.
>>Oded
>
>the only chips i hear of that can be is celeron but besides that.... the
>chip is not made to be over clocked..... if it was supposed to be over
>clocked it would RUN at THAT SPEED not 50-100 MHz slower



Actually this is not true at all it would mean that Intel had no plans for
upgrading the chip or improving the process.    Without going to the advance
processess in the first place for the Celeron, Intel would not have had the
oportunity to quickly add the on chip cache that makes the new Celerons such
standouts.

Economics come into play as the "faster" processes are also the processes
that produce the smaller dies.    This means of course more Celerons per
wafer which means cheap.    Since the product was initially targeted at the
low end of the market there is no reason to believe that the Celeron wasn't
designed from the beginning to run faster.    Intel just tried to rake a
little money in from the sucker market.

Dave




------------------------------

From: [EMAIL PROTECTED] (Kyler B. Laird)
Subject: Re: Red Hat's sick sense of humor (support)
Date: 26 Feb 1999 19:07:01 GMT

[EMAIL PROTECTED] (Martin ) writes:

>>So...I tried ordering the Red Hat Linux/Intel
>>package from Red Hat.  I told the person that I
>>didn't really want the package, I just wanted the
>>registration number.  No dice.  I'd have to wait
>>to receive the box.

>If you purchase the RedHat boxed set - which doesn't cost much more than $30 - 
>you get about one month's installation support included - seems to me that 
>your problem should fall within that scope...

The point that you chose to ignore is that I didn't
mind paying for the support.  I didn't even mind
paying for the entire boxed set.  I did *not* want
to wait for it to arrive, though.

--kyler

------------------------------

From: Jeff Dearmin <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Diald on Sparc Redhat 5.2
Date: Thu, 25 Feb 1999 17:41:21 -0500

I'm trying to get "diald" to work on a Sparc 5 running Redhat 5.2.   I
got the laatest source (0.16.5), and after commenting out undefined
signals from "diald.c", got it to compile successfully.  When the
program runs, it generates a LOT (thousands) of syslog messages saying
"Stray signal 10 ignored", and does not work.  A little investigation
revealed the offending line of code is in "firewall.c" in the
"check_firewall" routine:

 if (FW_TCP_STATE(term->offset))
                v = (lflags.tcp_flags >> term->shift) && term->mask;
            else
                v = (ntohl(*(int *)(&(FW_IN_DATA(term->offset)?data:pkt)

                                  [FW_OFFSET(term->offset)]))
                    >> term->shift) & term->mask;

I changed the code to:

if (FW_TCP_STATE(term->offset))
                v = (lflags.tcp_flags >> term->shift) && term->mask;
            else
                /*
                v = (ntohl(*(int *)(&(FW_IN_DATA(term->offset)?data:pkt)

                                  [FW_OFFSET(term->offset)]))
                    >> term->shift) & term->mask;
                */
                if (FW_IN_DATA(term->offset))
                {
                        v = (ntohl(data[FW_OFFSET(term->offset)])
                          >> term->shift) & term->mask;
                } else {
                        v = (ntohl(pkt[FW_OFFSET(term->offset)])
                          >> term->shift) & term->mask;
                }

The program runs and connects, does not generate the stray SIGBUS traps,
but now randomly crashes the system every few hours.  Any suggestions?
Anyone know of a Diald replaacement that does on demand PPP
connections?  I don't have a fixed IP address from the ISP, so PPPD
alone probably won't work.

Thanks!!

--
Jeff Dearmin
phone: (407)875-9480
fax:   (407)661-4715
email: [EMAIL PROTECTED]

------------------------------

From: [EMAIL PROTECTED] (Greg H.)
Subject: Re: Is Linux better than DOS/Win3.1 on 486?
Date: Fri, 26 Feb 1999 19:19:54 GMT

Walter Strong ([EMAIL PROTECTED]) wrote:

> I'm running Slack on a 486/66 with 16 meg ram on native ext2 partition 
> and it gives me excellent performance even with X going.  Just don't 
> forget to give yourself a good size swap file.  It gives better 

   Back in summer '94, I was running Slackware with X on a 486SX at 20 MHz
with 4 MB of RAM and 80 MB of harddrive space total.  It _can_ be done.
The X performance was decent, too.

   Greg H.

------------------------------

From: [EMAIL PROTECTED] (Robert Sexton)
Crossposted-To: 
comp.unix.questions,comp.unix.advocacy,comp.unix.misc,comp.unix.bsd.freebsd.misc
Subject: Re: Linux/FreeBSD compatability (Was Re: Best Free Unix? (why FreeBSD?))
Date: 26 Feb 1999 19:28:17 GMT

Donn Miller ([EMAIL PROTECTED]) wrote:
: Mike Willett LADS LDN X7563 wrote:

: > With all this talk about FreeBSD/Linux issues, I was thinking
: > about compatability issues with the two system.
: >

: Well, when I run FreeBSD, I sometimes compile "Linux-only" programs like
: iptraf, which hasn't been ported to FBSD yet.  It wouldn't compile, since it
: wanted to link with Linux specific network libs and headers which FreeBSD
: doesn't have.  So for FreeBSD, it would be nice to have certain Linux
: headers and libs in order to ease the porting from Linux to FreeBSD.  In
: fact, I think there's such a thing for FBSD, called the "Linux development
: kit".  No, not the standard Linux emulation stuff that runs on FBSD, but an
: actual kit that allows you to port Linux programs to FreeBSD and to
: cross-compile Linux programs on FreeBSD.

: > Or can we only get executable compatablity ?
: > (I assume this isn't a simple yes/no answer ? What level of compatability
: > is there and what effort is required to improve this ?)

: FreeBSD has Linux emulation.  Linux, I heard, can emulate FreeBSD through
: the IBCS compatibility module option in the kernel,  but it is old and
: untested.  I talked to other Linuxers, and they all said what's the point in
: emulating FreeBSD, since a larger subset of programs are written for Linux
: than for FBSD.  So FBSD is the OS that needs to emulate Linux, and not vice
: versa.  Hence, we have an IBCS compatibility module which doesn't work too
: much anymore (at least as far as running FBSD programs on Linux).

You're referrring to icbS2, which is the standard for AT&T system V v3,
and its derivatives (Sco, for one). It uses the COFF executable
format and the icbs2 system call interface (mostly POSIX, as I recall)

FreeBSD uses the Berkely libraries and aout or elf executable formats.

: It's a sad but true fact that there's a lot of Linux -only programs out
: there.  You very rarely see "FreeBSD-only" stuff out there.  The solution to
: this is, let's get some kind of add-on to FreeBSD that allows us to include
: Linux include files (no pun intended) and link with Linux libs on FreeBSD,
: in order to achieve greater interoperability between the two.  And let's fix
: that IBCS module so we can run FreeBSD programs on Linux. ;)

Don,
You are missing the core problem.  By definition, most things you
write for use on BSD Unix are portable to the largest portion of
the UNIX world, whether that be BSD, SunOS, Solaris, AIX, HP/UX,
Linux, Dynix, Ultrix or whatever flavor you like. Thats because BSD Unix
has really defined a great deal of the UNIX programming environment.
(I know because I spent a good bit of my career in the AT&T universe)

(I'm talking about market coverage here, not total
nosecount. ) There really aren't any FreeBSD-Only programs out
there.  Naturally, I have to declare an exception here for device
drivers and kernel extensions.  I also don't want to short shrift the
AT&T Unix heritage.  BSD is the direct descendant of the AT&T work.

Before Linux arrived, many of us were using this terrific software
from the GNU project.  Most of the software was better than what came
out of the box with a commercial UNIX.  BSD was a bit of an exception
on this front, because it evolved an improved faster than most of the
purely commercial flavors.  But I digress.  One of the most wonderful
things about the GNU stuff was that it was portable.  Portable down
deep.  The kind of portable that comes from thinking that way from the
outset.  Its not even all that hard.  You just have to code with it in
mind.

GNU was really the capstone of soemthing that went back a long way:
freely available unix software.  There is a rich heritage here, based
upon portablity.  Thats what got everybody excited.  I knew that my
software would RUN ANYWHERE.  A few tweaks of the header files, a
different libary of two, and I'm off to the races.  You can even
automate this stuff.  Hence autoconf.

I'd be willing to posit that a greater portion of BSD code is designed
compatibly with Linux than the other way around.  On one hand, BSD has
largely set the standards.  On the other, portability is part of the
culture. 

There is a lot of non-portable Linux code out there.
FreeBSD does not have /usr/include/FreeBSD, nor /usr/include/bsd, or
anything of the sort.  I suppose /usr/include/linux is a good way to
wall off the portable from the non portable, but anybody who goes in
there with any alternatives should have their fingers broken.

I'm still unsure as to why the Linux camp has made /proc into the
lazy programmers kernel interface.   I guess its easier to open a file
than to look up a system call.  Load averages are a case in point.

Anyway, back to the point.   Linux has harnessed a lot of brains, and
to some good result.  However, the goal here is to make a bigger pie.
Linux-specific software doesn't make the pie any bigger, it just
increases the size of YOUR share.  In that case somebody else loses.

In some ways, I'm glad to see Linux bring UNIX back into the
mainstream, and away from the speciality it's been in recent years.
The part that makes me sad is that so many people don't know their
history, that Linux is part of something bigger.  They see it as some
sort of crashless windows, and they write their software accordingly.

I fear that I will wake up one day, and Linux will be writ large on
the computer industry, and it won't be UNIX anymore.

\end{rant}


UNIX(TM) is a trademark of somebody, but I forget who currently owns
it :-)
--
Robert Sexton - [EMAIL PROTECTED], Cincinnati OH, USA
Kudra.com - New home of the hazardous data disposal service, 
featuring our exclusive Write Only Memory(TM) Technology!
Read the Newton FAQ! <http://www.kudra.com/newton/newton-faq>

------------------------------

From: Alex Holloway <[EMAIL PROTECTED]>
Crossposted-To: comp.lang.java.setup,comp.lang.java.api
Subject: Re: Any JDK for Linux?
Date: Fri, 26 Feb 1999 19:32:36 +0000

On Fri, 26 Feb 1999, James Tam wrote:
>Hi,
>
>I am looking for a JDK for Linux.
>Anyone know anything?
>
>Thanks
>               James

Yeah, try www.blackdown.org - they're doing the port of Sun's JDK


--
alex holloway

purpose: student
subject: computer science
place:   robinson college, cambridge
         ah281.robinson.cam.ac.uk


------------------------------


** 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 (and comp.os.linux.misc) via:

    Internet: [EMAIL PROTECTED]

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-Misc Digest
******************************

Reply via email to