[M3devel] religion, philosophy, Posix vs. Win32

[Mika Nystrom]

Well Unix compatibility is what Cygwin "is for"...  it's a little
odd to have Cygwin pretend to be Windows---if you want Windows, 
you don't need Cygwin!

The old PM3 doesn't use X for its Trestle.  It uses native Windows
things.

Note there is no contradiction between using Windows UI for Trestle
and what I said before.  Under Cygwin, you want to be able to use
X, just like on Unix.  But if M3 doesn't use it, well no matter.
Although if you want to extend wish lists, sure, it might be nice
to be able to build Trestle under X on Cygwin, too.  (I mean, why
not?  It ought to be "trivial", right?)

    Mika


Jay writes:
>--_5cf007a7-8622-47c5-8550-283b4142bb52_
>Content-Type: text/plain; charset="Windows-1252"
>Content-Transfer-Encoding: quoted-printable
>
>ok..well...some provision I guess eventually should be made for those who a=
>gree with you and Tony.
>What will come first is something else. :)
> A system that is Win32 as much as possible, EXCEPT that it uses the gcc-ba=
>sed backend, and as, and ld, and maybe throws around some forward slashes h=
>ere and there.
>=20
>I weakly propose:
>  NT386GNU be deleted
>  and NT386MINGWIN (NT386MING? I386_MINGWIN? X86_MINGWIN?)
>  and NT386CYGWIN (NT386CYG? I386_CYGWIN? X86_CYGWIN?)
> takes its place.
>=20
>but not right now.
>=20
>However, it still might be one target, and you edit the config file.
>It depends on what people define as a "target" and what it means for two "d=
>ifferent" things to be the same "target".
>The same target can link with same target. To a huge extent, NT386 can link=
> with NT386GNU.
>Different targets get installed to different pkg directories.
>=20
>SOLgnu and SOLsun might provide a similar set of questions/answers, not tha=
>t I know anything about it. :)
>=20
>Anyway, I'm in a practical mode now, so I'll stop rambling in email. :)
>=20
>There's also the matter of X Windows or not...
>=20
> - Jay
>
>> To: jayk123 at hotmail.com> CC: m3devel at elegosoft.com> Subject: Re: [M3devel=
>] religion, philosophy, Posix vs. Win32 > Date: Sun, 13 Jan 2008 16:40:17 -=
>0800> From: mika at async.caltech.edu> > Jay,> > I don't know if this is compl=
>etely responsive to your question, and> take it with a grain of salt becaus=
>e I have never programmed on a> "pure" Windows system (and God willing, I n=
>ever shall).> > I expect the NT386GNU target, Cygwin, Windows-POSIX, whatev=
>er you> call it, to look as much as Unix "as possible." (Obviously not,> bu=
>t as "reasonable".) The main things are (1) yes, /bin/sh, and> (2) Unix (PO=
>SIX?) libc---Modula-3 does a good job of covering up> the basic stuff so yo=
>u don't need to worry about it for the most> part, but sometimes, just some=
>times, you need to get, say, some> system information that doesn't have an =
>existing M3 interface (say> the amount of free diskspace or something like =
>that). Then you> should be able to call C code that is portable between Uni=
>x and> Windows.> > One thing that is bizarre about the old NT386GNU PM3 tha=
>t doesn't> conform to the above description: it uses Windows Time.T. It's a=
>ll> the more bizarre because Cygwin uses Unix time_t. (Time.Now and> time(3=
>) are off by a couple of hundred years on NT386GNU, whereas> they are the s=
>ame on all Unixes.) I am sure there are other things.> Oh yes, the open-fil=
>e (default) locking semantics are different> too. Those two are actually th=
>e only differences that I noticed> in the programming environments.> > Mika=
>> > Jay writes:> >--_d0659847-2f20-454a-9665-5df3b5ce2832_> >Content-Type: =
>text/plain; charset=3D"iso-8859-1"> >Content-Transfer-Encoding: quoted-prin=
>table> >> >The conclusion here is:> > Just try to convince me otherwise. :)=
>> >> >The gist of the question is:> > Just what characteristics of "Posix" =
>and "Win32" should "NT386GNU" have?> >> > Just what characteristics of "Pos=
>ix" do people notice and like?> >You can't say "the whole thing". :)> >> >R=
>estated:> >> >What is Posix? I don't mean, where is the standard document w=
>eb site.> >I mean what is it TO YOU?> >> >Is it open/read/write/close?> >Is=
> it fork?> >Is it sbrk?> > On these previous ones, they are interesting to =
>many programmers,> > but Modula-3 already layers over all this, so these ar=
>e unlikely defining> >characteristics here.> >> >Is it sh? Probably somewha=
>t. But if have you sat in Cygwin sh much vs. cmd?> > It is annoying. Comman=
>d line history is flaky. Paths are munged to say /cy=3D> >gdrive.> > You lo=
>se the critical F8 command line completion against history feature.> > You =
>lose a lot of editing capability, but do gain some alternate possibly=3D> >=
> "portable"> > similar.> >> >Is it Perl? No.> >Is it X Windows. I doubt it.=
>> >Is it pthreads? Probably not.> > Or at least vtalarm? Maybe?> > Or setjm=
>p/longjmp? Ok.> > Longjmp to the same buffer multiple times? Not likely.> >=
> Longjmp to a place that has already returned? Not likely.> >All full paths=
> starting with a forward slash? Maybe.> > Backward slash being a valid non-=
>seperator path character? Maybe.> >> >Think about that last one.> >It is va=
>lid and occurs in Win32 code to treat forward slash and backward sl=3D> >as=
>h the same.> >Some code does, some code does not.> >What if you just did th=
>at unconditionally, like even in Posix?> >What is the role of backward slas=
>h in Posix paths? It either just doesn't o=3D> >ccur,> >or is an escape cha=
>racter, or is a regular character. Something at a low le=3D> >vel> >could r=
>eplace the backward slashes with forward slashes.> >> >I would like to chan=
>ge the code in m3-sys/cm3/m3path.m3 to treat backward s=3D> >lashes> >and f=
>orward slashes the same. Backward slashes just won't occur on Posix, r=3D> =
>>ight?> >So far I have managed without this change, but not having it has b=
>itten me =3D> >a few times.> >I do have the analogous change to the Win32 p=
>ath code in m3core.> >> >It is the redundant filenames libfoo.a instead of =
>foo.lib and libfoo.so ins=3D> >tead of foo.so?> >Perhaps. This seems dumb, =
>redundant to me. Why have prefixes and suffixes?> >Don't suffixes suffice?>=
> >> >That all said, I remain convinced that a useful intermediate or probab=
>ly en=3D> >dpoint> >is a toolset that targets Windows using cm3cg, ld, as, =
>but otherwise looks =3D> >like> >NT386 as much as possible. MAYBE alter the=
> library names to libfoo.*.> >> >sh and make are needed for building cm3cg.=
>> >I did use Cygwin for that. MinGWin sounds promising here too but I haven=
>'t =3D> >tried.> >> > - Jay> >> >> >_______________________________________=
>__________________________> >Get the power of Windows + Web with the new Wi=
>ndows Live.> >http://www.windowslive.com?ocid=3D3DTXT_TAGHM_Wave2_powerofwi=
>ndows_012008=3D> >> >--_d0659847-2f20-454a-9665-5df3b5ce2832_> >Content-Typ=
>e: text/html; charset=3D"iso-8859-1"> >Content-Transfer-Encoding: quoted-pr=
>intable> >> ><html>> ><head>> ><style>> >.hmmessage P> >{> >margin:0px;> >p=
>adding:0px> >}> >body.hmmessage> >{> >FONT-SIZE: 10pt;> >FONT-FAMILY:Tahoma=
>> >}> ></style>> ></head>> ><body class=3D3D'hmmessage'>The conclusion here=
> is:<br> Just try to conv=3D> >ince me otherwise. :)<br><br>The gist o=
>f the question is:<br>  Just wh=3D> >at characteristics of "Posix" and=
> "Win32" should "NT386GNU" have?<br><br>&n=3D> >bsp;Just what characteristi=
>cs of "Posix" do people notice and like?<br>You =3D> >can't say "the whole =
>thing". :)<br><br>Restated:<br><br>What is Posix? I do=3D> >n't mean, where=
> is the standard document web site.<br>I mean what is it TO =3D> >YOU?<br><=
>br>Is it open/read/write/close?<br>Is it fork?<br>Is it sbrk?<br>&=3D> >nbs=
>p; On these previous ones, they are interesting to many programmers,<br>=3D=
>> > but Modula-3 already layers over all this, so these are unlikely d=
>efi=3D> >ning<br>characteristics here.<br><br>Is it sh? Probably somewhat. =
>But if ha=3D> >ve you sat in Cygwin sh much vs. cmd?<br> It is annoyin=
>g. Command line=3D> > history is flaky. Paths are munged to say /cygdrive.<=
>br> You lose the=3D> > critical F8 command line completion against his=
>tory feature.<br>  You=3D> > lose a lot of editing capability, but do =
>gain some alternate possibly "por=3D> >table"<br>  similar.<br><br>Is =
>it Perl? No.<br>Is it X Windows. I doub=3D> >t it.<br>Is it pthreads? Proba=
>bly not.<br>  Or at least vtalarm? Maybe=3D> >?<br>  Or setjmp/lo=
>ngjmp? Ok.<br>    Longjmp to the sam=3D> >e buffer multiple =
>times? Not likely.<br>    Longjmp to a pla=3D> >ce that has =
>already returned? Not likely.<br>All full paths starting with a=3D> > forwa=
>rd slash? Maybe.<br>  Backward slash being a valid non-seperator=3D> >=
> path character? Maybe.<br><br>Think about that last one.<br>It is valid an=
>=3D> >d occurs in Win32 code to treat forward slash and backward slash the =
>same.<=3D> >br>Some code does, some code does not.<br>What if you just did =
>that uncondi=3D> >tionally, like even in Posix?<br>What is the role of back=
>ward slash in Posi=3D> >x paths? It either just doesn't occur,<br>or is an =
>escape character, or is =3D> >a regular character. Something at a low level=
><br>could replace the backward=3D> > slashes with forward slashes.<br><br>I=
> would like to change the code in m3=3D> >-sys/cm3/m3path.m3 to treat backw=
>ard slashes<br>and forward slashes the sam=3D> >e. Backward slashes just wo=
>n't occur on Posix, right?<br>So far I have mana=3D> >ged without this chan=
>ge, but not having it has bitten me a few times.<br>I =3D> >do have the ana=
>logous change to the Win32 path code in m3core.<br><br>It is=3D> > the redu=
>ndant filenames libfoo.a instead of foo.lib and libfoo.so instead =3D> >of =
>foo.so?<br>Perhaps. This seems dumb, redundant to me. Why have prefixes =3D=
>> >and suffixes?<br>Don't suffixes suffice?<br><br>That all said, I remain =
>con=3D> >vinced that a useful intermediate or probably endpoint<br>is a too=
>lset that=3D> > targets Windows using cm3cg, ld, as, but otherwise looks li=
>ke<br>NT386 as =3D> >much as possible. MAYBE alter the library names to lib=
>foo.*.<br><br>sh and =3D> >make are needed for building cm3cg.<br>I did use=
> Cygwin for that. MinGWin s=3D> >ounds promising here too but I haven't tri=
>ed.<br><br> - Jay<br><br><br=3D> > /><hr />Get the power of Windows + =
>Web with the new Windows Live. <a href=3D> >=3D3D'http://www.windowslive.co=
>m?ocid=3D3DTXT_TAGHM_Wave2_powerofwindows_012008=3D> >' target=3D3D'_new'>G=
>et it now!</a></body>> ></html>=3D> >> >--_d0659847-2f20-454a-9665-5df3b5ce=
>2832_--
>_________________________________________________________________
>Put your friends on the big screen with Windows Vista=AE + Windows Live=99.
>http://www.microsoft.com/windows/shop/specialoffers.mspx?ocid=3DTXT_TAGLM_C=
>PC_MediaCtr_bigscreen_012008=
>
>--_5cf007a7-8622-47c5-8550-283b4142bb52_
>Content-Type: text/html; charset="Windows-1252"
>Content-Transfer-Encoding: quoted-printable
>
><html>
><head>
><style>
>.hmmessage P
>{
>margin:0px;
>padding:0px
>}
>body.hmmessage
>{
>FONT-SIZE: 10pt;
>FONT-FAMILY:Tahoma
>}
></style>
></head>
><body class=3D'hmmessage'>ok..well...some provision I guess eventually shou=
>ld be made for those who agree with you and Tony.<BR>
>What will come first is something else. :)<BR>
> A system that is Win32 as much as possible, EXCEPT that it uses the g=
>cc-based backend, and as, and ld, and maybe throws around some forward slas=
>hes here and there.<BR>
> <BR>
>I weakly propose:<BR>
>  NT386GNU be deleted<BR>
>  and NT386MINGWIN (NT386MING? I386_MINGWIN? X86_MINGWIN?)<BR>
>  and NT386CYGWIN (NT386CYG? I386_CYGWIN? X86_CYGWIN?)<BR>
> takes its place.<BR>
> <BR>
>but not right now.<BR>
> <BR>
>However, it still might be one target, and you edit the config file.<BR>
>It depends on what people define as a "target" and what it means for two "d=
>ifferent" things to be the same "target".<BR>
>The same target can link with same target. To a huge extent, NT386 can link=
> with NT386GNU.<BR>
>Different targets get installed to different pkg directories.<BR>
> <BR>
>SOLgnu and SOLsun might provide a similar set of questions/answers, not tha=
>t I know anything about it. :)<BR>
> <BR>
>Anyway, I'm in a practical mode now, so I'll stop rambling in email. :)<BR>
> <BR>
>There's also the matter of X Windows or not...<BR>
> <BR>
> - Jay<BR><BR>
><HR id=3DstopSpelling>
>> To: jayk123 at hotmail.com<BR>> CC: m3devel at elegosoft.com<BR>> Subj=
>ect: Re: [M3devel] religion, philosophy, Posix vs. Win32 <BR>> Date: Sun=
>, 13 Jan 2008 16:40:17 -0800<BR>> From: mika at async.caltech.edu<BR>> <=
>BR>> Jay,<BR>> <BR>> I don't know if this is completely responsive=
> to your question, and<BR>> take it with a grain of salt because I have =
>never programmed on a<BR>> "pure" Windows system (and God willing, I nev=
>er shall).<BR>> <BR>> I expect the NT386GNU target, Cygwin, Windows-P=
>OSIX, whatever you<BR>> call it, to look as much as Unix "as possible." =
>(Obviously not,<BR>> but as "reasonable".) The main things are (1) yes, =
>/bin/sh, and<BR>> (2) Unix (POSIX?) libc---Modula-3 does a good job of c=
>overing up<BR>> the basic stuff so you don't need to worry about it for =
>the most<BR>> part, but sometimes, just sometimes, you need to get, say,=
> some<BR>> system information that doesn't have an existing M3 interface=
> (say<BR>> the amount of free diskspace or something like that). Then yo=
>u<BR>> should be able to call C code that is portable between Unix and<B=
>R>> Windows.<BR>> <BR>> One thing that is bizarre about the old NT=
>386GNU PM3 that doesn't<BR>> conform to the above description: it uses W=
>indows Time.T. It's all<BR>> the more bizarre because Cygwin uses Unix t=
>ime_t. (Time.Now and<BR>> time(3) are off by a couple of hundred years o=
>n NT386GNU, whereas<BR>> they are the same on all Unixes.) I am sure the=
>re are other things.<BR>> Oh yes, the open-file (default) locking semant=
>ics are different<BR>> too. Those two are actually the only differences =
>that I noticed<BR>> in the programming environments.<BR>> <BR>> Mi=
>ka<BR>> <BR>> Jay writes:<BR>> >--_d0659847-2f20-454a-9665-5df3=
>b5ce2832_<BR>> >Content-Type: text/plain; charset=3D"iso-8859-1"<BR>&=
>gt; >Content-Transfer-Encoding: quoted-printable<BR>> ><BR>> &g=
>t;The conclusion here is:<BR>> > Just try to convince me otherwise. :=
>)<BR>> ><BR>> >The gist of the question is:<BR>> > Just w=
>hat characteristics of "Posix" and "Win32" should "NT386GNU" have?<BR>> =
>><BR>> > Just what characteristics of "Posix" do people notice and=
> like?<BR>> >You can't say "the whole thing". :)<BR>> ><BR>>=
> >Restated:<BR>> ><BR>> >What is Posix? I don't mean, where =
>is the standard document web site.<BR>> >I mean what is it TO YOU?<BR=
>>> ><BR>> >Is it open/read/write/close?<BR>> >Is it fork?=
><BR>> >Is it sbrk?<BR>> > On these previous ones, they are inte=
>resting to many programmers,<BR>> > but Modula-3 already layers over =
>all this, so these are unlikely defining<BR>> >characteristics here.<=
>BR>> ><BR>> >Is it sh? Probably somewhat. But if have you sat i=
>n Cygwin sh much vs. cmd?<BR>> > It is annoying. Command line history=
> is flaky. Paths are munged to say /cy=3D<BR>> >gdrive.<BR>> > =
>You lose the critical F8 command line completion against history feature.<B=
>R>> > You lose a lot of editing capability, but do gain some alternat=
>e possibly=3D<BR>> > "portable"<BR>> > similar.<BR>> ><BR=
>>> >Is it Perl? No.<BR>> >Is it X Windows. I doubt it.<BR>> =
>>Is it pthreads? Probably not.<BR>> > Or at least vtalarm? Maybe?<=
>BR>> > Or setjmp/longjmp? Ok.<BR>> > Longjmp to the same buffer=
> multiple times? Not likely.<BR>> > Longjmp to a place that has alrea=
>dy returned? Not likely.<BR>> >All full paths starting with a forward=
> slash? Maybe.<BR>> > Backward slash being a valid non-seperator path=
> character? Maybe.<BR>> ><BR>> >Think about that last one.<BR>&=
>gt; >It is valid and occurs in Win32 code to treat forward slash and bac=
>kward sl=3D<BR>> >ash the same.<BR>> >Some code does, some code=
> does not.<BR>> >What if you just did that unconditionally, like even=
> in Posix?<BR>> >What is the role of backward slash in Posix paths? I=
>t either just doesn't o=3D<BR>> >ccur,<BR>> >or is an escape ch=
>aracter, or is a regular character. Something at a low le=3D<BR>> >ve=
>l<BR>> >could replace the backward slashes with forward slashes.<BR>&=
>gt; ><BR>> >I would like to change the code in m3-sys/cm3/m3path.m=
>3 to treat backward s=3D<BR>> >lashes<BR>> >and forward slashes=
> the same. Backward slashes just won't occur on Posix, r=3D<BR>> >igh=
>t?<BR>> >So far I have managed without this change, but not having it=
> has bitten me =3D<BR>> >a few times.<BR>> >I do have the analo=
>gous change to the Win32 path code in m3core.<BR>> ><BR>> >It i=
>s the redundant filenames libfoo.a instead of foo.lib and libfoo.so ins=3D<=
>BR>> >tead of foo.so?<BR>> >Perhaps. This seems dumb, redundant=
> to me. Why have prefixes and suffixes?<BR>> >Don't suffixes suffice?=
><BR>> ><BR>> >That all said, I remain convinced that a useful i=
>ntermediate or probably en=3D<BR>> >dpoint<BR>> >is a toolset t=
>hat targets Windows using cm3cg, ld, as, but otherwise looks =3D<BR>> &g=
>t;like<BR>> >NT386 as much as possible. MAYBE alter the library names=
> to libfoo.*.<BR>> ><BR>> >sh and make are needed for building =
>cm3cg.<BR>> >I did use Cygwin for that. MinGWin sounds promising here=
> too but I haven't =3D<BR>> >tried.<BR>> ><BR>> > - Jay<B=
>R>> ><BR>> ><BR>> >______________________________________=
>___________________________<BR>> >Get the power of Windows + Web with=
> the new Windows Live.<BR>> >http://www.windowslive.com?ocid=3D3DTXT_=
>TAGHM_Wave2_powerofwindows_012008=3D<BR>> ><BR>> >--_d0659847-2=
>f20-454a-9665-5df3b5ce2832_<BR>> >Content-Type: text/html; charset=3D=
>"iso-8859-1"<BR>> >Content-Transfer-Encoding: quoted-printable<BR>&gt=
>; ><BR>> ><html><BR>> ><head><BR>> ><st=
>yle><BR>> >.hmmessage P<BR>> >{<BR>> >margin:0px;<BR>&=
>gt; >padding:0px<BR>> >}<BR>> >body.hmmessage<BR>> >{<=
>BR>> >FONT-SIZE: 10pt;<BR>> >FONT-FAMILY:Tahoma<BR>> >}<B=
>R>> ></style><BR>> ></head><BR>> ><body cl=
>ass=3D3D'hmmessage'>The conclusion here is:<br>&nbsp;Just try =
>to conv=3D<BR>> >ince me otherwise. :)<br><br>The gist of=
> the question is:<br>&nbsp; Just wh=3D<BR>> >at characteris=
>tics of "Posix" and "Win32" should "NT386GNU" have?<br><br>&amp=
>;n=3D<BR>> >bsp;Just what characteristics of "Posix" do people notice=
> and like?<br>You =3D<BR>> >can't say "the whole thing". :)<=
>br><br>Restated:<br><br>What is Posix? I do=3D<BR>>=
> >n't mean, where is the standard document web site.<br>I mean wha=
>t is it TO =3D<BR>> >YOU?<br><br>Is it open/read/write/cl=
>ose?<br>Is it fork?<br>Is it sbrk?<br>&=3D<BR>> &g=
>t;nbsp; On these previous ones, they are interesting to many programmers,&l=
>t;br>=3D<BR>> >&nbsp;but Modula-3 already layers over all this=
>, so these are unlikely defi=3D<BR>> >ning<br>characteristics h=
>ere.<br><br>Is it sh? Probably somewhat. But if ha=3D<BR>> &=
>gt;ve you sat in Cygwin sh much vs. cmd?<br>&nbsp;It is annoying.=
> Command line=3D<BR>> > history is flaky. Paths are munged to say /cy=
>gdrive.<br>&nbsp;You lose the=3D<BR>> > critical F8 command=
> line completion against history feature.<br>&nbsp; You=3D<BR>&gt=
>; > lose a lot of editing capability, but do gain some alternate possibl=
>y "por=3D<BR>> >table"<br>&nbsp; similar.<br><br&g=
>t;Is it Perl? No.<br>Is it X Windows. I doub=3D<BR>> >t it.<=
>br>Is it pthreads? Probably not.<br>&nbsp; Or at least vtalarm=
>? Maybe=3D<BR>> >?<br>&nbsp; Or setjmp/longjmp? Ok.<br&g=
>t;&nbsp;&nbsp;&nbsp; Longjmp to the sam=3D<BR>> >e buffer=
> multiple times? Not likely.<br>&nbsp;&nbsp;&nbsp; Longjm=
>p to a pla=3D<BR>> >ce that has already returned? Not likely.<br&g=
>t;All full paths starting with a=3D<BR>> > forward slash? Maybe.<b=
>r>&nbsp; Backward slash being a valid non-seperator=3D<BR>> > =
>path character? Maybe.<br><br>Think about that last one.<br&=
>gt;It is valid an=3D<BR>> >d occurs in Win32 code to treat forward sl=
>ash and backward slash the same.<=3D<BR>> >br>Some code does, s=
>ome code does not.<br>What if you just did that uncondi=3D<BR>> &g=
>t;tionally, like even in Posix?<br>What is the role of backward slash=
> in Posi=3D<BR>> >x paths? It either just doesn't occur,<br>or =
>is an escape character, or is =3D<BR>> >a regular character. Somethin=
>g at a low level<br>could replace the backward=3D<BR>> > slashe=
>s with forward slashes.<br><br>I would like to change the code =
>in m3=3D<BR>> >-sys/cm3/m3path.m3 to treat backward slashes<br>=
>and forward slashes the sam=3D<BR>> >e. Backward slashes just won't o=
>ccur on Posix, right?<br>So far I have mana=3D<BR>> >ged withou=
>t this change, but not having it has bitten me a few times.<br>I =3D<=
>BR>> >do have the analogous change to the Win32 path code in m3core.&=
>lt;br><br>It is=3D<BR>> > the redundant filenames libfoo.a i=
>nstead of foo.lib and libfoo.so instead =3D<BR>> >of foo.so?<br&gt=
>;Perhaps. This seems dumb, redundant to me. Why have prefixes =3D<BR>> &=
>gt;and suffixes?<br>Don't suffixes suffice?<br><br>That a=
>ll said, I remain con=3D<BR>> >vinced that a useful intermediate or p=
>robably endpoint<br>is a toolset that=3D<BR>> > targets Windows=
> using cm3cg, ld, as, but otherwise looks like<br>NT386 as =3D<BR>&gt=
>; >much as possible. MAYBE alter the library names to libfoo.*.<br&gt=
>;<br>sh and =3D<BR>> >make are needed for building cm3cg.<br=
>>I did use Cygwin for that. MinGWin s=3D<BR>> >ounds promising her=
>e too but I haven't tried.<br><br>&nbsp;- Jay<br><=
>br><br=3D<BR>> > /><hr />Get the power of Windows + We=
>b with the new Windows Live. <a href=3D<BR>> >=3D3D'http://www.win=
>dowslive.com?ocid=3D3DTXT_TAGHM_Wave2_powerofwindows_012008=3D<BR>> >=
>' target=3D3D'_new'>Get it now!</a></body><BR>> ></=
>html>=3D<BR>> ><BR>> >--_d0659847-2f20-454a-9665-5df3b5ce283=
>2_--<BR><BR><br /><hr />Put your friends on the big screen with Windows Vis=
>ta=AE + Windows Live=99. <a href=3D'http://www.microsoft.com/windows/shop/s=
>pecialoffers.mspx?ocid=3DTXT_TAGLM_CPC_MediaCtr_bigscreen_012008' target=3D=
>'_new'>Start now!</a></body>
></html>=
>
>--_5cf007a7-8622-47c5-8550-283b4142bb52_--