Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
Moin Willi! Willi Dyck schrieb am Tuesday, den 22. October 2002: > On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > > - broken home/end keys in bash in xterm (even in Woody) > > Ever heard of ^a and ^e? Using bash? > > ~$ man bash for more info's As stated before, that is not the point. There are distribution where this problem is fixed. If a user is used to use (sic!) the home and end keys, sHe wants to see them working. Currently, they do not on Woody. Gruss/Regards, Eduard. -- Alfie: Also gehen tut das Ding, die Fehler sehe ich doch beim starten? -- LnxBil über die Bedeutung von "use strict;" -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
Eduard Bloch <[EMAIL PROTECTED]> [2002-10-21 09:07:48 +0200]: > - broken home/end keys in bash in xterm (even in Woody) I have been reading about these. But in what way are they broken? They work fine for me in xterm in woody. What is broken about them? [HOME] produces ESC-O-H and [END] produces ESC-O-F for me. Everything seems to interpret these. (Except that emacs switches to the MS definition and makes that beginning/end of line instead of the previous definition of beginning/end of file. But I can adapt better than MS users, I guess. :-) Bob msg07638/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, 22 Oct 2002 17:08:12 -0400
"Justin F. Knotzke" <[EMAIL PROTECTED]> wrote:
> On Tue, Oct 22, 2002 at 03:46:37PM -0400, Gregory Seidman wrote:
>
> > It's being done, frighteningly enough. See http://freshrpms.net/
>
>According to their examples, being done quite well at that. Surely
> there must be limitations if RedHat aren't using it themselves?
The first thing I noted is the preponderance of the first person pronoun
("I") on the home page.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
On Wed, Oct 23, 2002 at 01:57:19AM -0700, Paul Johnson wrote: > On Tue, Oct 22, 2002 at 07:57:25PM -0400, Justin F. Knotzke wrote: > >I used RedHat a few years ago for about a month and rpm gave me such > > a headache that I bolted back to Debian. > > Heck, when I was a regular Red Hat user, we ended up moving the > machines over to Debian because it was simply easier to deal with > .debs. This was back when bo was current and before apt was standard. Me, too. Of course in those days Red Hat made it easy to switch, since after trying to update you usually had to wipe out the disk and install from scratch anyway. Bob -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Tue, Oct 22, 2002 at 11:32:29PM -0500, Steve Langasek wrote: > Oh, poppycock. Users don't report bugs to do us any *favors*, they > report bugs out of self-interest. How many users go out of their way to > look for bugs that don't actually impede their use of the software? Not many, however, the day-to-day, casual sid users do provide the important favor of giving Debian a workout while it's being developed. > Debian should work, but I as far as I can see, the only way for you to > change this is by becoming a maintainer to make sure there's enough > manpower to go around -- which may help reduce the work imposed upon While I can't currently commit to a date, I'm waiting until times get a little better and I don't have so much of my day spoken for that I can take on maintainership of a package orphaned or up for adoption. -- Baloo msg08598/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 07:57:25PM -0400, Justin F. Knotzke wrote: >I used RedHat a few years ago for about a month and rpm gave me such > a headache that I bolted back to Debian. Heck, when I was a regular Red Hat user, we ended up moving the machines over to Debian because it was simply easier to deal with .debs. This was back when bo was current and before apt was standard. -- Baloo msg08593/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Tue, Oct 22, 2002 at 07:31:24PM -0700, Paul Johnson wrote: > On Mon, Oct 21, 2002 at 05:35:45PM -0500, Steve Langasek wrote: > > > I don't see why...it takes a few seconds to pass on a bug. > > > > If it only takes a few seconds, then why do you object to being asked to > > do it yourself? ;) > Because frequently the developer knows what's expected out of the > other BTS's community and how that BTS works. My understanding is one > of the functions of the developers is to act as an embassador between > the Debian users and the upstream community. > Plus, users reporting bugs do Debian (and the upstream maintainers) a > favor, why make the users jump through *more* hoops? Oh, poppycock. Users don't report bugs to do us any *favors*, they report bugs out of self-interest. How many users go out of their way to look for bugs that don't actually impede their use of the software? I'm not saying that users' efforts in filing bugs aren't *appreciated*, but as noted, there's sometimes only so much a maintainer can do given the volume of bugs that have to be handled; and sometimes we have to depend on users for more than just sending us bug reports in order to get bugs fixed. I'm sorry if that doesn't fit your conception of how Debian should work, but I as far as I can see, the only way for you to change this is by becoming a maintainer to make sure there's enough manpower to go around -- which may help reduce the work imposed upon users in the general case, but doesn't seem to help you very much if your concern is making sure that *you* don't have to talk to upstream... :) Steve Langasek postmodern programmer msg08568/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 03:46:37PM -0400, Gregory Seidman wrote: > } Unless some people start a Debian-like project using rpm packages, the > } terraparsecs are going to stay. Imagine the consequences of a Redhat > } 9.0 system you can rpm-get to version 12.0. RPM and an XGUI-based > } installer is great for CD-installed systems. > > It's being done, frighteningly enough. See http://freshrpms.net/ This should get *quite* entertaining to watch, since they don't have the standardized package names for basic stuff that lots of things depend on between distros. -- Baloo msg08558/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Wed, Oct 23, 2002 at 02:26:29AM +0800, csj wrote: > Unless some people start a Debian-like project using rpm packages, the > terraparsecs are going to stay. Imagine the consequences of a Redhat > 9.0 system you can rpm-get to version 12.0. RPM and an XGUI-based > installer is great for CD-installed systems. CD-installed system that is never upgraded (even by a newer CD) and never modified by packages that didn't come specifically from the same distributor as your system. -- Baloo msg08557/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 08:59:11PM -0500, Steve Langasek wrote: > I assumed it was a given that a maintainer asking a bug submitter to > take such an active role would provide such guidance as needed. That's my understanding, too. And only one or two maintainers that I've run into seem to follow this concept, as well. If it's not already a standardized process, it should be, if only for the sake of good customer service. Let the maintainer clarify the bug, then upstream gets a really awesome bug report. It's not like there isn't a conversation log at b.d.o if anybody needs to refer back to it. -- Baloo msg08556/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 05:35:45PM -0500, Steve Langasek wrote: > > I don't see why...it takes a few seconds to pass on a bug. > > If it only takes a few seconds, then why do you object to being asked to > do it yourself? ;) Because frequently the developer knows what's expected out of the other BTS's community and how that BTS works. My understanding is one of the functions of the developers is to act as an embassador between the Debian users and the upstream community. Plus, users reporting bugs do Debian (and the upstream maintainers) a favor, why make the users jump through *more* hoops? -- Baloo msg08555/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 03:33:52PM -0700, nate wrote: > i haven't tried apt-get for rpm on redhat yet, but I tried it > on suse 7.3 about 4-5 months ago and was dissapointed in it. tried > to do some security updates, some completed fine then it errored > out and refused to continue(or provide any hints why it errored out). > I do have a really good redhat 7.3 box at my feet here which i could > try it out on, sounds interesting. I used RedHat a few years ago for about a month and rpm gave me such a headache that I bolted back to Debian. I am curious. How would one upgrade from V7 to V8.0 of RedHat? Is it a complete install requiring you purchase the entire CD set again or is there some sort of upgrade process and if there is, does it work well? I am awaiting a new ThinkPad and I'm in a bind. I know that the ThinkPad I am getting only works with XFree 4.2 which Debian doesn't yet support (at least in stable). Which means I'd have to point to another source list (who's address I forget). No real biggee. But what is, is IBM's support for only RPM. I work heavily with IBM's Websphere. On Linux that means you have to use rpm. You could try using alien but installing Websphere WITH rpm is enough of a nightmare; trying to resolve internal dependencies with alien is another I'd like to avoid. I might give RedHat 8 a try just for sh*ts and giggles but it really pains me to leave Debian. Justin. -- Justin F. Knotzke [EMAIL PROTECTED] http://www.shampoo.ca msg08533/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 05:08:12PM -0400, Justin F. Knotzke wrote: > On Tue, Oct 22, 2002 at 03:46:37PM -0400, Gregory Seidman wrote: > > > It's being done, frighteningly enough. See http://freshrpms.net/ > > According to their examples, being done quite well at that. Surely > there must be limitations if RedHat aren't using it themselves? yeah, you don't make as much money selling CDs every release :) -- Nathan Norman - Incanus Networking mailto:nnorman@;incanus.net I retract that silly statement. Somebody slap me. -- Roy Smith -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
Justin F. Knotzke said: > On Tue, Oct 22, 2002 at 03:46:37PM -0400, Gregory Seidman wrote: > >> It's being done, frighteningly enough. See http://freshrpms.net/ > > According to their examples, being done quite well at that. Surely > there must be limitations if RedHat aren't using it themselves? the biggest problem i can see is redhat probably wouldn't be able to get absolute control over the system. that is to be sure all packages available are fully tested & stable for the redhat systems. for optimal results they would need strict procedures in place to control what is made available, and how (much like debian's procedures probably). And for them to support it, it would be a big & expensive task to support hundreds or even thousands of additional packages that they already have to support. i haven't tried apt-get for rpm on redhat yet, but I tried it on suse 7.3 about 4-5 months ago and was dissapointed in it. tried to do some security updates, some completed fine then it errored out and refused to continue(or provide any hints why it errored out). I do have a really good redhat 7.3 box at my feet here which i could try it out on, sounds interesting. nate -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 03:46:37PM -0400, Gregory Seidman wrote: > It's being done, frighteningly enough. See http://freshrpms.net/ According to their examples, being done quite well at that. Surely there must be limitations if RedHat aren't using it themselves? J. -- Justin F. Knotzke [EMAIL PROTECTED] http://www.shampoo.ca msg08509/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
Gregory Seidman ([EMAIL PROTECTED]) wrote*: >} Unless some people start a Debian-like project using rpm packages, the >} terraparsecs are going to stay. Imagine the consequences of a Redhat >} 9.0 system you can rpm-get to version 12.0. RPM and an XGUI-based >} installer is great for CD-installed systems. > >It's being done, frighteningly enough. See http://freshrpms.net/ OK I admit I use RedHat, I have not read that slashdot review, but I have been using and making rpm packages for several years now. For RedHat 8.0 the rpm app got a major brain upgrade. I can tell you that before RH 8.0 version, rpm DID need the brains of apt to do anything slightly complicated. I use apt for rpm for that reason (yes, apt is awesome). However, NOW I can actually do something of a "dist upgrade" by dumping all rpm files in a directory and running "rpm -F *.rpm". Shockingly, this will work now. Previously, it would take hours to manually handle the deps from the rpm error messages, even though all necessary files WERE RIGHT THERE in that directory! Now it can handle that which I consider that finally it has some brains. My guess is that they want to make greater use of their "up2date" service, the front end to that is written in Python, so I assume they decided to put the real brains in the rpm app itself. Just a guess. In any case I still use apt for rpm and dig it greatly. -- That's "angle" as in geometry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
csj sez: } On Mon, 21 Oct 2002 04:56:10 -0700 } Paul Johnson <[EMAIL PROTECTED]> wrote: } } [snip: about a reviewer's unfavorable comments about Debian] } } > I have to seriously question whether or not he knows what he's } > talking about about RPM. I've used RPM recently. It's still painful } > to use and terraparsecs behind apt-get *still*. Even with urpmi. } } Unless some people start a Debian-like project using rpm packages, the } terraparsecs are going to stay. Imagine the consequences of a Redhat } 9.0 system you can rpm-get to version 12.0. RPM and an XGUI-based } installer is great for CD-installed systems. It's being done, frighteningly enough. See http://freshrpms.net/ --Greg -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
On Mon, 21 Oct 2002 04:56:10 -0700 Paul Johnson <[EMAIL PROTECTED]> wrote: [snip: about a reviewer's unfavorable comments about Debian] > I have to seriously question whether or not he knows what he's > talking about about RPM. I've used RPM recently. It's still painful > to use and terraparsecs behind apt-get *still*. Even with urpmi. Unless some people start a Debian-like project using rpm packages, the terraparsecs are going to stay. Imagine the consequences of a Redhat 9.0 system you can rpm-get to version 12.0. RPM and an XGUI-based installer is great for CD-installed systems. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
Hello Price,, On Oct 22, "Price, Erik" <[EMAIL PROTECTED]> wrote: | > You *have* tried xserver-3dlabs, right? | Um is that a driver, or a product? The answer is no to both, | though. That doesn't show up as one of the choices of driver when I | run `dpkg-reconfigure xserver-xfree86`. In fact just about all of | the choices are 5-character names or less. It's a Debian package: jason@inj ( jason ) 14:11$ apt-cache show xserver-3dlabs Package: xserver-3dlabs Priority: optional Section: x11 Installed-Size: 2188 Maintainer: Branden Robinson <[EMAIL PROTECTED]> Architecture: i386 Source: xfree86v3 Version: 3.3.6-44 Provides: xserver Depends: debconf (>= 0.5), xserver-common-v3 (>= 3.3.6-44), libc6 (>= 2.2.4-4), zlib1g (>= 1:1.1.3) Filename: pool/main/x/xfree86v3/xserver-3dlabs_3.3.6-44_i386.deb Size: 916426 MD5sum: 7100dd3688bcc688b144941b3995b331 Description: X server for 3DLabs GLINT and Permedia-based graphics cards xserver-3dlabs is an 8-bit PseudoColor, 16-bit TrueColor, 24-bit TrueColor, and 32-bit TrueColor X server suitable for use with some 3DLabs GLINT 500TX, GLINT MX, Permedia, and Permedia 2 graphic accelerator boards. 24-bit operation is supported only on the Permedia 2. -- Jason [EMAIL PROTECTED] http://wonka.hampshire.edu/~jason GPG key - 0EFB1DFE msg08466/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Tue, 22 Oct 2002 12:41:34 -0400 "Price, Erik" <[EMAIL PROTECTED]> wrote: > > You *have* tried xserver-3dlabs, right? > > Um is that a driver, or a product? Neither it's a Debian package: xserver-3dlabs - X server for 3DLabs GLINT and Permedia-based graphics cards -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
On Tue, Oct 22, 2002 at 12:41:34PM -0400, Price, Erik wrote: > Um is that a driver, or a product? The answer is no to both, > though. That doesn't show up as one of the choices of driver when I > run `dpkg-reconfigure xserver-xfree86`. In fact just about all of the > choices are 5-character names or less. It's a package: $ dpkg -p xserver-3dlabs Package: xserver-3dlabs Priority: optional Section: x11 Installed-Size: 2188 Maintainer: Branden Robinson <[EMAIL PROTECTED]> Architecture: i386 Source: xfree86v3 Version: 3.3.6-44 Provides: xserver Depends: debconf (>= 0.5), xserver-common-v3 (>= 3.3.6-44), libc6 (>= 2.2.4-4), zlib1g (>= 1:1.1.3) Filename: pool/main/x/xfree86v3/xserver-3dlabs_3.3.6-44_i386.deb Size: 916426 MD5sum: 7100dd3688bcc688b144941b3995b331 Description: X server for 3DLabs GLINT and Permedia-based graphics cards xserver-3dlabs is an 8-bit PseudoColor, 16-bit TrueColor, 24-bit TrueColor, and 32-bit TrueColor X server suitable for use with some 3DLabs GLINT 500TX, GLINT MX, Permedia, and Permedia 2 graphic accelerator boards. 24-bit operation is supported only on the Permedia 2. -- Colin Watson [[EMAIL PROTECTED]] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RE: Two Debian 3.0 reviews at Slashdot
> >> > >>I did a quick check on the xfree86 site. It doesn't look > >>like the intense3d > >>is supported. > >> > >>The one thing you can do to make sure that installing Linux > >>is easy is to make > >>sure all your hardware is fully supported. > > > > > > Hear that! I'm going to try to find a used video card. I > don't need anything fancy (I don't play games on this machine). > > > > > You *have* tried xserver-3dlabs, right? Um is that a driver, or a product? The answer is no to both, though. That doesn't show up as one of the choices of driver when I run `dpkg-reconfigure xserver-xfree86`. In fact just about all of the choices are 5-character names or less. If it *is* a driver, how can I incorporate it into my system? I'm very new to X11, so pardon me if I've missed something obvious. (I'm especially hopeful because I just got the big old "negatory" from the boss when I asked if I could purchase a supported video card with my own money so I could get X11 up and running. So if I can't use what I've got I may be SOL.) Erik -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > - broken home/end keys in bash in xterm (even in Woody) Ever heard of ^a and ^e? Using bash? ~$ man bash for more info's > - missing apt localisation extensions (who t.f. told we that we are going to >release in the next few weeks, again and again for almost 6 months?!) > - centralised "setup" tool which would reconfigure etherconf, pppoeconf, and >do sth. as gx-debconf does, but be more understandable. -- i can't stand this Debian vs XYZ thing everybody feel free to stick with XYZ! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Two Debian 3.0 reviews at Slashdot
Price, Erik wrote: -Original Message- From: David A. Rogers [mailto:darogers@;speakeasy.net] Sent: Monday, October 21, 2002 11:47 AM To: [EMAIL PROTECTED] Subject: RE: Two Debian 3.0 reviews at Slashdot I did a quick check on the xfree86 site. It doesn't look like the intense3d is supported. The one thing you can do to make sure that installing Linux is easy is to make sure all your hardware is fully supported. Hear that! I'm going to try to find a used video card. I don't need anything fancy (I don't play games on this machine). You *have* tried xserver-3dlabs, right? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 11:27:34PM -0400, Mark L. Kahnt wrote: > My understanding is that the original intention of the Debian BTS was to > be about packaging bugs - it has evolved beyond that partly as it is > handy to report any bugs, and for users that don't recognise the > difference between distributions and developers of specific software > (which *does* on occasion happen amongst Debian users,) it can be the > only address available to pass on problems, whether actually appropriate > or not. I think the original intent was to be about all bugs, actually. If you look back in the archives, some of the earliest bugs we still have records of are really nasty upstream problems, and many of those came from the original author of the BTS. As far as I can tell, it was always intended that we should help the free software community by acting as a middle-man for bug reports in the software we distribute. The exact mechanism doesn't really matter: it's OK if a maintainer says "yes, this is an upstream problem, but they prefer to talk to users directly, so I recommend that you report it upstream yourself using this address", as long as the maintainer is willing to help follow through. What's not acceptable, IMHO at least, is for bugs to be *dismissed* in that way. Fortunately it's only a small minority of developers who do the latter. -- Colin Watson [[EMAIL PROTECTED]] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, 21 Oct 2002 09:07:48 +0200 Eduard Bloch <[EMAIL PROTECTED]> wrote: > - broken home/end keys in bash in xterm (even in Woody) Sounds like a possible reason to use RXVT... never noticed that it's broken in xterm as I've been using RXVT since before I moved to Debian. However, I did just try it on my (mostly) Sarge system and Home and End work just fine in xterm *shrug* -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RE: Two Debian 3.0 reviews at Slashdot
> -Original Message- > From: Andrew Perrin [mailto:clists@;perrin.socsci.unc.edu] > Sent: Monday, October 21, 2002 10:40 AM > To: Jamin W.Collins > Cc: [EMAIL PROTECTED] > Subject: Re: Two Debian 3.0 reviews at Slashdot > > > I agree entirely. And I also think that most of the added > benefit of linux > over windows is not available if you just run some office > suite under X > instead of under windows. I use linux exclusively because the time it > took me to learn the real tools -- grep, wc, emacs, perl, latex, bash, > etc. -- has paid off in spades when it comes to day-to-day > efficiency. But > that logic is obscured when you just use, say, koffice under > X instead of > msoffice under windows. > I hear you -- and I agree entirely. But combining the ease of use of a GUI with the ease of use of power tools like the ones you describe is one of the things that attracts me to Linux most of all, and I still haven't figured out a way to get my Intense3D video card to be recognized correctly by XFree86. :( While emacs is cool, I use an IDE to do my development and also need a working X server to test -- I'm confined to working in Windows at the moment, unless I can somehow justify a different video card from the IT dept... Erik -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
RE: Two Debian 3.0 reviews at Slashdot
I did a quick check on the xfree86 site. It doesn't look like the intense3d is supported. The one thing you can do to make sure that installing Linux is easy is to make sure all your hardware is fully supported. dar On Mon, 21 Oct 2002, Price, Erik wrote: > > > > -Original Message- > > From: Andrew Perrin [mailto:clists@;perrin.socsci.unc.edu] > > Sent: Monday, October 21, 2002 10:40 AM > > To: Jamin W.Collins > > Cc: [EMAIL PROTECTED] > > Subject: Re: Two Debian 3.0 reviews at Slashdot > > > > > > I agree entirely. And I also think that most of the added > > benefit of linux > > over windows is not available if you just run some office > > suite under X > > instead of under windows. I use linux exclusively because the time it > > took me to learn the real tools -- grep, wc, emacs, perl, latex, bash, > > etc. -- has paid off in spades when it comes to day-to-day > > efficiency. But > > that logic is obscured when you just use, say, koffice under > > X instead of > > msoffice under windows. > > > > I hear you -- and I agree entirely. But combining the ease of use of a GUI with the >ease of use of power tools like the ones you describe is one of the things that >attracts me to Linux most of all, and I still haven't figured out a way to get my >Intense3D video card to be recognized correctly by XFree86. :( While emacs is cool, >I use an IDE to do my development and also need a working X server to test -- I'm >confined to working in Windows at the moment, unless I can somehow justify a >different video card from the IT dept... > > > > Erik > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] > > > -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
Hi, On Mon, Oct 21, 2002 at 01:23:26PM +0200, Josip Rodin wrote: > On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > > - broken home/end keys in bash in xterm (even in Woody) > > Say, is this the "application mode" thing? Cf. #133258. I don't know, but if it isn't, adding this to /etc/inputrc helped me a lot. OTOH, if libreadline should put the keypad in application mode but doesn't, then please consider this just a dirty hack. # Add PC movement keys $if term=xterm "\e[5~": backward-word "\e[6~": forward-word "\e[H": beginning-of-line "\e[F": end-of-line $endif Cheers, Emile. -- E-Advies / Emile van Bergen | [EMAIL PROTECTED] tel. +31 (0)70 3906153| http://www.e-advies.info msg08325/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, 2002-10-21 at 18:15, Branden Robinson wrote: > On Mon, Oct 21, 2002 at 01:21:00PM -0700, Paul Johnson wrote: > > On Mon, Oct 21, 2002 at 02:45:44PM -0500, Branden Robinson wrote: > > > Whining about Debian developers whining about upstream implies that you > > > expect Debian developers to fix every problem. For instance, I suppose > > > > My problem is some developers will tell the user to submit it > > upstream, when my understanding is the devel is supposed to flag it as > > upstream and go submit it upstream themselves, since they're supposed > > to have an idea what bugs are upstream anyway, and know how to provide > > more helpful input for upstream. > > And sometimes the Debian developer can't really be helpful. > > I, for instance, have *never* had any luck acting as a go-between for > Debian users and upstream XFree86 problems. The upstream XFree86 guys > *always* seem to want to deal with the user directly. > > This, and the lack of a bug tracking system for the XFree86 Project, is > why I don't mark XFree86 bugs as forwarded in the Debian BTS. > > I enjoy what I think is a good working relationship with XFree86 > upstream despite the fact that the Developers' Reference tells me to do > things differently than I do. It is better to be accomodating of > upstream than to mindlessly adhere to a Debian-specific best practices > document. In this case, our users are better served the way I'm doing > it, because that's how XFree86 wants to handle things. Who am I to tell > them to change their ways? > > -- > G. Branden Robinson| Convictions are more dangerous > Debian GNU/Linux | enemies of truth than lies. > [EMAIL PROTECTED] | -- Friedrich Nietzsche > http://people.debian.org/~branden/ | I would also interject that sometimes, particularly when dealing with package maintainers looking after numerous diverse packages, tracking the design details of every aspect of each package *may not* be a reasonable expectation, and particularly when the bug is unclear in its nature or source, it is better to have upstream ask the questions to get a clear description of what is happening, especially when a user hasn't *quite* explained it clearly. My understanding is that the original intention of the Debian BTS was to be about packaging bugs - it has evolved beyond that partly as it is handy to report any bugs, and for users that don't recognise the difference between distributions and developers of specific software (which *does* on occasion happen amongst Debian users,) it can be the only address available to pass on problems, whether actually appropriate or not. I don't fault the *two routes* approach to kicking bugs upstream. So long as it is effective, it is all part of the bazaar approach of Free & Open Source sorfware and the volunteer nature of Debian. -- Mark L. Kahnt, FLMI/M, ALHC, HIA, AIAA, ACS, MHP ML Kahnt New Markets Consulting Tel: (613) 531-8684 / (613) 539-0935 Email: [EMAIL PROTECTED] signature.asc Description: This is a digitally signed message part
RE: Two Debian 3.0 reviews at Slashdot
> -Original Message- > From: David A. Rogers [mailto:darogers@;speakeasy.net] > Sent: Monday, October 21, 2002 11:47 AM > To: [EMAIL PROTECTED] > Subject: RE: Two Debian 3.0 reviews at Slashdot > > > I did a quick check on the xfree86 site. It doesn't look > like the intense3d > is supported. > > The one thing you can do to make sure that installing Linux > is easy is to make > sure all your hardware is fully supported. Hear that! I'm going to try to find a used video card. I don't need anything fancy (I don't play games on this machine). Erik -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 09:30:01PM -0400, Matt Zimmerman wrote: > On Mon, Oct 21, 2002 at 05:35:45PM -0500, Steve Langasek wrote: > > On Mon, Oct 21, 2002 at 03:21:40PM -0700, Paul Johnson wrote: > > > I don't see why...it takes a few seconds to pass on a bug. > > If it only takes a few seconds, then why do you object to being asked to > > do it yourself? ;) > The submitter may not even know who to contact about upstream bugs; the > maintainer should always know the most appropriate place to send bug > reports. I assumed it was a given that a maintainer asking a bug submitter to take such an active role would provide such guidance as needed. Steve Langasek postmodern programmer msg08276/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 05:35:45PM -0500, Steve Langasek wrote: > On Mon, Oct 21, 2002 at 03:21:40PM -0700, Paul Johnson wrote: > > I don't see why...it takes a few seconds to pass on a bug. > > If it only takes a few seconds, then why do you object to being asked to > do it yourself? ;) The submitter may not even know who to contact about upstream bugs; the maintainer should always know the most appropriate place to send bug reports. -- - mdz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 10:21:26PM +0200, Robert Wilhelm Land wrote: > Michael Cardenas wrote: > >Isn't it the downstream packager's job to take upstream and configure > >it in a usable way for most users? It seems that if the patches > >discussed earlier in this thread work, then whining about upstream > >attitudes is just an excuse for not applying them. > > What I just had read in computer magazin a few days ago was > that lindows uses the debian GPL system to in turn buld up > the commercial lindows OS and not providing they're code to the > community - thus giving offence to the GPL. > > No idea how this can be defeated in a long-term way. > > Michael - your message was meant as a joke - wasn't it? If you're going to flame somebody, pick somebody better to flame. Michael's a Debian developer as well as a Lindows employee. Also, I'm reasonably sure Lindows aren't actually breaking the GPL, despite uninformed comments to the contrary. Cheers, -- Colin Watson [[EMAIL PROTECTED]] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 03:21:40PM -0700, Paul Johnson wrote: > On Mon, Oct 21, 2002 at 05:04:43PM -0500, Steve Langasek wrote: > > submitter to work with upstream directly, or the maintainer has to add > > the bug to the bottom of his own long TODO list. > I don't see why...it takes a few seconds to pass on a bug. If it only takes a few seconds, then why do you object to being asked to do it yourself? ;) The reality is that while it might only take a few seconds to blindly forward the bug report to the upstream developers, *following through* on the bug and ensuring that it gets *fixed* can sometimes be a very lengthy process -- *even when you already have a patch for the bug*. Sometimes upstream disagrees with the way the patch was done; sometimes upstream moves out from under you while the patch is being written, and you have to resubmit it; sometimes, it's just a matter of pinging upstream to remind them to apply the patch. There are lots of reasons why upstream bugs can -- and usually do -- take a lot longer than a few seconds to get fixed. If we as maintainers are concerned about *getting bugs fixed* instead of just passing the buck, sometimes the best thing we can do to help our users is put them in touch with upstream developers. Steve Langasek postmodern programmer msg08241/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 05:04:43PM -0500, Steve Langasek wrote: > submitter to work with upstream directly, or the maintainer has to add > the bug to the bottom of his own long TODO list. I don't see why...it takes a few seconds to pass on a bug. > If a bug submitter comes to me with a well-formed patch for an upstream > issue, there are definitely times that it helps if the submitter can talk > directly with upstream about it, rather than using me as a relay for > everything. Well, that's an obvious exception. -- Baloo msg08237/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 01:21:00PM -0700, Paul Johnson wrote: > On Mon, Oct 21, 2002 at 02:45:44PM -0500, Branden Robinson wrote: > > Whining about Debian developers whining about upstream implies that you > > expect Debian developers to fix every problem. For instance, I suppose > > My problem is some developers will tell the user to submit it > upstream, when my understanding is the devel is supposed to flag it as > upstream and go submit it upstream themselves, since they're supposed > to have an idea what bugs are upstream anyway, and know how to provide > more helpful input for upstream. And sometimes the Debian developer can't really be helpful. I, for instance, have *never* had any luck acting as a go-between for Debian users and upstream XFree86 problems. The upstream XFree86 guys *always* seem to want to deal with the user directly. This, and the lack of a bug tracking system for the XFree86 Project, is why I don't mark XFree86 bugs as forwarded in the Debian BTS. I enjoy what I think is a good working relationship with XFree86 upstream despite the fact that the Developers' Reference tells me to do things differently than I do. It is better to be accomodating of upstream than to mindlessly adhere to a Debian-specific best practices document. In this case, our users are better served the way I'm doing it, because that's how XFree86 wants to handle things. Who am I to tell them to change their ways? -- G. Branden Robinson| Convictions are more dangerous Debian GNU/Linux | enemies of truth than lies. [EMAIL PROTECTED] | -- Friedrich Nietzsche http://people.debian.org/~branden/ | msg08236/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 01:21:00PM -0700, Paul Johnson wrote: > On Mon, Oct 21, 2002 at 02:45:44PM -0500, Branden Robinson wrote: > > Whining about Debian developers whining about upstream implies that you > > expect Debian developers to fix every problem. For instance, I suppose > My problem is some developers will tell the user to submit it > upstream, when my understanding is the devel is supposed to flag it as > upstream and go submit it upstream themselves, since they're supposed > to have an idea what bugs are upstream anyway, and know how to provide > more helpful input for upstream. If the developer feels that the best way to get the bug fixed upstream is by having the bug submitter communicate with them, that too is valid, IMHO. I don't think this means that the time spent filing the bug with the Debian BTS is wasted, since there's still some benefit in having the package maintainer make that determination; but there are lots of bugs to go around, and maintainers can be stretched rather thin at times, so sometimes what it boils down to is that the maintainer asks the bug submitter to work with upstream directly, or the maintainer has to add the bug to the bottom of his own long TODO list. If a bug submitter comes to me with a well-formed patch for an upstream issue, there are definitely times that it helps if the submitter can talk directly with upstream about it, rather than using me as a relay for everything. Steve Langasek postmodern programmer msg08232/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Mon, 21 Oct 2002, Jamin W.Collins wrote: > Without a doubt. Why not educate your friends about how their PCs work? > A fundamental problem today is that people don't understand the "how" and > "why". To attempt to protect the user from how a PC operates is IMHO to > do much more harm than good. I agree entirely. And I also think that most of the added benefit of linux over windows is not available if you just run some office suite under X instead of under windows. I use linux exclusively because the time it took me to learn the real tools -- grep, wc, emacs, perl, latex, bash, etc. -- has paid off in spades when it comes to day-to-day efficiency. But that logic is obscured when you just use, say, koffice under X instead of msoffice under windows. -- Andrew J Perrin - http://www.unc.edu/~aperrin Assistant Professor of Sociology, U of North Carolina, Chapel Hill [EMAIL PROTECTED] * andrew_perrin (at) unc.edu -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
[I am not subscribed to -user.] On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > - broken home/end keys in bash in xterm (even in Woody) You'll have to persuade the upstream bash/libreadline cabal that actually supporting DEC VT100 and later terminal emulation is a worthwhile thing to do, instead of half-assing their way through it. > - missing apt localisation extensions (who t.f. told we that we are going to >release in the next few weeks, again and again for almost 6 months?!) It's a western Canadian conspiracy to have the whole world talking English, eh? > - centralised "setup" tool which would reconfigure etherconf, pppoeconf, and >do sth. as gx-debconf does, but be more understandable. Like the configlet-capplet frontend (cf. configlet-frontends)? -- G. Branden Robinson| Never underestimate the power of Debian GNU/Linux | human stupidity. [EMAIL PROTECTED] | -- Robert Heinlein http://people.debian.org/~branden/ | msg08190/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On Mon, Oct 21, 2002 at 12:51:33PM -0400, Joey Hess wrote: > echo expert >~/.dselect.cfg Holy schnikes, Batman! Why isn't this obviously documented? > What confused me is that both reviews chose to go into dselect, even > though the woody installer defaults to not running it if you choose to > run tasksel. Because some people (like me) don't like how incredably grainy tasksel is. -- Baloo msg08186/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 12:15:16PM -0700, Michael Cardenas wrote: > On Mon, Oct 21, 2002 at 12:49:34PM -0500, Branden Robinson wrote: > > On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > > > > > - broken home/end keys in bash in xterm (even in Woody) > > > > You'll have to persuade the upstream bash/libreadline cabal that > > actually supporting DEC VT100 and later terminal emulation is a > > worthwhile thing to do, instead of half-assing their way through it. > > Isn't it the downstream packager's job to take upstream and configure > it in a usable way for most users? It seems that if the patches > discussed earlier in this thread work, then whining about upstream > attitudes is just an excuse for not applying them. You need to know whereof you speak before you shoot off your mouth like this. * VT220+ terminal emualtion is a stateful thing. Certain keys have a one meaning in application-keypad mode, and another meaning in normal mode. Terminal control sequences can (and will) move the terminal in and out of application-keypad mode. /etc/inputrc doesn't know how to express "if the terminal is in application-keypad mode, issue this escape sequence when the Home key is pressed; else issue this other escape sequence". * It is not the downstream packager's job to fix major design flaws in upstream software. * There are times when shortsightedness makes it impossible to please all of the users, because they have conflicting expectations. * Sometimes Debian developers don't have the skills or knowledge necessary to fix a problem. Hypothetically, it may be the case that Debian's bash package maintainer isn't a wizard when it comes to the details of DEC terminal emulation, and/or cannot think of and implement a clever way to track arbitrary states for arbitrary keygroups in libreadline so as to achieve a general solution. Should we really expect so much of our package maintainers, or should we simply expect them to do a good job *packaging* software? I think not; not unless you're willing to underwrite their educational expenses. Whining about Debian developers whining about upstream implies that you expect Debian developers to fix every problem. For instance, I suppose you expect me as XFree86 package maintainer to "configure XFree86 in a usable way for most users" that don't even use video cards that XFree86 supports. If that is true, perhaps you'd like to arrange to get programming documentation from the manufacturers of such video cards, supply me with the hardware in question, and pay me enough money that it would it be a worthwhile use of my time to do such a thing. If *you're* offended by the home/end issue, perhaps *you* should stop complaining and put some of your effort to resolving the limitations of bash/libreadline when it comes to keeping track of the terminal's application-keypad state. Finally, you should be careful to ensure that your words don't get interpreted as coming from the domain identified in the From: line of your message, or the organization identified in your .signature, if your words aren't to be taken as position statements from that company. -- G. Branden Robinson|I'm sorry if the following sounds Debian GNU/Linux |combative and excessively personal, [EMAIL PROTECTED] |but that's my general style. http://people.debian.org/~branden/ |-- Ian Jackson msg08181/pgp0.pgp Description: PGP signature
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 02:45:44PM -0500, Branden Robinson wrote: > On Mon, Oct 21, 2002 at 12:15:16PM -0700, Michael Cardenas wrote: > > On Mon, Oct 21, 2002 at 12:49:34PM -0500, Branden Robinson wrote: > > > On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > > > > > > > - broken home/end keys in bash in xterm (even in Woody) > > > > > > You'll have to persuade the upstream bash/libreadline cabal that > > > actually supporting DEC VT100 and later terminal emulation is a > > > worthwhile thing to do, instead of half-assing their way through it. > > > > Isn't it the downstream packager's job to take upstream and configure > > it in a usable way for most users? It seems that if the patches > > discussed earlier in this thread work, then whining about upstream > > attitudes is just an excuse for not applying them. > > You need to know whereof you speak before you shoot off your mouth like > this. > > * VT220+ terminal emualtion is a stateful thing. Certain keys have a > one meaning in application-keypad mode, and another meaning in normal > mode. Terminal control sequences can (and will) move the terminal in > and out of application-keypad mode. /etc/inputrc doesn't know how to > express "if the terminal is in application-keypad mode, issue this > escape sequence when the Home key is pressed; else issue this other > escape sequence". Is it safe to say that most people are not concerned with the home and end keys when using keypad mode? If not, maybe a sensible approach would be to offer a debconf question about this issue. ... > * Sometimes Debian developers don't have the skills or knowledge > necessary to fix a problem. Hypothetically, it may be the case that ... > simply expect them to do a good job *packaging* software? I think > not; not unless you're willing to underwrite their educational > expenses. > notice that my comment was prepended with "if the patches discussed earlier in this thread work". > Whining about Debian developers whining about upstream implies that you > expect Debian developers to fix every problem. For instance, I suppose > you expect me as XFree86 package maintainer to "configure XFree86 in a > usable way for most users" that don't even use video cards that XFree86 > supports. > Well, since there was no patch offered in this thread to do so, no, I don't expect that. ... > > If *you're* offended by the home/end issue, perhaps *you* should stop > complaining and put some of your effort to resolving the limitations of > bash/libreadline when it comes to keeping track of the terminal's > application-keypad state. > I'm not, because I don't use xterm. I was speaking about the general case. > Finally, you should be careful to ensure that your words don't get > interpreted as coming from the domain identified in the From: line of > your message, or the organization identified in your .signature, if your > words aren't to be taken as position statements from that company. > Thanks, will do. michael -- michael cardenas | lead software engineer, lindows.com hyperpoem.net | GNU/Linux software developer people.debian.org/~mbc | encrypted mail preferred "How terrible to watch a man who has the Incomprehensible within his grasp, doesn't know what to do, and sits down playing with a toy called God." - Fyodor Dostoyevsky msg08172/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
Paul Johnson wrote: > On Mon, Oct 21, 2002 at 12:51:33PM -0400, Joey Hess wrote: > > echo expert >~/.dselect.cfg > > Holy schnikes, Batman! Why isn't this obviously documented? YM in the man page? Well, it is. You can also use dselect --expert of course. -- see shy jo msg08171/pgp0.pgp Description: PGP signature
Re: Two Debian 3.0 reviews at Slashdot
On 21 Oct 2002 13:26:46 +0800 Crispin Wellington <[EMAIL PROTECTED]> wrote: > So I would look at the commercial distro's dependency hell and say "This > is simple fundamental stuff, the kind of thing Debian had sorted out > back in 97 when no one knew what Linux even was." Absolutely. This and the drastic difference in default installation size are what lead me to Debian. > Is Debian really something you want to recommend to your friends? Maybe > other geeks, but most of my friends don't know the computers power > switch from the CD eject button! And I want to "recommend" them an OS > where they'll have to understand what a /dev/ttyS0 is? Bah! Without a doubt. Why not educate your friends about how their PCs work? A fundamental problem today is that people don't understand the "how" and "why". To attempt to protect the user from how a PC operates is IMHO to do much more harm than good. > When people ask me what computer they should use, I ask them, what are > you going to use it for? Again, I agree, but one part of helping them select the right system is to help them understand that system. -- Jamin W. Collins -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Make Debian better (Re: Two Debian 3.0 reviews at Slashdot)
On Mon, Oct 21, 2002 at 12:49:34PM -0500, Branden Robinson wrote: > [I am not subscribed to -user.] > > On Mon, Oct 21, 2002 at 09:07:48AM +0200, Eduard Bloch wrote: > > > - broken home/end keys in bash in xterm (even in Woody) > > You'll have to persuade the upstream bash/libreadline cabal that > actually supporting DEC VT100 and later terminal emulation is a > worthwhile thing to do, instead of half-assing their way through it. > Isn't it the downstream packager's job to take upstream and configure it in a usable way for most users? It seems that if the patches discussed earlier in this thread work, then whining about upstream attitudes is just an excuse for not applying them. michael -- michael cardenas | lead software engineer, lindows.com hyperpoem.net | GNU/Linux software developer people.debian.org/~mbc | encrypted mail preferred "The devils of truth steal the souls of the free." - Trent Reznor msg08154/pgp0.pgp Description: PGP signature
