Re: [Fink-devel] fink vs. the alternatives
On 11/10/06, Paul Mitchum <[EMAIL PROTECTED]> wrote: On Nov 10, 2006, at 1:27 AM, "Lars Rosengreen"<[EMAIL PROTECTED]> wrote:> How about making the unstable tree enabled by default on new installs? >Or, perhaps *move packages into the stable tree on occassion.*A lot of maintainers wait for some positive feedback from users before moving a package into stable. If you use a package and think it is ready for stable, please let the maintainer know. -- Lars Rosengreenhttp://www.phylopy.org/~lars - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
On Nov 10, 2006, at 4:18 AM, Paul Mitchum wrote: > > On Nov 10, 2006, at 1:27 AM, "Lars Rosengreen" > <[EMAIL PROTECTED]> wrote: > >> How about making the unstable tree enabled by default on new >> installs? >> > > > Or, perhaps *move packages into the stable tree on occassion.* > +1 - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
On 11/10/06, Charles Lepple <[EMAIL PROTECTED]> wrote: > On 11/10/06, Paul Mitchum <[EMAIL PROTECTED]> wrote: > > > > On Nov 10, 2006, at 1:27 AM, "Lars Rosengreen" > > <[EMAIL PROTECTED]> wrote: > > > > > How about making the unstable tree enabled by default on new installs? > > > > > > > > > Or, perhaps *move packages into the stable tree on occassion.* > > Testing packages for stable is harder than it appears at first glance. > If I want to move a GTK+-based package into stable, I first need to > check that the version of GTK+ itself will not cause any problems. > Then, it helps to actually compile in a real stable tree (so that > stable does not lose its value due to maintainers just adding whatever > packages they assume will work). > > -- > - Charles Lepple > It also helps if there is positive user feedback--especially on packages that a maintainer doesn't use themselves. So as a generality, a package should 1) have all of its dependencies (of appropriate versioning) in the stable tree 2) follow all other policies, e.g. the Shlibs poilicy to go into the stable tree. I -now- (there have been prior mis-commits) have a clean build setup for stable, so I use that to check whether my packages build to make sure that all the dependencies exist in stable and that it doesn't somehow generate a bad deb file (e.g. no executable) when compilied from scratch as is done in generating the binary distribution. Also, maintainers without commit access need to put out requests to move their packages to stable more often. We've kicked around the prospects of a 3-tiered system on the IRC channel: stable--as it has been testing--the better packages from unstable unstable--this might even be set up to allow commits from registered maintainers without giving everybody full access. These packages would be moved to testing by someone with wide commits access once they've been verified to pass criteria (TBD). -- Alexander K. Hansen Fink Documenter (still) - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
On 11/10/06, Paul Mitchum <[EMAIL PROTECTED]> wrote: > > On Nov 10, 2006, at 1:27 AM, "Lars Rosengreen" > <[EMAIL PROTECTED]> wrote: > > > How about making the unstable tree enabled by default on new installs? > > > > > Or, perhaps *move packages into the stable tree on occassion.* Testing packages for stable is harder than it appears at first glance. If I want to move a GTK+-based package into stable, I first need to check that the version of GTK+ itself will not cause any problems. Then, it helps to actually compile in a real stable tree (so that stable does not lose its value due to maintainers just adding whatever packages they assume will work). -- - Charles Lepple - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
On Nov 10, 2006, at 1:27 AM, "Lars Rosengreen" <[EMAIL PROTECTED]> wrote: > How about making the unstable tree enabled by default on new installs? > Or, perhaps *move packages into the stable tree on occassion.* - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
On 11/9/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: I believe that the term "unstable" scares many people away from using the unstable tree. They think unstable means "might blow up your system", when it really means "changes several times a day". Sure, any new package version might have a bug, but there is a good chance it will be fixed within a day or so. If a package in the binary stable distribution has a bug (or the program that the package installs has a bug), it will be months before it is fixed. In fact, given that the newer program versions in the unstable tree probably have bug fixes that aren't in the stable tree, many users will find that the unstable tree works better. How about making the unstable tree enabled by default on new installs? MacPorts doesn't have anything like a stable tree, and they seem to get along fine without it. I think more new users would benefit from the greater numbers and more up to date packages in the unstable tree than would be hurt by the frequency of updates or brokenness. In those specific cases where a user needs a more conservative approach, they can continue to use the stable tree. A more neutral name for the unstable tree, as others have suggested, might be a good idea too -- how about "standard" or "default" or "current" :) -- Lars Rosengreenhttp://www.phylopy.org/~lars - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
> > "static" and "dynamic", maybe?
>
> or stable and testing, since that's pretty much exactly what they are...
stable, {latest,current,development,bleeding-edge}?
Even more than one non-stable tree might work, but that might incur more
work on the part of maintainers.
I've also heard from colleagues who were under the false impression that
'unstable' is something the average user should avoid. Most people aren't
even aware of it. (I can't live without it!) Perhaps fink-configure
could add a prompt to enable the unstable tree? (Instead of: "Edit
/sw/etc/fink.conf as root using your favorite text editor, and insert the
following magic words on line...")
Fang
David Fang
Computer Systems Laboratory
Electrical & Computer Engineering
Cornell University
http://www.csl.cornell.edu/~fang/
-- (2400 baud? Netscape 3.0?? lynx??? No problem!)
-
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
___
Fink-devel mailing list
Fink-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Nov 9, 2006, at 11:12 PM, [EMAIL PROTECTED] wrote: > > "static" and "dynamic", maybe? or stable and testing, since that's pretty much exactly what they are... - -chris zubrzycki - - -- PGP public key: http://homepage.mac.com/beren/publickey.txt ID: 0xA2ABC070 Fprint: 26B0 BA6B A409 FA83 42B3 1688 FBF9 8232 A2AB C070 A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing on usenet and in e-mail? -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.5 (Darwin) Comment: Please sign reply-http://www.gnupg.org iEYEARECAAYFAkVT/ukACgkQ+/mCMqKrwHAPegCePbVIsTThX+BRlsHT64rjudag rN8AoNVHhslJnrCFlC9rpDv8XrgBiK9w =ai5q -END PGP SIGNATURE- - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
"static" and "dynamic", maybe? On Thu, 9 Nov 2006 19:24:02 -0800 (PST) [EMAIL PROTECTED] wrote: > > - Original Message >From: William Scott <[EMAIL PROTECTED]> > > I wonder if this is something we should worry about, or >whether we > should be like my grad student friend? I noticed rather >few had > anything of substance to say, but there were references >to fink being > more cumbersome, the "unstable" branch being the only >one up to date, > etc. > > Is it worth assessing how fink is presented to the user >community, or > the demarcation between stable and unstable (or even >whether the name > "unstable", which I assume is inherited > from debian, really conveys > the intended message)? > > === > I believe that the term "unstable" scares many people >away from using the unstable tree. They think unstable >means "might blow up your system", when it really means >"changes several times a day". Sure, any new package >version might have a bug, but there is a good chance it >will be fixed within a day or so. If a package in the >binary stable distribution has a bug (or the program that >the package installs has a bug), it will be months before >it is fixed. In fact, given that the newer program >versions in the unstable tree probably have bug fixes >that aren't in the stable tree, many users will find that >the unstable tree works better. > > Rather than "stable" and "unstable", I believe we should >find some other names that will be less confusing to >those who do not have a Debian background. How about >"out_of_date" and "up_to_date"? ;) > > Another thing - we need to speed up the process of >moving packages from the unstable to the stable trees, >and release > binary distributions more often. Right now we don't >move packages until we get sufficient positive feedback. > But, people generally never think to give positive >feedback. However, if a package is broken, there is a >very good chance that people will complain. So the >absence of negative feedback should be sufficient to >allow a package to be moved to the stable tree. If no >one complains two weeks after a package is revised in the >unstable tree, then it should be moved to stable. New >binary distributions should be done as often as possible >(monthly?). > > > > > > Kevin Horton > Ottawa, Canada > - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
- Original Message From: William Scott <[EMAIL PROTECTED]>I wonder if this is something we should worry about, or whether we should be like my grad student friend? I noticed rather few had anything of substance to say, but there were references to fink being more cumbersome, the "unstable" branch being the only one up to date, etc.Is it worth assessing how fink is presented to the user community, or the demarcation between stable and unstable (or even whether the name "unstable", which I assume is inherited from debian, really conveys the intended message)?===I believe that the term "unstable" scares many people away from using the unstable tree. They think unstable means "might blow up your system", when it really means "changes several times a day". Sure, any new package version might have a bug, but there is a good chance it will be fixed within a day or so. If a package in the binary stable distribution has a bug (or the program that the package installs has a bug), it will be months before it is fixed. In fact, given that the newer program versions in the unstable tree probably have bug fixes that aren't in the stable tree, many users will find that the unstable tree works better.Rather than "stable" and "unstable", I believe we should find some other names that will be less confusing to those who do not have a Debian background. How about "out_of_date" and "up_to_date"? ;)Another thing - we need to speed up the process of moving packages from the unstable to the stable trees, and release binary distributions more often. Right now we don't move packages until we get sufficient positive feedback. But, people generally never think to give positive feedback. However, if a package is broken, there is a very good chance that people will complain. So the absence of negative feedback should be sufficient to allow a package to be moved to the stable tree. If no one complains two weeks after a package is revised in the unstable tree, then it should be moved to stable. New binary distributions should be done as often as possible (monthly?).Kevin HortonOttawa, Canada- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
>Hi folks: Hello Bill. >Periodically, I see posted on other mailing lists or bulletin boards >a question like "should I use fink or Darwinports or ..."? I'm not >sure what it says about my state of mind, but I always read these and >come away somewhat disappointed, because fink often seems to take a >beating. From a marketing point of view this should concern me. Somehow it does not. MOst of the comments and statements I have read are either incorrect or they work from the wrong assumptions. For example not understanding how our stable tree really works and that there is a good reason for it to be so 'outdated'. >Of the replies, I seem to be the only person who likes fink. Probably because most do not bother to answer any more. Fink is still the one project that offers most packages. Sooner or later most end up with the system that we emply for the community. >(When I first started grad school a thousand years ago, I met someone >who told me he was from Brooklyn. I said I had never been there. His >reply was "That's ok. You don't need Brooklyn, and Brooklyn doesn't >need you.) Unfortunately that attitude is something we cannot cultivate. There are many things we could do for better Public relations, unfortunately the fink core team is a bit slow in the decision making process when it comes to deciding how to further FDN and get the necessary to fund such efforts. >I wonder if this is something we should worry about, or whether we >should be like my grad student friend? I noticed rather few had >anything of substance to say, but there were references to fink being >more cumbersome, the "unstable" branch being the only one up to date, >etc. No, I would not worry about it. I do not see that these sporadic message have any say in tipping the scale on the opinion one of the majority might have about Fink. However if we can identify the key criticsm then we can counter act with information management. >Is it worth assessing how fink is presented to the user community, or I think I have a pretty good picture of that. >the demarcation between stable and unstable (or even whether the name >"unstable", which I assume is inherited from debian, really conveys >the intended message)? Yes this is indeed something that we might review. >Sorry in advance if I am just being neurotic about this. You are most welcome and I do love to see such concerns coming up from the community, usually I am the one and I was alone with it :) -d - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
Re: [Fink-devel] fink vs. the alternatives
Bill, I think you've raised a valid concern. The big problem, as I see it, is that the fink project doesn't have the manpower to put fink into the kind of better shape which would make users happier about it. And, not to put too fine a point on it, the fact that not only users, but also developers, have a choice between fink and macports doesn't help this situation -- the very existence of their project is a drain on manpower resources for ours and vice versa. Is it time for a merger or for one project to throw in the towel? Probably not. But in the long run, I don't see how the community will be able to sustain both. -- Dave On Nov 8, 2006, at 5:29 PM, William Scott wrote: > > Hi folks: > > Periodically, I see posted on other mailing lists or bulletin boards > a question like "should I use fink or Darwinports or ..."? I'm not > sure what it says about my state of mind, but I always read these and > come away somewhat disappointed, because fink often seems to take a > beating. > > The most current example is a thread on the TextMate mailing list. > (TextMate is a nice editor for OS X, btw). > http://murl.info/14677 > > Of the replies, I seem to be the only person who likes fink. > > (When I first started grad school a thousand years ago, I met someone > who told me he was from Brooklyn. I said I had never been there. His > reply was "That's ok. You don't need Brooklyn, and Brooklyn doesn't > need you.) > > I wonder if this is something we should worry about, or whether we > should be like my grad student friend? I noticed rather few had > anything of substance to say, but there were references to fink being > more cumbersome, the "unstable" branch being the only one up to date, > etc. > > Is it worth assessing how fink is presented to the user community, or > the demarcation between stable and unstable (or even whether the name > "unstable", which I assume is inherited from debian, really conveys > the intended message)? > > Sorry in advance if I am just being neurotic about this. > > Bill > > > > > -- > --- > Using Tomcat but need to do more? Need to support web services, > security? > Get stuff done quickly with pre-integrated technology to make your > job easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache > Geronimo > http://sel.as-us.falkag.net/sel? > cmd=lnk&kid=120709&bid=263057&dat=121642 > ___ > Fink-devel mailing list > Fink-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/fink-devel - Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 ___ Fink-devel mailing list Fink-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-devel
