From andrew.buck at ndsu.edu Wed Sep 3 10:12:11 2008 From: andrew.buck at ndsu.edu (Andrew Buck) Date: Wed Sep 3 10:12:09 2008 Subject: [ale] Wiki article deletion Message-ID: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> Someone nominated the wikipedia article on ale (http://en.wikipedia.org/wiki/Anti-Lamenessing_Engine) for deletion. Thought you would probably like to know. -Buck From pipatron at gmail.com Wed Sep 3 10:32:35 2008 From: pipatron at gmail.com (Anders Andersson) Date: Wed Sep 3 10:32:45 2008 Subject: [ale] Wiki article deletion In-Reply-To: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> Message-ID: <69cdc2b80809031032i1063f69ex1463d34f1cf4a7b5@mail.gmail.com> This might possibly be because there's no links to ALE what-so-ever on google. Each time I want to show it to my friends, I have to dig up the bookmarks. I don't really know why this is. Makes me wonder how I found ALE the first time. // Anders On Wed, Sep 3, 2008 at 7:12 PM, Andrew Buck wrote: > Someone nominated the wikipedia article on ale > (http://en.wikipedia.org/wiki/Anti-Lamenessing_Engine) for deletion. > Thought you would probably like to know. > > -Buck > > > _______________________________________________ > ALE mailing list > ALE@ventricle.dyndns.org > http://ventricle.dyndns.org/cgi-bin/mailman/listinfo/ale > From gmaxwell at gmail.com Wed Sep 3 10:36:08 2008 From: gmaxwell at gmail.com (Gregory Maxwell) Date: Wed Sep 3 10:36:18 2008 Subject: [ale] Wiki article deletion In-Reply-To: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> Message-ID: On Wed, Sep 3, 2008 at 1:12 PM, Andrew Buck wrote: > Someone nominated the wikipedia article on ale > (http://en.wikipedia.org/wiki/Anti-Lamenessing_Engine) for deletion. > Thought you would probably like to know. In fairness, the article really does not meet Wikipedia's criteria for inclusion ? but there are thousands of other pieces of software documented in Wikipedia which also do not. Since Wikipedia is largely without central authority many things are done inconsistently, so "but you have an article on FOO and FOO is even less important" is not recognized as a compelling argument (http://en.wikipedia.org/wiki/Wikipedia:Other_stuff_exists). I, personally, have found the ALE article useful because I forget the darn URL and forget how to spell "lamenessing". So I tossed in some support for the article, but I recognize that my only non-personal reason for wanting the article is that there are so many others like it... so my position should rightly be discounted in the final decision. I think Wikipedia need better coverage of many of the techniques of ALE such as Irani Peleg iterated back-projection, which I think is not covered at all. Such articles are more clearly suited to Wikipedia's purpose than software cataloging, but almost all of them would rightly cite ALE as an example of the technique (it's far more useful to cite ALE than some other pieces of software because ALE is Free Software). Of course, ... someone else here might have some good references why ALE is widely significant. If you have them, please provide them. (for example, examples of ALE being an important component in a high profile project). I'm sure such use exists, but it's probably all private and not known publicly, which isn't good enough for Wikipedia. Such references would be good cause to keep the article. Figuring out why Google seems to punish the ALE main page might also be helpful. Were it not for freshmeat I never would have found the package. From dhilvert at gmail.com Wed Sep 3 12:32:11 2008 From: dhilvert at gmail.com (David Hilvert) Date: Wed Sep 3 12:32:24 2008 Subject: [ale] Naming and site presentation In-Reply-To: References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> Message-ID: <20080903143211.20e394c7.dhilvert@gmail.com> On Wed, 3 Sep 2008 13:36:08 -0400 "Gregory Maxwell" wrote: > I, personally, have found the ALE article useful because I forget the > darn URL and forget how to spell "lamenessing". Proposed solutions to this problem, including renaming, are welcome. > Figuring out why Google seems to punish the ALE main page might also > be helpful. Were it not for freshmeat I never would have found the > package. Proposed solutions would be welcome here, as well; now might be a good time to make changes, given recent work on GLSL and library code. I have no particular preferences about web presentation at this point; I would be quite happy to just push software and documentation edits to git repositories. From shadow at gmail.com Wed Sep 3 13:20:17 2008 From: shadow at gmail.com (Derrick Brashear) Date: Wed Sep 3 13:20:56 2008 Subject: [ale] Naming and site presentation In-Reply-To: <20080903143211.20e394c7.dhilvert@gmail.com> References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> <20080903143211.20e394c7.dhilvert@gmail.com> Message-ID: On Wed, Sep 3, 2008 at 3:32 PM, David Hilvert wrote: > On Wed, 3 Sep 2008 13:36:08 -0400 > "Gregory Maxwell" wrote: > >> I, personally, have found the ALE article useful because I forget the >> darn URL and forget how to spell "lamenessing". > > Proposed solutions to this problem, including renaming, are welcome. > >> Figuring out why Google seems to punish the ALE main page might also >> be helpful. Were it not for freshmeat I never would have found the >> package. > > Proposed solutions would be welcome here, as well; now might be a good time to > make changes, given recent work on GLSL and library code. I have no particular > preferences about web presentation at this point; I would be quite happy to > just push software and documentation edits to git repositories. a non-dyndns domain name and (if you want free something like hosted google) plus one of the open repo sites might help. google code seems to work nicely. From dhilvert at gmail.com Wed Sep 3 13:49:19 2008 From: dhilvert at gmail.com (David Hilvert) Date: Wed Sep 3 13:49:35 2008 Subject: [ale] Naming and site presentation In-Reply-To: References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> <20080903143211.20e394c7.dhilvert@gmail.com> Message-ID: <20080903154919.667e3116.dhilvert@gmail.com> On Wed, 3 Sep 2008 16:20:17 -0400 "Derrick Brashear" wrote: > a non-dyndns domain name and (if you want free something like hosted > google) plus one of the open repo sites might help. google code seems > to work nicely. A different domain name wouldn't be a problem (I think I own zythepsary.com, but this might not be the most easily recalled). On further investigation, I suspect that difficulty in locating the page via Google specifically might stem from the page being generally poorly written rather than anything being wrong with the domain. (E.g., 'lamenessing' returns the page as first out of 1K or so hits, placing it even above Wikipedia's page). As for Google code in particular, I think they're opposed to AGPL, which libale (the new library code, including a perhaps more GPGPU-friendly architecture) is currently licensed under. Sourceforge seems to make downloads irritating for users, so I'd rather avoid them, I think. From shadow at gmail.com Wed Sep 3 14:06:21 2008 From: shadow at gmail.com (Derrick Brashear) Date: Wed Sep 3 14:07:01 2008 Subject: [ale] Naming and site presentation In-Reply-To: <20080903154919.667e3116.dhilvert@gmail.com> References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> <20080903143211.20e394c7.dhilvert@gmail.com> <20080903154919.667e3116.dhilvert@gmail.com> Message-ID: On Wed, Sep 3, 2008 at 4:49 PM, David Hilvert wrote: > On Wed, 3 Sep 2008 16:20:17 -0400 > "Derrick Brashear" wrote: > >> a non-dyndns domain name and (if you want free something like hosted >> google) plus one of the open repo sites might help. google code seems >> to work nicely. > > A different domain name wouldn't be a problem (I think I own zythepsary.com, > but this might not be the most easily recalled). > > On further investigation, I suspect that difficulty in locating the page via > Google specifically might stem from the page being generally poorly written > rather than anything being wrong with the domain. (E.g., 'lamenessing' returns > the page as first out of 1K or so hits, placing it even above Wikipedia's > page). > > As for Google code in particular, I think they're opposed to AGPL, Hadn't considered that; You are correct. > which libale > (the new library code, including a perhaps more GPGPU-friendly architecture) is > currently licensed under. Sourceforge seems to make downloads irritating for > users, so I'd rather avoid them, I think. Hm. BerliOS is the only other option which seems at all plausible in this space, I'd bet. From sopues at gmail.com Wed Sep 3 21:29:36 2008 From: sopues at gmail.com (=?ISO-8859-1?Q?John_B=E4ckstrand?=) Date: Wed Sep 3 21:29:46 2008 Subject: [ale] Wiki article deletion In-Reply-To: References: <1220461931.13004.2.camel@numbercruncher.domain.actdsltmp> Message-ID: Ah, I defended ALE previously when it was flagged for deletion. I think I created the article to begin with. On Wed, Sep 3, 2008 at 19:36, Gregory Maxwell wrote: > On Wed, Sep 3, 2008 at 1:12 PM, Andrew Buck wrote: > > Someone nominated the wikipedia article on ale > > (http://en.wikipedia.org/wiki/Anti-Lamenessing_Engine) for deletion. > > Thought you would probably like to know. > > In fairness, the article really does not meet Wikipedia's criteria for > inclusion ? but there are thousands of other pieces of software > documented in Wikipedia which also do not. Since Wikipedia is > largely without central authority many things are done inconsistently, > so "but you have an article on FOO and FOO is even less important" is > not recognized as a compelling argument > (http://en.wikipedia.org/wiki/Wikipedia:Other_stuff_exists). > > I, personally, have found the ALE article useful because I forget the > darn URL and forget how to spell "lamenessing". So I tossed in some > support for the article, but I recognize that my only non-personal > reason for wanting the article is that there are so many others like > it... so my position should rightly be discounted in the final > decision. > > I think Wikipedia need better coverage of many of the techniques of > ALE such as Irani Peleg iterated back-projection, which I think is not > covered at all. Such articles are more clearly suited to Wikipedia's > purpose than software cataloging, but almost all of them would rightly > cite ALE as an example of the technique (it's far more useful to cite > ALE than some other pieces of software because ALE is Free Software). > > Of course, ... someone else here might have some good references why > ALE is widely significant. If you have them, please provide them. (for > example, examples of ALE being an important component in a high > profile project). I'm sure such use exists, but it's probably all > private and not known publicly, which isn't good enough for Wikipedia. > Such references would be good cause to keep the article. > > Figuring out why Google seems to punish the ALE main page might also > be helpful. Were it not for freshmeat I never would have found the > package. > > _______________________________________________ > ALE mailing list > ALE@ventricle.dyndns.org > http://ventricle.dyndns.org/cgi-bin/mailman/listinfo/ale > > -- John B?ckstrand From dhilvert at gmail.com Sat Sep 6 12:01:15 2008 From: dhilvert at gmail.com (David Hilvert) Date: Sat Sep 6 12:01:28 2008 Subject: [ale] libale guile backend runs a simple client Message-ID: <20080906140115.c0bc36b7.dhilvert@gmail.com> The libale guile backend now runs a simple client. To try this, do git-clone git://repo.or.cz/libale.git cd libale ./bootstrap ./configure make ./beer An unfinished GLSL backend can also be run with: ./beer --glsl Note that the bootstrap script requires the following: autoconf-archive autoconf automake libtool swig These can be installed with 'apt-get install ' on most apt-enabled systems. Other packages required for build are listed in the README. If I have forgotten anything, or if anything breaks, please send bug reports or (if possible) patches to ale@ventricle.dyndns.org . From zouookmoetenwerken at gmail.com Tue Sep 9 02:37:02 2008 From: zouookmoetenwerken at gmail.com (max power) Date: Tue Sep 9 02:37:12 2008 Subject: [ale] libale dependencies Message-ID: Hello, I tried to build the git version of libale today, and it seems one more dependency should be added to the list. Initially, configure gave the following error: checking for glewInit in -lGLEW... yes checking libguile.h usability... no checking libguile.h presence... no checking for libguile.h... no configure: error: Usable guile library was not found. (libguile.h not found) After which I installed the guile-1.8-dev package. With this package, the configuration worked, but it emitted a few warnings: checking libguile.h usability... yes checking libguile.h presence... yes checking for libguile.h... yes ./configure: line 23680: GUILE_PROGS: command not found ./configure: line 23681: GUILE_FLAGS: command not found checking whether scm_from_locale_string is declared... yes configure: creating ./config.status config.status: creating Makefile The actual build however seems to fail on some scm_ components. perhaps I am using the wrong version of the guile library? (cd .libs && rm -f libale.la && ln -s ../libale.la libale.la) /bin/sh ./libtool --tag=CC --mode=link gcc -I./include -g -O2 -Wall -o beer beer.o libale.la mv -f .deps/libale-guile.Tpo .deps/libale-guile.Plo /bin/sh ./libtool --tag=CC --mode=link gcc -I./include -g -O2 -Wall -o libale-guile.la -rpath /usr/local/lib libale-guile.lo gcc -I./include -g -O2 -Wall -o .libs/beer beer.o ./.libs/libale.so ./.libs/libale.so: undefined reference to `scm_init_guile' ./.libs/libale.so: undefined reference to `scm_from_uint64' ./.libs/libale.so: undefined reference to `scm_string_append' ./.libs/libale.so: undefined reference to `scm_eval_string' ./.libs/libale.so: undefined reference to `scm_from_int32' ./.libs/libale.so: undefined reference to `scm_list_3' ./.libs/libale.so: undefined reference to `scm_from_double' ./.libs/libale.so: undefined reference to `scm_append' ./.libs/libale.so: undefined reference to `scm_to_double' ./.libs/libale.so: undefined reference to `scm_list_2' ./.libs/libale.so: undefined reference to `scm_to_locale_string' ./.libs/libale.so: undefined reference to `scm_call_4' ./.libs/libale.so: undefined reference to `scm_from_locale_string' ./.libs/libale.so: undefined reference to `scm_to_int32' ./.libs/libale.so: undefined reference to `scm_c_eval_string' ./.libs/libale.so: undefined reference to `scm_to_uint64' ./.libs/libale.so: undefined reference to `scm_list_1' ./.libs/libale.so: undefined reference to `scm_primitive_eval' ./.libs/libale.so: undefined reference to `scm_c_define_gsubr' collect2: ld returned 1 exit status make[1]: *** [beer] Error 1 make[1]: *** Waiting for unfinished jobs.... gcc -shared .libs/libale-guile.o -Wl,-soname -Wl,libale-guile.so.0 -o .libs/libale-guile.so.0.0.0 (cd .libs && rm -f libale-guile.so.0 && ln -s libale-guile.so.0.0.0 libale-guile.so.0) (cd .libs && rm -f libale-guile.so && ln -s libale-guile.so.0.0.0 libale-guile.so) ar cru .libs/libale-guile.a libale-guile.o ranlib .libs/libale-guile.a creating libale-guile.la (cd .libs && rm -f libale-guile.la && ln -s ../libale-guile.la libale-guile.la) make[1]: Leaving directory `/lhome/jorik/build/software/image_processing/ale-dev/libale' make: *** [all] Error 2 kind regards, Jorik Blaas From dhilvert at gmail.com Sat Sep 13 13:02:22 2008 From: dhilvert at gmail.com (David Hilvert) Date: Sat Sep 13 13:02:42 2008 Subject: [ale] libale dependencies In-Reply-To: References: Message-ID: <20080913150222.687a176d.dhilvert@gmail.com> On Tue, 9 Sep 2008 11:37:02 +0200 "max power" wrote: > Hello, > > I tried to build the git version of libale today, and it seems one more > dependency should be added to the list. > > Initially, configure gave the following error: > > checking for glewInit in -lGLEW... yes > checking libguile.h usability... no > checking libguile.h presence... no > checking for libguile.h... no > configure: error: Usable guile library was not found. (libguile.h not found) > > After which I installed the guile-1.8-dev package. > > With this package, the configuration worked, but it emitted a few warnings: > > checking libguile.h usability... yes > checking libguile.h presence... yes > checking for libguile.h... yes > ./configure: line 23680: GUILE_PROGS: command not found > ./configure: line 23681: GUILE_FLAGS: command not found > checking whether scm_from_locale_string is declared... yes > configure: creating ./config.status > config.status: creating Makefile Please try running './bootstrap -r' with guile-1.8-dev installed and re-configuring. The guile-1.8-dev package includes a file that must be present at the time of bootstrap; otherwise, configure and build will fail. Indeed, this package should be added to the list of packages required at bootstrap time. Thanks, David Hilvert From silfreed at silfreed.net Wed Sep 24 06:36:08 2008 From: silfreed at silfreed.net (Douglas E. Warner) Date: Wed Sep 24 06:36:33 2008 Subject: [ale] Patch to support GCC 4.3 Message-ID: <200809240936.14247.silfreed@silfreed.net> Skipped content of type multipart/mixed-------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part. Url : http://ventricle.dyndns.org/pipermail/ale/attachments/20080924/7c5aaf5b/attachment.pgp From silfreed at silfreed.net Wed Sep 24 06:50:38 2008 From: silfreed at silfreed.net (Douglas E. Warner) Date: Wed Sep 24 06:50:51 2008 Subject: [ale] Patch to support GCC 4.3 In-Reply-To: <200809240936.14247.silfreed@silfreed.net> References: <200809240936.14247.silfreed@silfreed.net> Message-ID: <200809240950.43423.silfreed@silfreed.net> On Wednesday 24 September 2008 09:36:08 Douglas E. Warner wrote: > I've been sitting on this patch for awhile and don't think I sent it to the > list; it adds a missing include needed to support GCC 4.3 (used on Fedora 9 > and up). ?It's a patch against ALE 0.9.0.1. Mailing list stripped my patch, so I've sent it to David and HJ from the authors files. -Doug -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part. Url : http://ventricle.dyndns.org/pipermail/ale/attachments/20080924/eaa2b4d8/attachment.pgp From dhilvert at gmail.com Thu Sep 25 09:54:57 2008 From: dhilvert at gmail.com (David Hilvert) Date: Thu Sep 25 09:55:12 2008 Subject: [ale] Patch to support GCC 4.3 In-Reply-To: <200809240950.43423.silfreed@silfreed.net> References: <200809240936.14247.silfreed@silfreed.net> <200809240950.43423.silfreed@silfreed.net> Message-ID: <20080925115457.d7e11cdf.dhilvert@gmail.com> On Wed, 24 Sep 2008 09:50:38 -0400 "Douglas E. Warner" wrote: > Mailing list stripped my patch, so I've sent it to David and HJ from the > authors files. Thanks for the patch; I've applied this to git HEAD. It should be possible to include patches in-line as message text, rather than as attachments; I suspect this would be preferable.