From tanner.burson at gmail.com Thu Mar 2 12:18:41 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Thu, 2 Mar 2006 11:18:41 -0600 Subject: [Instantrails-developers] Build Script Updates Message-ID: Hello All, Sorry for the delay in getting this out to everyone. I've got a couple of links for you all, one is to a tgz file containing my full build directory for IR to this point. This includes the ruby sources, the XAMPP sources, and all the dependencies I needed, as well as the script to build it to this point. I've got a second tgz file that is JUST the script file, and a README (which I'll post here as well). The build script is not fully functional at this point, but it will build ruby, gems, rails, scgi, and put them in the proper places in the XAMPP install. The README contains a lot more information on this, as well as my TODO list for the script. I know there are several of you out there waiting to dig in, so here's a place to start. At this point the script is really a rough port and rewrite of the rorox compile.sh script. If you have questions on where the next place to go is, or what is missing, you can check that script (available in the download from http://rubyforge.org/projects/rorox). Post any problems or questions to this list...Thanks! Full install http://www.tannerburson.com/ir/ir_full_build.tar.gz - 91 MB (Not fully uploaded at the time of this posting, wait about an hour before downloading) Script only http://www.tannerburson.com/ir/ir_build.tar.gz - 2.4 KB README only http://www.tannerburson.com/ir/README -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... From tanner.burson at gmail.com Fri Mar 3 09:59:47 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Fri, 3 Mar 2006 08:59:47 -0600 Subject: [Instantrails-developers] Build script progress Message-ID: Hello All, Sorry for the delay in getting this out to everyone. (And sorry if you're getting this again, I never got a copy the first time) I've got a couple of links for you all, one is to a tgz file containing my full build directory for IR to this point. This includes the ruby sources, the XAMPP sources, and all the dependencies I needed, as well as the script to build it to this point. I've got a second tgz file that is JUST the script file, and a README (which I'll post here as well). The build script is not fully functional at this point, but it will build ruby, gems, rails, scgi, and put them in the proper places in the XAMPP install. The README contains a lot more information on this, as well as my TODO list for the script. I know there are several of you out there waiting to dig in, so here's a place to start. At this point the script is really a rough port and rewrite of the rorox compile.sh script. If you have questions on where the next place to go is, or what is missing, you can check that script (available in the download from http://rubyforge.org/projects/rorox). Post any problems or questions to this list...Thanks! Full install http://www.tannerburson.com/ir/ir_full_build.tar.gz - 91 MB (Not fully uploaded at the time of this posting, wait about an hour before downloading) Script only http://www.tannerburson.com/ir/ir_build.tar.gz - 2.4 KB README only http://www.tannerburson.com/ir/README -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060303/ce410bb3/attachment.htm From curt.hibbs at gmail.com Mon Mar 6 12:52:11 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Mon, 6 Mar 2006 11:52:11 -0600 Subject: [Instantrails-developers] AJAX Scaffold Generator Message-ID: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> I haven't tried it yet, but this looks like it could really help us get started, the AJAX Scaffold Generator: HTTP://ajaxscaffold.height1percent.com/ajax_scaffold_demo - Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060306/9842fd00/attachment.htm From mortonda at dgrmm.net Mon Mar 6 12:52:36 2006 From: mortonda at dgrmm.net (David Morton) Date: Mon, 06 Mar 2006 11:52:36 -0600 Subject: [Instantrails-developers] AJAX Scaffold Generator In-Reply-To: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> References: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> Message-ID: <440C76E4.7060105@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > I haven't tried it yet, but this looks like it could really help us get > started, the AJAX Scaffold Generator: > > HTTP://ajaxscaffold.height1percent.com/ajax_scaffold_demo > maybe, maybe not. In my experince, the scaffolding doesn't last very long. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEDHbkSIxC85HZHLMRAkbWAKCGIETKSdyw5UqqDWF0wvQvemgrbwCgiHPX Oeb4BrbDwUmwOqnKz+f8BzY= =OQr/ -----END PGP SIGNATURE----- From tanner.burson at gmail.com Mon Mar 6 13:36:20 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Mon, 6 Mar 2006 12:36:20 -0600 Subject: [Instantrails-developers] AJAX Scaffold Generator In-Reply-To: <440C76E4.7060105@dgrmm.net> References: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> <440C76E4.7060105@dgrmm.net> Message-ID: On 3/6/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Curt Hibbs wrote: > > I haven't tried it yet, but this looks like it could really help us get > > started, the AJAX Scaffold Generator: > > > > HTTP://ajaxscaffold.height1percent.com/ajax_scaffold_demo > > > > maybe, maybe not. In my experince, the scaffolding doesn't last very > long. > > > - -- > David Morton > Maia Mailguard - http://www.maiamailguard.com > Morton Software Design and Consulting - http://www.dgrmm.net > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.2 (MingW32) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFEDHbkSIxC85HZHLMRAkbWAKCGIETKSdyw5UqqDWF0wvQvemgrbwCgiHPX > Oeb4BrbDwUmwOqnKz+f8BzY= > =OQr/ > -----END PGP SIGNATURE----- > _______________________________________________ > Instantrails-developers mailing list > Instantrails-developers at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-developers > Completely off topic here, but did my previous posts about the build script actually hit the list? Looking in my history I never actually got them back from the list. -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060306/1cfc72c4/attachment.htm From curt.hibbs at gmail.com Mon Mar 6 13:55:47 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Mon, 6 Mar 2006 12:55:47 -0600 Subject: [Instantrails-developers] AJAX Scaffold Generator In-Reply-To: References: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> <440C76E4.7060105@dgrmm.net> Message-ID: <31d15f490603061055o4a19a880kae1a8f7578d5e145@mail.gmail.com> On 3/6/06, Tanner Burson wrote: > > > Completely off topic here, but did my previous posts about the build > script actually hit the list? Looking in my history I never actually got > them back from the list. > > I received them. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060306/04a04c44/attachment-0001.htm From tanner.burson at gmail.com Mon Mar 6 13:58:17 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Mon, 6 Mar 2006 12:58:17 -0600 Subject: [Instantrails-developers] AJAX Scaffold Generator In-Reply-To: <31d15f490603061055o4a19a880kae1a8f7578d5e145@mail.gmail.com> References: <31d15f490603060952n7a10f670j708b4b21a9961fe4@mail.gmail.com> <440C76E4.7060105@dgrmm.net> <31d15f490603061055o4a19a880kae1a8f7578d5e145@mail.gmail.com> Message-ID: On 3/6/06, Curt Hibbs wrote: > > On 3/6/06, Tanner Burson wrote: > > > > > > Completely off topic here, but did my previous posts about the build > > script actually hit the list? Looking in my history I never actually got > > them back from the list. > > > > > I received them. > Thanks. Curt > > _______________________________________________ > Instantrails-developers mailing list > Instantrails-developers at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-developers > > -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060306/bd4bf9ab/attachment.htm From curt.hibbs at gmail.com Sat Mar 18 01:48:52 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Sat, 18 Mar 2006 00:48:52 -0600 Subject: [Instantrails-developers] I'm speaking about Instant Rails at RailsConf Message-ID: <31d15f490603172248h91ae11ah78beef6ca6f3eddb@mail.gmail.com> My talk proposal for RailsConf in June was accepted, its titled "Inside Instant Rails". Its about how Instant Rails works (people are particularly curious about the "no touch" installation). I'll mainly talk about Instant Rails 1, but I'd also like so say as much as I can about Instant Rails 2. So whatever we can get accomplished between now and June I'll cover (as well as future plans). This will be a great opportunity to get some exposure for IR 2! Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060318/54536f8b/attachment.htm From mortonda at dgrmm.net Tue Mar 28 10:07:19 2006 From: mortonda at dgrmm.net (David Morton) Date: Tue, 28 Mar 2006 09:07:19 -0600 Subject: [Instantrails-developers] rails 1.1 Message-ID: <44295127.4070504@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Since it should be fairly easy, I figure we should package the rails 1.1 release and release an IR 1.1 pretty soon, shouldn't we? - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKVEnSIxC85HZHLMRAnctAJ49YVfr2cTROHEc8NnMELCTO6hYwACeOIBm YuqRFKEYShz80fRHIUUlv2c= =vs5E -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Tue Mar 28 10:50:19 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Tue, 28 Mar 2006 09:50:19 -0600 Subject: [Instantrails-developers] rails 1.1 In-Reply-To: <44295127.4070504@dgrmm.net> References: <44295127.4070504@dgrmm.net> Message-ID: <31d15f490603280750r199a24carc7689454786e72ef@mail.gmail.com> On 3/28/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Since it should be fairly easy, I figure we should package the rails 1.1release > and release an IR 1.1 pretty soon, shouldn't we? Yes, in fact I'd like to get it out today or tomorrow. I want to package it with the One-Click Ruby Installer for the Ruby runtime, but we I need to fix one problem in the installer that keeps it from being used with the RDT eclipse plugin. David, if I get an updated one-click installer out today (say by mid-day), can you handle deleting the old ruby tree out of SVN, replacing it with the installer, installing all of the rubygems, and then checking it in? This would be impossible for me to do during the day because I can't get at SVN through our corporate firewall and have to use my cellphone as a dialup modem. That's fine for small stuff, but for something this big it wouldn't work. Anyway, if you could do that then I could do the actual release tonight when I get home (about 10pm central). Does this work for you? If not, then I can do all that when I get home, and we can release the next day. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060328/86c0ed13/attachment.htm From mortonda at dgrmm.net Tue Mar 28 11:11:30 2006 From: mortonda at dgrmm.net (David Morton) Date: Tue, 28 Mar 2006 10:11:30 -0600 Subject: [Instantrails-developers] rails 1.1 In-Reply-To: <31d15f490603280750r199a24carc7689454786e72ef@mail.gmail.com> References: <44295127.4070504@dgrmm.net> <31d15f490603280750r199a24carc7689454786e72ef@mail.gmail.com> Message-ID: <44296032.4090404@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > David, if I get an updated one-click installer out today (say by > mid-day), can you handle deleting the old ruby tree out of SVN, > replacing it with the installer, installing all of the rubygems, and > then checking it in? I can do some work on it, but I'm not sure if we can get it by tonight. Depends on if my son will let me work or if he is "clingy". (He's teething) I did the simple gem update for rails on the existing code base, and it seems to work well. Go ahead and work on the one-click, and I'll start cleaning out the svn tree to get it ready... I assume you want to do like last time and completely remove the ruby tree and then replace it with fresh sources? Also, watch the iconv stuff, the add-on zip file put the dll in lib, but it seems everyone needs it in bin to work. That's going to be part of the one-click installer, isn't it? - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKWAySIxC85HZHLMRAgSKAJwP5nUfvX/GfeP6ZaHcicJK1fVNxQCfSX/i Qr0feT/zcPwJEjxQd1RUIQI= =gq7a -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Tue Mar 28 11:25:28 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Tue, 28 Mar 2006 10:25:28 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: References: <31d15f490603271337l6a1201ffh7aa6f77dcb9fedfc@mail.gmail.com> Message-ID: <31d15f490603280825x19dd15a2n4e1b4e806bd37297@mail.gmail.com> Thanks, Zed. I've been following your work on Mongrel, and it looks really good. I originally wanted to use Lighttpd, but canned that when I realized I'd have to do a native windows port myself (it only ran under cygwin at the time). IR-Dev Team: I've been meaning to pose this question and hadn't gotten around to it, but Zed has given me the nudge here (see Zed's message below). For Instant Rails 2 would it make sense for use to use Mongrel, and if so, then how? It could be as an SCGI replacement with Apache proxying requests to it (we could also allow standalone use just like WEBrick). We could replace Apache entirely and just use Mongrel (this would be simpler, but less flexible). Or we could just stick with SCGI. What do you think? Curt On 3/28/06, Zed Shaw wrote: > > Hi Curt, > > I haven't really ran into anything like this, but I've been head down in > Mongrel code lately. I think I'm pretty close to spending a weekend porting > some of the mongrel stuff over to SCGI, but I was wondering if you'd be > interested in working on getting instant rails working with just Mongrel? > Could solve all these problems and also has the win32 service stuff which > should make for a very nice clean setup. > > To debug this problem though I'd start trying to rule out whether it's his > application, the machine, or just SCGI. Have him make a blank test > application and see how fast that runs. If that runs slow then it's > probably SCGI. If it's fast then his application is probably doing > something funky. > > Also, see if he can post more particulars of the setup like machine speed > etc. I've seen some platforms for weird reasons go incredibly slow on > windows, and then others are fast and snappy. > > Zed > > > On 3/27/06 4:37 PM, "Curt Hibbs" wrote: > > Steve, I don't know what is going on here, but I copied Zed Shaw on this > (aka Mr. SCGI), because he might be able to help. > > Curt > > On 3/27/06, *Debaun, Steve* < Steve.Debaun at seminis.com > > wrote: > > Hi all, > > So far, I'm loving the package... kudos to Curt for such a great > contribution! Currently, however, I'm having a problem with getting SCGI > to > behave. > > InstantRails 1.0 > > Using a simple test app, I see response times like: > - Webrick: 3 seconds > - SCGI: 15 seconds > > Further oddities... the log/scgi.log is just being populated with > start-event lines, no request handling is being logged at all. In fact, > if > I don't start the SCGI handler, the page still responds (and just as > slowly). This leads me to believe that the SCGI handler isn't even being > called, and that 'something else' is handling the requests very slowly. > > The :port: setting in scgi.yaml is set to 7001, and the Apache conf has a > VirtualHost with: > SCGIMount /dispatch.fcgi 127.0.0.1:7001 . > > > SCGI mode is production. I have no hosts entry because the hostname is a > real dns name. > > Has anyone run into any problems with extreme SCGI slowness? Can anyone > explain the things that are confusing me (no log entries, page renders > despite scgi not being started)? > > sd > This e-mail message may contain privileged and/or confidential > information, > and is intended to be received only by persons entitled to receive such > information. If you have received this e-mail in error, please notify the > sender immediately. Please delete it and all attachments from any servers, > hard drives or any other media. Other use of this e-mail by you is > strictly > prohibited. > All e-mails and attachments sent and received are subject to monitoring, > reading and archival by Seminis. The recipient of this e-mail is solely > responsible for checking for the presence of "Viruses" or other "Malware". > > Seminis accepts no liability for any damage caused by any such code > transmitted by or accompanying this e-mail or any attachment. > _______________________________________________ > Instantrails-users mailing list > Instantrails-users at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-users > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060328/46cb8b33/attachment-0001.htm From tanner.burson at gmail.com Tue Mar 28 11:56:02 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Tue, 28 Mar 2006 10:56:02 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: <31d15f490603280825x19dd15a2n4e1b4e806bd37297@mail.gmail.com> References: <31d15f490603271337l6a1201ffh7aa6f77dcb9fedfc@mail.gmail.com> <31d15f490603280825x19dd15a2n4e1b4e806bd37297@mail.gmail.com> Message-ID: On 3/28/06, Curt Hibbs wrote: > > Thanks, Zed. > > I've been following your work on Mongrel, and it looks really good. I > originally wanted to use Lighttpd, but canned that when I realized I'd have > to do a native windows port myself (it only ran under cygwin at the time). > > IR-Dev Team: > > I've been meaning to pose this question and hadn't gotten around to it, > but Zed has given me the nudge here (see Zed's message below). For Instant > Rails 2 would it make sense for use to use Mongrel, and if so, then how? > > It could be as an SCGI replacement with Apache proxying requests to it (we > could also allow standalone use just like WEBrick). We could replace Apache > entirely and just use Mongrel (this would be simpler, but less flexible). Or > we could just stick with SCGI. > > What do you think? > I'm all for removing SCGI. It's one of the most complicated dependencies, and configurations in IR. Given that we're planning on basing this on XAMPP as the base for the build, we get Apache itself more or less for free with the package. Mod_Proxy is also pretty simple to configure in Apache, and definitely gives us some serious flexibility. But I'm not entirely sold on needing all of that flexibility for IR. But if we're still planning on including php,phpMyAdmin, etc, we may as well include mod_proxy support, as we'll need Apache anyway. But that's just my two cents. Curt > -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060328/eaf508c5/attachment.htm From mortonda at dgrmm.net Tue Mar 28 13:11:46 2006 From: mortonda at dgrmm.net (David Morton) Date: Tue, 28 Mar 2006 12:11:46 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: References: <31d15f490603271337l6a1201ffh7aa6f77dcb9fedfc@mail.gmail.com> <31d15f490603280825x19dd15a2n4e1b4e806bd37297@mail.gmail.com> Message-ID: <44297C62.1040906@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Tanner Burson wrote: > I'm all for removing SCGI. It's one of the most complicated > dependencies, and configurations in IR. Given that we're planning on huh? it's like 3 scripts. It's far less complicated than fastcgi. The only thing that is more complicated is the fact that scgi runs as a seperate process and thus requires an additional program to start, but this is actually a security feature, as I have mentioned before. The thing most people don't stop to think about is that rails is not like php; it is not just a scripting language. To run best (ie, anything excpet cgi), it is an application server running all the time, to which the webserver proxies requests. SCGI seems to be the simplest way to host that application server. I'm not familiar with mongrel, but at first glance, I have a few reservations about it; I really doubt it can compete with something like lighttpd for speed. Looking at the mongrel website, it looks like it is an alternative to webrick, which is certainly not the fastest way to do things. Remember, the web server is responsible for serving static content too (caching!), so a pure ruby webserver I think would lag behind in this department. All in all, SCGI still looks like the winner to me. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKXxiSIxC85HZHLMRAuVMAJ97JMD8Da4HWgjDXTNlajPKYFLTzACeLwYT NJfp6a8E3FQPdB7U3Ey2lnQ= =gLbt -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Tue Mar 28 13:41:03 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Tue, 28 Mar 2006 12:41:03 -0600 Subject: [Instantrails-developers] Mongrel in Instant Rails Message-ID: <31d15f490603281041v1fba2be2mf372cd5c50909be@mail.gmail.com> Zed, I temporarily added you to the Instant rails developers ML just for this thread, so that you can answer and we can get the facts strainght. See below... On 3/28/06, David Morton wrote: > > Tanner Burson wrote: > > > I'm all for removing SCGI. It's one of the most complicated > > dependencies, and configurations in IR. Given that we're planning on > > huh? it's like 3 scripts. It's far less complicated than fastcgi. The > only > thing that is more complicated is the fact that scgi runs as a seperate > process > and thus requires an additional program to start, but this is actually a > security feature, as I have mentioned before. > > The thing most people don't stop to think about is that rails is not like > php; > it is not just a scripting language. To run best (ie, anything excpet > cgi), it > is an application server running all the time, to which the webserver > proxies > requests. SCGI seems to be the simplest way to host that application > server. > > I'm not familiar with mongrel, but at first glance, I have a few > reservations > about it; I really doubt it can compete with something like lighttpd for > speed. > Looking at the mongrel website, it looks like it is an alternative to > webrick, > which is certainly not the fastest way to do things. Zed can elucidate here, but from what I have been reading, Mongrel is very, very fast -- not at all comparable to WEBrick. Remember, the web server is responsible for serving static content too > (caching!), so a pure ruby webserver I think would lag behind in this > department. > > All in all, SCGI still looks like the winner to me. > Zed, can you weigh in here. What would be the trade off between proxying requests from apache to mongrel vs. just using SCGI? Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060328/4d2d6ca9/attachment.htm From mortonda at dgrmm.net Tue Mar 28 23:18:24 2006 From: mortonda at dgrmm.net (David Morton) Date: Tue, 28 Mar 2006 22:18:24 -0600 Subject: [Instantrails-developers] rails 1.1 In-Reply-To: <31d15f490603280750r199a24carc7689454786e72ef@mail.gmail.com> References: <44295127.4070504@dgrmm.net> <31d15f490603280750r199a24carc7689454786e72ef@mail.gmail.com> Message-ID: <442A0A90.10907@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 One showstopper: rails 1.1 breaks typo. 3 options: 1) remove typo 2) wait until they fix typo 3) include another copy of rails 1.0 in typo's vendor directory, like a freeze_edge - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKgqQSIxC85HZHLMRArNdAJ0QQfXnP8YQcaN40qV8i6GL90dkCACfZulR 5P6BzPXoOl+aGj6LLURH+Kk= =2bZv -----END PGP SIGNATURE----- From zedshaw at zedshaw.com Wed Mar 29 09:58:03 2006 From: zedshaw at zedshaw.com (Zed Shaw) Date: Wed, 29 Mar 2006 09:58:03 -0500 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: <31d15f490603280825x19dd15a2n4e1b4e806bd37297@mail.gmail.com> Message-ID: Hey folks, I?d be really keen on getting IR moved over to Mongrel. Mongrel has full win32 services support and has the ability to assign itself to CPU with affinity which is actually more than what it can do on POSIX systems. I?m thinking it could really clean up a lot of the stuff you had to do in instant rails to get SCGI to show up. I?ve also chatted with people and it seems there?s two levels of need for something like instant rails: Newbie ? The target for this would be classes where the instructor just hands out CDs and no matter what platform people are able to get instant rails going with a few clicks. In this case complete faithfulness to the RoR platform isn?t as important as speed of setup. If you could get it all to run without install onto their computer that might be even better, but the main thing is that this would be a very light setup. My thinking was Ruby one-click plus mongrel/ackbar/kirbybase and you?d be set. Serious ? This is someone who?s used the newbie setup in a class and is now trying to write or deploy a serious rails application. These folks are going to need something closer to what you have now. My thinking on this is basically take what you?ve got now and simply swap out Mongrel, but include a services management piece for their applications. And if you guys are doing this I?ll chip in as I know a couple of people who?ve asked for at least the Newbie part. Zed On 3/28/06 11:25 AM, "Curt Hibbs" wrote: > Thanks, Zed. > > I've been following your work on Mongrel, and it looks really good. I > originally wanted to use Lighttpd, but canned that when I realized I'd have to > do a native windows port myself (it only ran under cygwin at the time). > > IR-Dev Team: > > I've been meaning to pose this question and hadn't gotten around to it, but > Zed has given me the nudge here (see Zed's message below). For Instant Rails 2 > would it make sense for use to use Mongrel, and if so, then how? > > It could be as an SCGI replacement with Apache proxying requests to it (we > could also allow standalone use just like WEBrick). We could replace Apache > entirely and just use Mongrel (this would be simpler, but less flexible). Or > we could just stick with SCGI. > > What do you think? > > Curt > > On 3/28/06, Zed Shaw wrote: >> Hi Curt, >> >> I haven't really ran into anything like this, but I've been head down in >> Mongrel code lately. I think I'm pretty close to spending a weekend porting >> some of the mongrel stuff over to SCGI, but I was wondering if you'd be >> interested in working on getting instant rails working with just Mongrel? >> Could solve all these problems and also has the win32 service stuff which >> should make for a very nice clean setup. >> >> To debug this problem though I'd start trying to rule out whether it's his >> application, the machine, or just SCGI. Have him make a blank test >> application and see how fast that runs. If that runs slow then it's probably >> SCGI. If it's fast then his application is probably doing something funky. >> >> Also, see if he can post more particulars of the setup like machine speed >> etc. I've seen some platforms for weird reasons go incredibly slow on >> windows, and then others are fast and snappy. >> >> Zed >> >> >> On 3/27/06 4:37 PM, "Curt Hibbs" wrote: >> >> Steve, I don't know what is going on here, but I copied Zed Shaw on this (aka >> Mr. SCGI), because he might be able to help. >> >> Curt >> >> On 3/27/06, Debaun, Steve < Steve.Debaun at seminis.com >> > wrote: >> Hi all, >> >> So far, I'm loving the package... kudos to Curt for such a great >> contribution! Currently, however, I'm having a problem with getting SCGI to >> behave. >> >> InstantRails 1.0 >> >> Using a simple test app, I see response times like: >> - Webrick: 3 seconds >> - SCGI: 15 seconds >> >> Further oddities... the log/scgi.log is just being populated with >> start-event lines, no request handling is being logged at all. In fact, if >> I don't start the SCGI handler, the page still responds (and just as >> slowly). This leads me to believe that the SCGI handler isn't even being >> called, and that 'something else' is handling the requests very slowly. >> >> The :port: setting in scgi.yaml is set to 7001, and the Apache conf has a >> VirtualHost with: >> SCGIMount /dispatch.fcgi 127.0.0.1:7001 >> . >> >> >> SCGI mode is production. I have no hosts entry because the hostname is a >> real dns name. >> >> Has anyone run into any problems with extreme SCGI slowness? Can anyone >> explain the things that are confusing me (no log entries, page renders >> despite scgi not being started)? >> >> sd >> This e-mail message may contain privileged and/or confidential information, >> and is intended to be received only by persons entitled to receive such >> information. If you have received this e-mail in error, please notify the >> sender immediately. Please delete it and all attachments from any servers, >> hard drives or any other media. Other use of this e-mail by you is strictly >> prohibited. >> All e-mails and attachments sent and received are subject to monitoring, >> reading and archival by Seminis. The recipient of this e-mail is solely >> responsible for checking for the presence of "Viruses" or other "Malware". >> Seminis accepts no liability for any damage caused by any such code >> transmitted by or accompanying this e-mail or any attachment. >> _______________________________________________ >> Instantrails-users mailing list >> Instantrails-users at rubyforge.org >> http://rubyforge.org/mailman/listinfo/instantrails-users >> >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060329/ec341f43/attachment-0001.htm From curt.hibbs at gmail.com Wed Mar 29 10:24:42 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Wed, 29 Mar 2006 09:24:42 -0600 Subject: [Instantrails-developers] Fwd: Mongrel in Instant Rails In-Reply-To: References: <31d15f490603281041v1fba2be2mf372cd5c50909be@mail.gmail.com> Message-ID: <31d15f490603290724t65eebe7al9a452ab278fbaf7c@mail.gmail.com> Zed, I think you hit "reply" instead of "reply all" because this came only to me and not the Instant Rails Dev ML -- so I'm forwarding this. Also, the way I quoted the original stuff I sent you buried the fact that most of the questions you address were from David Morton, not Tanner. So guys: below where Zed says "Tanner", please read it as "David". :-) Anyway, thanks Zed for clarifying the facts. Also, I do like the idea of a super simple "Instant Rails Light". Curt ---------- Forwarded message ---------- From: Zed Shaw Date: Mar 29, 2006 9:08 AM Subject: Re: Mongrel in Instant Rails To: curt at hibbs.com On 3/28/06 1:41 PM, "Curt Hibbs" wrote: > Zed, I temporarily added you to the Instant rails developers ML just for this > thread, so that you can answer and we can get the facts strainght. See > below... > > On 3/28/06, David Morton < mortonda at dgrmm.net> wrote: >> Tanner Burson wrote: >> >>> I'm all for removing SCGI. It's one of the most complicated >>> dependencies, and configurations in IR. Given that we're planning on >> >> huh? it's like 3 scripts. It's far less complicated than fastcgi. The only >> thing that is more complicated is the fact that scgi runs as a seperate >> process >> and thus requires an additional program to start, but this is actually a >> security feature, as I have mentioned before. >> >> The thing most people don't stop to think about is that rails is not like >> php; >> it is not just a scripting language. To run best (ie, anything excpet cgi), >> it >> is an application server running all the time, to which the webserver proxies >> requests. SCGI seems to be the simplest way to host that application server. >> >> I'm not familiar with mongrel, but at first glance, I have a few reservations >> about it; I really doubt it can compete with something like lighttpd for >> speed. >> Looking at the mongrel website, it looks like it is an alternative to >> webrick, >> which is certainly not the fastest way to do things. > > Zed can elucidate here, but from what I have been reading, Mongrel is very, > very fast -- not at all comparable to WEBrick. > Hi Tanner, I actually wrote both Mongrel and SCGI just in case that isn't clear. The main advantage of Mongrel isn't so much pure raw ruthless speed but ease of deployment. For win32 Mongrel actually is *easier* than SCGI. It's got services support; a nice little configtool that lets you install, start, stop, restart, services; and the ability to do CPU affinity on SMP machines. With this you could probably simplify a lot of how Instant Rails works. You can keep the Apache setup and just move to mod_proxy instead of mod_scgi (which means no more compile). But, this also means that since it's HTTP you can just go with IIS or Microsoft's proxy server (whatever it's called). Even better, since it is called "instant" Rails you can do a mini version (the newbie version from a previous post of mine) that has just Mongrel for people who just want to get up right *now* and code with no fuss. I think the final thing that should be attractive for Mongrel is that is much more extensible than SCGI. You are able to add your own handlers, create your own commands, and even create them as gems so others just install and run with minimal configuration. Finally, the speed of Mongrel is about the same as SCGI, but it's getting more love and attention (and money from VeriSign) so it's going to get faster and more solid. It also has a dedicated Win32 developer who takes care to keep things working. So that's my pitch. >> Remember, the web server is responsible for serving static content too >> (caching!), so a pure ruby webserver I think would lag behind in this >> department. >> >> All in all, SCGI still looks like the winner to me. > Tanner, you might be confusing the approach proposed by Curt. In my mind you could literally go with what you have and just swap out Mongrel. That'd be the low effort solution. What you get for this is services for real which instant rails can manage (no more command windows). Next level would be my proposal for a "newbie" instant rails for people who just want rails up real quick for a class on any platform. In this case, a quick install and just mongrel would be perfect. I know of a few people doing classes who'd be very interested in this. The third level would be to let the user go from the newbie level for quick setups and installs and then "productionize" their newbie setup under apache. The main thing to keep in mind though is that Mongrel is an attempt to be just as fast as SCGI/FastCGI yet still as easy to run and deploy as webrick. This opens a wide range of web servers including many of the ones for Win32 and various proxy servers. What do you think? Zed -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060329/0299ab42/attachment.htm From mortonda at dgrmm.net Wed Mar 29 10:30:01 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 09:30:01 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: References: Message-ID: <442AA7F9.1090308@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Zed Shaw wrote: > Hey folks, > > I?d be really keen on getting IR moved over to Mongrel. Mongrel has > full win32 services support and has the ability to assign itself to CPU > with affinity which is actually more than what it can do on POSIX > systems. I?m thinking it could really clean up a lot of the stuff you > had to do in instant rails to get SCGI to show up. That does sound nice. There really isn't a lot of stuff to do to get scgi running any more, now that the bugs are fixed. :) I'll give mongrel a try though. > > I?ve also chatted with people and it seems there?s two levels of need > for something like instant rails: This is an interesting thought. Actually, for the newbie level, the easiest thing would be to include radrails, and let it set up the webrick server. This is actually how I do most of my development. I wonder if you could get mongrel included with radrails? Then for small site installations, we could use apache/mongrel for production. I still wonder about performance, as this link shows: http://weblog.textdrive.com/article/201/rails-with-zeus-and-mongrel-or-fcgi I'd like to see a comparison with SCGI in there too, but it looks like the lighttpd/fcgi combo has a very fast implementation. I also wonder if something was configured differently for those benchmarks, like file sessions vs memory sessions. In any case those perfomance issues are only a problem if you have a lot of traffic, and at that point, InstantRails is probably not the best hosting solution. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKqf5SIxC85HZHLMRAhcnAJ9Kwky61/AoZgW+yzM0d0NjU97sNgCfVy3w PHxOUS3QqrCR+tl8OYcplSM= =vWx2 -----END PGP SIGNATURE----- From mortonda at dgrmm.net Wed Mar 29 10:42:27 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 09:42:27 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: <442AA7F9.1090308@dgrmm.net> References: <442AA7F9.1090308@dgrmm.net> Message-ID: <442AAAE3.6040700@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 David Morton wrote: > Then for small site installations, we could use apache/mongrel for production. > I still wonder about performance, as this link shows: > http://weblog.textdrive.com/article/201/rails-with-zeus-and-mongrel-or-fcgi Whoops, following a link on that page yields this page: http://weblog.textdrive.com/article/200/playing-with-lighttpd-litespeed-fcgi-and-mongrel where they say: Mongrel as a proxy performs just as well as fcgi does, either with lighttpd or litespeed, and could easily become an Adios FCGI thing. Impressive! I think the problem I saw on the previous link had to do with zeus. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKqrjSIxC85HZHLMRAlANAJsHklf2b1Yp4NoYj7Et1kNOlz+PRACeMlyy SQr7hf5Fu8sH6u+xHLb8bzw= =x2+p -----END PGP SIGNATURE----- From mortonda at dgrmm.net Wed Mar 29 10:52:06 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 09:52:06 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: <442AA7F9.1090308@dgrmm.net> References: <442AA7F9.1090308@dgrmm.net> Message-ID: <442AAD26.9090205@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 David Morton wrote: > That does sound nice. There really isn't a lot of stuff to do to get scgi > running any more, now that the bugs are fixed. :) I'll give mongrel a try though. OOF! That's fast! I'm sold! I pulled up an app I'm working on, and when run by webrick, you can see all the images render one at a time. with mongrel, it just popped the entire page in at once. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKq0mSIxC85HZHLMRAvp/AKCRlRyxbjfEbQT+Rhz2fX1h7DOmsACghJnl ZwWaj0BfzNLiDYmHJeNCXqE= =KVM7 -----END PGP SIGNATURE----- From tanner.burson at gmail.com Wed Mar 29 11:01:44 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Wed, 29 Mar 2006 10:01:44 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: <442AAD26.9090205@dgrmm.net> References: <442AA7F9.1090308@dgrmm.net> <442AAD26.9090205@dgrmm.net> Message-ID: On 3/29/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > David Morton wrote: > > That does sound nice. There really isn't a lot of stuff to do to get > scgi > > running any more, now that the bugs are fixed. :) I'll give mongrel a > try though. > > OOF! That's fast! I'm sold! > > I pulled up an app I'm working on, and when run by webrick, you can see > all the > images render one at a time. with mongrel, it just popped the entire page > in at > once. You mentioned in the other thread about Rails really needing an application server, as it is fundamentally different than PHP. I think Mongrel is a much bigger step (a nd could eventually get all the way ) to being THE Rails application server. Much more so than SCGI and FCGI in my opinion. -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060329/d893761d/attachment.htm From mortonda at dgrmm.net Wed Mar 29 11:47:03 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 10:47:03 -0600 Subject: [Instantrails-developers] [Instantrails-users] SCGI Slowness Problem In-Reply-To: References: <442AA7F9.1090308@dgrmm.net> <442AAD26.9090205@dgrmm.net> Message-ID: <442ABA07.9040307@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Tanner Burson wrote: > You mentioned in the other thread about Rails really needing an > application server, as it is fundamentally different than PHP. I think > Mongrel is a much bigger step (a nd could eventually get all the way ) > to being THE Rails application server. Much more so than SCGI and FCGI > in my opinion. I'm beginning to think so too. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKroHSIxC85HZHLMRAva/AKCRuHVUI9/s4mccgC7bvHkv2UJlVgCfXdik P7wouLZ4G/JTqh9B3BLwCv0= =NogM -----END PGP SIGNATURE----- From mortonda at dgrmm.net Wed Mar 29 14:17:49 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 13:17:49 -0600 Subject: [Instantrails-developers] uploading one-click installer into trunk Message-ID: <442ADD5D.1060707@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 FYI, I removed the ruby directory and am now uploading the new one-click version into svn... I expect it to take some time. After this, we have a bunch of gems to reinstall. I plan to install both scgi and mongrel for now, so we can evaluate both. If we don't see a solution to typo soon, I think we should just free edge on it, and we can always change it once they get it working with 1.1. At that point we could release a 1.1.1 or something. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKt1dSIxC85HZHLMRAhdgAJoD2gZUoPHRZmmOvZAcWYGB4BhCfACfblf9 dPyfeJft3UIlZRRK7pEFAcI= =3amc -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Wed Mar 29 14:54:13 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Wed, 29 Mar 2006 13:54:13 -0600 Subject: [Instantrails-developers] uploading one-click installer into trunk In-Reply-To: <442ADD5D.1060707@dgrmm.net> References: <442ADD5D.1060707@dgrmm.net> Message-ID: <31d15f490603291154w45913d1h9d0535da68f13f91@mail.gmail.com> On 3/29/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > FYI, I removed the ruby directory and am now uploading the new one-click > version > into svn... I expect it to take some time. After this, we have a bunch > of gems > to reinstall. I plan to install both scgi and mongrel for now, so we can > evaluate both. > > If we don't see a solution to typo soon, I think we should just free edge > on it, > and we can always change it once they get it working with 1.1. At that > point > we could release a 1.1.1 or something. > Excellent! Let me know when you're done, and I'll test everything out and prepare a release. If things look good with Mongrel I'll make appropriate mods to the manager for a later release. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060329/ee937e02/attachment.htm From curt.hibbs at gmail.com Wed Mar 29 16:24:21 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Wed, 29 Mar 2006 15:24:21 -0600 Subject: [Instantrails-developers] uploading one-click installer into trunk In-Reply-To: <31d15f490603291154w45913d1h9d0535da68f13f91@mail.gmail.com> References: <442ADD5D.1060707@dgrmm.net> <31d15f490603291154w45913d1h9d0535da68f13f91@mail.gmail.com> Message-ID: <31d15f490603291324y6bf52191xa896992e2c556543@mail.gmail.com> On 3/29/06, Curt Hibbs wrote: > > On 3/29/06, David Morton wrote: > > > > -----BEGIN PGP SIGNED MESSAGE----- > > Hash: SHA1 > > > > FYI, I removed the ruby directory and am now uploading the new one-click > > version > > into svn... I expect it to take some time. After this, we have a bunch > > of gems > > to reinstall. I plan to install both scgi and mongrel for now, so we > > can > > evaluate both. > > > > If we don't see a solution to typo soon, I think we should just free > > edge on it, > > and we can always change it once they get it working with 1.1. At that > > point > > we could release a 1.1.1 or something. > > > > Excellent! > > Let me know when you're done, and I'll test everything out and prepare a > release. > > If things look good with Mongrel I'll make appropriate mods to the manager > for a later release. > > Curt > David, did you also upgrade Rake -- the new version of Rails requires the new version of Rake. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060329/f6f41314/attachment.htm From mortonda at dgrmm.net Wed Mar 29 17:29:31 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 16:29:31 -0600 Subject: [Instantrails-developers] uploading one-click installer into trunk In-Reply-To: <31d15f490603291324y6bf52191xa896992e2c556543@mail.gmail.com> References: <442ADD5D.1060707@dgrmm.net> <31d15f490603291154w45913d1h9d0535da68f13f91@mail.gmail.com> <31d15f490603291324y6bf52191xa896992e2c556543@mail.gmail.com> Message-ID: <442B0A4B.50004@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > David, did you also upgrade Rake -- the new version of Rails requires > the new version of Rake. I think you already had it in the 1 click installer. rake-0.7.0 anyway, I think everything is uploaded... I haven't done any testing yet. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKwpLSIxC85HZHLMRAk8XAJ9DtNJPHAM13ujgKsdNP/gkQIZarwCgl3Ul 0lL4pXYSQbDlhIT0WZdLysg= =/k0v -----END PGP SIGNATURE----- From mortonda at dgrmm.net Wed Mar 29 19:02:22 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 18:02:22 -0600 Subject: [Instantrails-developers] uploading one-click installer into trunk In-Reply-To: <442B0A4B.50004@dgrmm.net> References: <442ADD5D.1060707@dgrmm.net> <31d15f490603291154w45913d1h9d0535da68f13f91@mail.gmail.com> <31d15f490603291324y6bf52191xa896992e2c556543@mail.gmail.com> <442B0A4B.50004@dgrmm.net> Message-ID: <442B200E.1040006@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 David Morton wrote: > anyway, I think everything is uploaded... I haven't done any testing yet. ok, well typo doesn't work, as expected, but the cookbook app does, and another app I am working on seems fine. mongrel is installed, it starts as documented: mongrel_rails start -d it defaults to port 3000, like webrick. I don't know what the configuration is for apache mod_proxy to make it work with mongrel... but that's small fry... :) test away, I think it is ready. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEKyAOSIxC85HZHLMRAo8GAJ9uCA9LBW5XqB2Z7aWywHdjXu5bJACfYd1Y bbZszWbYZ54m5g/QeGzum5s= =m/Pz -----END PGP SIGNATURE----- From mortonda at dgrmm.net Wed Mar 29 23:47:14 2006 From: mortonda at dgrmm.net (David Morton) Date: Wed, 29 Mar 2006 22:47:14 -0600 Subject: [Instantrails-developers] typo is frozen Message-ID: <442B62D2.8010506@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have typo working with rails 1.0 frozen in the vendor folder. Once typo has a fix, we can reverse this easy enough, just delete the rails folder in vendor. :) (uploading to svn now) - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEK2LSSIxC85HZHLMRAjGZAJ9knQg0zK7o/0QggEiNHsWqLhdekQCbBmpD ZdMSiH55QD97u0WWiJWPu0k= =WG1m -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Thu Mar 30 10:27:59 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 09:27:59 -0600 Subject: [Instantrails-developers] typo is frozen In-Reply-To: <442B62D2.8010506@dgrmm.net> References: <442B62D2.8010506@dgrmm.net> Message-ID: <31d15f490603300727l32f74f49tf5776cc64cea89d2@mail.gmail.com> On 3/29/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I have typo working with rails 1.0 frozen in the vendor folder. Once typo > has a > fix, we can reverse this easy enough, just delete the rails folder in > vendor. :) > > (uploading to svn now) > In Zed Shaw's release announcement for Mongrel 0.3.12 he says: * Initial support for Rails 1.1. Remember *if you use Typo it is broken not Mongrel*. Typo is being frantically fixed so be patient. I never got an answer from my query to Tobias. Do any of you follow the Typo ML? If Typo will be fixed soon, I'd prefer to wait. In any case, while we try to get some answers, I'll start testing this out... Thanks David. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/5f94e1af/attachment.htm From mortonda at dgrmm.net Thu Mar 30 10:42:08 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 09:42:08 -0600 Subject: [Instantrails-developers] typo is frozen In-Reply-To: <31d15f490603300727l32f74f49tf5776cc64cea89d2@mail.gmail.com> References: <442B62D2.8010506@dgrmm.net> <31d15f490603300727l32f74f49tf5776cc64cea89d2@mail.gmail.com> Message-ID: <442BFC50.3000608@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > On 3/29/06, *David Morton* > wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > I have typo working with rails 1.0 frozen in the vendor > folder. Once typo has a > fix, we can reverse this easy enough, just delete the rails folder > in vendor. :) > > (uploading to svn now) > > > In Zed Shaw's release announcement for Mongrel 0.3.12 he says: > > * Initial support for Rails 1.1. Remember *if you use Typo it is broken not > Mongrel*. Typo is being frantically fixed so be patient. It does work with mongrel when frozen to 1.0, so it works in the current svn trunk > I never got an answer from my query to Tobias. Do any of you follow the > Typo ML? If Typo will be fixed soon, I'd prefer to wait. Looking through the mailing list, I hadn't seen anything substantial until this morning... 10 hours ago they made a 1.1 branch and have it working. I don't see yet when it will be released, but I do at least see some improvement. Let's wait another day and see. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEK/xQSIxC85HZHLMRAjWHAJ9AEZr3PCEj9aj8E4MxU4oGyAK46gCfWIYs eBUFKTDHMX3qxsf8tp6UEqY= =pard -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Thu Mar 30 11:48:05 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 10:48:05 -0600 Subject: [Instantrails-developers] Fwd: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails In-Reply-To: <195a01c65411$7c5dff30$6501a8c0@dp2000> References: <5440c2a90603290942y40c7f133yd1f0bc6e64e5a948@mail.gmail.com> <6cf2a94f0603291025gb7c22bbi2ab92e1bcc35f625@mail.gmail.com> <3a3d58af0603291327g665c9ap756cc73609002ad6@mail.gmail.com> <195a01c65411$7c5dff30$6501a8c0@dp2000> Message-ID: <31d15f490603300848j5287473aveb4656136245450c@mail.gmail.com> I think the idea of a streamlined "light" version of Instant Rails is good. The main issue is having the manpower to do it. IR Team: take a look at what Bill has written below and offer your opinion. Thanks, Curt ---------- Forwarded message ---------- From: Bill Walton Date: Mar 30, 2006 9:49 AM Subject: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails To: curt at hibbs.com Cc: file Hi Curt, Curt Hibbs wrote: > Sometime in the next week or so we > will be releasing Instant Rails 1.1 > which will include Rails 1.1. I've got a request I'd like you and the rest of the core team to consider (please forward this as appropriate). My request is similar to a recent post on the list to which Tanner responded positively, so I thought I'd throw mine in as well. I'll make it, and then explain my motivation. I'm hoping that you and the rest of the team will give both the request and the motivation some thought. I'll understand completely if both are ultimately rejected. Request: A version of IR, conceptually IR-Basic, that includes only the bare bones: MySQL, Rails, Ruby, WEBrick, and the IR app. MySQL-Front is so easy to obtain that I don't see including it in the distribution as necessary at all. And since it's free use expires after 30 days, I personally feel the difference in licensing argues for its exclusion from the package. The minimized content of the distro reduces the chance that folks will create problems for themselves. Motivation: I've been toying with the idea of doing an article or two targeting a new audience for RoR and am soliciting your (the team's) feedback and input before moving forward. While the reaction among developers to RoR has been positive, there are some significant barriers to its widespread adoption within the corporate development community. Chief among those, IME, is that 1) the customer doesn't care what tools the developers use and so don't typically actively support switching to new ones, and 2) the Ops group does care about tools to the extent that change creates complexity in their world and they, more often than not, resist if not outright *oppose* the use of new stuff that impacts them. I believe there's a huge uplift for RoR waiting in the form of it's use by requirements analysts, UI designers, and testers in the corporate world. RoR, and particularly IR, gives them capabilities that, based on my limited experience working with a couple, they find absolutely transformational. >From their perspective... Requirements analysts can develop working prototypes to suplant a large portion of the documentation they currently produce. The productivity of RoR puts this capability well within the time / cost constraints they face (requirements definition is typically allotted 10-15% of the schedule / budget in the corporate IT world) and gives them something the users will find much easier to "sign-off on." Working prototypes give UI designers a medium that allows them to try out multiple, different approaches early in the process. It's not uncommon today for UI designers to be relegated to a "make it pretty" role that may not even involve users beyond a review / judgment interaction. Their training, however, tells them their job is to improve the user experience. That involves much more than "putting lipstick on the pig." Testers responsible for UAT would have a vehicle for defining Acceptance tests before development begins. The definition of Acceptance tests prior to development remains an elusive goal for agile methodologies. The use of IR for prototyping could eliminate much of this barrier to the adoption of Agile within shops that want to move that way but can't figure out how to do so in small steps. As long as these folks are working with IR on their laptops, there'll be no resistance from the Ops groups. It's easy to anticipate, though, that customers will want to be able to share the prototypes among geographically dispersed teams. That will create an opportunity for the Ops folks to get some experience with RoR in a way that's not operationally critical. Most of the Ops folks I've worked with would welcome that approach to the introduction of new technology. WRT to the developers themselves... there are, of course, already supporters within their ranks who would welcome support from other groups around them. There is some historic resistance to prototyping within the development ranks that deserves attention. I have seen this first hand within the Agile community in particular. OTOH, faced with the choice between a stack of documents and a working prototype I've never met a developer who'd choose the former over the latter. Moreover, there is no threat to the adoption of an Agile approach to development inherent in this. Rather, the 'industrialization' of a prototype is an activity that's inherently well-suited to an incremental appoach. So while this strategy doesn't make the move to Agile in one big jump as many of the Agile community would like, it provides a path to Agile for those companies that simply do not, will not, take risks of that scale. The core of the strategy is simple. Create a demand-pull model for adoption that minimizes sources of potential resistance. I have no intention of moving forward with promoting this strategy unless I have some indication that it will, at least, not be opposed by you folks. I would appreciate and look forward to the team's feedback and input on these thoughts. Best regards, Bill -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/66b0abe6/attachment-0001.htm From tanner.burson at gmail.com Thu Mar 30 11:55:11 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Thu, 30 Mar 2006 10:55:11 -0600 Subject: [Instantrails-developers] Fwd: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails In-Reply-To: <31d15f490603300848j5287473aveb4656136245450c@mail.gmail.com> References: <5440c2a90603290942y40c7f133yd1f0bc6e64e5a948@mail.gmail.com> <6cf2a94f0603291025gb7c22bbi2ab92e1bcc35f625@mail.gmail.com> <3a3d58af0603291327g665c9ap756cc73609002ad6@mail.gmail.com> <195a01c65411$7c5dff30$6501a8c0@dp2000> <31d15f490603300848j5287473aveb4656136245450c@mail.gmail.com> Message-ID: On 3/30/06, Curt Hibbs wrote: > > I think the idea of a streamlined "light" version of Instant Rails is > good. The main issue is having the manpower to do it. > > IR Team: take a look at what Bill has written below and offer your > opinion. > > Thanks, > Curt > > ---------- Forwarded message ---------- > From: Bill Walton > Date: Mar 30, 2006 9:49 AM > Subject: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails > To: curt at hibbs.com > Cc: file > > Hi Curt, > > Curt Hibbs wrote: > > > > Sometime in the next week or so we > > will be releasing Instant Rails 1.1 > > which will include Rails 1.1. > > I've got a request I'd like you and the rest of the core team to consider > (please forward this as appropriate). > > My request is similar to a recent post on the list to which Tanner > responded > positively, so I thought I'd throw mine in as well. I'll make it, and > then > explain my motivation. I'm hoping that you and the rest of the team will > give both the request and the motivation some thought. I'll understand > completely if both are ultimately rejected. > > Request: > A version of IR, conceptually IR-Basic, that includes only the bare bones: > MySQL, Rails, Ruby, WEBrick, and the IR app. > > MySQL-Front is so easy to obtain that I don't see including it in the > distribution as necessary at all. And since it's free use expires after > 30 > days, I personally feel the difference in licensing argues for its > exclusion > from the package. > > The minimized content of the distro reduces the chance that folks will > create problems for themselves. > > Motivation: > I've been toying with the idea of doing an article or two targeting a new > audience for RoR and am soliciting your (the team's) feedback and input > before moving forward. > > While the reaction among developers to RoR has been positive, there are > some > significant barriers to its widespread adoption within the corporate > development community. Chief among those, IME, is that 1) the customer > doesn't care what tools the developers use and so don't typically actively > support switching to new ones, and 2) the Ops group does care about tools > to > the extent that change creates complexity in their world and they, more > often than not, resist if not outright *oppose* the use of new stuff that > impacts them. > > I believe there's a huge uplift for RoR waiting in the form of it's use by > > requirements analysts, UI designers, and testers in the corporate world. > RoR, and particularly IR, gives them capabilities that, based on my > limited > experience working with a couple, they find absolutely transformational. > From their perspective... > > Requirements analysts can develop working prototypes to suplant a large > portion of the documentation they currently produce. The productivity of > RoR puts this capability well within the time / cost constraints they face > > (requirements definition is typically allotted 10-15% of the schedule / > budget in the corporate IT world) and gives them something the users will > find much easier to "sign-off on." > > Working prototypes give UI designers a medium that allows them to try out > multiple, different approaches early in the process. It's not uncommon > today for UI designers to be relegated to a "make it pretty" role that may > not even involve users beyond a review / judgment interaction. Their > training, however, tells them their job is to improve the user experience. > That involves much more than "putting lipstick on the pig." > > Testers responsible for UAT would have a vehicle for defining Acceptance > tests before development begins. The definition of Acceptance tests prior > to development remains an elusive goal for agile methodologies. The use > of > IR for prototyping could eliminate much of this barrier to the adoption of > > Agile within shops that want to move that way but can't figure out how to > do > so in small steps. > > As long as these folks are working with IR on their laptops, there'll be > no > resistance from the Ops groups. It's easy to anticipate, though, that > customers will want to be able to share the prototypes among > geographically > dispersed teams. That will create an opportunity for the Ops folks to get > some experience with RoR in a way that's not operationally critical. Most > > of the Ops folks I've worked with would welcome that approach to the > introduction of new technology. > > WRT to the developers themselves... there are, of course, already > supporters > within their ranks who would welcome support from other groups around > them. > There is some historic resistance to prototyping within the development > ranks that deserves attention. I have seen this first hand within the > Agile > community in particular. OTOH, faced with the choice between a stack of > documents and a working prototype I've never met a developer who'd choose > the former over the latter. Moreover, there is no threat to the adoption > of > an Agile approach to development inherent in this. Rather, the > 'industrialization' of a prototype is an activity that's inherently > well-suited to an incremental appoach. So while this strategy doesn't > make > the move to Agile in one big jump as many of the Agile community would > like, > it provides a path to Agile for those companies that simply do not, will > not, take risks of that scale. > > The core of the strategy is simple. Create a demand-pull model for > adoption > that minimizes sources of potential resistance. > > I have no intention of moving forward with promoting this strategy unless > I > have some indication that it will, at least, not be opposed by you folks. > > I would appreciate and look forward to the team's feedback and input on > these thoughts. > > Best regards, > Bill > I completely agree. This is something that Zed brought up as well, and I think it's a really great idea. I also think it warrants some real discussion as to the whole distribution concept we've come up with. I think it could be worth looking into building a light version, and building the rest of IR on top of it, instead of extracting the light version from the full version. But that is obviously a large task, and has some implications on ease of cross platform support. I'd really like to hear everyone's opinions on this, and what direction we think we should be going. -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/575e1a54/attachment.htm From curt.hibbs at gmail.com Thu Mar 30 12:11:47 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 11:11:47 -0600 Subject: [Instantrails-developers] Fwd: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails In-Reply-To: <19c501c65419$f0a15ba0$6501a8c0@dp2000> References: <5440c2a90603290942y40c7f133yd1f0bc6e64e5a948@mail.gmail.com> <6cf2a94f0603291025gb7c22bbi2ab92e1bcc35f625@mail.gmail.com> <3a3d58af0603291327g665c9ap756cc73609002ad6@mail.gmail.com> <195a01c65411$7c5dff30$6501a8c0@dp2000> <31d15f490603300848j5287473aveb4656136245450c@mail.gmail.com> <19c501c65419$f0a15ba0$6501a8c0@dp2000> Message-ID: <31d15f490603300911y432f24cka65e5cbe61d59f3@mail.gmail.com> Bill, I added you to the Instant Rails Developer ML to make it easier for you to participate in this dicussions, so you can now respond directly. IR Team: See Bill's clarification below. Curt ---------- Forwarded message ---------- From: Bill Walton Date: Mar 30, 2006 10:49 AM Subject: Re: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails To: curt at hibbs.com Cc: file Hi Curt, Thanks for the quick response. On rereading what I sent, I realize I'm looking for two responses and may not have made that as clear as I could have. 1) IR-light 2) promoting IR as a prototyping tool #2 does not depend on #1. IR-light just makes it less likely that folks responding to #2 will screw up and not give IR a fair shake. Thus, the feedback I'm *really* interested in getting is in response to #2. Thanks, Bill ----- Original Message ----- *From:* Curt Hibbs *To:* Bill Walton ; Instant Rails DEV *Sent:* 2006-03-30 10:48 AM *Subject:* Fwd: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails I think the idea of a streamlined "light" version of Instant Rails is good. The main issue is having the manpower to do it. IR Team: take a look at what Bill has written below and offer your opinion. Thanks, Curt ---------- Forwarded message ---------- From: Bill Walton Date: Mar 30, 2006 9:49 AM Subject: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails To: curt at hibbs.com Cc: file > Hi Curt, Curt Hibbs wrote: > Sometime in the next week or so we > will be releasing Instant Rails 1.1 > which will include Rails 1.1. I've got a request I'd like you and the rest of the core team to consider (please forward this as appropriate). My request is similar to a recent post on the list to which Tanner responded positively, so I thought I'd throw mine in as well. I'll make it, and then explain my motivation. I'm hoping that you and the rest of the team will give both the request and the motivation some thought. I'll understand completely if both are ultimately rejected. Request: A version of IR, conceptually IR-Basic, that includes only the bare bones: MySQL, Rails, Ruby, WEBrick, and the IR app. MySQL-Front is so easy to obtain that I don't see including it in the distribution as necessary at all. And since it's free use expires after 30 days, I personally feel the difference in licensing argues for its exclusion from the package. The minimized content of the distro reduces the chance that folks will create problems for themselves. Motivation: I've been toying with the idea of doing an article or two targeting a new audience for RoR and am soliciting your (the team's) feedback and input before moving forward. While the reaction among developers to RoR has been positive, there are some significant barriers to its widespread adoption within the corporate development community. Chief among those, IME, is that 1) the customer doesn't care what tools the developers use and so don't typically actively support switching to new ones, and 2) the Ops group does care about tools to the extent that change creates complexity in their world and they, more often than not, resist if not outright *oppose* the use of new stuff that impacts them. I believe there's a huge uplift for RoR waiting in the form of it's use by requirements analysts, UI designers, and testers in the corporate world. RoR, and particularly IR, gives them capabilities that, based on my limited experience working with a couple, they find absolutely transformational. >From their perspective... Requirements analysts can develop working prototypes to suplant a large portion of the documentation they currently produce. The productivity of RoR puts this capability well within the time / cost constraints they face (requirements definition is typically allotted 10-15% of the schedule / budget in the corporate IT world) and gives them something the users will find much easier to "sign-off on." Working prototypes give UI designers a medium that allows them to try out multiple, different approaches early in the process. It's not uncommon today for UI designers to be relegated to a "make it pretty" role that may not even involve users beyond a review / judgment interaction. Their training, however, tells them their job is to improve the user experience. That involves much more than "putting lipstick on the pig." Testers responsible for UAT would have a vehicle for defining Acceptance tests before development begins. The definition of Acceptance tests prior to development remains an elusive goal for agile methodologies. The use of IR for prototyping could eliminate much of this barrier to the adoption of Agile within shops that want to move that way but can't figure out how to do so in small steps. As long as these folks are working with IR on their laptops, there'll be no resistance from the Ops groups. It's easy to anticipate, though, that customers will want to be able to share the prototypes among geographically dispersed teams. That will create an opportunity for the Ops folks to get some experience with RoR in a way that's not operationally critical. Most of the Ops folks I've worked with would welcome that approach to the introduction of new technology. WRT to the developers themselves... there are, of course, already supporters within their ranks who would welcome support from other groups around them. There is some historic resistance to prototyping within the development ranks that deserves attention. I have seen this first hand within the Agile community in particular. OTOH, faced with the choice between a stack of documents and a working prototype I've never met a developer who'd choose the former over the latter. Moreover, there is no threat to the adoption of an Agile approach to development inherent in this. Rather, the 'industrialization' of a prototype is an activity that's inherently well-suited to an incremental appoach. So while this strategy doesn't make the move to Agile in one big jump as many of the Agile community would like, it provides a path to Agile for those companies that simply do not, will not, take risks of that scale. The core of the strategy is simple. Create a demand-pull model for adoption that minimizes sources of potential resistance. I have no intention of moving forward with promoting this strategy unless I have some indication that it will, at least, not be opposed by you folks. I would appreciate and look forward to the team's feedback and input on these thoughts. Best regards, Bill -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/46fea45b/attachment-0001.htm From curt.hibbs at gmail.com Thu Mar 30 12:15:25 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 11:15:25 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] Message-ID: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> On 3/30/06, Tanner Burson wrote: > > > > On 3/30/06, Curt Hibbs wrote: > > > > I think the idea of a streamlined "light" version of Instant Rails is > > good. The main issue is having the manpower to do it. > > > > IR Team: take a look at what Bill has written below and offer your > > opinion. > > > > Thanks, > > Curt > > > > ---------- Forwarded message ---------- > > From: Bill Walton < bill.walton at charter.net> > > Date: Mar 30, 2006 9:49 AM > > Subject: OFFLINE - Re: [Rails] Upgrading to Rails 1.1 in InstantRails > > To: curt at hibbs.com > > Cc: file > > > > Hi Curt, > > > > Curt Hibbs wrote: > > > > > > > Sometime in the next week or so we > > > will be releasing Instant Rails 1.1 > > > which will include Rails 1.1. > > > > I've got a request I'd like you and the rest of the core team to > > consider > > (please forward this as appropriate). > > > > My request is similar to a recent post on the list to which Tanner > > responded > > positively, so I thought I'd throw mine in as well. I'll make it, and > > then > > explain my motivation. I'm hoping that you and the rest of the team > > will > > give both the request and the motivation some thought. I'll understand > > completely if both are ultimately rejected. > > > > Request: > > A version of IR, conceptually IR-Basic, that includes only the bare > > bones: > > MySQL, Rails, Ruby, WEBrick, and the IR app. > > > > MySQL-Front is so easy to obtain that I don't see including it in the > > distribution as necessary at all. And since it's free use expires after > > 30 > > days, I personally feel the difference in licensing argues for its > > exclusion > > from the package. > > > > The minimized content of the distro reduces the chance that folks will > > create problems for themselves. > > > > Motivation: > > I've been toying with the idea of doing an article or two targeting a > > new > > audience for RoR and am soliciting your (the team's) feedback and input > > before moving forward. > > > > While the reaction among developers to RoR has been positive, there are > > some > > significant barriers to its widespread adoption within the corporate > > development community. Chief among those, IME, is that 1) the customer > > doesn't care what tools the developers use and so don't typically > > actively > > support switching to new ones, and 2) the Ops group does care about > > tools to > > the extent that change creates complexity in their world and they, more > > often than not, resist if not outright *oppose* the use of new stuff > > that > > impacts them. > > > > I believe there's a huge uplift for RoR waiting in the form of it's use > > by > > requirements analysts, UI designers, and testers in the corporate world. > > RoR, and particularly IR, gives them capabilities that, based on my > > limited > > experience working with a couple, they find absolutely transformational. > > > > From their perspective... > > > > Requirements analysts can develop working prototypes to suplant a large > > portion of the documentation they currently produce. The productivity > > of > > RoR puts this capability well within the time / cost constraints they > > face > > (requirements definition is typically allotted 10-15% of the schedule / > > budget in the corporate IT world) and gives them something the users > > will > > find much easier to "sign-off on." > > > > Working prototypes give UI designers a medium that allows them to try > > out > > multiple, different approaches early in the process. It's not uncommon > > today for UI designers to be relegated to a "make it pretty" role that > > may > > not even involve users beyond a review / judgment interaction. Their > > training, however, tells them their job is to improve the user > > experience. > > That involves much more than "putting lipstick on the pig." > > > > Testers responsible for UAT would have a vehicle for defining Acceptance > > > > tests before development begins. The definition of Acceptance tests > > prior > > to development remains an elusive goal for agile methodologies. The use > > of > > IR for prototyping could eliminate much of this barrier to the adoption > > of > > Agile within shops that want to move that way but can't figure out how > > to do > > so in small steps. > > > > As long as these folks are working with IR on their laptops, there'll be > > no > > resistance from the Ops groups. It's easy to anticipate, though, that > > customers will want to be able to share the prototypes among > > geographically > > dispersed teams. That will create an opportunity for the Ops folks to > > get > > some experience with RoR in a way that's not operationally > > critical. Most > > of the Ops folks I've worked with would welcome that approach to the > > introduction of new technology. > > > > WRT to the developers themselves... there are, of course, already > > supporters > > within their ranks who would welcome support from other groups around > > them. > > There is some historic resistance to prototyping within the development > > ranks that deserves attention. I have seen this first hand within the > > Agile > > community in particular. OTOH, faced with the choice between a stack of > > > > documents and a working prototype I've never met a developer who'd > > choose > > the former over the latter. Moreover, there is no threat to the > > adoption of > > an Agile approach to development inherent in this. Rather, the > > 'industrialization' of a prototype is an activity that's inherently > > well-suited to an incremental appoach. So while this strategy doesn't > > make > > the move to Agile in one big jump as many of the Agile community would > > like, > > it provides a path to Agile for those companies that simply do not, will > > not, take risks of that scale. > > > > The core of the strategy is simple. Create a demand-pull model for > > adoption > > that minimizes sources of potential resistance. > > > > I have no intention of moving forward with promoting this strategy > > unless I > > have some indication that it will, at least, not be opposed by you > > folks. > > > > I would appreciate and look forward to the team's feedback and input on > > these thoughts. > > > > Best regards, > > Bill > > > > I completely agree. This is something that Zed brought up as well, and I > think it's a really great idea. I also think it warrants some real > discussion as to the whole distribution concept we've come up with. I think > it could be worth looking into building a light version, and building the > rest of IR on top of it, instead of extracting the light version from the > full version. But that is obviously a large task, and has some implications > on ease of cross platform support. I'd really like to hear everyone's > opinions on this, and what direction we think we should be going. > I'd be in favor of this. Besides all of the benefits already mentioned, we can actually get something out the door more quickly. Perhaps it should be Mongrel based rather than WEBrick based. Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/c015a348/attachment.htm From mortonda at dgrmm.net Thu Mar 30 12:47:06 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 11:47:06 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] In-Reply-To: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> Message-ID: <442C199A.6090109@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > I completely agree. This is something that Zed brought up as well, > and I think it's a really great idea. I also think it warrants some > real discussion as to the whole distribution concept we've come up > with. I think it could be worth looking into building a light > version, and building the rest of IR on top of it, instead of > extracting the light version from the full version. But that is > obviously a large task, and has some implications on ease of cross > platform support. I'd really like to hear everyone's opinions on > this, and what direction we think we should be going. > > > I'd be in favor of this. Besides all of the benefits already mentioned, > we can actually get something out the door more quickly. > > Perhaps it should be Mongrel based rather than WEBrick based. Perhaps I'm not understanding the issue here... IR is pretty light relatively, the only difficult part is probably configure hosts file and apache vhost. Ok, well, I don't do that during my development process anyway. But I do find phpmyadmin infintely helpful, so I wouldn't want to do away with apache, unless there's another webserver that can do it. Unless we find a rails base replacement for phpMyAdmin.... that would be nifty. Mongrel is quite nice, it can be used the same way as webrick, only it's faster. - From the viewpoint of making it quick and easy for developers to try rails, there's no faster way than IR + RadRails. It can set up the webrick server from within the IDE, so the only other step needed is to start mysql and create a database. I would strongly vote for including radrails if possible. Then, for your next version up, include apache and/or lighttpd and methods to proxy to mongrel, for a more production ready environment. But then once you have all this I don't see any benefit from maintaining two or more releases. The tools can all fit together in one distro, it's just a matter of what is presented to the new user. So the workflow I'd propose using one master distro: Fire up IR management tool, which starts mysql and mongrel for any configured apps. It then starts up radrails. At this point we have a fully working environment to develop with. Also in the IR management app, maybe under an advanced menu, are tools to configure apache, hosts, vhosts, etc. This way, we have one tool that is easy for beginners or people who just want to dink around, and it also provides the tools to at least evaluate a simple production environment. I don't think we are shooting for a final "enterprise" production hosting environment, are we? The is no substitute for knowledgeable sysadmin at that point, and rails *isn't* really that hard to install. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELBmaSIxC85HZHLMRApZ/AJ992FHWffN3JGNVuE8YIC1Dv/iSdwCfdOHc 1wIO+q6vmOs7jOe5yIpa1u0= =3j9G -----END PGP SIGNATURE----- From bill.walton at charter.net Thu Mar 30 13:16:40 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 12:16:40 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net> Message-ID: <1a3c01c65426$17a35b70$6501a8c0@dp2000> Hi David, David Morton wrote: > > Perhaps I'm not understanding the issue here... > - From the viewpoint of making it quick and easy > for developers to try rails, The viewpoint I'm advocating is making is quick and easy for NON-developers to try Rails. I believe they, and their organizations, will find value in it for the reasons I articulated in my original email. I believe that Rails will find value (i.e., increased rate of adoption in the corporate IT environment) as a result of their support. To that end, making is as hard as possible for the non-developers to make a mis-step is important. HTH, Bill From mortonda at dgrmm.net Thu Mar 30 13:55:51 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 12:55:51 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] In-Reply-To: <1a3c01c65426$17a35b70$6501a8c0@dp2000> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000> Message-ID: <442C29B7.7040402@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bill Walton wrote: > The viewpoint I'm advocating is making is quick and easy for NON-developers > to try Rails. I believe they, and their organizations, will find value in > it for the reasons I articulated in my original email. I believe that Rails > will find value (i.e., increased rate of adoption in the corporate IT > environment) as a result of their support. > > To that end, making is as hard as possible for the non-developers to make a > mis-step is important. I guess I still don't see it.... Rails is a tool for developers (I include designers here too). I don't see an accountant playing with it... INo matter what we do, my mom will never be able to use it. :) IR, as it is, plus radrails, provides the easiest way I can see to using rails. How does that not fit your need? - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELCm3SIxC85HZHLMRAq4xAJ4lqBeWGErHQHQ8RigW8tmAfLERugCfYuFr 3Wgk+qD290V70QTyzkTG+Fo= =LHS+ -----END PGP SIGNATURE----- From bill.walton at charter.net Thu Mar 30 13:42:25 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 12:42:25 -0600 Subject: [Instantrails-developers] Instant Rails Light References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com><442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000> Message-ID: <1a8201c65429$b1e8e3a0$6501a8c0@dp2000> As luck would have it, a post that just came in on another list (AgileProjectManagement) that provides a very good, real-life example of what I was getting at in my original post re: Motivation. IMO, the use of RoR as a prototyping tool by non-developers could help these folks a lot. hth, Bill -------------------------- We are lucky enough to have them co-located with us much of the time... I think it really goes to a misunderstanding of the approach. The customer is thinking waterfall, but working in "agile" and we are thinking agile and being held against a waterfall approach. I am trying to think of a good way to illustrate this point... IMO it seems like the customer side of the team (customer, design, all of the higher ups etc) are unsure about the specifics of what they want. Of course, they cannot know it till they see it, but we might spend 8 weeks out of 10 getting to a point where they can tell us that feature X is not really needed after all, or we have completely reconceived of how it should work. We always have a timebox for the release, and we do not usually get time on the project after the release date. So, what ends up happening is the customer is frustrated because we have burned through 8 weeks and they still do not have what they want and we are frustrated because we are not sure that any of the work we do will "stick" in the product. To answer Ron's question from a previous message, we are using timeboxes, planning games, prioritization lists, backlogs, pair programming, TDD, continuous integration and co-location in our Agile approach. After reading all of the great posts to this topic, I think I get the issue that is happening in our team. It seems like the team is working in agile, but the communications is still "waterfall". Its like half agile. ----- Original Message ----- From: "Bill Walton" To: ; Sent: 2006-03-30 12:16 PM Subject: Re: [Instantrails-developers] Instant Rails Light [was: ] > Hi David, > > David Morton wrote: > > > > > > Perhaps I'm not understanding the issue here... > > > - From the viewpoint of making it quick and easy > > for developers to try rails, > > The viewpoint I'm advocating is making is quick and easy for NON-developers > to try Rails. I believe they, and their organizations, will find value in > it for the reasons I articulated in my original email. I believe that Rails > will find value (i.e., increased rate of adoption in the corporate IT > environment) as a result of their support. > > To that end, making is as hard as possible for the non-developers to make a > mis-step is important. > > HTH, > Bill > _______________________________________________ > Instantrails-developers mailing list > Instantrails-developers at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-developers From mortonda at dgrmm.net Thu Mar 30 14:40:33 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 13:40:33 -0600 Subject: [Instantrails-developers] Instant Rails Light In-Reply-To: <1a8201c65429$b1e8e3a0$6501a8c0@dp2000> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com><442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000> <1a8201c65429$b1e8e3a0$6501a8c0@dp2000> Message-ID: <442C3431.6040805@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bill Walton wrote: > what I was getting at in my original post re: Motivation. IMO, the use of > RoR as a prototyping tool by non-developers could help these folks a lot. The only step I can think besides what we are doing is to include several examples to illustrate what RoR can do... and maybe have an interface in the IR manager to launch them. This is why we have the cookbook app and typo included... In the example you gave it looks like they have the tools, but were stuck in the wrong mentality, of trying to specify everything before implementing. The agile method is to "just do it", to steal from Nike's slogan. Another approach is simply to have good documentation and tutorials, which I'm all for including, but the scope of that is larger than just IR. :) - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELDQxSIxC85HZHLMRAmspAKCRHGdYagfNURGGZiLNw4dDHqehUACgkgsU Dqb1ad2BEs7f7pmpHtdjTYU= =wzjH -----END PGP SIGNATURE----- From bill.walton at charter.net Thu Mar 30 14:28:26 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 13:28:26 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net><1a3c01c65426$17a35b70$6501a8c0@dp2000> <442C29B7.7040402@dgrmm.net> Message-ID: <1ac901c65430$1e4e01f0$6501a8c0@dp2000> Hi David, David Morton wrote: > > I guess I still don't see it.... Rails is a tool > for developers (I include designers here too). The testing community today also has a large number of folks doing "development" work using test automation tools. > I don't see an accountant playing with it... Many, if not most, of the accountants I've worked with as requirements analysts were *quite* capable developers WRT tools like VBA. Another nice thing about accountants is they're very results-focused. And I've never come across a development tool that delivers results like RoR. I think they'd see its value immediately. > INo matter what we do, my mom will never > be able to use it. :) Let's leave mothers out of it ;-) > IR, as it is, plus radrails, provides the easiest > way I can see to using rails. > How does that not fit your need? It does. My request for IR-Lite was based on an experience-set that says "if you can avoid a potential problem, avoid it." IR-Lite is, for my purposes, of secondary importance at the moment. Nice to have, but not necessary to achieve my short-term goal. What's most important for me to know at this point is this. If I were to begin promoting IR (Lite or otherwise) as a tool that can provide value to organizations by using it for prototyping, will this team public oppose that type of usage? I ask because I've run into Agilists who were/are adamantly opposed to prototyping. Thanks for your feedback. Best regards, Bill From mortonda at dgrmm.net Thu Mar 30 14:50:42 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 13:50:42 -0600 Subject: [Instantrails-developers] Instant Rails Light [was: ] In-Reply-To: <1ac901c65430$1e4e01f0$6501a8c0@dp2000> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net><1a3c01c65426$17a35b70$6501a8c0@dp2000> <442C29B7.7040402@dgrmm.net> <1ac901c65430$1e4e01f0$6501a8c0@dp2000> Message-ID: <442C3692.9090308@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Bill Walton wrote: > It does. My request for IR-Lite was based on an experience-set that says > "if you can avoid a potential problem, avoid it." IR-Lite is, for my > purposes, of secondary importance at the moment. Nice to have, but not > necessary to achieve my short-term goal. Ok, well, I think that can be managed by having the initial steps be the easiest, and then have more sophisticated things a menu down or something... or have a novice view and an advanced view. > If I were to begin promoting IR (Lite or otherwise) as a tool that can > provide value to organizations by using it for prototyping, will this team > public oppose that type of usage? > > I ask because I've run into Agilists who were/are adamantly opposed to > prototyping. Heh, ok, well, I have no problem... use the tool for what it's good at. :) If you look at "Agile Web Development with Rails", it gets right into prototyping the shopping cart right away, and promotes incremental changes that you can show the customer as you build it. So if anything, I would think this is more or less the official Rails philosophy. :) - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELDaSSIxC85HZHLMRAuo3AJ0TDP/gEMaE2FzrARrL5miNyOVvXwCfbG0w lbAgtXKW6rlQAlfXYKMCajs= =Cefn -----END PGP SIGNATURE----- From mortonda at dgrmm.net Thu Mar 30 15:26:33 2006 From: mortonda at dgrmm.net (David Morton) Date: Thu, 30 Mar 2006 14:26:33 -0600 Subject: [Instantrails-developers] typo and rails 1.1 Message-ID: <442C3EF9.3090701@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 After spending some time on irc, it looks like typo 2.6.0 will never work with rails 1.1, so we might as well leave it frozen. That said, they are nearing completion of typo 4.0. Piers Cawley indicated they may be releasing a "4.0.0_pre1-daredevils-only" version real soon. What we could do then is include the daredevils-only app too, to run in 1.1 mode, and then when it's all done do a new minor point release to update it. I ran it briefly, and it seems quite usable. This thread with Bill has underscored a need for some more up to date tutorials and examples, so it wouldn't hurt to come up with some simple examples to include. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELD74SIxC85HZHLMRAqRTAKCa2Aw/bGC7ByzEOWILu6pG+LWrdACeLz8L 7QjRGDgUsqwi56qosl5n2TA= =Jjfm -----END PGP SIGNATURE----- From bill.walton at charter.net Thu Mar 30 15:12:25 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 14:12:25 -0600 Subject: [Instantrails-developers] Instant Rails Light References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net><1a3c01c65426$17a35b70$6501a8c0@dp2000> <442C29B7.7040402@dgrmm.net><1ac901c65430$1e4e01f0$6501a8c0@dp2000> <442C3692.9090308@dgrmm.net> Message-ID: <1b0401c65436$43973cf0$6501a8c0@dp2000> David Morton wrote: > > Bill Walton wrote: > > > It does. My request for IR-Lite was based on an experience-set that says > > "if you can avoid a potential problem, avoid it." IR-Lite is, for my > > purposes, of secondary importance at the moment. Nice to have, but not > > necessary to achieve my short-term goal. > > Ok, well, I think that can be managed by having the initial steps be the > easiest, and then have more sophisticated things a menu down or something... or > have a novice view and an advanced view. That would totally resolve my 'issue'. "You folks want to go exploring... fine. We assume you know what you're doing." Just don't want them tripping over things if they stay around the campsite. Thanks. > > > If I were to begin promoting IR (Lite or otherwise) as a tool that can > > provide value to organizations by using it for prototyping, will this team > > public oppose that type of usage? > > > > I ask because I've run into Agilists who were/are adamantly opposed to > > prototyping. > > Heh, ok, well, I have no problem... use the tool for what it's good at. :) If > you look at "Agile Web Development with Rails", it gets right into prototyping > the shopping cart right away, and promotes incremental changes that you can show > the customer as you build it. So if anything, I would think this is more or > less the official Rails philosophy. :) Excellent. Thanks. Both for your position statement and for your patience working through this today. I appreciate both. Still waiting to hear if there are any dissenting voices. I definitely would like to understand the boundary conditions. Thanks again to all. Best regards, Bill From tanner.burson at gmail.com Thu Mar 30 15:30:53 2006 From: tanner.burson at gmail.com (Tanner Burson) Date: Thu, 30 Mar 2006 14:30:53 -0600 Subject: [Instantrails-developers] Instant Rails Light In-Reply-To: <1b0401c65436$43973cf0$6501a8c0@dp2000> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000> <442C29B7.7040402@dgrmm.net> <1ac901c65430$1e4e01f0$6501a8c0@dp2000> <442C3692.9090308@dgrmm.net> <1b0401c65436$43973cf0$6501a8c0@dp2000> Message-ID: On 3/30/06, Bill Walton wrote: > > David Morton wrote: > > > > Bill Walton wrote: > > > > > It does. My request for IR-Lite was based on an experience-set that > says > > > "if you can avoid a potential problem, avoid it." IR-Lite is, for my > > > purposes, of secondary importance at the moment. Nice to have, but > not > > > necessary to achieve my short-term goal. > > > > Ok, well, I think that can be managed by having the initial steps be the > > easiest, and then have more sophisticated things a menu down or > something... or > > have a novice view and an advanced view. > > That would totally resolve my 'issue'. "You folks want to go exploring... > fine. We assume you know what you're doing." Just don't want them > tripping > over things if they stay around the campsite. Thanks. I think that's definitely something we can all agree on, the simple things SHOULD be simple, and the rest should be as simple as we can make it. > > > > If I were to begin promoting IR (Lite or otherwise) as a tool that can > > > provide value to organizations by using it for prototyping, will this > team > > > public oppose that type of usage? > > > > > > I ask because I've run into Agilists who were/are adamantly opposed to > > > prototyping. > > > > Heh, ok, well, I have no problem... use the tool for what it's good at. > :) > If > > you look at "Agile Web Development with Rails", it gets right into > prototyping > > the shopping cart right away, and promotes incremental changes that you > can show > > the customer as you build it. So if anything, I would think this is > more > or > > less the official Rails philosophy. :) > > Excellent. Thanks. Both for your position statement and for your > patience > working through this today. I appreciate both. > > Still waiting to hear if there are any dissenting voices. I definitely > would like to understand the boundary conditions. Thanks again to all. I don't think you'll hear any dissent for using IR in a situation like that. Rails is great for prototyping (and making full applications as well) and therefore IR should be good at prototyping as well. Glad to get some comments, and see some interest from others. Best regards, > Bill > -- ===Tanner Burson=== tanner.burson at gmail.com http://tannerburson.com <---Might even work one day... -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/0ce789a5/attachment.htm From bill.walton at charter.net Thu Mar 30 15:22:32 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 14:22:32 -0600 Subject: [Instantrails-developers] Instant Rails Light References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com><442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000><1a8201c65429$b1e8e3a0$6501a8c0@dp2000> <442C3431.6040805@dgrmm.net> Message-ID: <1b0c01c65437$ad4bfa90$6501a8c0@dp2000> David Morton wrote: > The only step I can think besides what we > are doing is to include several examples to > illustrate what RoR can do... and maybe have > an interface in the IR manager to launch them. > This is why we have the cookbook app and typo > included... Just wanted to follow up here to make sure it's clear to all that I am NOT asking for this kind of work from you. I'm absolutely happy with the examples you currently ship and the current launch capability in IR manager is easy to understand and use. > Another approach is simply to have good > documentation and tutorials, which I'm > all for including, but the scope of that is > larger than just IR. :) Both of which I'm happy, excited really, to help with. Best regards, Bill From bill.walton at charter.net Thu Mar 30 16:42:34 2006 From: bill.walton at charter.net (Bill Walton) Date: Thu, 30 Mar 2006 15:42:34 -0600 Subject: [Instantrails-developers] typo and rails 1.1 References: <442C3EF9.3090701@dgrmm.net> Message-ID: <1b5101c65442$db2caa30$6501a8c0@dp2000> > This thread with Bill has underscored a need > for some more up to date tutorials and > examples, so it wouldn't hurt to come up with > some simple examples to include. Just as an FYI... I'll be redoing Curt's Cookbook tutorial in the next few weeks to illustrate some of the new 1.1 functionality. The tutorial will be IR-specific (i.e., no install stuff). Also, per our conversation today, I'll be doing a new tutorial shortly after Cookbook that targets the new audience I'm after. I haven't given it much thought yet but do expect it will be something corporate-oriented. Any suggestions on either one, especially on what new stuff to highlight in Cookbook, is welcomed and appreciated. Bill From curt.hibbs at gmail.com Thu Mar 30 17:08:10 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 16:08:10 -0600 Subject: [Instantrails-developers] typo and rails 1.1 In-Reply-To: <1b5101c65442$db2caa30$6501a8c0@dp2000> References: <442C3EF9.3090701@dgrmm.net> <1b5101c65442$db2caa30$6501a8c0@dp2000> Message-ID: <31d15f490603301408g5144df5cjbecc6ff07aaf7387@mail.gmail.com> On 3/30/06, Bill Walton wrote: > > > This thread with Bill has underscored a need > > for some more up to date tutorials and > > examples, so it wouldn't hurt to come up with > > some simple examples to include. > > Just as an FYI... > > I'll be redoing Curt's Cookbook tutorial in the next few weeks to > illustrate > some of the new 1.1 functionality. The tutorial will be IR-specific (i.e > ., > no install stuff). Yeah, I put Bill in touch with O'Reilly and he's going to rework my original articles to bring them up to date and O'Reilly is going to republish these new versions. David, were also suggesting that we include some more sample apps -- ones that are smaller than Typo? Its not a bad idea. Curt Also, per our conversation today, I'll be doing a new tutorial shortly after > Cookbook that targets the new audience I'm after. I haven't given it much > thought yet but do expect it will be something corporate-oriented. > > Any suggestions on either one, especially on what new stuff to highlight > in > Cookbook, is welcomed and appreciated. > > Bill > _______________________________________________ > Instantrails-developers mailing list > Instantrails-developers at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-developers > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/a7c0fbf6/attachment.htm From curt.hibbs at gmail.com Thu Mar 30 17:18:21 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Thu, 30 Mar 2006 16:18:21 -0600 Subject: [Instantrails-developers] Instant Rails Light In-Reply-To: <1b0c01c65437$ad4bfa90$6501a8c0@dp2000> References: <31d15f490603300915o7a94dcc2p2a0b7eb20f962fc7@mail.gmail.com> <442C199A.6090109@dgrmm.net> <1a3c01c65426$17a35b70$6501a8c0@dp2000> <1a8201c65429$b1e8e3a0$6501a8c0@dp2000> <442C3431.6040805@dgrmm.net> <1b0c01c65437$ad4bfa90$6501a8c0@dp2000> Message-ID: <31d15f490603301418r53c527d5i79bd8a6b33cf1c61@mail.gmail.com> It looks like the consensus here is that there is really nothing for us to do that we wouldn't have done anyway -- continue to make the process as simple as possible. On a related note (inspired by David's comments), maybe we should think about including RadRails all preconfigured and ready to work with existing Instant Rails pieces. On the plus side that would really make the entire development environment setup a no-brainer. The downside is hat we'd end up with a really huge download, and a lot of people with have their preferred IDE that won't be RadRails. Or, to eliminate that last problem, distibute a RadRails addon that you could download separately and merely unzip into you Instant Rails directory. The Manager could detect its presence and include it in the new-directory-detection to relocate (if needed) any of its config files. Is this worth doing? Curt On 3/30/06, Bill Walton wrote: > > David Morton wrote: > > > The only step I can think besides what we > > are doing is to include several examples to > > illustrate what RoR can do... and maybe have > > an interface in the IR manager to launch them. > > This is why we have the cookbook app and typo > > included... > > Just wanted to follow up here to make sure it's clear to all that I am NOT > asking for this kind of work from you. I'm absolutely happy with the > examples you currently ship and the current launch capability in IR > manager > is easy to understand and use. > > > > > Another approach is simply to have good > > documentation and tutorials, which I'm > > all for including, but the scope of that is > > larger than just IR. :) > > Both of which I'm happy, excited really, to help with. > > Best regards, > Bill > _______________________________________________ > Instantrails-developers mailing list > Instantrails-developers at rubyforge.org > http://rubyforge.org/mailman/listinfo/instantrails-developers > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060330/45976ba4/attachment.htm From curt.hibbs at gmail.com Fri Mar 31 17:06:01 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Fri, 31 Mar 2006 16:06:01 -0600 Subject: [Instantrails-developers] Instant Rails 1.1 -- things to fix Message-ID: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> I just went through all of the functionality of the Instant Rails 1.1 that David put together. For the most part everything works just fine. I just need to make changes for the following: - Update release notes: - verson # = 1.1 - update rails & rake version #s - include mongrel (with note about why) - explain typo: still rails 1.0 - Help>fxri not working I plan to do that this evening and then release Instant Rails 1.1. Thanks for doing all the work David! Curt -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060331/8add5e4f/attachment.htm From mortonda at dgrmm.net Fri Mar 31 21:10:18 2006 From: mortonda at dgrmm.net (David Morton) Date: Fri, 31 Mar 2006 20:10:18 -0600 Subject: [Instantrails-developers] Instant Rails 1.1 -- things to fix In-Reply-To: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> References: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> Message-ID: <442DE10A.6090908@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > I just went through all of the functionality of the Instant Rails 1.1 > that David put together. For the most part everything works just fine. I > just need to make changes for the following: > > * Update release notes: > o verson # = 1.1 > o update rails & rake version #s > o include mongrel (with note about why) Sounds good, and this way maybe we can get some feedback from users about mongrel... though I have no doubt it is the future. > o explain typo: still rails 1.0 As soon as typo has a new version, which may be soon, we can add it and release a 1.1.1 or 1.1-2 or something. > Thanks for doing all the work David! My pleasure! - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELeEKSIxC85HZHLMRAubEAJ9US4i2he0OhrsBGCL96dRySkECKQCgiZWN kFVYpRxx0BOyl1B8RyTjNCw= =pSDp -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Fri Mar 31 22:06:36 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Fri, 31 Mar 2006 21:06:36 -0600 Subject: [Instantrails-developers] Instant Rails 1.1 -- things to fix In-Reply-To: <442DE10A.6090908@dgrmm.net> References: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> <442DE10A.6090908@dgrmm.net> Message-ID: <31d15f490603311906p525ffd40l7fb2c365dd29c8f@mail.gmail.com> I'm going to remove the BlueCoth rubygem because that was where we were getting false virus hits. In was included because Typo will offer it as one of the markup choices, but its not really necessary and I'd rather avoid the anti-virus hits. Curt On 3/31/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Curt Hibbs wrote: > > I just went through all of the functionality of the Instant Rails 1.1 > > that David put together. For the most part everything works just fine. I > > just need to make changes for the following: > > > > * Update release notes: > > o verson # = 1.1 > > o update rails & rake version #s > > o include mongrel (with note about why) > > Sounds good, and this way maybe we can get some feedback from users about > mongrel... though I have no doubt it is the future. > > > o explain typo: still rails 1.0 > > As soon as typo has a new version, which may be soon, we can add it and > release > a 1.1.1 or 1.1-2 or something. > > > Thanks for doing all the work David! > > My pleasure! > > - -- > David Morton > Maia Mailguard - http://www.maiamailguard.com > Morton Software Design and Consulting - http://www.dgrmm.net > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.2 (MingW32) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFELeEKSIxC85HZHLMRAubEAJ9US4i2he0OhrsBGCL96dRySkECKQCgiZWN > kFVYpRxx0BOyl1B8RyTjNCw= > =pSDp > -----END PGP SIGNATURE----- > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060331/c59fb125/attachment.htm From mortonda at dgrmm.net Fri Mar 31 22:08:37 2006 From: mortonda at dgrmm.net (David Morton) Date: Fri, 31 Mar 2006 21:08:37 -0600 Subject: [Instantrails-developers] Instant Rails 1.1 -- things to fix In-Reply-To: <31d15f490603311906p525ffd40l7fb2c365dd29c8f@mail.gmail.com> References: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> <442DE10A.6090908@dgrmm.net> <31d15f490603311906p525ffd40l7fb2c365dd29c8f@mail.gmail.com> Message-ID: <442DEEB5.8020107@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Curt Hibbs wrote: > I'm going to remove the BlueCoth rubygem because that was where we were > getting false virus hits. In was included because Typo will offer it as > one of the markup choices, but its not really necessary and I'd rather > avoid the anti-virus hits. Also, if you look, the blucloth, redcloth, and rubypants are all pre-installed in typo's vendor directory, so we don't really need to distribute them in gems... Unless we want to keep them around in case users want to use them... - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELe61SIxC85HZHLMRAtWeAKCfJJwi0NQEyLNH3hU1huYxZ6JnRQCfcURE 2Be8T3gbwt8fmSLRdDkZRMI= =D1wB -----END PGP SIGNATURE----- From curt.hibbs at gmail.com Fri Mar 31 22:14:14 2006 From: curt.hibbs at gmail.com (Curt Hibbs) Date: Fri, 31 Mar 2006 21:14:14 -0600 Subject: [Instantrails-developers] Instant Rails 1.1 -- things to fix In-Reply-To: <442DEEB5.8020107@dgrmm.net> References: <31d15f490603311406y3c1277a4g227cd7a6644edb1f@mail.gmail.com> <442DE10A.6090908@dgrmm.net> <31d15f490603311906p525ffd40l7fb2c365dd29c8f@mail.gmail.com> <442DEEB5.8020107@dgrmm.net> Message-ID: <31d15f490603311914w3c2a0001wc7661e969fdea058@mail.gmail.com> I didn't realize that. Ok then, I'm getting rid of all three. Curt On 3/31/06, David Morton wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Curt Hibbs wrote: > > I'm going to remove the BlueCoth rubygem because that was where we were > > getting false virus hits. In was included because Typo will offer it as > > one of the markup choices, but its not really necessary and I'd rather > > avoid the anti-virus hits. > > Also, if you look, the blucloth, redcloth, and rubypants are all > pre-installed > in typo's vendor directory, so we don't really need to distribute them in > gems... > > Unless we want to keep them around in case users want to use them... > > > > - -- > David Morton > Maia Mailguard - http://www.maiamailguard.com > Morton Software Design and Consulting - http://www.dgrmm.net > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.2 (MingW32) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFELe61SIxC85HZHLMRAtWeAKCfJJwi0NQEyLNH3hU1huYxZ6JnRQCfcURE > 2Be8T3gbwt8fmSLRdDkZRMI= > =D1wB > -----END PGP SIGNATURE----- > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://rubyforge.org/pipermail/instantrails-developers/attachments/20060331/5644af2b/attachment-0001.htm From mortonda at dgrmm.net Fri Mar 31 22:49:09 2006 From: mortonda at dgrmm.net (David Morton) Date: Fri, 31 Mar 2006 21:49:09 -0600 Subject: [Instantrails-developers] possible better fix for typo Message-ID: <442DF835.6080701@dgrmm.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I don't suppose it matters too much at this point, but instead of placing rails 1.0 in the vendor tree of typo, it could be installed as gems, and then typo runs with this patch: http://www.typosphere.org/trac/attachment/ticket/748/typo-2_6_0-with-rails-1_0-gem.patch I think this should be the default behavior of all rails apps.... just specify in the evironment.rb which version of rails you want. - -- David Morton Maia Mailguard - http://www.maiamailguard.com Morton Software Design and Consulting - http://www.dgrmm.net -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFELfg1SIxC85HZHLMRAtZGAJ9pHvLf4nYnWyTZv2pvj7wfdQ1/agCffgQV vsrBW2fqC7/VcU4J7QNL6qY= =6+/I -----END PGP SIGNATURE-----