Thread: [Audacity-devel] gsoc thanks
A free multi-track audio editor and recorder
Brought to you by:
aosiniao
From: Michael B. <mic...@gm...> - 2008-04-21 22:12:19
|
Hello, Though I was not selected, I wanted to thank you all for your help with my application. I certainly know more about how Audacity and digital sound work now than I did two months ago. I will definitely consider applying again next year. Do any of you who read my application have any last advice on what I could work on to make myself a more likely GSoC candidate in general? Thank you, Michael |
From: Prashant V. <p.v...@ja...> - 2008-04-21 22:28:25
|
Audacity dev team: Thanks also for the discussions and help on my proposal (audio diff), specially James Crook. I wasn't selected either, but in the process I downloaded and built the CVS head and am now already working towards sending a first patch (mfcc display) :-D For the coming few weeks I want to be making minor modifications, until early June when I should start working on the proposal I submitted. I'm mostly free this summer, so it should be a lot of fun. Cheers, Prashant 2008/4/22, Michael Brooks <mic...@gm...>: > > Hello, > Though I was not selected, I wanted to thank you all for your help with > my application. I certainly know more about how Audacity and digital > sound work now than I did two months ago. I will definitely consider > applying again next year. > > Do any of you who read my application have any last advice on what I > could work on to make myself a more likely GSoC candidate in general? > > Thank you, > Michael > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > _______________________________________________ > Audacity-devel mailing list > Aud...@li... > https://lists.sourceforge.net/lists/listinfo/audacity-devel > -- ____________________________________________________ PRASHANT VAIBHAV, Class of 2008 (EE/CS) School of Engineering and Science, Jacobs University Bremen Postal address: Phone: +49 421 200 5714 College Ring 7, Box 83 Apartment: College III B202 Bremen 28759, Germany ____________________________________________________ |
From: James C. <cr...@in...> - 2008-04-22 12:43:51
|
Prashant Vaibhav wrote: > Audacity dev team: > > Thanks also for the discussions and help on my proposal (audio diff), > specially James Crook. I wasn't selected either, but in the process I > downloaded and built the CVS head and am now already working towards > sending a first patch (mfcc display) :-D > > For the coming few weeks I want to be making minor modifications, > until early June when I should start working on the proposal I > submitted. I'm mostly free this summer, so it should be a lot of fun. > > Cheers, > Prashant Prashant, It's wonderful to hear your enthusiasm in spite of not getting a slot. Tell us more about how you'd like to do this. The 'how' here is organisational, meaning do you want to try and track the GSoC structure, or more freewheeling? It's an interesting situation which I've not quite adjusted to yet! If it's the audio side, Roger/Chris can send you any number of pdfs with background. If it's the GUI side, we can co-edit a wiki page on the ideal GUI, - perhaps easier for it not being done in GSoC as it matters less 'who does what'. One additional problem is that we are 'stretched' as mentors, less able to help you over any roadbumps. Whatever you plan, you need to take that into account too. If it starts not to be fun, tell us early (off list probably) what we're doing wrong, so that we can correct it in time. This is our first GSoC. We're learning too. --James. |
From: James C. <cr...@in...> - 2008-04-22 12:05:32
|
Michael Brooks wrote: > Do any of you who read my application have any last advice on what I > could work on to make myself a more likely GSoC candidate in general? > Next year from coursework and any other programming work you do you will have an extra year's coding experience, plus you'll also have the experience of having applied this year, so you know what's expected in an application. If you want advice for GSoC in general, mine is to not just 'try for GSoC', but to have fun doing it too. It will come across to the mentoring orgs. They will notice. If it's fun, you'll do more, and if you don't get selected you still end up feeling "I'm still glad I did that". LH says that many of the best/most successful projects were student's own ideas. Given that she probably has the best view of anyone as to what happens in GSoC, I'd say that's good advice. I'd add that to get the best ideas would usually mean spending some time using the latest CVS/SVN version of the code for something real. I'm not sure that happened that much for us this year, but as a general principle it seems a good one. --James. |
From: Alexandre P. <ale...@gm...> - 2008-04-22 14:06:13
|
On Tue, Apr 22, 2008 at 2:12 AM, Michael Brooks com> wrote: > Do any of you who read my application have any last advice on what I > could work on to make myself a more likely GSoC candidate in general? Avid users and contrubutors always have most chances to participate ;-) E.g. Inkscape has two GSoC2007 students this year again, its 2007 student is in Blender this year, and in 2007 a GSoC2006 student participated. Go figure ;-) Alexandre |
From: Richard A. <ri...@au...> - 2008-04-22 20:23:24
|
On Mon, 2008-04-21 at 18:12 -0400, Michael Brooks wrote: > Do any of you who read my application have any last advice on what I > could work on to make myself a more likely GSoC candidate in general? Grasp of "the bib picture" of how audacity (or any program) is put together is certainly one thing I would pick up on. The envelope class wasn't the right place to try and implement fading in and out (because it's used in many other places, as well as the issues you identified), it would have been good if you had either asked or spotted this. Another worry was whether the implications of trying to implement dynamic loop point changes were really clear to you - this section of the code has two if not three threads running in it, so the design is non-trivial (comments at the top of AudioIO.cpp expand in detail, the main issue is keeping the different contexts connected to each other reliably). The more you know about the project and it's background can only help - issues like making sure it is usable on low-specification machines, at 800x600 resolution and so on are regular themes around audacity, and any project will have some attitudes like this that underpin the way the project community thinks about ideas. The other think that raises students (as opposed to proposals) is if they have done something relevant outside of courses - especially contributions to other projects because it shows you can work as part of a team on a project where you aren't the sole director. Your application was better than some in this regard, but I would definitely make sure you write up what you did this year in any future applications. If you can separate out something simple from the work that is complete enough to be actually useful, and contribute that as patch, then even better! Richard Ash |
From: Michael B. <mic...@gm...> - 2008-04-25 22:01:28
|
I agree that my discussion of the loop-endpoints feature was not sufficient. I looked more closely at this feature in the week before decisions were made, and realized it would actually be pretty complicated to implement. I could see that the way the interface was built in the region of the Ruler was not easy to understand immediately. I was also aware that any changes like this would probably involve dealing with the multiple threads. Still, I never addressed any of these problems in my proposal. I am interested in pursuing the "Click Removal" feature, so I may work on that over the summer anyway. If anyone else has thought about best ways to implement this already, I'd like to hear suggestions. Thanks, Michael On Tue, Apr 22, 2008 at 4:23 PM, Richard Ash <ri...@au... <mailto:ri...@au...>> wrote: On Mon, 2008-04-21 at 18:12 -0400, Michael Brooks wrote: > Do any of you who read my application have any last advice on what I > could work on to make myself a more likely GSoC candidate in general? Grasp of "the bib picture" of how audacity (or any program) is put together is certainly one thing I would pick up on. The envelope class wasn't the right place to try and implement fading in and out (because it's used in many other places, as well as the issues you identified), it would have been good if you had either asked or spotted this. Another worry was whether the implications of trying to implement dynamic loop point changes were really clear to you - this section of the code has two if not three threads running in it, so the design is non-trivial (comments at the top of AudioIO.cpp expand in detail, the main issue is keeping the different contexts connected to each other reliably). The more you know about the project and it's background can only help - issues like making sure it is usable on low-specification machines, at 800x600 resolution and so on are regular themes around audacity, and any project will have some attitudes like this that underpin the way the project community thinks about ideas. The other think that raises students (as opposed to proposals) is if they have done something relevant outside of courses - especially contributions to other projects because it shows you can work as part of a team on a project where you aren't the sole director. Your application was better than some in this regard, but I would definitely make sure you write up what you did this year in any future applications. If you can separate out something simple from the work that is complete enough to be actually useful, and contribute that as patch, then even better! Richard Ash ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Audacity-devel mailing list Aud...@li... <mailto:Aud...@li...> https://lists.sourceforge.net/lists/listinfo/audacity-devel |
From: Richard A. <ri...@au...> - 2008-04-25 22:14:09
|
On Fri, 2008-04-25 at 18:00 -0400, Michael Brooks wrote: > I am interested in pursuing the "Click Removal" feature, so I may work > on that over the summer anyway. If anyone else has thought about best > ways to implement this already, I'd like to hear suggestions. Dig around in the list archives - someone else did a GSoC proposal for this, so there was some recent discussion which harked back to some previous stuff. The Repair effect is relevant here as a way of repairing the clicks once it has found them. Richard |
From: Martyn S. <mar...@go...> - 2008-04-26 22:44:51
|
Hi Michael When play starts, what happens? That would be a good place to start. At some point Mix mixes all the tracks together and that may be where a short fade in should be, if it is requested. Try a breakpoint at the start of Mixer::MixSameRate, look at the call stack. That's what I'd do anyway. HTH Martyn Michael Brooks wrote: ... > I am interested in pursuing the "Click Removal" feature, so I may work > on that over the summer anyway. If anyone else has thought about best > ways to implement this already, I'd like to hear suggestions. > > Thanks, > Michael ... |
From: Michael B. <mic...@gm...> - 2008-04-26 22:47:25
|
Thanks. I have already given this some thought, and found a couple of places in the code where a fade could be added. More what I meant was, what is the best way for me to make a change like this? How do I get it approved for inclusion, etc? Michael Martyn Shaw wrote: > Hi Michael > > When play starts, what happens? That would be a good place to start. > At some point Mix mixes all the tracks together and that may be where > a short fade in should be, if it is requested. Try a breakpoint at > the start of Mixer::MixSameRate, look at the call stack. That's what > I'd do anyway. > > HTH > Martyn > > Michael Brooks wrote: > ... > >> I am interested in pursuing the "Click Removal" feature, so I may work >> on that over the summer anyway. If anyone else has thought about best >> ways to implement this already, I'd like to hear suggestions. >> >> Thanks, >> Michael >> > ... > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > _______________________________________________ > Audacity-devel mailing list > Aud...@li... > https://lists.sourceforge.net/lists/listinfo/audacity-devel > |
From: Martyn S. <mar...@go...> - 2008-04-26 23:18:54
|
Michael -devel is the place to post an idea/proposal. Most people seems to prefer patches, personally I prefer a link to a web site where I can get a new xxx.cpp (etc) file that I can replace my current one with to try (not very feasible if it's a lot of files, but it probably isn't). (I guess it depends on if reviewers are command-line oriented or drag-and-drop oriented, how many files are affected, if they can work 'patch' quickly and easily etc.) Another idea is to post an .exe on a web site, but that is prone to error (as I proved recently), excludes people on a different OS, and is impossible to code-review. Also 'big' attachments to this list could get stuck in some sort of 'review' process that does not work well / happen (?). Sending a link to a web site with all this stuff on is definitely a better bet. HTH Martyn Michael Brooks wrote: > Thanks. I have already given this some thought, and found a couple of > places in the code where a fade could be added. More what I meant was, > what is the best way for me to make a change like this? How do I get it > approved for inclusion, etc? > > Michael > > Martyn Shaw wrote: >> Hi Michael >> >> When play starts, what happens? That would be a good place to start. >> At some point Mix mixes all the tracks together and that may be where >> a short fade in should be, if it is requested. Try a breakpoint at >> the start of Mixer::MixSameRate, look at the call stack. That's what >> I'd do anyway. >> >> HTH >> Martyn >> >> Michael Brooks wrote: >> ... >> >>> I am interested in pursuing the "Click Removal" feature, so I may work >>> on that over the summer anyway. If anyone else has thought about best >>> ways to implement this already, I'd like to hear suggestions. >>> >>> Thanks, >>> Michael >>> >> ... >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference >> Don't miss this year's exciting event. There's still time to save $100. >> Use priority code J8TL2D2. >> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone >> _______________________________________________ >> Audacity-devel mailing list >> Aud...@li... >> https://lists.sourceforge.net/lists/listinfo/audacity-devel >> |
From: Vaughan J. <va...@au...> - 2008-04-27 04:19:23
|
Send us a patch. Thx, Vaughan Michael Brooks wrote: > Thanks. I have already given this some thought, and found a couple of > places in the code where a fade could be added. More what I meant was, > what is the best way for me to make a change like this? How do I get > it approved for inclusion, etc? > > Michael > > Martyn Shaw wrote: >> Hi Michael >> >> When play starts, what happens? That would be a good place to start. >> At some point Mix mixes all the tracks together and that may be where >> a short fade in should be, if it is requested. Try a breakpoint at >> the start of Mixer::MixSameRate, look at the call stack. That's what >> I'd do anyway. >> >> HTH >> Martyn >> >> Michael Brooks wrote: >> ... >> >>> I am interested in pursuing the "Click Removal" feature, so I may work >>> on that over the summer anyway. If anyone else has thought about best >>> ways to implement this already, I'd like to hear suggestions. >>> >>> Thanks, >>> Michael >>> >> ... >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference >> Don't miss this year's exciting event. There's still time to save $100. >> Use priority code J8TL2D2. >> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone >> _______________________________________________ >> Audacity-devel mailing list >> Aud...@li... >> https://lists.sourceforge.net/lists/listinfo/audacity-devel >> > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > ------------------------------------------------------------------------ > > _______________________________________________ > Audacity-devel mailing list > Aud...@li... > https://lists.sourceforge.net/lists/listinfo/audacity-devel > |
From: Vaughan J. <va...@au...> - 2008-04-22 21:23:45
|
Michael, very gracious of you. And likewise Prashant and Jen. We're very happy you two are planning to work on Audacity anyway! - Vaughan Michael Brooks wrote: > Hello, > Though I was not selected, I wanted to thank you all for your help with > my application. |