On Sun, Apr 18, 2010 at 12:20 AM, Faré <span dir="ltr"><<a href="mailto:fahree@gmail.com" target="_blank">fahree@gmail.com</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
PS: after 2.0 is out, I'd like to resign from ASDF maintainership, to<br>
focus on XCVB (help welcome on that, too). Is anyone volunteering to<br>
replace me on ASDF?<br></blockquote></div><br clear="all">After thinking about it, I am willing to step up as a possible replacement. The reasons why I would go that far are<br>* My own personal interest in ASDF working properly with ECL and other implementation philosophies.<br>
* My belief that ASDF can and should be improved along the directions I have repeatedly mentioned: declarative system files, simpler user interface for common tasks (testing, building, packing, installing...).<br>* My will to make it simpler to systematically download, build and test existing software with different implementations [1]<br>
<br>Reasons why I am still doubting:<br>
* I have the feeling my points of view do not find a strong resonance here.<br>* If other implementations remain silent, as mostly so far, this may end up a futile effort.<br>* Total unfamiliarity with the ASDF development process, with launchpad and<br>
* in particular with the history of past decisions (though this might be a plus)<br>* The previous point might also mean inadequacy to solve some existing bugs (TRAVERSE).<br>* Limited exposure to other implementations myself (SBCL, CLISP and stop counting)<br>
* ECL is and will continue to be my first priority in terms of time and resources.<br>* Very intermittent availability: CL is my hobby, not my job.<br><br>Among my goals would be to implement a faster development model that is based on a more agile dialogue<br>
* Feature is proposed with a model implementation<br>* Arguments in pro/cons in the list<br>* Implementation committed only to an asdf-devel branch or to a personal repo.<br>* Verification that all relevant software builds<br>
* Iterate from (2)<br>* Repeat until enough features grant a push to master and a release.<br><br>Ingredients needed<br>* Help.<br>* Dialogue and feedback here: developers, implementors and users.<br>* A short term roadmap and an agreement on long term goals.<br>
* A test system that works on the asdf-devel branch. I can provide computational power and several operating systems if somebody helps me identifying the perfect list of software and setting it up -- <a href="http://ecls.sourceforge.net/logs.html" target="_blank">http://ecls.sourceforge.net/logs.html</a><br>
* Involvement from the community at large, perhaps with periodic calls for evaluation at c.l.l and other forums, setting up a wiki for wish lists and collaboration with particular efforts (libcl, clbuild, lisp in a box?)<br>
* Connecting the test system with an enlarged version of the ASDF
analyzer, to monitor evolution of software in real time and how
and whether ASDF features get widely accepted.<br>
<br>All this can not be done without help. I am willing to learn and to listen to everybody's point of views, but I would expect the same, as well as understanding of my verbosity, my lack of communication skills and my occasional emotional outbursts. <br>
<br>Juanjo<br>
<br>[1] Seen <a href="http://174.129.66.148:8000/systems" target="_blank">http://174.129.66.148:8000/systems</a> I failed to do the same with ECL's current extensions due to the limitations of the current system.<br>
<br>-- <br>
Instituto de Física Fundamental, CSIC<br>
c/ Serrano, 113b, Madrid 28006 (Spain) <br><a href="http://tream.dreamhosters.com" target="_blank">http://tream.dreamhosters.com</a><br>