From: Drew M. <dre...@ya...> - 2004-01-06 22:52:28
|
Recently there was a discussion on comp.lang.lisp about how hard it was to start using Lisp. Maybe some of you remember it; it was more productive than most such discussions. At some point one of the participants mentioned that it was very hard to understand the CLOCC front page (http://clocc.sourceforge.net/), and someone else concurred. I remember feeling that way myself when I first saw it. Under "Getting CLOCC" it says "nightly snapshots" and two references to PORT and CLLIB that don't go anywhere useful. I eventually figured out that the "nightly snapshots" was a good place to download from, but the phrase "nightly snapshots" tends to indicate instability. I think we should reorganize the page so that after the advertising comes a clearly delineated part about how the beginning user can download pieces of or all of CLOCC. In addition, there is a section on how to get started with each piece. The phrase "runs out of the box" in the advertising becomes very annoying after you poked around and discovered there is no box, so to speak. After that section could come the stuff about mailing lists and bug reports. Finally would come pointers to other sites -- repositories, faqs, or whatever. By my count there are several different ways to get CLOCC: Download the latest build (the "snapshots") Anonymous or nonanonymous CVS Use debian packages The prominently displayed options "CVS browsing" and "[tarball]" are _not_ ways of getting CLOCC, although I didn't realize that for a long time. (I'm a relatively newcomer to CVS, and lots of remain relative newcomers forever.) The tarball is especially misleading. If you download this you discovery it's the current repository. Why would anyone want that? The CVS browser could be useful to someone (although certainly not a beginner); I suppose one might download the tarball so he could browse on his local machine, although it's hard to see why someone would want up-to-minute archive information about an archive frozen at a particular time (July, 2003?). Anyway, it should be clearly explained what the pros and cons are of the true downloading options. Then exactly what you do to download and install something should be spelled out. It would also be good if the web page could make clear that CLOCC consists of several pieces that are usually independent of each other, and make the dependencies clear. (I know there are tools out there that allow one to download a software chunk and all the things it depends on, but I'm not familiar with them.) At the very least the page should have a table indicating, for each piece, which versions are current. (From a glance at my own contribution, ytools, it looks like the nightly snapshot includes the main branch of each CLOCC component. Information on downloading using CVS might mention the existence of other branches.) Somewhere on the page it should mention the procedure for making new contributions to CLOCC. Obviously, anyone who submits stuff to CLOCC is responsible for maintaining the pointer on the web site to the latest version of the stuff. I will just go ahead and create a new front page, and send it around. If anyone has comments on my proposal, I'd like to hear them. I'm not sure what my powers are as a developer to edit everything at the CLOCC site. One thing that needs to be deleted is the bogus 'src' directory below /cvsroot/clocc. I accidentally created this when trying to upload ytools the first time. I'm sure its existence has confused people who got close to finding CLOCC and were misdirected at the last minute. Could someone who knows how to delete the 'src' directory do so? Thanks. -- -- Drew McDermott Yale University CS Dept. |
From: Sam S. <sd...@gn...> - 2004-01-06 23:25:08
|
> * Drew McDermott <qerj.zpqrezbgg@lnyr.rqh> [2004-01-06 17:52:25 -0500]: > > two references to PORT and CLLIB that don't go anywhere useful. they go to the respective README files which, in turn, point to ZIP distributions - what's wrong with that? > I will just go ahead and create a new front page, and send it around. > If anyone has comments on my proposal, I'd like to hear them. great! please make sure that it is valid XHTML strict, just like the current page is. > I'm not sure what my powers are as a developer to edit everything at > the CLOCC site. I suggest that we put the web page under the CVS - then everyone will be able to edit it. would you like a top-level (parallel to "clocc" and "CVSROOT") directory "www" or 2nd-level (parallel to "etc", "src", "bin") directory "doc"? please vote. > One thing that needs to be deleted is the bogus 'src' > directory below /cvsroot/clocc. <https://sourceforge.net/tracker/index.php?func=detail&aid=872022&group_id=1&atid=200001> -- Sam Steingold (http://www.podval.org/~sds) running w2k <http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/> <http://www.mideasttruth.com/> <http://www.honestreporting.com> Beauty is only a light switch away. |