Greetings: I want to add here that Paul's project certainly has my blessings. We all know how badly the site needs redesigned, and Paul is aware that I'm not going to put the time into learning a better content management system. I'd like to, but my time is severely limited and I have to be doing other things. IIRC, we also discerned that one of the first tasks would be the establishment of a backend database from the existing links and categories (which I would like to see redefined too). The URLs would need culled and organized: I don't think this would be a terribly difficult first step, maybe a good Perl script can do it ? Is PW not a Perlmeister ?? ;) At any rate, if anyone else wants to redefine the Linux soundapps pages, please contact me about it. This message is simply to state that PW and I have corresponded about such a development, that there is a historical context to his quoted remarks, and that I support the effort. Best regards, dp PS: I released a new edition a couple of days ago. Too busy/lazy to announce it... Paul Winkler wrote:

Hi all, For a template to possibly follow or borrow from for a new linux sound apps site, this is really nicely designed - needs more linux apps listed too: http://www.databaseaudio.co.uk/ (you get a lot of hits from interested non-linux people too ;)) cheers, dave -- ................................. www.pawfal.org/nebogeo

On Sat, Jan 31, 2004 at 05:14:12 -0500, Dave Phillips wrote: RDF! (rdf, rdf, rdf, ...) http://www.w3.org/RDF/ Full disclosure: You may want to take this with a pinch of salt due to favourite hammer syndrome; I currently justify my existence by developing a large-scale (GPL'd of course) RDF server. Good points: * this is exactly the kind of thing RDF is designed for * nice query syntax (like SQL but better) * very fashionable * interoperability (metadata can be used, and partly understood, by other tools because RDF has inherent semantics) Bad points: * RDF/XML syntax is horrific (but you hardly ever deal with it) * not famialiar to people who know LAMP systems * a bit more resource intensive than PHP+MySQL * RDF tools not as mature as relational databases You wouldn't want to use RDF for things like user comments and reviews, but its ideal for metadata-y things. I could help by providing things like web form -> RDF/XML CGI scripts and RDF knowledge-base hosting for betas. - Steve

Steve Harris wrote: topic maps! topic maps! topic maps! XTM? (XML Topic Maps) http://xml.coverpages.org/topicMaps.html Cheers, Andreas

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dave Phillips wrote: [snip] [snip] I'd be interested in doing this first step. Can I propose a simple csv file for now? That or db - I have a little time and could write the scripts to parse your website. - -- Robin Y. Millette (aka Lord D. Nattor) http://rym.waglo.com/

On Sun, Feb 01, 2004 at 02:10:02 +0100, Frank Barknecht wrote: RDF servers tend to be databases underneath anyway (mines built on mysql) for speed and persistence reasons. You might be confusing RSS and RDF, RSS uses one RDF schema (and quite a popular one) but you generally use others in addition for domain specific things (like application and person metadata in this case). - Steve

On Sun, Feb 01, 2004 at 04:02:37 +0100, Frank Barknecht wrote: Heh, yes :) I think an RSS feed is pretty much essential, but wether we use RDF for the storage as well depends on the implementor. I knocking up a quick RDF version of Daves site to show what the data would look like. - Steve

On Sun, Feb 01, 2004 at 05:48:25 +0100, Frank Barknecht wrote: The RDF Primer is OK: http://www.w3.org/TR/rdf-primer/ You can see the data here: http://triplestore.aktors.org/~swh/apps/rdf/ Its a bit b0rken cos I just used a perl script to make it, but you can see the idea. I'l hack together a quick PHP script to browse it. - Steve

On Sun, Feb 01, 2004 at 02:10:02PM +0100, Frank Barknecht wrote: My mock-up is visually crude and is only meant to give an idea of the information each page would contain. Design help will be welcomed. -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's ALTO PEASANT VON ALPHA! (random hero from isometric.spaceninja.com)

On Sat, Jan 31, 2004 at 11:42:38PM +0000, Dave Griffiths wrote: It is pretty nice. A bit busy at first glance, but I think removing the right nav box (which contains only stuff we don't need) will fix that. And we will have potentially a lot more information for each listed app. I do like the mini-screenshots, that's a nice feature and wouldn't be very resource-intensive. Also, I hope someone can suggest a less dismal color scheme than gray, gray, and more gray ;-) -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's COMBUSTION REACTION BLADE! (random hero from isometric.spaceninja.com)

Hallo, Paul Winkler hat gesagt: // Paul Winkler wrote: I recently programmed, but not designed (I'm no desinger) this site, whose colors you *might* like: http://normalmailorder.de ;) ciao -- Frank Barknecht _ ______footils.org__

On Sunday 01 February 2004 10:22 am, Steve Harris wrote: They were so much better when they were still Military Surplus. Ah those were the days ... eh? tim hall

On Mon, Feb 02, 2004 at 09:08:05AM +0000, Steve Harris wrote: ok. I have to say that's pretty damn weird. near zero I do web development with Zope. My site code never depends on the data store :-) That is pretty interesting. I hadn't considered that *I* might be the 3rd party :-) -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's THE SPAMSTRANGLING FOOD! (random hero from isometric.spaceninja.com)

On Mon, Feb 02, 2004 at 02:55:02PM +0000, Steve Harris wrote: I'm not sure what "schema independence" means, so I will assume it means the ability to extend or modify the schema after you've already got live instances. I'm curious what languages you've used with an OODB. I imagine it would be a major pain if you didn't have as much runtime flexibility as python (e.g. ruby would be just fine, C would be insane, dunno about java). It's been a non-issue in the 3 years I've been doing Zope. Content types in zope are generally defined with classes. Adding an attribute to a class requires me to add a couple of lines to the class definition, including one to provide a default class-level value, then I tell zope to reload the class by clicking a button in the zope management interface. Live instances immediately see the default value. This requires no deep magic, it's just how python works. An untested-but-should-work example (feel free to skip the rest of this message): from ZODB import Persistence class SoftwareResource(Persistence.Persistent): meta_type='Software Resource' def __init__(self, name='', author='', download_url=''): self.name=name self.author=author def setAuthor(self, author): # setter methods are totally optional in python, # but we usually use them in zope so we can # declare security restrictions on them, # which I won't bother with in this example. self.author = author def setName(self, name): self.name = name Now let's add a release_number attribute with a default value of '': class SoftwareResource(Persistence.Persistent): release_number = '' ... def __init__(self, name='', author='', release_number=''): ... self.release_number=release_number ... def setRelease(self, release_number): self.release_number=release_number Reload the class and you're done. Changing the 'author' attribute to 'author_name' is only barely more difficult. You can embed self-upgrading code in the class, but such code has a way of sticking around long after it should be deleted. Another way to do it is with a trivial upgrade script like this, to be run once after reloading the relevant class changes: # A bit like find . -type ... for obj in context.ZopeFind(meta_type='Software Resource'): if not hasattr(obj, 'author_name'): obj.setAuthor(obj.author) del(obj.author) -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's THE FISHY ANUS! (random hero from isometric.spaceninja.com)

On Mon, Feb 02, 2004 at 05:31:39PM +0000, Steve Harris wrote: Splitting and joining classes? You lost me there. Refactoring? -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's VOODOO FARMER! (random hero from isometric.spaceninja.com)

On Mon, Feb 02, 2004 at 06:42:47PM +0000, Steve Harris wrote: How is this done in the RDF world? sorry to be a pest, it's just new concepts to me and I'm slow to grasp the benefits. -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's THE FINALE! (random hero from isometric.spaceninja.com)

On Tue, Feb 03, 2004 at 09:56:03AM +0000, Steve Harris wrote: okey-doke. Now, how do you, as a third party, make use of all this stuff? Do I have to provide a query engine at some URL or do you just walk through my pile of RDF documents parsing stuff? Did you mean to say that the other way around? i.e. "any queries you have on ColouredThings will be satisfied by instances of GreenThings". -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's MY CAT PUMA! (random hero from isometric.spaceninja.com)

On Tue, Feb 03, 2004 at 03:37:21PM +0000, Steve Harris wrote: OK. I doubt i will do the latter. It might be pretty easy to provide something like your "view as RDF" link from each page. Thanks for all the answers! -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's SUPERFLOUS FORNICATOR KATANA! (random hero from isometric.spaceninja.com)

Hallo, Eric Dantan Rzewnicki hat gesagt: // Eric Dantan Rzewnicki wrote: As bibliographies are used a lot in science, it might be good to use a standardized more format. Bibtex comes to mind, which is used in Physics *everywhere*. If you look at a random Citeseer entry like: http://citeseer.nj.nec.com/lanneer95ches.html you'll see the Bitex reference on every page. You could of course generate bibtex entries from other sources, like the mentioned RDF or store things in a database. But I would say, that a bibtex-import is important as well. If you look at the wonderful Pybliographer you get a lot of Python code to copy and paste. ciao -- Frank Barknecht _ ______footils.org__

On Tue, Feb 03, 2004 at 03:23:38 -0500, Eric Dantan Rzewnicki wrote: Bittex is great for citation mangement, but its not a good format for storage of references, its quite hard to recover data from it for other purposes, but that may not be an issue. I really ought to be able to recommend to article annotation software, its something I've done research in, but sadly I dont know of anything thats particularly useful :-/ - Steve