Older blog entries for RyanMuldoon (starting at number 28)

It's been a long time since I've written. Oops. I have been pretty busy with school, and my life in general, to do much computer type stuff. I have been trying to go on MonkeyTalk every day and help out a little bit - it's nice because I can help a bunch of people with minimal effort, as most of their questions are pretty easy. MonkeyTalk is a great idea, as simple as the implementation may be.

I'm writing here to basically vent my frustration. Slashdot has an article slamming Ximian, Eazel, the FSF, and GNOME. These are among the few organizations that I have any amount of respect for in the whole linux "community." And it bothers me to see them slammed so unfairly. What makes it worse is that I can imagine how shitty it must be to be those individuals that such slams are directed at. Working really hard on something, and being proud of your work, then watching everyone rip it to pieces for no good reason can't be a very rewarding experience. As I posted on slashdot, I am becoming increasingly saddened by all of this. People care less and less about Freedom, and more and more in entitlement and things being no cost and convenient. It is becoming rather self-destructive. I wonder if it is people that have been (at least peripherally) involved in the Free Software world for a while getting somehow disillusioned, or if it is a new batch of people that never took the time to understand the tenets of Free Software so they don't understand how the dynamic is supposed to work. I think it is the latter. Ah well. I'm just going to try and let it not bother me so much.

Iain: Have you listened to any Grandaddy? They are kind of Radiohead-esque, and are pretty good. An excellent song by them is "So you'll aim toward the sky." But, as for Amnesiac, I must say that I enjoy "Knives out" quite a bit. But The Bends is still their best album in my opinion. Amnesiac has a little more of the vintage Radiohead feel when compared to Kid A, which is kind of nice. And I enjoy the more tame horn accompaniments this time around....kind of a cool jazz feel.

As for my little Nautilus patch, I am having a hard time actually compiling Nautilus. It's complaining about me not having gtk and glib headers, even though I do. Hopefully I'll figure out what the problem is. It's annoying to not be able to do run-time testing now that I have my patch "complete" and without compiler errors. Once I have time I'll figure it out, and submit the patch. Then, hopefully I'll take on another little task that isn't too time-critical. At some point I'll have to write down a list of little things that annoy me about the software I use, and actively try and fix the problems.

I've been reading the available material on the Reef project - it sounds damn cool. I'm glad that it is clearly intended as a community project. It is great to see the ideals of the free software community, and their potential benefits, well-represented in web-enabled services. Once the semester is over I'm going to see if I can play with it a bit, and perhaps contribute some. My goal is to be a reasonably regular contributor to the GNOME project by the time GNOME 2 rolls around. We'll see how it goes.

I've been working on my first patch for Nautilus.....a right-click menu for the "Up" button in the toolbar. I am really enjoying the quality of the code in Nautilus - at least what I've looked at. It is all really easy to read and understand. Hopefully my changes follow suit. I'm pretty much done with the first pass at the patch, except I need to figure out how I can get a NautilusScalableIcon for a given uri. Once I have that, I'm pretty much set I think. So hopefully on Monday or Tuesday I'll have completed my patch, assuming I find some coding time. It should be a nice little feature though. And hopefully something of a jumping off point for other small little fixes/features in Nautilus. We'll see how it goes.

jaq:I'm glad to read that you're working on a GNOME version of Gnutella. I'd suggest that you *not* use the same UI as gtk_gnutella. It is pretty hideous. I was thinking about spending some time reworking the UI for gnapster and gtk_gnutella, as I don't really like either too much. One big improvement in both would be to use the shortcut bar that is in evolution for switching between sections. And in gtk_gnutella's case, actually having preferences in a preference dialog would be better. Also, some means to auto-grab active gnutella servers would be useful.....similar to how gnapster has a listing of OpenNap servers. Anyway, if you're interested in more ideas (and maybe a little code) email me.

It's been a long time since my last diary entry. A lot has happened since then, so I'll only talk about recent events. I am extremely happy with Nautilus 1.0.....I have a few problems with it, but they are already being worked on, which is great. I am now very much wanting to learn how to write views and sidebars for Nautilus, and bonobo components in general. The RSS viewer thing that is in the works is a really excellent idea. I am deeply saddened by Eazel having to lay people off......that is really unfortunate. But, I think that people are jumping to the conclusion that Eazel's failed - which is obviously not the case. They haven't really tried to sell anything yet. But, as soon as they do offer services for sale, I am most definitely going to get them. I want to support Eazel and Ximian. I think that they do great work. I hope that I will be able to contribute to their projects, and GNOME in general soon. Time is a hard thing to come by, but so many people here seem to manage it, so I'll make an extra effort.

I really place most of the blame on the fact that the US has become obsessed with the stock market. The concept of daytrading has really just messed up the function of the stock market, as well as normal business practices. I don't think that it is a problem unique to Free Software companies, or Linux companies, or tech companies. It is just investors not realizing that an investment is supposed to be long-term. Believe in the company, and support them. But our economy is now on a slide. Why? Because it is in the interests of the bush administration to push us into a recession, so he can gain support for a tax cut that only really benefits the wealthy. I can't say that I am particularly looking forward to the reincarnation of 80's class warfare, powered by trickle-down (oops, I mean supply-side) economics. And so legitimately good companies that I think are going to not only have excellent profit potential a little down the road, but are able to do so with the good of the public in mind are getting hurt. That is really unfortunate. I am glad that there has been something of a shakeout to get rid of companies just interested in capitalizing on a buzzword, but I hate to see problems extending to companies with a strong commitment to their work, and the community.

I am enjoying reading the "Moneyflow" article discussion. The more I think about it, the more certain I am of a few things. First, the Internet is kind of at a crossroads right now. It has the potential to start getting really cool, or keep getting less and less useful. Second, I am fed up with copyright law, and I am planning on starting to write my various governmental representatives. Third, I think a lot of the problems with payment on the Internet, as well as nifty peer to peer filesharing techniques, can be resolved with a robust, standardized system for Metadata. How else can you find the little guy? How else can you know who to pay? We can't work under the assumption that the consumer will go to the creator's website to get whatever there is to offer. Building copy protection and payment mechanisms into protocols and file formats is a really bad idea. It institutionalizes the middle man, and hurts Free Software. It also encourages piracy rather than diminishing it. Systems that will work are ones that don't treat people like criminals, and allow individuals to pick a proper "reward" for content creators. Furthermore, I refuse to believe that artists create just to get paid. If they do, the really are just not artists. Whenever I choose to create something, there is not much profit motive there. Some of the greatest works ever were created without any notion of copyright. I do believe in the original ideals of copyright law, but as it stands now, it is vastly contorted and rewritten to favor distributors rather than the people that matter - artists and appreciators.

I am becoming more and more interested in figuring out ways to leverage public domain works for the public good. There is a lot of absolutely incredible stuff out there. I have the feeling that a lot of people don't take advantage of it, either because it is hard to find, or they don't know they can. I want to figure out a way to unify a lot of the "virtual library" type projects out there, so people can search and access this stuff using something with more of a napster feel. Websites are all well and good, but we should be thinking of them more as "leisure" sorts of things. We need better searching so we can find what we want initially, and then can choose if we want to bother with the website or not. Sometimes I just want to browse through some Van Gogh. Other times I want to read about each painting, and find out about their relationships. We need ways to facilitate this. We need solid metadata systems. We need means of cross-referencing large bodies of work on the fly. We need ways to tip the people that make this all available for us. And we need it all to be in open file formats that are designed for searching and portability. Documents should be in well-structured XML. Audio should be in open formats like Ogg Vorbis. Images should be in JPG or PNG. Ok, enough ranting on this. One more rant to go:

On the topic of people complaining about various features (or lack of features) of advogato: Really, I think that the current structure of advogato is what makes it a unique and well-defined community. The diaries allow everyone to see what various community members are working on, and the public conversations are interesting to follow. The articles have no real need to be threaded - there are typically few responses to articles, so we may as well keep them as open conversations. Threading can be nice, but at the same time I see it as limiting conversation styles. The other thing that I have been kind of annoyed at is people asking for certification to a given level. It seems kind of antithetical to the idea of *trust* metrics that people ask for a given level of certification. The idea is that you prove yourself in some manner. While I'd like to be ranked higher than Apprentice, I realize that since I haven't been ranked higher, there is probably a reason. So I am perfectly content with my Apprenticeship, and am confident that when I actually deserve it, I'll be ranked higher. I certify people as a result of taking part in discussion with them. To me, that is the best indicator of where they should be in the trust metric system. The trust metric is a pretty damn cool system, and I see it as having a ton of useful applications. But it becomes pretty pointless once it is no longer about trust. So if you're not being certified up, it is probably because you haven't shown that you should be yet. Ok, that's the end of my rant.

slef: I don't think that we'd have to have a regress of metametadata......if the metadata format were standardized, and there were a query language built on top of it, there would be no need for additional description. I guess though, that a schema for the metadata itself is in a way metametadata, but it should all work out nicely. ;-)

Last night I got fed up with XMMS and how it displayed track information in a tasklist, so I changed it from being "XMMS - tracknum. track (time)" to being "track (time) - tracknum - XMMS", which lets you get a good deal more information at a quick glance. I sent a patch the the xmms people....hopefully it is incorporated.

The thought of extending metadata to services is cool. It has a lot of potential.

Quote of the day:
"The mind of man is capable of anything - because everything is in it, all the past as well as all the future. What was there after all? Joy, fear, sorrow, devotion, valour, rage - who can tell? - but truth - truth stripped of its cloak of time. Let the fool gape and shudder - the man knows, and can look on without a wink."
--Joseph Conrad, Heart of Darkness

Ankh: I'd definitely be interested in getting in touch with some people with similar goals. I think that it is ultimately essential to the health of the Internet that public domain works are made easily available, and also that searching is vastly improved.

I'll take this opportunity to rant a bit. First, I am becoming increasingly disillusioned by the world wide web. I used to think that it was the coolest thing since sliced bread. But, overcommercialization is killing it. It is becoming harder and harder to find the actual information that I want, because searching was tacked on as an afterthought. I think that the world-wide web can be broken down into 4 categories:

  • Community sites: Things like Advogato, Slashdot, K5, etc. Sites that are too broad or with too great a userbase are really showing it. Slashdot used to be great, but now it is barely worth going to, except that it is habit. Advogato in my mind strikes a great balance. I think the 2 big reasons for that are that it has a focus (free software developers), and the trust metric. The trust metric is great, and has a ton of uses.
  • Services: Things like expedia.com, ticketmaster.com, and buy.com are all pushing a "web application" and are pretty useful. But I think that they can probably be revised in terms of metadata to be much cooler.
  • News sites: salon.com, cnn.com, etc. These are all useful, but it is annoying that I have to go to the front page to see if there are even any articles I want to read. Another candidate for metadata magic, or client pull.
  • Research Material/Public domain works:This stuff was the original purpose of the WWW. It is really lacking though, because the nature of research is that you have to be able to find it. But, as I said before, searching is kind of weak right now. This is also a huge candidate for metadata magic.

The problem is that people are trying to make the desktop more like the web, where I think the opposite should be true. Web sites should be seen like any document or application. Mozilla should not be an evironment for me to do anything. It should be a rendering engine for the content that I asked for. I think that a nice unified search system should be how I find what content I want. Same with things I want to buy. News should be client-pulled for me and put into my desktop environment (like a "News" subdirectory in the gnome menu). Why "browse" unless I'm trying to kill time? It seems kind of dumb.

Now my rant will break out from just technological complaints to general intellectual property complaints. I completely agree with Ankh that people should be focusing their investments on Museums, Libraries, and other public repositories, rather than hogging important works to themselves. I can understand the joy of owning an original painting, or a first edition (and would definitely love to be in a position to be able to afford such things one day), but I'd like to think that it would be better to give or loan such things to museums, and just buy the print for my own enjoyment. Some things are too important to be held privately. However, Museums and libraries need to shape up. They don't display anywhere near 20% of their holdings. What isn't on display is packed in crates where no one can enjoy it, research it, or do anything with it. This isn't in line with the function of a museum. I think that they have an obligation to supply electronic versions of everything they have. Imagine the boon to research that this would represent. Or even just personal enrichment. It would be an admittedly enormous task, but even doing things piece by piece would be beneficial. The arguments that this would discourage people from actually seeing the real thing is foolish. I am thrilled that I can go to webmuseum and look at Van Gogh's amazing paintings, but that just makes me want to see the real things even more. And, when I do get the chance to see them, I appreciate it that much more. All of this stuff should be readily available.

I am thrilled to see organizations fund projects like ibiblio.org - it is an excellent collection of knowledge. But, while browsing it yesterday, I couldn't help but think how great it would be if all of that information had accompanying metadata. And then the development of a distributed filesharing system that has places like ibiblio.org as permenant nodes. It would be truly great. It frustrates me that the technology is there, but it is just not happening yet. Hopefully I can help make it happen a bit faster. Incidentally, a filesharing system that uses servers like ibiblio.org as permenant nodes would be virtually impossible to stop - the government couldn't help but fund such an effort eventually. It would be a quantum leap in the usefulness of computers and the internet. Being able to do crossreferencing on the fly would be cool as well, but I could live if that were a later feature. All a project like this needs is a lot of people willing to spend a little time adding metadata to things. After a while, it will be easy to maintain. To some extend, computers would be able to generate some of the metadata for us, leaving us to fill in the blanks as we have time. A guy can dream. ;-)

I took a look at www.canonicaltomes.org - it is a very cool idea. It reminded me of a project I wanted to do 3 or 4 years ago, but I still have yet to get off the ground, or see anyone else really do. The project would be a central compilation of every public domain work that has been digitized. The goal would be to provide a nice, navigable, searchable interface to all the extremely useful research materials out there.

It would have to have the following features:

  • A Yahoo-like category interface for browsing casually
  • Each work would have extensive metadata, covering a standard like Dublin Core
  • A search interface that lets you perform all the basic searches, but also searches by metadata, so you can dynmically regroup works to your liking
  • A nice gdict-like desktop application that is a search gateway
  • Palm/WAP interface?
  • Cross-referencing
  • Text prettification, so you don't get stuck with ascii if you don't want to....some nice HTML or XML with stylesheets would be nice

I'd imagine that the technical side would be the easy part. It would basically just have to be a big database with a well thought-out schema. The hard part is definitely organizing the content, attaching the metadata, and finding it all. Also, it would be good to be mirrored. Eventually it should be able to act as part of a distributed filesharing system. It would be an invaluable research tool.

With things like GNUpedia, and other similar efforts to create free-license encyclopedias, it seems like a much more worthwhile effort is to work on something like I describe above. An encyclopedia is only useful after there is a collection of works to reference. This would probably go further to accomplish what RMS wanted to get done: there is already no copyright on this material, so no competing interest can do anything about it. Once there is a community around it, it can be extended in all sorts of directions.

Of course, I think that the Library of Congress should provide such a resource, but the person running it seems to disagree with me. Ah well. Maybe one day when I have free time I'll try and get something like this started.

Ankh: Thanks for the suggestions for query languages. After spending a bit of time looking into it, I have found the following query languages that I will want to review: XPath, XQuery, SQL, OQL, and WebQL. One potential problem that I will have to review is any licensing issues and patents. WebQL looks like part of a proprietary product. I want to make sure what I do is unencumbered by patents. It will be beneficial to look at it for ideas though, I'm sure. The XPath and XQuery systems look very interesting - the only thing that I would be concerned about if I chose one of them explicitly is that I want to treat the fact that the metadata is stored as XML as an implementation detail. I would like seamless transition to a method that can support resource forks in filesystems, as well as files that store their own metadata. But, I do want to use the XML DOM as how I deal with the data itself. It seems well-designed. Another thing to consider is the XML Fragment Interchange spec, which looks ideal for simple metadata exchange systems. I'll definitely need to do a lot of reading on this. I understand XML well, but there are a lot of related technologies that I need to familiarize myself with. I really want to do this correctly......leveraging as many standards as possible. I need a sane public interface, a well-defined metadata set, and a well-defined query language. I really want to be able to support things like subqueries and unions. The ultimate goal is to provide a solid foundation for things like a much more powerful peer to peer filesharing system, superior search engines, and virtual folders on your desktop. I think if I can get the 3 things specced out right, the rest will fall into place for anyone to pick up and build on top of. But before I can get there, I need to keep doing a lot of research. And hopefully the other people involved in this project will bring as much to the table as I think they will. We'll see how it goes.

19 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!