Older blog entries for Raphael (starting at number 12)

Awww... I just saw the new article about Advogato meta-issues. There is a lot of overlap with the article that I have been slowly preparing about moderation and certification. I was too slow... Now I will have to re-work my article again. :-(

Well, the replies to that article contain a lot of interesting comments that I can now integrate in my article. I will try to have it ready next week. In the meantime, I will be off the 'net for a couple of days (even my mail server will be down.) I hope that I will be able to catch up with the discussions and the diary entries when I come back.

Hmmm... I see an interesting new account: esr. Real or fake?

I decided to update my home page. I created a new template yesterday, then I decided to rewrite it using WML and a set of Makefiles so that I can maintain the site easily (hopefully). In the spirit of release early, release often (which does not really apply to web pages, but I did it anyway), I already converted some pages to the new template. Of course, this means that there is even less interesting content than before because many pages have not been converted yet and there are no links to them, but this should be fixed in a couple of days. I am not really happy with the quality of the images I created for the tabs. I should be able to do much better than that. But hey, it's a first draft...

I am cleaning up my Advogato certifications. I try to keep only the certifications for the developers I met in real life and those who are not certified by many others (and thus need my certification to stay at their level). This means that I removed my certification for alan, for example, because he did not really need it. There were also a couple of accounts that I rated as Apprentice or Journeyer when they joined (based on a best guess from the info available on their page) and who have been certified as Master by others since then. In these cases, I removed my certification because it was only there to allow these people to get in.

I read the Romeo&Juliet article. The article itself is moderately funny, but the comments are really depressing. We do not need OS flamewars here... (FYI, my girlfriend uses Windows and I don't blame her for that.) Also, I do not think that it is a good idea to use the certification system as a way to influence what gets posted or not. Certification (of people) and moderation (of articles) should be two separate things.

It looks like more and more diary entries (ab)use HTML tricks to achieve some special effects. Let's see how long it takes before some script kiddie posts an article containing a huge background picture or some JavaScript alert() bombs...

advogato is getting crowded. The People page is now at 263 Kb. Also, it becomes more and more difficult to look at the list of diary updates on the front page: if you do not check the front page often enough (e.g. if you are away for two days), then you miss some entries and you have no way to know what you have missed. But the list cannot be much longer than it is now, unless it is moved to a separate page. Hmmm.. That's an idea.

It looks like Advogato is approaching its first (minor) scalability hurdle. In a way, this is good news, because this shows that the site is successful. Maybe I should have a look at the code and see what I can do to help the site to grow further...

I just read raph's diary entry for Wednesday (29 Mar 2000) in which he talks about Slashdot being slow. Indeed, this is something that has changed significantly in the last year(s), even if this is not the only thing that has affected the quality of Slashdot. The obvious question is: why? If Slashdot has some problems, we should make sure that Advogato and other sites do not suffer from the same problems later.

I think that two of the main reasons for the long delays in posting articles (which are not "news" anymore) are volume and noise level. If Slashdot has more than 1,000 articles in the input queue and there is no tool to help the editors to pre-filter and prioritize the submissions, then it becomes very difficult for them to pick up the good stuff. It gets even worse if the queue is mostly FIFO: it may take a couple of hours or days before enough old garbage is thrown away and an interesting article gets noticed by the editors (especially if it does not have a subject line that would immediately draw the attention of the editors and have them read the article earlier). By the time it is posted, the article may be outdated.

If you run a news service that has to be attractive to both casual and regular visitors, then you have to make a selection and post less than 20 articles per day (otherwise they do not fit on a reasonable front page). If I refer to the figures published by CmdrTaco some time ago, this means that Slashdot has to keep less than 1% of the submissions. Not an easy task, especially if you do not have good tools to help you.

We should pay attention to these problems, because they can certainly hit any site that becomes very popular. A system that works well for a few thoudsand users may suffer from scalability problems when the number of users increases by two orders of magnitude. I am not sure about how much content could be automatically pre-filtered. The naive approach of looking for keywords in the submissions would probably not work very well, because most people who submit stories think that their article is the most important (and besides, some weenies would certainly try to fool the filters on purpose). Filtering on the sender's name (Slashdot karma or Advogato trust metric) has other problems too, because sometimes the best contributions come from newcomers. So we need something more than that, perhaps a combination of several filters.

I'll think about it...
Note to self: I still haven't finished and posted my article on moderation, and I'm already drifting on another topic. Hmm...

On Saturday, I got my photos from GUADEC. Not too bad. I have been off the net since then and I haven't had the time to scan them and post them yet, but I will do that soon...

Here are more random tidbits that I forgot to add in my GUADEC summary:

  • While working on some icons for gnumeric, Tigert had set his desktop background to display an ocean image (I think it was this one or something similar). On top of that, there was xfishtank using the shape extension. Well, I could not believe how many people came and asked "hey, you have a cool background, what program did you use to do that?"
  • Wacom has probably sold a few more tablets by now, after some people saw how easy it is to use the GIMP with these nice tools (and Tigert is very good at demonstrating that). The support for pressure and tilt really makes a difference.
  • The presentation of Pango and the plans for GTK+ 1.4 were very interesting. As a kind of joke, I asked Owen if GTK+ would support UNICODE for the name of its internal signals. The answer was no, of course. Think about how you would read some source code containing signal names in Japanese, Greek, Arabic and so on... Wouldn't that be fun? (err... no!)
  • Nautilus can be extended to support more file formats (for creating thumbnails or viewing some files inside the Nautilus shell). I think it uses bonobo components for that. It would be very cool if the GIMP plug-ins could be used in Nautilus. Many plug-ins have been written to support exotic file formats and it would be nice to be able to use them in other GNOME applications. But bonobizing the plug-ins should not be considered until after the GIMP 1.2 release.
  • Some people use gmc and like it. Really!
  • Considering how much some hackers were drinking at the HelixCode party, it is good that none of them were drowned in the Seine (the party was on a boat). Or maybe the corpses haven't been found yet...
Hmm... I will probably add more later, when I remember...

It's about time for me to write a short report on the GUADEC... I'm probably the last one to do it.

I am glad that I decided to go to the GUADEC even if that was a last-minute decision, because otherwise I would have missed a great event and many opportunities to talk to some very cool guys. The discussions in the corridors, in the hacking room and around some restaurant tables were as important as the scheduled talks (if not more). I talked to some hackers that I had met before (Sven and Mitch) and many others that I met for the first time: Tigert (he got many requests for creating new icons and did some amazing things), Lauri Alanko (he reminded me of my Quake days), Raph Levien (he was very busy but he had some nice examples of his new printing stuff), Tim Janik and Owen Taylor (the GTK+ guys), Keith Packard, Ville Pätsi, David Monniaux and many others who will hopefully forgive me for not mentioning their names here.

Some random notes about what happened while I was there:

  • I had my camera with me but since it is not a digital one, it will take a few days before I can scan the photos and put them on some web site. I wonder if Larry Ewing will be happy to see a picture of himself sleeping in the hacking room. I also wonder what others will think of Miguel de Icaza after seeing his various facial expressions. I hope that the photos will be OK - the film is not developed yet.
  • On Thursday evening, we went to a small restaurant for dinner. There was an interesting discussion between Owen Taylor and Keith Packard about the way X Window handles scrolling and backing store (Owen: it sucks; Keith: it's not my fault, and you should not use it like that anyway).
  • On Friday and Saturday morning, the agenda said that the hacking room would be open from 8am until noon. I was not really surprised to see that the room was almost empty in the early morning (people started to come after 10am). Around noon, the room was filled with hackers glued to their laptops.
  • It was funny to see tigert and drc arguing about who should get the pen and/or the mouse for drawing some icons and other stuff. There were not enough computers in the room.
  • While tigert was using the GIMP, I wrote down all the bugs and other undocumented features that we discovered. I will try to fix some of them before 1.2 is released.
  • I liked the presentation of Evolution. I even won a small plush bonobo for guessing that Matt Loper (one of the authors) was using Outlook as his e-mail client. Despite the boos and bahs that came from the audience, I think that it is important to have a groupware tool that can be used easily by non-hackers. There is still some work to be done, but Evolution seems to be a good answer to that problem.
  • The presentation of Nautilus was also great. If they manage to include all of the features that were demoed as well as the ones that are planned into something that is stable and reasonably small and fast, then GNOME will have a very good file manager. Andy surprised some people with some quick hacks (i.e. being able to drag some colors into the window resulting in a nice gradient for the background) but there were many more important things that do not always look cool but are helpful for the user. For example, I liked the zoom levels giving more and more information about the files (similar to what EPOC does on the Psion 5). I also like the fact that the icons are trying to show (a part of) the contents of the files, and that the users can customize many display options. The different modes were also a good idea, so that the novice users are not lost in the menus and options. Oh, and of course the "view as music" option was very cool.
  • After seeing the demos, it looks like the GNOME applications such as dia and gnumeric have evolved a lot since I had a look at them. I think that I will grab a recent version and install them on my PC.
  • The HelixCode party on Saturday night was nice, except that the music was a bit too loud (for some reason, many hackers prefered to talk outside instead of going on the dance floor). This is also the first time that I go to a party in which the bar presents a selection of appetizers, drinks, CD-ROMs and plush bonobos. I didn't try to eat the CDs.

Well, what else can I say, except that I am looking forward to GUADEC 2...

Today is the last day of the GUADEC meeting. I came to the conference room in the morning. Looks like everybody was asleep because the room was almost empty (except for Keith Packard who seems to be always among the first to wake up). I saw that Sven is trying to recover from his late night activities. Lots of nice things here. Nice demos of Evolution and Nautilus. More about that when I come back on Monday...

It is a bit late, but I just decided that I would go to the GUADEC tomorrow. So I quickly called some hotels to find a room, tried to find some train tickets to go to Paris, and registered for the workshops and talks. I'm not sure that I will be able to stay until the Helixcode party, but I hope that I will have a good time there and lots of interesting talks with other developers.

Darn! Almost one month since I said that I would post a new story "soon". I have collected many ideas about moderation and content rating, but I still have to organize them in a structured document that is suitable for posting here. That would be a good complement to the discussion about content syndication. Hmm... I'll have to re-shuffle some things on my TODO list... So many things to do, so little time...

3 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!