Random Developments

Ok, I know I complained about not receiving many copies of previous viruses, but this time I think I already got my share. One of my mailboxes is receiving a few hundred messages a day (all without the payload, removed by the mail server, but still annoying). And that is not counting the messages saying that I sent an infected message that was blocked somewhere.

By the way, if you make anti-virus software that blocks e-mail messages, do not return the messages you block. The chance of the user who actually sent it receiving the message back, nowadays, is exactly zero.

I guess I’m still getting off somewhat easy. Some people I know whose usernames are in the main list used by the virus are receiving in excess of 2,000 messages an hour.

Come on, this is getting ridiculous. 225 messages were blocked as spam yesterday in one of my e-mail accounts, and not one of them was a false positive. A few spams (two or three) were not blocked, but almost all of them were. By contrast, that same mailbox received 146 non-spam messages from Friday night until now. That’s 225 spams in 24 hours, and 146 non-spams in 48 hours.

The good news: Bayesian filters still work, and they are responsible for the nearly flawless spam-blocking I enjoy. The bad news: there is more spam coming in each day. Weird as it may sound, I do hope Bill Gates is right.

The second lander and rover are on the ground, and they’re sending amazing pictures. Check them out at the JPL site.

Also, it looks like Spirit might recover from its current problems, so we can actually have two working rovers on Mars at the same time. And let’s hope that Opportunity doesn’t get hit by the same problems as Spirit.

It turns out that there is one major flaw with gatewaying (is this a real word?) RSS feeds to an e-mail app using POP3: if you read more that a handful of feeds, creating filters to sort all of them into folders gets boring real quickly.

The way I implemented the gateway, the generated e-mail has a header line (X-Rsstomail-Channel:) with the URL of the feed, so you can easily filter on it. However, this still leaves you with having to create the folder to receive the messages, and somehow creating the filter (easy on Becky, Alt+drag-and-drop; very hard on Thunderbird, it’s actually easier to write a small Perl script to create the filters than to do it by hand).

The correct way, of course, would be not to use POP3, but IMAP4, and automatically sort the messages into folders on the server side. It turns out that there is a RSS-to-IMAP4 service out there, the Blogstreet Info Aggregator; however, either I did something wrong, or they don’t do the obvious thing (sorting messages into folders), they just dump everything in your inbox. Why using IMAP, then? Beats me.

I didn’t work on this project as much as I wanted to lately, so it is not much different from last time. I will still go the way of the POP3 server at first, simply because it is easier to implement than IMAP4, but I expect to have IMAP4 at some point in the future as well.

No, I won’t talk about Dean.

A long time ago, I had a cool idea (at least I thought it was cool) and started writing a sci-fi story. I wrote a few paragraphs and promptly forgot about it. The story was called “Election Day”, hence the title of this entry.

Now, I found it again, but I really don’t have the time (or the inspiration) to continue it. So, I’ll post it as-is, so that at least maybe someone else will see it and tell me what they think of it.

If you want to read it, click on the link below and check it out. Bear in mind that it is incomplete and very unpolished (that is, it could use an editor). If you do read it, leave a comment telling me what you think of it.

Continue Reading »

Nokia is giving out some quite interesting advice on the manuals for their mobile phones… “If you use your mobile phone in a fast-food restaurant, keep your voice down so you won’t annoy people. And order a salad … Quit smoking. Drink less.” They also tell you to turn your phone off on weekends and forget about work, so you will live longer.

Yahoo! News has the story. Nokia has, of course, the user guide for the phone mentioned in the story, but the text is not there.

This weblog is changing servers. The URL and everything else will remain the same, so you shouldn’t even notice it.

If you are reading this post, you are seeing the new server.

The project I’m working on, which I mentioned previously, is a RSS-to-POP3 gateway. It’s a personal project, of the type I saw someone describing as a “itch-scratching” project. That is, I had something that was bothering me, and I decided to do something about it.

Looking around on the net I saw a few other projects doing approximately the same thing, but none did exactly what I wanted to do. I want to join the best things about server-side aggregators with the ease of use of regular e-mail software. So, my idea is to have a piece of software that will periodically go through a list of feeds, retrieve the updates, and store the data (the posts) in a database; that’s your basic server-side aggregator, minus the interface. And the other piece of software, which would provide a sort of interface, would be a POP3 daemon which would read the database looking for posts the users hasn’t seen yet and return them as e-mail messages.

This way, you can create a single account on your e-mail software which will retrieve all the new posts in all the feeds you read, and you can (if you want to) filter them into folders according to information added to the e-mail header by the gateway. Good things about this:

• since the server-side aggregator is always updating the feeds, you won’t miss posts if you stop reading the feeds for a long time
• if several people use the same installation and read the same feeds, the data is only retrieved once per user (that’s also a very good thing for feed publishers)
• it’s easy to save posts you want to keep (or read later)
• you can read the posts from your favourite e-mail software
• and possibly others

This is where I am so far: I have the software to go through the list of feeds and update a local database with the new posts, and I also have code to go through the database and generate a mailbox file with all the new posts for a given user. That’s enough for a “proof of concept”, allowing me to see the result and to ensure that the core code is working (although it still doesn’t like some feeds out there). What’s missing: basically, the POP3 daemon and an interface where users can register and add feeds to their account. Also, I need to make the code deal with error conditions in a more elegant way than what I have now. And the code to control what the user has already read needs some changes to work well with the kind of thing POP3 clients can do.

I’ll release the code when it is presentable; if anyone is interested, drop me a line or add a comment to this entry.

And it has sent pictures! The one you see on the right is one of them. NASA has a small gallery with all the pictures the lander has sent so far. I believe we’ll see many more of them over the next weeks. Full information on both missions can be found at the JPL Mars site.

After what looks like a complete loss of the British Beagle 2 lander late last month, we get a second chance later tonight, with the first of the two NASA rovers, Spirit, scheduled to land at 11:35pm ET (that’s 04:35 GMT tomorrow, January 04). NASA has a step-by-step guide describing the (scary) entry process, and NASA TV will have live coverage that can be watched online.

Also, in case Spirit does not make it to the surface intact, there will be another chance with Opportunity, on January 24. Let’s hope at least one of them is able to replicate Pathfinder’s success story.

First of all, happy new year to anyone who should happen to be reading this while the year is still new
For a project I’m starting to work on (more on this later), I spent some time today reading on the several RSS flavors out there (there are a few, and that’s not even including Atom, which I shall deal with later). However, it’s amazing how many different sites “teach” you how to parse RSS files by using the Perl module XML::RSS. So that’s what I set out to do, even though at first I was planning on using either Python or Java to do that.

First step: actually getting XML:RSS on my machine. Sounds easy, right? You do a

then you sit back and relax. Well, not so easy. This is a Windows machine. It has Cygwin, which does make things easier, and which allowed me to begin by doing just what I wrote above and have most things working just as I wanted them to, but XML:RSS needs a set of other modules, and it turned out that LWP::Simple wouldn’t install (tar complained of an invalid utime in a file). Downloading the module by hand and doing that old “perl Makefile.PL / make / make test / make install” dance should make the trick, but then I had to go after the pre-requisites by hand as well.

After spending a few minutes trying to convince the CPAN module to retrieve and install the URI module, I finally got it and was able to run my first RSS parser:

Sure, not very elegant or fail-safe, but it’s a beginning. I’ll keep you posted on how things progress from there.