Present Perfect

This week, after receiving a Google API key from a kind mecenas (which probably violates the EULA and thus she shall go unnamed) I implemented another one of my wishlist features for moap: looking up your project's home page using Google or Yahoo.

This is part of my preparation for my GUADEC talk on Practical Project Maintenance. Especially in the early days of a project, you need to make sure people can find your project easily, not only on the name of your project, but also on keywords related to what your project does. People are out there googling for keywords that should lead to your particular project if it's any good.

Since Yahoo's web API is completely open, I'm defaulting to yahoo's search for now. I must say I was always down on Yahoo like everyone else. But Google has gone from a search engine to a an ad company and made their API less useful, while Yahoo - who made the transition of search -> ads a long time before - actually trumps Google on accessible API for the search part.

Anyways, a curious thing has happened. My project page used to be hit number 5 on Google for the keyword "moap". After reading some page recently on Google optimization, telling me I should have as many key words as possible in the first paragraph and such, I expanded the text on the start page a little. I did not yet add keywords - as I need to figure out how to do that with Trac. I also added a logo and a favourite icon - which is unrelated to the optimization I want to do but I wanted to make it look a little nicer than the default page.

Now, today I run the search again, and I'm not even in the top 100 anymore. Did Google not like my logo ? Did Google know I was going to shuffle a complaint in this blog post before I even wrote it ? I don't know ! I don't understand this whole search engine thing.

Luckily, Yahoo has the project page at hit number 3. Another reason to like Yahoo...

I'm going to do another release of moap soon.

Feel free to leave tips on optimizing my entry page for search results !

This week, after receiving a Google API key from a kind mecenas (which probably violates the EULA and thus she shall go unnamed) I implemented another one of my wishlist...

I want to like beagle, I really do.  I want to use it, and I want it to work for me.  Some of the smartest people in free software have hacked on it, and I can only say good stuff about them.

So where's the disconnect ? I have been going crazy for the last week trying to figure out why my home machine is consuming 100% CPU when it should be idling because I'm not doing anything on it.  I can tell it's doing stuff because the fans are blowing like crazy.  When I move my mouse to wake up the screens and X, I can see my CPU monitor suddenly going from 50% (on this HT machine) back to 0%.

It wazzz in my PC.  Eatsing my CPU.

But I couldn't figure out what it was, as it went away when I checked, and I had no second computer in the house to log into the machine.

Tonight I brought my laptop back home, and it's poor old beagled.  It's probably been Doing Stuff for the last week, and it has had 100% of one hyperthread for more than 75% of the time to it in.

But it is Not Yet Done.

And it goes skulking off into the corner whenever the X session wakes up - typically the sort of nifty hack Joe or Jon would have come up with.

Who can tell me What It Is Doing ?

Here's a textdump of beagle-status output if it helps:

Every 5.0s: beagle-info --status                        Fri May 18 22:15:40 2007
Scheduler:
Count: 1755041513
Status: Finding next task to execute

Pending Tasks:
1 Delayed 0 (5/18/2007 10:15:41 PM)
File Crawler

2 Maintenance 100 (5/18/2007 8:55:15 AM)
Final Flush for FileSystemIndex

3 Maintenance 0 (5/18/2007 8:55:18 AM)
Optimize FileSystemIndex

Any tips on debugging beagle - from a user perspective - are welcome too.

I want to like beagle, I really do.  I want to use it, and I want it to work for me.  Some of the smartest people in free software have...

So, my GUADEC proposal got accepted, and now I'm left with the stress of actually gathering enough content for my talk on Practical Project Maintenance.  I'm trying to find my random collection of notes I have collected ideas in for this talk.  As Andy puts it, "You are the perfect target for Beagle".  He's probably right, but somehow I always end up turning off Beagle after another instance of beagled using 100% CPU on something.  It is one of those cool technologies that you actually need to put effort into to learn enough of it so you can troubleshoot yourself out of a corner - like I had to do with SELinux for example.

But I digress.  One of the topics I wanted to bring up in the talk is how to get your project known.  As a good engineer, I want to measure before I change and affect, so I wanted to write a simple program to figure out where some of my projects turn up in Google if you search for relevant keywords.

Sadly, it turns out that the Google Web SOAP API (for which there is a python module) is deprecated, and no API keys are being handed out anymore.  The only interface left is the AJAX API (which I guess you can only sensibly use from a web page in the first place).  If I understand correctly Google's EULA also does not allow me to screenscrape.  Doing a simple wget confirms this - though it's easy to fool around with the user agent of course.

I'm no expert on business but I'm going to guess that Google closing this SOAP API is part of the move from search as their core business to advertisement.

I'm going to play around with various alternatives - probably no simple program using BeautifulSoup to screenscrape (I'm saying "no" because my lawyer tells me to), using this amusing EvilAPI website, or using Yahoo Search API instead.  I want to be practical, so anything that gets the job done.

If any of you have a Google API key left that they're not using, feel free to hook me up.  I promise I will Do No Evil with it...

So, my GUADEC proposal got accepted, and now I'm left with the stress of actually gathering enough content for my talk on Practical Project Maintenance.  I'm trying to find my...

A new window manager written in Haskell.  And here I thought sawfish was hard to hack on for picking a relatively lesser-known language to be implemented in.

I don't know whether to be in awe or in tears.  Let's read the home page.

Number of times use is used as a verb: 2

Number of times utilize is being utilized: 3

Ok, it failed the utility test - I'm in tears.

A new window manager written in Haskell.  And here I thought sawfish was hard to hack on for picking a relatively lesser-known language to be implemented in. I don't know...

A new moap release is out ! This release adds support for Darcs to the moap changelog prepare/commit/diff and moap ignore commands, and generation of iCal data based on the releases in DOAP files.

I've been wanting to add iCal and RSS feed generation based on this for a long time, so I can automatically update feeds when I do a release of any piece of software. RSS feeds are an excellent way to keep updated about project releases. And I like seeing the GNOME iCal spread all over my Evolution calendar.

Now, with Dates getting better on my N800 with every release Opened Hand is rocking out on, I wanted to get the GStreamer releases on there too. It was pretty hard to figure out why Evolution was refusing my ics files when subscribed to on the web. Importing them worked fine. In the end, it turned out I was missing a UID field for each VENTRY. And evolution-data-server-1.8 was telling me so ... on its stdout.

(Sure, it is nice for Evo to be all component-based and stuff, but if the end result is that your code is writing to stdout without anything being connected to stdout - because you are some background server process - then you need a better way to get warnings to the main application. Now the components in eds just litter silly output to stdout, showing that nobody ever sees that stuff because nobody bothers keeping it clean. Like I really need to know what memory pointer the currently selected calendar uses ? Enough ranting though.)

Anyway, now I have lovely Dates showing lovely GStreamer releases. Maybe now I should actually make some !

HELP

I am looking for people interested in making me a nice image or icon. Hey, it works for the Opened Hand guys, why can't I use the intarweb to shake down for icons ?

I am also looking for people to package moap for their favourite distro. Come on Gentoo hackers - be agile ! Come on Ubuntu people - care about maintenance !

Shed bike

This is your opportunity to bikeshed ! I have a quick-and-dirty implementation of an RSS feed feature ready to get commited using templating with Cheetah. Users will want to customize their RSS feeds so allowing them to template the feed makes a lot of sense. I am also planning to get the release mail be templated, and do atom support ,and possibly more stuff.

Cheetah was easy to get started with, in about an hour I had the meat of the feature programmed. I am also going to look at Nevow, just because I am ridiculously reverent of anything Twisted - even though it seems Nevow is more suited for XML-like output.

Here's your chance to bike shed ! Which (Python) template system should I use and why ? Bear in mind that I want to be able to generate at least plain text, HTML, RSS, and Atom. Answers on a post card or in the comments !

A new moap release is out ! This release adds support for Darcs to the moap changelog prepare/commit/diff and moap ignore commands, and generation of iCal data based on the...