dBera released Beagle 0.3.2 today. It fixes a handful of bugs found in the previous 0.3.x releases and adds a few new features as well. Among them are using HAL for detecting whether the system is on battery, support for gzipped and bzip2′d man pages and a system-wide man page crawler, a new Empathy backend, and indexing file sizes of on-disk files, email attachments, and files inside archives. As usual, get it from the Beagle home page.
I’ve been negligent in my blogging duties. Two weeks ago I released Beagle 0.3.0 and Beagle-Xesam 0.1. We followed it up quickly with a 0.3.1 bug fix release on Monday, dBera’s first ever release as co-maintainer!
The 0.3 series is a new major release, and it has a bunch of great new stuff:
- Date queries — the infrastructure for this has been around forever, but we can now do date ranges using keywords in your search.
- All new Thunderbird backend — this completely revamped backend uses a Thunderbird extension to extract email. While you do have to be running Thunderbird to index your mail, the upside is that we have a foolproof way to index Thunderbird email with no memory impact.
- All new Firefox and Epiphany extensions — Greatly improves the code quality, and includes some new features like indexing of pages on demand.
- Network searches — You can now search across Beagle daemons and access your remote files. Other systems can be found automatically using Avahi. This is an experimental feature in 0.3.0, and will continue to prove over time.
- Web user interface — Built on top of the networking code, you can now access your files through a web browser in addition to the existing rich GUI options. Also experimental.
- Nautilus metadata backend — Not a super new whizbang feature, but we now index Nautilus emblems and notes. But the real improvement here is in our infrastructure. The Nautilus metadata backend is our first to index metadata stored completely separately from the data itself. This makes integration with, for example, tagging systems much easier.
- Improved text cache implementation — A complaint of many people in the past is the amount of disk space used by the text cache, a data store that makes retrieval of snippets fast from complex documents. We’ve reimplemented it and on average it uses about half the disk space it previously used.
- Plenty of other fixes, optimizations, and additions. Read the release announcement for details.
Meanwhile, Beagle-Xesam is our Xesam adaptor, which Arun Raghavan wrote as part of Beagle’s participation in the Google Summer of Code program this past summer. Developing as a separate project allowed us to maintain stability of the core search code, learning how to implement the Xesam spec, and getting something up an running sooner rather than later.
Speaking of the Summer of Code program, it’s been very good to us. In addition to the Xesam adaptor, the new Thunderbird backend and the new Firefox and Epiphany extensions were all Summer of Code projects from this year, and they were all quickly integrated into the tree. Also, the network search support is based on code from two SoC projects from last year which finally made it into the tree thanks to Lukas’s hard work.
A couple people didn’t quite understand my parody post referring to an off-the-cuff remark by Federico nine years ago. I happened to come across it just mindlessly browsing it the other day, and thought it would be funny to suggest that the GNOME and Novell folk have been cooking up this idea to sell out the Linux community since 1998.
The other posts actually do hit on the point that this community split is dumb and that we should give credit to Microsoft where it’s truly due. And sure, I agree with all of this.
But in reading those posts, one is from a new member of the GNOME community and the other is from a KDE guy who reads Planet GNOME. I know what you’re thinking: why aren’t these people digging up and memorizing my life story? But some people are just lazy these days. So I added an “about me” page to my blog, to hopefully give my bad jokes more context in the future.
Interesting article in the New York Times today. Junk food costs only $1.76 per 1000 kcal, whereas healthy food costs $18.16. This is in part because junk food is much heavier in sugars and other carbohydrates. But as Michael Pollen wrote about in April, a large part of it is because of the massive subsidies for a specific few crops in the American farm bill. The end result? A package of carrots, an unprocessed food and rather straightforward crop, costs significantly more than a package of Twinkies, a massively produced mixture of food (mostly corn)-dervied chemicals.
GNOME CO-FOUNDER AND NOVELL EMPLOYEE SAYS EXCEL IS “CERTAINLY A NICE PIECE OF SOFTWARE.”
When will those Novell guys stop suckling at the Microsoft teat? It makes me sick.
Update: Sigh. This is why we can’t have nice things.
Before hockey, Jacob went to pick me up a burrito from El Pelon and found it boarded up. Apparently there was a fire there this morning. I love that place — when Ximian was in the Landmark Center we would eat there four times a week. Jacob led in the burrito club standings for a long time, and I wasn’t far behind. Here’s hoping they open again soon.
From Valleywag, my favorite tech gossip rag:
Take poor Joe Shaw from UC-Davis for example. “I bought some ‘enlargement’ pills online,” he writes in the group’s forum. “The next thing you know, boom, it shows up on my news-feed for all my friends to see. The worst part about it: they were breast-enlargement pills!” What a pickle!
I assure you this is not me I have never even been to Davis let alone enrolled at UC there please believe me my breasts don’t need enlarging
The following companies have begun advertising Christmas stuff to me before Thanksgiving, thereby earning themselves a blacklist from my patronage:
- Lowe’s
- Best Buy
- Garmin
- Dunkin’ Donuts
On the other hand, Nordstrom’s is awesome.
My second day at ITA has been a little bit of a challenge so far:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
21136 root 25 0 1168g 2.9g 2.9g R 38.6 86.6 14:08.40 yum
It’s a good thing this is a 64-bit machine, or else we would have run out of memory a long time ago, before yum mapped 1 terabyte of memory and kept 2.9 gigabytes of it resident.
(That’s about 5 orders of magnitude more memory than it needs to map in, and 2 or 3 orders of magnitude less than it needs to actually use.)
But package upgrades aside, things have been fun so far. Stay tuned.
Dear Lazyweb,
One of the ways Microsoft threatens the Linux community is by telling customers they’re liable for patent infringement if they run it in their data centers, on workstations, etc. Can someone explain to me how and why customers are liable for patent infringement when another party provided them with the infringing product? It would seem to me that since money changed hands from the customer to the entity selling the product, the liability exists solely with the selling entity. Does this mean if I patent a physical widget and a company infringes on that patent while building a dingus and sells that dingus to a million customers then I can sue those customers for violating my patent? Or is software (or more specifically, open source software through its licenses) somehow special?
Please only email me if you know and cite sources. Do not email me about Microsoft, or whether they have a leg to stand on. Shut up, I don’t care about your opinion.
Update: Adam Williamson is my hero. He says:
http://www.law.cornell.edu/patent/35uscs271.html
U.S. Patent Act
..Part III. Patents and Protection of Patent Rights
….Chapt. 28. Infringement of Patents
“Except as otherwise provided in this title [35 USCS Sects. 1 et seq.],
whoever without authority makes, uses or sells any patented invention,
within the United States during the term of the patent therefor,
infringes the patent.”
Note: “uses” is included in that paragraph.
