Month: September 2005

There’s been a lot of noise over in Livejournal by the user base complaining about Google’s new Blogsearch indexing tool. Most users are reacting by changing their entire journal over to ‘friends only’, and casting aspersions upon ‘the evil google search engine’.
This post is primarily aimed at that audience.
Folks, Livejournal is a blogging tool. It’s purpose is to post your entries onto the public net for you. What it can and cannot do is made pretty clear. People seem to have an assumption that once you post something on the net, somehow you have total control over it, and can control what happens to it, and who can access it. This is a notion people should get over.
Livejournal automatically creates RSS feeds for all your journals. Those RSS feeds are publically available, and have been forever. There is no ‘protocol’ on the net that restricts who can read an RSS feed, therefore ‘robots.txt’ and the like will not affect it. Google’s blogsearch system reads RSS feeds and indexes them, just as any other RSS news aggregator on the planet can do.
Google is doing nothing wrong, and not breaking any rules, and Livejournal is doing nothing wrong, though they should have the option of providing RSS feeds or not. My point here is casting aspersions upon Google is inappropriate. Having a realistic view of what a blog host, and what blogging software does is more appropriate.
Livejournal DOES give a small bone to the community though. You can cut down the amount of public information fed into the RSS feeds. See this helpdesk ticket for more information.
My point here is twofold.

1. Yelling at Google for doing what they do, index public information, is a foolish endeavour. They are not doing anything wrong, they are indexing information that has been made public.
2. Putting your faith, thoughts, and information on someone elses server means you are automatically signing on to their practices and policies. You’ve already partially lost control of your information, don’t be surprised when something happens you didn’t expect.

This post was posted on my own blog installation, on my own servers, in software I control. I fully recognize that by hitting [Save], I lose the possibility of controlling who and what has access to this content, and where it goes. I accept that even in my own environment. Don’t expect more in someone elses system.
Such is the nature of the Net.

Well, it’s been about two weeks since I got an iPod nano for my birthday, courtesy of my sister. A couple observations, and I’ll let this whole thing fade off into history…

• It’s SMALL!
  No, really, it’s small. It’s easy to lose. I have a nice Ogio laptop backpack with an ‘mp3 player pouch’ on the top of it. It fits nicely into that, which gives me a good place to ‘always keep it’.

• It goes FOREVER
  I have not had an opportunity to even come close to killing the battery in it. I’m not a power-listener by any stretch, but noting “Hm, it’s been 3-4 days. Maybe I should plug it in again” is a nice feeling.

• The headphones are ‘eh’
  The stock white headphones are nothing to write home about. I have a set of Shure E2C earbuds that are mighty sexy, but take longer to just ‘pop in’. And because they’re sound insulating, its hard to just put them in and leave them in. Durned checkout drones are hard to understand when they say “For here or to go?”.

• It works fine with Linux
  Thanks to the wonder of GTKpod, syncing from the laptop is a breeze.

• Yes, the screen scratches easily
  Do not put these things in your pocket with your keys. The screen will scratch up terribly, as reported in The Register. Fortunately the carry spot in my bag has a felt lining, but when I’m carrying in my pocket, I choose a spot that has nothing else in it.

• 2 gig is not enough
  I’m a disk space hog. 2 gig is not quite enough for me. I think the 4 gig version would be okay, but I could see myself doing the same thing there. “Not enough!” I see the attraction to a 40gig ipod.

Final thoughts? A mighty cool piece of equipment. Thanks sis!

While finally getting into some more serious coding, I had this thought. THings you don’t hear very often anymore, but were pretty common back in the day.
“A snack? Sure, let me just start this compile running, it’ll hopefully finish by the time we get back.”
Compiles I’m running now usually take less than 6 seconds, sometimes a larger build and deploy and initialize may run for 10-15 seconds.
Ah the good old… days?

… a successful off-site backup.
Now, to flesh out that commentary – most of our services are colocated at various Undisclosed Locations around the US (usually one door down from Dick Cheney). This means it’s sort of difficult to trot over to the machine, pop in a backup tape, make a copy, and toss the tape in the storage vault. Sure, colo facilities provide this capability, but it tends to be painful to work with, not to mention costing extra money.
With broadband now as widely available as sand, it’s possible to take on the concept really put forward 10+ years ago. “Why not back things up over the net?” When I was first approached with the idea (oh, 1995-ish), network connectivity was just edging out of dialup, and if you were SUPER lucky, you could have a T1 line, but it would most likely run you $800 a month. That sort of connectivity at home? Not likely.
Nowadays folks have DSL and Cable modems that have huge amounts of bandwidth. We’re no different here at Chez Geek, happily slurping at the nozzle from Comcast.
Last night I finished noodling a 4-5 line backup script that synchronizes various Important Directories [tm, reg us pat off] on our server (things like, oh, say, /home), and a spare 160gig USB drive I picked up at Microcenter for $80.
What does this have to do with Linux, I hear you cry? Because all of this was done with free software, and it did it fast and efficiently. Using the rsync utility, it takes one command to synchronize one set of dirs on one machine with another. It even compresses and optimizes the transfers, only copying those files that need to be updated.
So we have free software, cheap bandwidth, cheap disk space, and cheap hardware. It’s Geek Nirvana.
For the detail-oriented folks in the crowd, here’s the end result:

sent 3756414 bytes received 7763330071 bytes 530665.56 bytes/sec
total size is 13301911159 speedup is 1.71

That’s some nice throughput.