sjh - mountain biking running linux vegan geek spice - mtb / vegan / running / linux / canberra / cycling / etc

Steven Hanley hackergotchi picture Steven
Hanley

About

email: sjh@svana.org

web: https://svana.org/sjh
twitter: https://twitter.com/sjhmtb
instagram: https://instagram.com/sjhmtb

Other online diaries:

Aaron Broughton,
Andrew Pollock,
Anthony Towns,
Chris Yeoh,
Martijn van Oosterhout,
Michael Davies,
Michael Still,
Tony Breeds,

Links:

Linux Weekly News,
XKCD,
Girl Genius,
Planet Linux Australia,
Bilbys,
CORC,

Canberra Weather: forecast, radar.

Subscribe: rss, rss2.0, atom

March
Mon Tue Wed Thu Fri Sat Sun
       
19 20 21 22 23 24
25 26 27 28 29 30 31

2024
Months
Mar
Apr May Jun
Jul Aug Sep
Oct Nov Dec

Categories:

Archive by month:

Wed, 24 May 2006

Made a few hackergotchis - 16:34
So I noticed a while back that Mikal had stopped using the crown of flowers hackergotchi I created for him Mikal Crown of Flowers Hackergotchi, this got me thinking, maybe I should make a few more hackergotchis, after all Michael Davies did suggest it would be good to have at least most of the LA committee gotchiid up.

So I made one for Rusty the other day, however he may not like it seen in the wild so I wont put it here. Then I made one for Stewart Stewart enjoying a beer at Tridge's which means two more of the LA type people may have one (Stewart said he may as well use this as it is as good as any).

I have a few photos of jk, cyeoh, Hugh and even a fairly good one of Alli (one I hope she would not kill me for suggesting the use of).

I was wondering who else to do (rather than make the hackergotchis of the above mentioned people) and it occurred to me I have a lot of photos of Michael (Crash) Carden available from various cycling events and such. So I grabbed three photos I had taken that could make reasonable hackergotchis and created them.

Crash about to enjoy a beer at Polaris 05 Crash riding the Corn Trail Crash looking startled in New Zealand

He likes the first one the most (probably because the images with helmets make him look like my hackergotchi or something) so hopefully Michael Davies will use that one soon.

I added alt tags to all the img tags, though I do not know if they will show up on a mouseover or not. I had someone point out to me I need to use the title tag in the img to get the string of text that comes up on mouseover, this seems to be duplicating some of the alt purpose to some extent, ahh well.

[/comp/internet] link

Mon, 13 Feb 2006

Connected once more. - 20:18
Ahhhh, that sweet Internet connection is once more available at home. Two days before leaving for Victoria at the start of my recent holiday my Internet connection at home failed. I had no time to deal with it and there seemed nothing obvious wrong. The pppoe logs however showed PADI packets being sent out and nothing coming back.

My guess eventually was that the ADSL modem had died, however I could not do anything much about it until I got home. Today I purchased a new ADSL modem (D-LINK DSL-502T) and what do you know, everything works again now.

[/comp/internet] link

Sun, 15 Jan 2006

The slashdot effect vs digg - 22:52
On Friday Kottke wrote a good analysis of the slashdot effect as it applied to a recent hit on his own website. Comparing it with the hit from a website called Digg (of which I have never heard).

This is somewhat reminiscent of the slashdot effect t-shirt the Canberra Ozlabs crew made up a few years ago (I was unable to find a photo of any of them wearing it after a quick search), however it is interesting some the conclusions Jason draws here and some of the numbers from his own logs.

I can see what he means about the editorial backing behind slashdot keeping it relevant and interesting to many people, heck Rob Malda weighs in on the Digg style of voting for stories in the slashdot FAQ. Personally I stopped reading slashdot regularly sometime in late 1998 as I decided I did not personally like the editorial line, accuracy (or lack there of) and for some items lag in getting on there. However I can see why is popular still to such an extent, and there is occasionally some interesting comments on there. Also I will read interviews or the odd other item if someone points me toward it.

However the bit Jason points out about stories lasting there longer, the editorial control rather than voting or similar, and a few other points definitely help show some of the reasoning behind slashdot being as long lived as it is. The analysis is worth a read IMO, though it is interesting to note that Rob Malda dealt with a lot of the conclusions in the FAQ back in the year 2000 or so already. I am happy enough with LWN for my geek news though.

[/comp/internet] link

Wed, 04 Jan 2006

First non Latin alphabet google logo. - 19:24
I must say this Louis Braille google logo is fricking neat I think. I could not remember any time previously they have used a non Latin Alphabet based logo, so I had a look to be sure and my suspicion appears to be correct. Ignoring a few April fools day (logo) logos (logo) that could be interpreted as something completely different anyway.

Anyway, rather cool I think.

[/comp/internet] link

Wed, 28 Dec 2005

Making you stop and think again - 17:12
I just finished reading the book The Cuckoo's Egg by Clifford Stoll, it really does make you stop and think a bit. As a large part of net lore (100 greatest moments in Internet history article, a good read) I am somewhat surprised I had not come across this book in the past. The copy I read was loaned to me by a friend who is not a self proclaimed geek to a huge extent who had recommended it as a good read.

I should have been doing a lot of other things today but spent most of it reading the book, ahh well. As mentioned above it makes me stop and think, should I go and recheck my nagios setup, should I run some more log analysis tools, etc etc. Consider that this all happened back in 1986 and 1987, compare the scale of the Internet and the number of people behaving maliciously online back then to today and you start to wonder are you doing enough with security concerns?

[/comp/internet] link

Wed, 06 Jul 2005

Creepy crawlies at NLA - 11:44
I suspect at least someone was listening to Mikal (or put some more thought into the problem) over at the National Library archive. A quick glance through logs for svana.org starting in September 2000, the first occurrence of "http://pandora.nla.gov.au/crawl.html" in the logs is June 16 2005.

Since that first hit I have had 1715 hits from the crawler, downloading everything from ride photos, race reports, results, diary entries and many other things sitting around in the various subdirectories linked on svana.org. Someone may have noticed the owner of svana.org resides in Australia, it is far more effective than crawling any of my domains ending in .au as I don't use them currently.

[/comp/internet] link

Tue, 14 Jun 2005

My very own Hackergotchi - 15:15
Earlier today MRD mentioned he had added Hackergotchi support to Planet Linux Australia and requested that people email him their Hackergotchi and IRC nickname (similar to Planet Gnome).

I looked through a bunch of photos and eventually chose one taken of me at last year's Mont Australian 24 Hour race. I followed Wouter's instructions and made my own hackergotchi. MRD asked for them in 70x74, which did not quite fit with mine, so I have two, one in 57x74 and one in 70x90.

[/comp/internet] link

Tue, 29 Mar 2005

Do we really have to use it for good causes? - 12:47
After Mikal put the effort into a high ranking in google for a certain phrase. (he is now the top 4 hits for that phrase in google) It appears other people are keen to be recognised by google and thus the Internet world for for interesting activities, at least if Chris' entry is anything to go by.

The title of this entry is of course in reference to using your google juice for good or evil.

[/comp/internet] link

Sun, 20 Mar 2005

Never underestimate the bandwidth - 19:40
The title of course comes from the line in Tanenbaum's computer networks book, "Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway."

One of my house mates is running ICQ or similar and it appears to happily use all available bandwidth all the time. This means the latency to Internet sucks for ssh and other interactive stuff. Web browsing even is slowed, and it uses a lot more than half the available bandwidth if I start a large download.

I need to use some form of bandwidth management or queueing on my NAT box, probably something akin to the traffic conditioner mentioned in section 15.8 of the Linux Advanced Routing and Traffic Control HOWTO (which is a little bit out of date now but most of it probably still applies).

I have thought about setting this sort of thing up for a while, so ssh has low latency, however my NAT machine is running an old 2.4 kernel and I do not have the appropriate modules compiled in, the last time I tried a newer kernel out it only got half way through the boot up process. It is time to try another new kernel, this time a 2.6 kernel I reckon, however it will take forever to download 33MB of kernel bzip2 with the current issues on the network. This got me to thinking, it would be faster for me to ride into work, ftp the latest kernel image onto usb memory stick or similar and ride home than it would be for me to download one through the degraded bandwidth into the house.

I would hope I have avoided under estimating the bandwidth of a usb memory stick in a pocket while riding a bike to and from work :)

[/comp/internet] link

Tue, 15 Mar 2005

Are technorati tags doomed to go the way of html meta tags? - 23:09
Has anyone else noticed Mikal has been blogging with great frequency over the past 5 days? Anyway he noticed that technorati tags have been polluted by spammers already. A lot of other people have been wondering when this would happen also. About a month ago I read something written by Anil Kumar (I have no memory of where I found the link to his diary originally)

As he points out, google no longer gives Meta tags any ranking due to spammers, and once google (or any other search engine) starts giving any note to technorati tags spammers will use them. Well it happened, if you search for some topic Mikal has blogged about the top few links almost always point to the technorati category he put the diary entry in. I am sure this is happening to many others, thus spammers are interested.

It looks like technorati tags may well (as many have predicted) be doomed.

[/comp/internet] link

Write like no nobody's watching - 22:35
Okay so I stole the title from the Satchel Paige quote "Work like you don't need the money. Love like you've never been hurt. Dance like nobody's watching.", the sentiment applies here however.

Mikal once more mentioned some of the reasoning behind why he started blogging. It is good to see he knows his reasoning and is still happy with this. It appears many people seem to want to do it for strange reasons that almost seem doomed to failure.

Looking at various somewhat prolific bloggers suggest some reasons or methods on how to blog (2005 Bloggie winning entry), why some blogs succeed and some don't and how blogging may help your career (found on Anil Dash) (some of which Mikal and his co-workers appear to have been discussing recently also).

An interesting point, at least to note if a Blog is popular, the people who write them tend to enjoy or love writing. This is upheld in a lot of the above. Blog if you want to and want to write, and want to have something to say. Other reasons will probably not satisfy you. Mikal fits in here as he blogs because he wants to to satisfy his reason for doing so.

As an obvious example, Heather Armstrong won 4 of the 2005 Bloggies, she updates Dooce daily, and puts effort into writing it, she also appears to enjoy writing it. Like free software with Release early, release often as a mantra. Write like nobody's watching, write early, write often, etc. Or at least satisfy your own reasons for blogging.

[/comp/internet] link

Tue, 15 Feb 2005

That google juice thing - 14:42
So Andrew has been surprised by the way google juice affects stuff in the past. I must admit I just got a big surprise too. I have known for years I tend to have reasonably high google juice, but this one still surprised me. Last week I posted on my diary some stuff about Louise Werner, lead singer of the British group Sleeper. Today I noticed some google hits in my apache logs from searches for Louise Werner and Sleeper, curious I googled for "Sleeper Louise Werner" and simply "Louise Werner" I am now the top two google hits on that subject.

When I think about it I can see why, the band do not appear to have their own website (they split up anyway AFAIR) and there do not appear to be many fan sites online any more and as mentioned I seem to have some google juice, but it does kind of annoy me, I do not have much information about the band or the lead singer Louise Werner on my website anywhere and thus do not give them a fair treatment if someone is looking for information. Oh well I hope no one is offended by the lack of relevant information if they happen to come here from those searches.

[/comp/internet] link

Fri, 11 Feb 2005

Unable to access svana.org - 21:49
So around 2:30pm this afternoon something broke on the link between Comindico and Apex (where svana.org is hosted) Apex have their downlinks through transact and when I spoke to them on the phone around 4:30pm they said they still had not heard back from Transact as to what was causing the problem. Now, after 9pm something is still broken. From ANU I get the following traceroute output.

[21:54:04] 1 diver sjh ~>traceroute svana.org
traceroute to svana.org (203.20.62.76), 30 hops max, 38 byte packets
 1  sneakya (150.203.160.1)  0.371 ms  0.305 ms  0.226 ms
 2  hanhub.anu.edu.au (150.203.164.1)  0.548 ms  0.436 ms  0.450 ms
 3  huxhubrb.anu.edu.au (150.203.202.1)  0.557 ms  0.438 ms  0.456 ms
 4  carnohubb.carno.net.au (203.22.212.65)  0.557 ms  0.443 ms  0.454 ms
 5  ATM1-0-1.cn1.optus.net.au (202.139.39.245)  1.307 ms  1.303 ms  1.077 ms
 6  GigEth12-0-0.mn1.optus.net.au (202.139.188.131)  16.171 ms  14.794 ms  15.102 ms
 7  ComindicoInternational.mn1.optus.net.au (202.139.138.198)  15.485 ms  17.360 ms  15.313 ms
 8  ge6-2.cor02-dryb-mel.comindico.com.au (203.194.56.60)  28.724 ms  29.073 ms  28.620 ms
     MPLS Label=354 CoS=3 TTL=1 S=0
 9  pos2-0.cor01-kent-syd.comindico.com.au (203.194.1.209)  29.312 ms  28.825 ms  28.473 ms
     MPLS Label=350 CoS=3 TTL=1 S=0
10  pos9-0-0.cor01-nort-cbr.comindico.com.au (203.194.1.90)  28.359 ms  28.577 ms  28.593 ms
     MPLS Label=442 CoS=3 TTL=1 S=0
11  ge1-0.wsr01-nort-cbr.comindico.com.au (203.194.57.54)  28.346 ms  28.592 ms  28.327 ms
12  * * *

From pretty much everywhere else I get the same endpoint stopping traffic, ge1-0.wsr01-nort-cbr.comindico.com.au, which at a guess is the comindico north Canberra gateway or something.

I have written in the past about how annoying it is to be disconnected from my email when svana.org is inaccessible. So far with the downtime of around 7 hours it really is starting to get ridiculous. Yes I openly admit to being an Internet addict <g>, and I want my email. That this takes the Bilbys website and a whole bunch of other sites offline on calyx along is just more annoying.

[/comp/internet] link

Fri, 04 Feb 2005

The wayback machine and the font of all knowledge - 23:16
There is a lot of assumed knowledge among user groups and communities. Whether it be in jokes or simply something you expect someone else to know if they do are a member of some specific group. Among avid web users irrelevant of their personal specialities there are a few sites I would expect many people know about today. The most obvious of these today is Google, two others I expect most heavy web users would know about these days are Wikipedia and the Internet Archive.

I suppose due to the fact I count these as assumed knowledge by heavy web users today I was somewhat surprised to hear my sister Jane only found out about the Internet Archive in the last week and until I mentioned had still not actually consciously heard of Wikipedia (I have written about it here previously so she probably saw mention of it without taking in the information).

The Internet Archive is basically what it says, the web is changing daily, sites that once existed either change or disappear. Since 1996 the people running the Internet Archive have been archiving data from the web. Disk space is getting progressively cheaper so why the heck not, on the site itself you can access entire snapshots of sites or whatever from any time since the archive started. I have previously linked to the archive from a diary entry, referring to it by the colloquial name "The Wayback Machine" (they use this themselves, the term originally comes from Rocky and Bullwinkle) when I wanted to reference a website Jim Trail used to maintain for Triple J that is no longer online at ABC. This like google is yet another fantastic use of gobs of disk space on cheap x86 computers running linux. Reading the Wayback Machine FAQ is a good plan to learn more about it.

Wikipedia on the other hand is an online massive collaborative encyclopaedia. Wikipedia itself provides a good definition of what a Wiki is. This is lightly moderated and relies on the accuracy of the data added to it. Like any other source of data it should not be relied upon entirely, simply use it as yet another source of data on some given topic. I have commented on Wikipedia in the past (search google for "site:svana.org wikipedia" if you want, I can not be bothered linking to all the entries here). Others I know well (Martin Pool, Rusty Russell, Chris Yeoh for example) have all commented on the reliability issue in the past. As have other people such as Danah Boyd and Cory at BoingBoing to name two. WikiPedia has many advantages of traditional encyclopaedia's, one of which that quickly becomes obvious is on pop culture and recent events. I can almost guarantee Dr Who, Star Trek, The Simpsons and other cult tv phenomena do not get anywhere near this much coverage in any traditional encyclopaedia. Nor will there be coverage of events the day after they happen, or even as they are happening as there often is on WikiPedia. WikiPedia is not alone either, another good example of a massive collaborative online encyclopaedia is Everything 2.

There you go Jane, and anyone else who had not heard of these two rather cool sites, go have some fun.

[/comp/internet] link


home, email, rss, rss2.0, atom