It’s been a good year in that regard, we have done a lot.  I managed to work with a couple of great candidates standing in a raft of elections.  I supported Tim Dobson in Gorton, Loz Kaye in Oldham and most recently Graeme Lambert in Bury.  It’s been fun, it’s been hard work and most of all it has been an opportunity to really see what works and what doesn’t.

It has also meant doing research, polling and analysing the work that has been done, something that really has spurred me on.  There is nothing quite like doing something, getting feedback on how well it has gone and collecting evidence to show how it can be done better.

In any case I might now have an opportunity, with Peter Brett standing down as the Party’s campaigns officer, to do the same in an official capacity and hopefully with a few more resources and more support,  so I am standing for the post.

So now more than ever it is time for me to think about ‘evidence based’ campaigning, working out the best way of pushing out our message and connecting with people in a more than superficial way and building a future for a party that is so important in the current world.

I am laying out what I think needs to be done and how it should be done over in the Pirate Party forums and hopefully answering questions as they come up, so do take a look at the discussion and get involved.

Oh and if you are interested in getting involved, especially if you want to get out and campaign, make some noise, or simply help get the party message out, then do.  Join the party, or get in touch with me or one of my colleagues and you can help make a difference. Oh and if you are already a member of the party have a look at the various campaigns officer nominations and make sure you vote for the one you want.

There is a strange but constant narrative in British politics, that revolves around politicians being out of touch with the population.  Many reasons are suggested as the root cause of this, everything from wealth and education to isolation from the public and ideology.  Personally I have come to a slightly different (although not new) conclusion.

Quite simply politicians are no more disconnected from the general public than any other group.  If you took a group of IT professionals and put them into a room and asked them about the general population the responses would probably be somewhat negative, comments about people not bothering to understand, failing to appreciate what is done for them and so on.  I would bet that applies to essentially every group from retail workers (it was certainly true when I worked my first job in Woolworth’s way back when..) through to doctors.

More importantly of course the reverse is true too, ask a random selection of people about any given group that they are not a part of and you will find that they feel that they are different. Those asked will assume that members of the group in question don’t really understand what they want or need, “they don’t really understand us”.

The class system, racial tension, regional rivalry and a whole host of other factors play into it too, how can a southerner understand the problems in the post-industrial north? What does someone in Oldham understand about street crime in Farnborough?  There are so many legitimate differences and so many more related to (often flawed) perception alone.

The difference of course is that unlike IT professionals or retail workers, politicians are supposed to be representing us (or trying to get to a position where they can).  Our individual MP’s are certainly people who should be in touch with us, our government should broadly understand what is important to us and act on that basis.  Unlike doctors and IT professionals, their level of ‘connectedness’ is deemed central to their ability to do their jobs.

And yet striving to understand the issues that we all face and finding solutions, either due to an idealistic wish to serve or in pursuit of power, makes them different.  Just like the activist who believes in an issue and works to change the world to deal with it, they separate themselves from the average person in their desire to identify and fix what they see as the problems.  Granted, they often compound that perception of difference by claiming solidarity with everything and everyone, by pretending that they are like everyman, by doing things that are so blatantly false that it is cringe inducing to watch.  But in theory at least they are doing it for the right reasons (or maybe they started doing it for the right reasons..).

Politicians do live in a bubble, they do apply different priorities to different issues and they do get it very wrong on a regular basis.  We should however not forget that they are not alone.  If I look at my twitter feed, or my discussions on reddit and slashdot or listened to my (rather rare..) discussions in the pub, you would be forgiven for thinking that the country was full of people who cared passionately about electoral reform, see the DE bill as the some of the worst legislation ever written and is composed largely of people who both hold and advocate and ideology as well as being engaged in the political process.

Of course you would be wrong.

So what is the solution? I would say that for the politicians it is all about real communication, talking to people and really engaging, not just before an election or during a campaign, not just in areas where there are votes to be won and not just on political matters.  Getting rid of the gimmicks would help, ceasing the headlong rush to appear to be ‘like’ the people they seek to represent in every aspect.  But then again, what do I know? I can hardly claim that I am in touch with people who are significantly different from me, people in significantly different situations than I am, all I can say is that I do try and listen and that I might be able to understand some of their issues on some level.

Maybe there is also something in the idea of greater unity, solidarity and tolerance, but again that is something lots of government and political parties have striven for with varying degrees of success and no doubt almost always against a resisting public for a long time.

Given that the goal of the site is for people to take the survey, it seems clear that the site is doing it’s job.  It is interesting however, to look at the slightly more raw figures.  I the last 24 hours the site has seen 3,167 pageviews from 2,296 unique visitors. Of course that doesn’t include visitors running things like noscript (which can be somewhat high.. the discrepancy between analytics and server logs on some of the more techy sites I have can run as high as 20-25%), but we will ignore that for now.

The browser break-down is also interesting:

I’ve often seen stats that suggest a preference for Firefox, but rarely one that had Internet Explorer (all versions….) down in fourth place, I certainly haven’t seen Chrome rank as highly.  Unfortunately, when we look at operating systems things aren’t nearly as interesting, with the vast majority of visitors using machines that claimed to be Windows, 22% Macs and 16% Linux.

As to when the traffic came in, well that was somewhat predictable, but I have graphed it below anyway, it seems that the lunchtime spike for this kind of site really isn’t myth…

So, 2,296 visitors in 24 hours.  We know that 159 actually tweeted the fact that they had taken the survey, but how many actually took it? Well we can see that of the 2,296 visitors we saw, 1,827 actually looked at the survey and 1116 decided to take the survey.  Of that 1116, 968 ended up being defined as pirates, the remaining 148 taking the high road and declaring as non pirate.  Of the 968, barely 16% tweeted the result.

Now obviously tweeting the result requires people to both actually have a twitter account and be comfortable tweeting something that could be seen as controversial… The former is probably the biggest barrier to that particular goal and I don’t have any idea what kind of take up twitter has in the demographic that looked at the site.  However I can say that the absolute percentage of visitors seeing the siteand later tweeting it stands at 6%, not a bad conversion rate.

As to referrers, well that would be telling, but twitter pushed more than half the traffic to the site (not surprising given the nature of it) and the increase was fairly exponential once it started, other hits came from reddit, thenextweb, facebook, stumbleupon, and here…

The question seems almost absurd, unless you spend your time wearing an eye patch whilst attacking others with swords, use the words ‘arrr’ regulary, or enjoy taking over the odd supertanker, the chances are that you don’t think you are… Of course the term pirate is also used to describe those vexatious villains that breach copyright.  In the UK this has been somewhat immortalised by the ‘Would you steal a handbag’ commercials (possibly aided by the parody of the same on the ‘IT Crowd’ that follows. )

Now Whilst the average person would probably assume that ‘pirates’ are either guilty of some sort of nautical mis-bihaviour, or at the very least mass copyright infringement for substantial personal profit, they tend to overlook the fact that any unlawful copy is as bad as any other under the law…

This post is reply to Mark Pack’s post “Election imprints in the online world: what should the rules say?” and originated, for me at least,with a Tom Watson MP blog post entitled “Spoof videos will not be policed in the general election”.

Essentially Tom Watson (on the back of a BBC report on the Election Commission being unable to police the expected explosion in spoof videos at the next election and discussion in other areas) suggested that a set of guidelines could be put together by the Internet community at large, or at least that is how I read his challenge:

“I’m going to write to Jack Straw to ask him to sort it out. It strikes me that between us, we could knock up a set of guidelines quite quickly. I nominate Mark to get the ball rolling with a draft. Anyone else got ideas?”

Since the initial request, there has been a little discussion on both Mark and Toms blogs, in addition Mark put out an initial set of guidelines yesterday to which I thought I might add a little commentary. 

Overall principles
1.The logic for having imprints on printed literature – making clear who it is from, informing people who is legally responsible for the contents etc. – applies as well to online material as offline.

Absolutely, any imprint should establish the source and who it is for.

2. Online imprints should be easy to access and should not be so burdensome as to make it impossible or impractical to use a particular technology. If that is the case, the imprint information should instead be readily and clearly available from elsewhere.

I would take this one step further. An imprint in digital media should be available in two ways, firstly in a manner that is obvious to someone using the media in the normal way (listening to a pod-cast, watching a video, seeing an image or reading written text) and should, where possible be included in any metadata so it becomes machine readable too… This is easily achievable when creating such media and could prevent issues whereby such media is taken out of context

3. The information should be readily available, e.g. not requiring someone to watch a 10 minute film to find it.

Agreed, I would say that in the case of audio or video the imprint should be seen or heard before the content.

Information that should be provided in online imprints

1.The same trio of information should appear as for printed material – the electronic equivalent of the printer, the promoter and who they have promoted it on behalf of.

2. The “printer” for web services is the firm who provides the service, e.g. Facebook or YouTube. For distribution services, such as email, the “printer” is the firm whose technology does the distribution, e.g. the firm running the SMTP server used.

I partly disagree with this. The information I would say is required is the source (or author) and the person or group who it has been created for. The problem I see with including the “printer” is that it will vary if something has been reused.  For email it is a potential minefield for anyone without a dedicated SMTP server, you could hardly list BT’s or Virgin Media’s just because that happened to be where it was sent through.  For other media I see issues where if an image has been created and first issued on Facebook or Youtube, but is then moved and embedded in a blog, or mashed up with other information to create a derivative, the initial “printer” is no longer important, the focus should be on the source and  the person who it was produced for.

Audiovisual material

1. In order for the imprint information to be readily available to people, it should be available at the start of any such material or immediately adjacent to the material (such as on the packaging of a box that a DVD comes in).

I would simply say that should be included at the start (unless length restrictions make it impossible).

2. Where audiovisual material can be republished or embedded elsewhere, only having the imprint adjacent to the audiovisual material in its original location is not sufficient.

Material with length limits

1.If, and only if, a technology imposes a length limit on content which means that a full imprint would take up more than 15% of a message, the imprint need not be included with the message.

In these cases (and I can’t think of many) there should be a link to the relevant information, unless that is impossible.

2. Where the message automatically contains a link back to a profile or user page, such as tweets, then the full imprint information should be provide on that profile or user page.

Definitely.

3. Where there are no such links, such as with text messages, either it must be possible to reply to the message, or the message must contain contact details, and either way any reply requesting the imprint information should be responded to within 48 hours.

The only think I would add is that in terms of the information required, a real address (as in brick and mortar) would be preferable than an electronic one for most things,although both would probably be better.

One thing that does strike me is that we should be using a wiki or similar for this kind of collaborative work, although it may not be ideal (and a public wiki would potentially be a nightmare) it would make collaboration generally easier.  If it seems like a good idea I could put together something at www.crossparty.co.uk intended as a party neutral discussion area, although I suppose it would have to be by invitatin only, at least initially, indeed policy collaboration was what I intended the domain to be used for initially, however it fell into the pile of  my unstarted projects because there seemed to be no interest… That said, I’m not sure if I would be duplicating effort, so if something already exists, please, someone point it out.

In any case, Virgin recently announced a new “Advanced Network Error Search” a system whereby erroneously entered URL’s, rather than returning the relevant DNS response code to such a search (whether it be unassigned or otherwise…) returns a nice neat record, not related to the search, but rather a redirect to virgins own, Yahoo powered search pages.

Virgin’s own description of the service is fairly transparent:

We all make mistakes when we type in website addresses. Perhaps we miss a few letters, or the website doesn't exist any longer. If an address you enter doesn’t locate a site, this handy feature will convert the incorrect address into a web search, so instead of an error message you will get a list of our closest matches, plus some additional related links."


Obviously for certain users this makes a certain amount of sense, but given that not everyone wants to be faces with a list of search results when entering a URL, or indeed may have applications that depend on the correct response to function properly, Virgin quite sensibly offer an opt out. More than that it is apparently a proper opt out, those that opt out get the proper DNS responses (rather than a cookie based mess returning a false page for the browser but a valid DNS response…).

So I opted out. That is I ticked the box that said “No – I would not like to use the advanced network error search”. Fairly clear, or so I thought…

Opt Out is shown as being active.

All has been well since then until today, I should be clear that nothing has changed (honest) my cable modem not rebooted since the initial opt out and nothing else should matter. Indeed on virgins own opt out page, the opt out is still showing as active.

However as of today, I have started to see virgins search pages when, on occasion, I have entered a malformed URL. Initially it was inconsistent, sometimes I would, sometimes I wouldn’t, but as of 1900 (GMT) every such request has plonked me in front of a virgin search page. It should be said that the results aren’t exactly wonderful either.

The result when mistyping reddit.com (reddit.cm)

Just to be clear, my DNS servers haven’t changed. They are 194.168.8.100 and 194.168.4.100. I have now rebooting my router, curious to see if I would receive a different set of servers, but alas no, they remain the same (and as I said my opt out is still showing…).  I even tried to opt out again, but alas,  as I’ve already opted out, I can’t do that again…

Once opted out, you can't opt out again. Which makes sense.

So lets see what is happening.

Process.

I thought I would take a quick look to see if anything unusual was happening, making a valid and then invalid request whilst sniffing my traffic with wireshark didn’t indicate anything nefarious (or at least nothing I hadn’t expected) but it does let me describe the process that is occurring:

First, as a user types in address into search bar, in my case reddit.cm, a DNS request is made to virgins DNS server (in this case 194.168.8.100). Essentially what happens is that my machine goes and asks a question of the DNS server I have specified. It asks just one question, can the DNS server give it the Host address for reddit.cm. Given my opt out I should get …. However I don’t. I get a non-authoritative response from Virgins DNS server informing me that a valid host address for reddit.cm is 81.200.64.50. Obviously it isn’t.

I can check that using nslookup:

username computer:pts/1 (~)
Wed,16 Sep @ 19:07 $ nslookup reddit.cm 100%(1:55:44)
Server: 194.168.8.100
Address: 194.168.8.100#53

Non-authoritative answer:
Name: reddit.cm
Address: 81.200.64.50

This works for anything that looks like a valid domain (just in case anyone has doubts about the existence of reddit.cm…:

username computer:pts/1 (~)
Wed,16 Sep @ 19:09 $ nslookup www.this.is.not.a.real.network.address.really.it.isnt.but.i.will.still.get.a.valid.ip.from.virgin.dns.servers.even.with.my.opt.out.com
Server: 194.168.8.100
Address: 194.168.8.100#53

Non-authoritative answer:
Name: www.this.is.not.a.real.network.address.really.it.isnt.but.i.will.still.get.a.valid.ip.from.virgin.dns.servers.even.with.my.opt.out.com
Address: 81.200.64.50

In fact if we do a quick reverse whois of the address returned we find, as expected that it is in fact the IP address belonging to advancedsearch.virginmedia.com.

username computer:pts/1 (~)
Wed,16 Sep @ 19:07 $ nslookup 81.200.64.50 100%(1:55:44)
Server: 194.168.8.100
Address: 194.168.8.100#53

Non-authoritative answer:
50.64.200.81.in-addr.arpa name = advancedsearch.virginmedia.com.

Of course what should be happening assuming the opt out worked… is the following (which is what my non-public DNS server supplies) :

username computer:pts/1 (~)
Wed,16 Sep @ 19:30 $ nslookup reddit.cm 172.16.200.2 99%(0:01:47)
Server: 199.7.83.42
Address: 199.7.83.42#53

Non-authoritative answer:
*** Can’t find reddit.cm: No answer

So back to the process Virgin is using to get us to their search pages and return relevant results:
Now that my machine thinks it knows where reddit.cm is, it can proceede, first establishing a TCP connection, then sending a HTTP GET request. The request goes something like this:

!/GET / HTTP/1.1
Host: reddit.cm
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-GB; rv:1.9.1.3pre)
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-gb,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive

The virgin server that recieves it knows that I am expecting to go to reddit.cm, it is in the GET request, it can now pass me a lovely page with Yahoo search results relating to the term (no doubt making virgin a little cash per request). And it returns its search result.  Here is one for debian.og.

A search for debian.og gives the same search page albeit with slightly better results.

Tech support

One thing I will say is that Virgin have decent first line tech support, it may not be able to solve your issues, but the guys and girls working for Virgin appear fairly informed and polite. When I called I the technician I spoke to was decent enough to state that he had no idea about the issue and that he hadn’t seen it before and that he was in fact very interested to find out what was going wrong. That made two of us.

So a quick escalation to Virgin’s second level of support was called for. Sadly that was as far as we got. After the initial comment that Virgin don’t support Linux, we decided that I knew how to restart my machine, flush my DNS cache and check to see if my DNS servers were correctly configured and that I wasn’t an idiot and fairly well clued in, so my comments and thoughts on the matter could be used, it was determined that no one had any idea what was going on.

Basically, Virgin would look into it, I should call them back in a couple of days if the problem persisted. I informed the first line tech that I would do so and in the meantime I would switch DNS servers, so my problem at least was solved, even if the larger issue wasn’t.

End Result,

Of course Virgins “Advanced Network Error Search” isn’t unique, many ISPs offer such a service and many won’t let you opt out (we will see where this goes with Virgin of course…). Indeed virgins search pages are considerably less objectionable than others (if we ignore the general breakage…). ‘DNS made easy’ has a similar system with a rather less slick end result (see http://205.234.170.218/).

username computer:pts/1 (~)
Wed,16 Sep @ 19:30 $ nslookup reddit.cm 205.234.170.215 99%(0:01:47)
Server: 205.234.170.215
Address: 205.234.170.215#53

Non-authoritative answer:
Name: reddit.cm
Address: 205.234.170.218

Even openDNS a popular alternative to ISP DNS servers amongst the more technical implements the same system, presumably to get the revenue required to provide a free DNS system in the first place. So this kind of quasi DNS-hijack is common but not yet universal, the more memorable DNS servers (such as 4.2.2.1…) and others I can remember (like those at blueyonder, or whatever it is now) still do things properly, of course so your mileage will vary.

Thanks and excuse the ramble. I will update this in a few days when I have more, hopefully either a fix or a response from Virgin.

In this case the problem occurred during an upgrade, the identities of the guilty will be obscured, but the lesson may be useful regardless.

The problem started with something quite innocuous. A problem with an upgrade.

Unpacking replacement coreutils
dpkg: error processing /var/cache/apt/archives/coreutils_7.4-2_sparc.deb (--unpack):
trying to overwrite `/usr/share/man/man1/arch.1.gz', which is also in package util-linux
dpkg-deb: subprocess paste killed by signal (Broken pipe)
Errors were encountered while processing:
/var/cache/apt/archives/coreutils_7.4-2_sparc.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)
server-002:/home/tech-003#

So, the installation of the latest version of coreutils cannot overwrite a man page also provided by util-linux. There are a number of ways to get around it, however the one chosen wasn’t one of the greatest ideas…

The tech in question decided that the best approach would be to remove both offending packages and reinstall them…

server-002:/home/tech-003# dpkg -r coreutils
dpkg: error processing coreutils (--remove):
This is an essential package - it should not be removed.
Errors were encountered while processing:
coreutils

Dpkg quite sensibly refuses.

Not to be thwarted however, our diligent tech decides that a little force will solve the problem…

server-002:/home/tech-003# dpkg --force-all -r coreutils
dpkg: warning: overriding problem because --force enabled:
This is an essential package - it should not be removed.
dpkg: coreutils: dependency problems, but removing anyway as you requested:
dpkg depends on coreutils (>= 5.93-1).
linux-image-2.6.18-6-sparc64-smp depends on coreutils (>= 5.96).
ucf depends on coreutils (>= 5.91).
debianutils depends on coreutils (>= 4.5.8-1).
(Reading database ... 35142 files and directories currently installed.)
Removing coreutils

So coreutils is removed, with dpgk screaming about it not being a good idea. Turns out dpkg was right…

server-002:/home/tech-003# apt-get install coreutils
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed
coreutils
0 upgraded, 1 newly installed, 0 to remove and 352 not upgraded.
Need to get 0B/3814kB of archives.
After this operation, 11.2MB of additional disk space will be used.
WARNING: The following packages cannot be authenticated!
coreutils
Install these packages without verification [y/N]? y
Selecting previously deselected package coreutils.
(Reading database ... 34831 files and directories currently installed.)
Unpacking coreutils (from .../coreutils_7.4-2_sparc.deb) ...
dpkg: error processing /var/cache/apt/archives/coreutils_7.4-2_sparc.deb (--unpack):
trying to overwrite `/usr/share/man/man1/arch.1.gz', which is also in package util-linux
dpkg-deb: subprocess paste killed by signal (Broken pipe)
dpkg (subprocess): failed to exec rm for cleanup: No such file or directory
dpkg: error while cleaning up:
subprocess rm cleanup returned error exit status 2
Errors were encountered while processing:
/var/cache/apt/archives/coreutils_7.4-2_sparc.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)

The key here of course is not that arch.1.gz cannot be overwritten, but rather that ‘rm’ cannot be executed (it is provided by coreutils after all….

Rather than giving up, or calling for help, our diligent tech decides that removing util-linux will solve the problem…

Server-002:/home/tech-003# dpkg --force-all -r util-linux
dpkg: warning: overriding problem because --force enabled:
This is an essential package - it should not be removed.
dpkg: util-linux: dependency problems, but removing anyway as you requested:
util-linux-locales depends on util-linux (>= 2.12r-0).
util-linux-locales depends on util-linux (< < 2.12r.0-0).
util-linux-locales depends on util-linux (>= 2.12r-0).
util-linux-locales depends on util-linux (< < 2.12r.0-0).
(Reading database ... 34830 files and directories currently installed.)
Removing util-linux ...
install-info(ipc): could not backup /usr/share/info/dir in /var/backups/infodir.bak: No such file or directory

Hmm, so no coreutils and no util-linux. The next step, would have been to install the pair again...

server-002:/home/tech-003# apt-get install coreutils util-linux
Reading package lists... Done
Building dependency tree
Reading state information... Done
You might want to run 'apt-get -f install' to correct these:
The following packages have unmet dependencies.
util-linux-locales: Depends: util-linux (< 2.12r.0-0) but 2.15.1~rc1-1 is to be installed
E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution).
server-002:/home/tech-003#

Of course now we have a rather broken system so a repair is needed to fix the varios dependency issues...

Server-002:/home/tech-003# apt-get -f install
Reading package lists... Done
Building dependency tree
Reading state information... Done
Correcting dependencies...Done
The following packages were automatically installed and are no longer required:
sensible-utils
Use 'apt-get autoremove' to remove them.
The following extra packages will be installed:
coreutils sensible-utils util-linux util-linux-locales
Suggested packages:
dosfstools
The following NEW packages will be installed
coreutils sensible-utils util-linux
The following packages will be upgraded:
util-linux-locales
1 upgraded, 3 newly installed, 0 to remove and 350 not upgraded.
Need to get 4442B/5701kB of archives.
After this operation, 13.9MB of additional disk space will be used.
Do you want to continue [Y/n]? y
WARNING: The following packages cannot be authenticated!
coreutils util-linux-locales util-linux sensible-utils
Install these packages without verification [y/N]? y
Get: 1 http://debian.virginmedia.com unstable/main sensible-utils 0.0.1 [4442B]
Fetched 4442B in 0s (25.0kB/s)
dpkg (subprocess): failed to exec rm for cleanup: No such file or directory
dpkg: error processing /var/cache/apt/archives/coreutils_7.4-2_sparc.deb (--unpack):
subprocess rm cleanup returned error exit status 2
Errors were encountered while processing:
/var/cache/apt/archives/coreutils_7.4-2_sparc.deb
E: Sub-process /usr/bin/dpkg returned an error code (1)
server-002:/home/tech-003#

Ah, still no ‘rm’… so a failure on clean up.

Confused our diligent tech decides to look for ‘rm’…

server-002:/home/tech-003# rm
rmail rmmod rmt rmt-tar

Which isn’t there…

server-002:/home/tech-003# ls
bash: ls: command not found

Actually there are quite a few things missing…

server-002:/home/tech-003# mv
bash: mv: command not found
server-002:/home/tech-003# cp
bash: cp: command not found
server-002:/home/tech-003# touch
bash: touch: command not found

Time to start panicking…

Now, happily this server was also an NFS server, happily up and running doing it’s NFS thing. Our diligent technician finally went for help. So first things first, it was rather necessary to get some of those core userland tools cp, mv, ls and rm back in place. The easiest way it seemed was to grab (from another system) the relevant binaries, for sparc, and put them somewhere accessible to the still logged in root user.

The relevant package was downloaded from the local apt repository and unpacked onto the NFS share.

Then we could start copying the tools we needed back to somewhere useful…

server-002:/srv/disk1/bin# ./mv ./rm /bin/

We tried first with ‘rm’

server-002:/srv/disk1/bin# rm
rm rmail rmmod rmt rmt-tar

Which worked nicely and as expected…
So on with everything else:

server-002:/srv/disk1/bin# ./mv ./* /bin/

Now we have a more functional system back…

server-002:/srv/disk1/bin# cd /root/
server-002:~# ls
172.db printers.conf dhcpd.conf

All that was left to do was fix the dependency issues:

server-002:~# apt-get install -f
Reading package lists... Done
Building dependency tree
Reading state information... Done
Correcting dependencies...Done
The following packages were automatically installed and are no longer required:
sensible-utils
Use 'apt-get autoremove' to remove them.
The following extra packages will be installed:
coreutils sensible-utils util-linux util-linux-locales
Suggested packages:
dosfstools
The following NEW packages will be installed
coreutils sensible-utils util-linux
The following packages will be upgraded:
util-linux-locales
1 upgraded, 3 newly installed, 0 to remove and 350 not upgraded.
Need to get 0B/5701kB of archives.
After this operation, 13.9MB of additional disk space will be used.
Do you want to continue [Y/n]? y
WARNING: The following packages cannot be authenticated!
coreutils util-linux-locales util-linux sensible-utils
Install these packages without verification [y/N]? y
(Reading database ... 34756 files and directories currently installed.)
Unpacking coreutils (from .../coreutils_7.4-2_sparc.deb) ...
Replacing files in old package mktemp ...
Setting up coreutils (7.4-2) ...
(Reading database ... 35072 files and directories currently installed.)
Preparing to replace util-linux-locales 2.12r-19etch1 (using .../util-linux-locales_2.15.1~rc1-1_all.deb) ...
Unpacking replacement util-linux-locales ...
Selecting previously deselected package util-linux.
Unpacking util-linux (from .../util-linux_2.15.1~rc1-1_sparc.deb) ...
Setting up util-linux (2.15.1~rc1-1) ...
Installing new version of config file /etc/init.d/hwclock.sh ...
Selecting previously deselected package sensible-utils.
(Reading database ... 35216 files and directories currently installed.)
Unpacking sensible-utils (from .../sensible-utils_0.0.1_all.deb) ...
Replacing files in old package debianutils ...
Setting up util-linux-locales (2.15.1~rc1-1) ...
Setting up sensible-utils (0.0.1) ...
server-002:~#

Then we could go back to doing the upgrade…

What a fun 15 minutes.

PDA with OPIE.

People, notably those who don’t use and haven’t tried Linux, tend to ask me why,  after all my routers worked fine with their stock firmware, my laptops came with Windows XP preloaded, my servers with Solaris installed and my PDAs with windows mobile. Why go through the hassle of putting Linux on them?

Price?

Obviously for most of the devices listed in the previous paragraph the cost of the operating system doesn’t really come into it. I couldn’t have acquired my PDA or Laptop cheaper without Windows on (in retrospect I realise I could have asked for a refund on the OS, but I didn’t), my Servers came with a copy of Solaris, so the operating system, in each case was essentially free anyway…

So the price of the OS is fairly immaterial, it is a small fraction of the total hardware cost, certainly not enough to make a huge decision when purchasing. So if not the OS, maybe there are cost savings somewhere else.

Package management with adept

The assumption tends to be that I don’t want to pay for the other software that I need to make those devices useful. Things like Office (about £250 for the full version) and Photoshop (something like £800) are expensive acquisitions, I also do sound editing, video editing, manage my music collection, take notes (on my tablet) and program using an IDE, all tasks that could be done using expensive software.

Of course many people (if not most) who use these pieces of software haven’t paid for them anyway.  In any case I do own a copy of Office Professional (2003 as it happens) and Paint Shop Pro (from way back when), I even bought a boxed version of Windows 2000 at one point.

So I’m not shy about buying software if I feel it brings value, in fact the last piece of software I bought was a copy of Crossover Office, you can imagine my disappointment when not long after they decided to give a version away free for a day, even more so given that after the first six months of use I found that I have pretty much no use for it now.

So it isn’t about cost as such. What it is about is the cost of software versus the value I get it.

On the desktop I can do everything I could under Windows without paying a

KDE4.1 on Debian

penny, moreover I can do it more efficiently, faster and usually with better results. There are some penalties (some things aren’t as portable as I’d like) but they are more than made up for by the benefits.

From a small business point of view it gets even better, I don’t need to spend thousands of pounds to get a functional mail, web, database, file or print server (If I did I wouldn’t have been able to start my business). I don’t need to spend a massive amount on a backup solution, or on anti-virus or even on a Firewall and IDS system. Everything is there, easy to install, provided by one provider, trustworthy and very functional, more functional indeed than many of the non-free stuff out there, this is why Linux is king on the server after all.

How many small businesses can boast a full back office environment, a full development and test system, fully fledged security systems, internal mail, web and application servers and still claim that they are 100% compliant with their licensing requirements? I’d bet not many.

This, however is verging on the enterprise, most home users don’t need, want or care about any of it. They aren’t bothered too much by licensing concerns (they may not even be aware that they exist). So why would I want to use Linux at home on my personal machines?

Linux is… better

Well, mainly because Linux, even on the Desktop is better.

Annotating a PDF

I can do anything I used to be able to do, and rather a lot more with Linux on my desktop and laptops than I could with Windows. I can do all those things without worrying about cost, licensing, DRM issues, install keys, mal-ware, viral infestation and disaster recovery too much.

Most Linux distributions, and especially my distribution of choice, provides applications that are immediately available and can do pretty much anything you want or need to do.

If I need to quickly create an icon I can, liquid rescale an image? no problem, convert a directory of videos? Done.  The tools are all there and most are very powerful.  It is also often true that you will have not only an application with a nice GUI available to you, but also the ability to use the same (or a similar) application from the command line.  Suddenly automation is easy, repetitive tasks become unnecessary (write a script once and keep hold of it) using the computer becomes more fun and less of a chore.

Car Ablaze

At a little before 23:00 today the usual quiet tranquility (well, general calm in any case) in my little part of the world was broken with something that seemed to be an explosion.  A loud bang and the general boom of a pressure wave, a small one, but still enough to rattle the door. My initial reaction was that someone in the house had either fallen out of bed, dropped something heavy or slammed a door, those are after all the most common causes of loud noises at night.

After checking taking a look around the house and finding nothing out of place, I was all but ready to assume that the neighbour’s had somehow made a loud noise, not something I can ever remember having noticed before (I hope they can say the same for me, although I have a feeling I may be wrong…), but I thought a brief look outside would be sensible, just in case something had fallen off (it is a couple of hundred years old after all).

Still Ablaze

To my surprise, rather than finding a couple of slates on the floor I noticed a glow down the road, and then a rather audible pop.  Dashing through the garden a number of thoughts ran through my head, was there a house on fire (although the initial explosion hadn’t seemed quite that large) or maybe one of the factories or gas storage facilities in the valley had caught light (far enough away for any noise to be muted).

As I rounded the garden wall and looked down the street I saw something that I hadn’t expected, a large ball of fire engulfing a large car about 200 yards down the road.

A Fire Fighter Tackles The Blaze

My initial reaction was to call the Fire Service and Police, my second was to grab the half charged batteries out of their charging cradle, scrabble for a CF card, grab the camera and join the 10-15 or so other people who had by now gathered to watch to see if I could get some decent photographs of what was happening.  As the car had burned the windows had blown out, scattering glass easily 100 yards up the road, it had also rolled further down the hill, leaving burned rubber as testament to where it had sat initially.

The prevailing theory amongst those that gathered, without any police confirmation at the time, was that the car had been dumped after it had been stolen and used for joyriding.  Why they chose to dump the car on our street is unclear (although yesterday the police did chase a joyrider down the same street…).  I just hope that it isn’t an indication that the area will become more popular for thieves to dump their stolen vehicles.

Burned Out Interior

I was impressed quite how fast the fire was brought under control, even if the Fire Brigade appeared to have managed to turn up to the wrong street… and how fast the burned out wreckage was recovered (2 hours is quite snappy in my book).  Although I am not sure that the police gave the incident as much attention as they might have, spending most of their time in their car and then conversing with the firefighters.  Of course I suppose there isn’t much they could do.  No one seems to have seen the car arrive, nor see anyone leave it, a shame considering that given how close to houses the vehicle was dumped (a matter of a pavement width)  and the obvious danger it could have posed to the occupants.

Car Recovered

So allin all an eventful evening, not one that I would hope to see repeated soon, but at least it has reminded me that I should have my camera and other gear ready and waiting all the time, rather than just when I am planning to use it.

For those interested there are some high-res images of the event on my flickr photostream.

Strange Google Adverts

I wonder if anyone else has seen anything similar (I was consider blocking the domain www.yoururlgoeshere.com in the Adsense configuration – which appears to be a link spam page in any case) and whether anyone had any idea what was going on. A message to Google has, so far at least, not generated a reply.

Well I guess it highlights one of the issues that people who maintain sites, whether it be a personal blog or a larger site, if you include adverts you are exposing your user base, the community that you have managed to build, to content that is not really directly under your control. Worse you are doing it for profit (or in my case an attempt to offset hosting costs, but the end result is the same). I suppose the only real upside is that most of the readers I would expect to see on the site will have adblock enabled…

Well thanks for reading.