Final report for “Inquiry into Improving Access to Victorian Public Sector Information and Data” released

The Victorian Government has been running an inquiry into access to the data that it generates, and they’ve finally tabled their report (PDF). I’ve only had a chance for a quick scan of it so far but its three main recommendations are as follows.

Firstly – this info should be made available and it should be cheap (ideally free!):

The Committee has proposed three key recommendations for access to and re-use of Government information. First, the Committee recommends that the Victorian Government develop an Information Management Framework for the purpose of facilitating access to and re-use of Victorian Government information by government, citizens and businesses. The default position of the framework should be that all PSI produced by Victorian Government departments from now on be made available at no or marginal cost.

Secondly – they should use Creative Commons licensing wherever possible!

The second key recommendation of the Committee is that the Victorian Government make use of the Creative Commons licensing model for the release of PSI. The Committee was told Creative Commons licences can be appropriately used for up to 85 per cent of government information and data, providing a simple to understand and widely used system for the re-use of PSI. Remaining Victorian Government PSI should either not be released, or released under licences tailored specifically for restricted materials.

Thirdly – and least excitingly – there should be a portal for this info..

The Committee’s third key recommendation is that the Victorian Government establish an on-line directory, where the public can search for and obtain information about PSI held by the Victorian Government. Depending on the access conditions Government has attached to specific PSI, people will be able to download information and data directly, or make contact with people in the Victorian Government to discuss access conditions.

They also have a recommendation and finding relating to state government purchasing of software related to open source:

The Committee also considers the use of open source software (OSS) within and by the Victorian Government. One of the Committee’s recommendations is that the Government ensure tendering for software is neither licence specific nor has proprietary software-specific requirements, and that it meet the given objectives of Government.

Finding 23: There is sufficient evidence of cost-competitiveness between open source software and proprietary software for government to carefully consider both options during software procurement and development.

They also consider the licensing of software developed by the government:

As noted in section 10.4.3.2 below, current Victorian Government policy is to allocate IP rights in software produced for it to the software developer, with certain restrictions to ensure the Government’s interests are protected. This means that there is nothing to restrict people who develop software for the government from subsequently releasing it as OSS.

Unfortunately it looks like MS Word stuffed up their references and headings for them – what irony! There is no section 10.4.3.2 in the PDF, it’s probably referring to section 10.3.3, which is followed by section 10.3.4 which in turn is followed by 10.3.3.1 – er ?

Even more interesting is when they talk about file formats:

Recommendation 42: That the Victorian Government require, as part of its whole-of-government ICT Procurement Policy, that software procured by the Government be capable of saving files in open standard formats, and that wherever possible, the software be configured to save in open standard formats by default.

There’s heaps more there, but I’ve run out of time to read it tonight! 🙂

(Found via OpenAustralia on Twitter)

ODF Plugfest

After the noise over whether or not the implementation of ODF (Open Document Format) in SP2 for Microsoft Office 2007 was deliberately broken for monopolistic purposes or incompetently implemented (or a combination of both) it’s nice to see that there is active interoperability work going on between vendors and developers at the ODF PlugFest, and the KOffice developers Jos van den Oever and Sven Langkamp attended and contributed to an article on the KDE DOT news website and Sven blogged about his positive experiences at the workshop.

It was first time I was going to such a workshop and I had expected that there would be fights between the different vendors like it happened in some blogs before the workshop. It was a pleasant surprise for me that the athmosphere was very friendly and productive. It was really nice to meet other people projects/companys, put the competition aside for some time, work and drink some beer together.

One neat feature mentioned there is the OfficeShots website which lets you submit an ODF document and then get back renderings of it (PDF, screenshot, ODF) from various ODF implementations. There are 8 listed at present (including KOffice), but sadly MS Word or Google Docs aren’t amongst them (yet).

ST31500341AS can be jumpered for 1.5Gb/s SATA

I’ve been struggling getting a 1.5TB Seagate ST31500341AS (CC1H firmware) to work on my ageing Mythtbuntu MythTV box (Gigabyte GA-8PE667 Ultra 2) to store TV programs on, it would just stop after the SATA BIOS for the onboard Silicon Image, Inc. SiI 3112 card had ID’d it correctly, before returning its size. So I went to Jaycar and bought a two port Silicon Image, Inc. SiI 3512 SATA card instead which seemed to work fine – the kernel booted and ID’d the card OK and I could partition it but then found that when I tried to make a filesystem I got lots of these:

ata3.00: exception Emask 0x0 SAct 0x0 SErr 0x2400000 action 0x0
ata3.00: BMDMA2 stat 0x86c1009
ata3: SError: { Handshk UnrecFIS }
ata3.00: cmd 25/00:08:28:7b:a8/00:00:ae:00:00/e0 tag 0 dma 4096 in
res 51/04:00:2f:7b:a8/00:00:ae:00:00/e0 Emask 0x1 (device error)
ata3.00: status: { DRDY ERR }
ata3.00: error: { ABRT }
ata3.00: configured for UDMA/100
ata3: EH complete

which caused I/O errors and the mkfs aborted. 🙁 Some nice people on the LUV list suggested jumpering the card for 1.5Gb/s operation, but I found a post on the Seagate forums saying:

The manual for Seagate SATA drives pretty clearly states (on page 22): “1.5Gbits jumper block only applies to ST3320613AS, ST3320813AS and ST3160813AS models” So it would seem that you can’t run your ST31500341AS in forced-1.5Gb mode (though the drive does support auto-negotiation – but I guess your controller does not).

I thought I was out of luck, and then I stumbled a post elsewhere that said:

I just called to Seagate. They insist that even though the product manual (At their website) says that using the jumpers to force SATA1(150mb/s) speed not supported for my HDD(ST31500341AS) and also on top of the HDD never print the label anymore, you still set the jumper and it’ll force SATA1 speed. According to the tech support dude.

So, having a jumper to hand I decided to give it a go, and it does appear to work! Not only can I finally successfully make a filesystem, but I can also run bonnie++ on it continuously without issues and then rsync all 200GB of TV shows onto it.

Iranian Internet Controls – Targeting Flash and Email ?

Some very interesting investigations done by the Arbor Networks security folks looking into Iranian traffic engineering and filtering from the time of the Iranian presidential election onwards. They have both a preliminary investigation showing a dramatic fall in traffic at the time of the election and a follow up deeper look demonstrating that they appear to be specifically targeting streaming media (flash, et. al) and email, as graphically demonstrated by this graph:

Graph of video streaming bandwidth used by Iran around the time of the election.

Web and other traffic have been left relatively unscathed, prompting this comment:

Perhaps games provide a possible source of covert channels (e.g. “Bring your elves to the castle on the island of Azeroth and we’ll plan the next Ahmadinejad protest rally?”)

WordPress 2.8 upgrade – beware SVN!

Just had a rather nasty experience trying to upgrade to WP 2.8 using svn switch to move to the 2.8 tag – it resulted in the whole thing going pear shaped during the upgrade process and complaining about:

Fatal error: Class ‘WP_Widget’ not found in $wordpress/wp-includes/default-widgets.php on line 15

Turns out that the files in SVN don’t match what’s in the 2.8 tarball, and the 2.8 tag in SVN doesn’t work. The fix was to rsync the files from the tarball over SVN. Very very weird (and the first time I’ve had such problems!).