FixMyStreet - AU

[Daniel O'Connor]

Hey guys,
I know there are JIRAs and what have you about this concept, but it hasn't quite taken off since 2009 - Heck, the ACT government managed to beat Open Australia to the punch.

Unfortunately the only other open platform I knew of was It's Buggered Mate - source on github (https://github.com/lonelyplanet/its-buggered-mate - ruby), but dead and gone at the moment.

I've got a bit of time on my hands; and one choice to make:

Do I fork: https://github.com/visiblegovernment/django-fixmystreet (python) or https://github.com/mysociety/fixmystreet (perl)

Both are maintained, but the perl version is probably driven by more momentum.

Since I'm likely to churn out a bunch of code in a short time then go quiet, for this to work I'd need collaborators - what do we have more of in the OA community? Python, Ruby or Perl? Or even, heaven forfend, PHP.

Who would be interested in this? 

What other code (iphone app mockups in JIRA?) already exists for this?

My immediate objectives would be:

• Get the model of LGAs done right. I'm sure there's data from ABS and planning alerts, plus freebase - it just needs cleaning and importing into whatever structure is in those projects.
• Start a giant spreadsheet or similar of who to contact or how to report problems to your local council - email reports + a link are no doubt supported already by fixmystreet, and worst case it's easy to make a fax cover sheet and a 'print now' button.
• Refactor the address/street autocomplete - google or open street map should have adequate replacements
• Ingest geonames or australia post or something else to relate addresses/geolocations to LGAs
• Look around at other, non council authorities. I have a broken bus stop near my house which is bugging me immensely and driving me to fix this.

[kat]

Hi Daniel

I am delighted in that case that your bus stop is broken. Unfortunately I can't answer any of your techie questions, but I expect three will be along at once who can soon! 

hurrah for you!

Kat

--
You received this message because you are subscribed to the Google Groups "OpenAustralia Community" group.
To post to this group, send email to openaust...@googlegroups.com.
To unsubscribe from this group, send email to openaustralia-...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/openaustralia-dev?hl=en.

--

0424 563 654

skype: kat.ska

twitter: @katska

[Henare Degan]

On 3 June 2012 16:38, kat <kat...@gmail.com> wrote:

hurrah for you!

Hurrah indeed, it's long overdue! :)

On 3 June 2012 16:33, Daniel O'Connor <daniel....@gmail.com> wrote:

I've got a bit of time on my hands; and one choice to make:

Do I fork: https://github.com/visiblegovernment/django-fixmystreet (python) or https://github.com/mysociety/fixmystreet (perl)

Both are maintained, but the perl version is probably driven by more momentum.

I'd strongly recommend using the mySociety codebase. There's a lot of subtle lessons learned that have been integrated into that platform over the years.

Furthermore mySociety have recently kicked off an effort to make projects such as FMS work in different jurisdictions: http://code.fixmystreet.com/

• Start a giant spreadsheet or similar of who to contact or how to report problems to your local council - email reports + a link are no doubt supported already by fixmystreet, and worst case it's easy to make a fax cover sheet and a 'print now' button.
  

This spreadsheet from Tweet My Council (http://tweetmycouncil.herokuapp.com/) might help: https://docs.google.com/spreadsheet/ccc?key=0AppM8FtCInYXdGxRZGpCY0t6eDV5aFNwSEFPWnJqV3c#gid=0

Cheers,

Henare

[Kieren Diment]

On 03/06/2012, at 4:52 PM, Henare Degan wrote:

> On 3 June 2012 16:38, kat <kat...@gmail.com> wrote:
>
>> hurrah for you!
>
>
> Hurrah indeed, it's long overdue! :)
>
> On 3 June 2012 16:33, Daniel O'Connor <daniel....@gmail.com> wrote:
>
>> I've got a bit of time on my hands; and one choice to make:
>>
>> Do I fork: https://github.com/visiblegovernment/django-fixmystreet (python)
>> or https://github.com/mysociety/fixmystreet (perl)
>>
>> Both are maintained, but the perl version is probably driven by more
>> momentum.
>>
>
> I'd strongly recommend using the mySociety codebase. There's a lot of
> subtle lessons learned that have been integrated into that platform over
> the years.
>
> Furthermore mySociety have recently kicked off an effort to make projects
> such as FMS work in different jurisdictions: http://code.fixmystreet.com/
>

Plus I keep seeing mysociety advertising on jobs.perl.org.

And finally fix my street is a Catalyst application, and so you have access to one of the Catalyst core team developers if you need it (albeit I am only the documentation lead). Perly/catalyst questions, or discrete reasonably well defined tasks you'd like to divvy out to me for porting the FMS app then please let me know - email or kd` on #openau...@irc.freenode.net

[Rich Larcombe]

Would like to contribute with a couple points to ponder (having been integral to the creation of Neatstreets a couple of years ago i've been through the pain).

For a successful system, the work effort is largely going to be related to the daily support of issues that arise from public reports - i.e unrelated to the technology. It's usually down to council process / privacy / boundary issues etc. All of which need to be treated on a per issue basis, applying the same logic for all councils is generally not possible, although am hopeful things will standardise over time.

So taking a long term view, getting a system up and running i would say is 25% of the work (especially given the options for inheriting a code base), the question of who is going to volunteer to do the other 75% of daily triaging and specific problem solving and report chasing is the question you really should answer before planning development - or at the latest going live.

The beauty of SnapSendSolve is that there is no triaging overhead because there's no intermediary platform, but then people will argue that exposing all issues in one place to the scrutiny of public is main purpose or power of the system.

Myself and another colleague went to great lengths to attempt to convince the management to open source neatstreets when we were working there. Obviously we didn't succeed, but looking back i no longer see the openness of the neatstreets source code as being nearly as important as the openness and accessibility of the data. 

Ideally all councils should chip in to fund the ongoing running costs of whatever system wins out, and the public should win by having a single reliable service with full access to all data.  

Please keep the group updated with how you travel with this, i may be able to chip in from time to time.

Regards

Rich

p.s i would try out both these systems for your bus stop issue, even if you decide to press on with a fixmystreet solution.

 

On 3 June 2012 16:33, Daniel O'Connor <daniel....@gmail.com> wrote:

[Daniel O'Connor]

On Tue, Jun 5, 2012 at 12:11 PM, Rich Larcombe <richla...@gmail.com> wrote:

Would like to contribute with a couple points to ponder (having been integral to the creation of Neatstreets a couple of years ago i've been through the pain).

Thanks! Valuable insight. I was only vaguely aware of neatstreets, thinking it was Vic focused only, but clicking around shows its' a bit more open than that.

 

The beauty of SnapSendSolve is that there is no triaging overhead because there's no intermediary platform, but then people will argue that exposing all issues in one place to the scrutiny of public is main purpose or power of the system.

This was strongly my desired outcome - forget red tape, just complain openly and let it waft it's way to the right spot. Good to see it's already solved fairly strongly.

 

Myself and another colleague went to great lengths to attempt to convince the management to open source neatstreets when we were working there. Obviously we didn't succeed, but looking back i no longer see the openness of the neatstreets source code as being nearly as important as the openness and accessibility of the data. 

Ideally all councils should chip in to fund the ongoing running costs of whatever system wins out, and the public should win by having a single reliable service with full access to all data.  

Please keep the group updated with how you travel with this, i may be able to chip in from time to time.

I'm bogged down in setup - fixmystreet has a lot of dependencies to wade through before you can even load in council geo data. I'm narrowing my focus to getting geodjango working and a way to ingest the LGA information repeatedly/clearly documented over the next few days - if nothing else it's then a generic spatial webservice that complements geonames/osm/etc.

Re open data, what do you see as the key parts to add that neatstreets isn't currently fulfilling? APIs? Ease of integrating into your Super Shiny Council System? Reuse of data?

 

p.s i would try out both these systems for your bus stop issue, even if you decide to press on with a fixmystreet solution.

Noodling around right now :)

[Jeffery Candiloro]

Hi all,

I know it is somewhat old but www.geo2gov.com.au will provide you LGA info for any location as JSON. As far as I know there haven't been any LGA boundary changes in the last few years so the data should be accurate.

Cheers

Jeffery

[Rich Larcombe]

Re open data, what do you see as the key parts to add that neatstreets isn't currently fulfilling? APIs? Ease of integrating into your Super Shiny Council System? Reuse of data?

 

Yep - think you've covered to fundamentals.

• Greater Access via APIs (they do provide some clunky soap read only web services for certain councils, but public should have access to these as would be a great asset for mashups)
• Front end search/filtering functionality (so public can for instance just pull up all issues in their neighbourhood - fairly common usage i would of thought)
• RSS feeds
• Registrations and subscriptions for public to provide and receive updates to issues/areas/categories etc
• Restful interface for councils to automate the data integration with their own systems.

We've been toying with the idea of launching our own RubyOnRails platform for a year or so now, We have ios + android mobile apps ready to go, with a server test instance handling and displaying the reports. All well and good but we stalled at the last hurdle of identifying the business model to support the many years triaging and maintenance of servers.

That's the thing with running your own dev shop, No fear of failure, just fear of going bankrupt.

Like i say we'd be happy to assist where possible.

Good luck 

Rich

    

[Daniel Hoolihan]

I read an interesting story in my local paper this morning over coffee about how our new mayor is unhappy with the current state of council IT and social media, openness etc..

I then got up to pay for the coffee and who should I run into, but the lady herself having coffee with the local federal member.. She was intrigued by my 30sec explanation of what fixmystreet (and the whole openaustralia movement) was all about and I promised to send through some more detail..

Since i have opened the channel and the audience seems receptive, i'm curious if there is any preprepared comms that i could use/steal/hack rather than starting from scratch..  Kat? Matt?

I would also be interested in the main pain points experienced when trying to integrate with councils (though i suspect they will be human rather than technological as Rich pointed out) since if I can get someone on the inside who is keen to make a difference it could help to build a model for other councils about how to work..

Still formulating a plan of attack at this stage, so any thoughts/opinions/ideas on how to capitalise on the opportunity are welcome..

Cheers

Daniel

[Rich Larcombe]

Wow - that was freaky, but even freakier than that - i thought initially you were talking about my incident last week with our local major.

http://www.mmg.com.au/local-news/deniliquin/police-called-to-remove-man-from-council-meeting-1.19268
But then on further reading it sounds as though our mayors are actually poles apart. (btw i've only lived in country NSW for 9 months - and still commute often to Melbourne)

Anyway back on track with your info, there's not a great deal a available on the usual suspects:

http://www.snapsendsolve.com/council/Snap_Send_Solve_for_Councils.html

http://www.neatstreets.com.au/FAQ

http://www.fixmystreet.com/for-councils

The only other literature i have access to which may be of some help (and also publicly available) is this. 

ns powerpoint

And don't let me put you off, cunning use of technology can help solve a lot of the triaging related issues we experienced 2+ years ago, you just need to remember councils are a slow moving beast, often many years behind the times (as my front page story helps highlight - but elections are due soon which may have an impact) 

My advice: 

The council are obliged to respond to issues from the public (regardless of which channel they come in on, it doesn't need to be 'integrated' - i.e costly and time consuming), so why focus on them - unless of course they offer either $$ or free promotion! If your app has the critical mass public usage (i.e you're app is just better and has a more open and community minded future), the council will naturally follow. The challenge is affording to build, host, promote and maintain the awesome open system that we all know is needed. 

Also whilst i think on it, i had a skype session with Tom from MySociety almost a year ago now but have remembered he said the new (at the time) fixmytransport app was going to be ruby, he said that he'd been in contact with Matthew from OA, maybe he knows more as to whether that would be a better starting point? (obviously i have a bias, but then you might have 2 extra devs on board if you go ruby!)

 

Hope that helps.

Rich