Seeking industry comment on Government Content Management System – GovCMS Draft Statement of Requirements

Author: 
John Sheridan - AGCTO & PC
Category: 

Finance intends approaching the market in May 2014 through a Request for Proposal (RFP) for a Government Content Management System – with the working title of GovCMS.

The Government Content Management System (GovCMS) is envisaged as an important service offering for Australian Commonwealth Government agencies. It represents a tangible implementation of the Government’s policy for eGovernment and the Digital Economy, which sets the direction to “Simplify Government ICT and eliminate duplicated, fragmented and sub-scale activities across agencies by requiring use of shared or cloud services where minimum efficient scale hurdles are not met,”

Our preference is for Software-as-a-Service on Public Cloud, using Open Source Drupal software.

As part of Finance’s consultations, I invite feedback and comments from industry on the draft version of the GovCMS Statement of Requirements

We’ll take feedback via this blog post and email (OSBTenders@finance.gov.au) over the next 2 weeks. We’ll close the discussion at 2.00pm (AEST) Wednesday 21st  May 2014.

I look forward to your feedback.

UPDATE: As requested, the evaluation report of why Drupal was chosen is now published - DoFD CMS Review Report - V2 2 FINAL.pdf

Comments (63)

This is really exciting. As a long time user of Drupal I wholeheartedly endorse this approach. Have just downloaded the draft statement of requirements to take a look at the detail.

Mandating the technology stack is counter-intuitive when a page view pricing model under SaaS is requested.
The selection of Drupal merely serves to eliminate a large number of digital agencies who operate on other platform and stifle creativity.

When calling a taxi one doesn't state the model of car.

John

I am not a user of Drupal but my understanding is that it requires some work to keep it running and to add new features. Why not opt for other solutions that would give you everything that you are looking for yet are far less painful?

Respectfully
Kris M

I'm pleased to see that we are already receiving useful comments on this post. They will help us move ahead with this procurement.

@Anonymous - calling a taxi is a limited tender. For various reasons, one has made a decision to choose a particular provider and is thus, to an extent, subject to the vehicle (solution) that they offer. We, on the other hand, are making a more general call - yes, to torture the analogy further, we want transport but of a particular sort (in this case, Drupal).

@Kris M - We have two primary requirements - an open source solution and the use of the public cloud. Both are directed by policy. Leaving the latter alone as it doesn't appear to be in dispute, let me address the former. We (Finance and agencies more generally) have invested significantly in Drupal in accordance with the open source policy. Continuing in this way offers benefits that appear, at this stage, to outweigh the costs. While other open source platforms exist, indeed we use WordPress for our GovSpace offering, Drupal appears to offer the best enterprise-level as shown by our previous extensive analysis. A change would introduce significant complexities in the acquisition and deployment, delaying the outcome.

We've also received some email comments already and we'll be replying publicly and generically to these in due course.

Thanks again

We have been asked whether GovCMS can be hosted in data centers in the US?

A solution hosted offshore would not be excluded from consideration.

We have also been asked what is involved in a security assessment "as an unclassified system, and what paperwork is required to meet this requirement.

The content we expect to have in GovCMS is publicly available data. So the formal Australian Government security classifications don’t apply, therefore it is “an unclassified system”.

However, the confidence level we need for GovCMS  is to ensure  that industry better practice processes are in place. An indication of the these better practices are the Top 35 Risks for web intrusion mitigation strategies  http://www.asd.gov.au/publications/Mitigation_Strategies_2014.pdf  and also the generic controls outlined in the Australian Government’s Information Security Manual http://www.asd.gov.au/infosec/ism/. In addition, we understand that Cloud Computing has additional risks, and our approach is discussed in the Australian Signals Directorate paper at http://www.asd.gov.au/publications/csocprotect/Cloud_Computing_Security_Considerations.pdf.

The assessment would involve the company providing evidence to an independent assessor regarding these mitigation strategies and generic controls.

Another question asks are the project and hosting budgets separate for GovCMS?

Our current plan is that Work Orders will be raised for initial set up of the system to our requirements and hosting of the initial web sites. Additional Work Orders for project based work to onboard additional websites and host them will be issued over time. We are interested in industry views on this approach.

Great work John and co - good to see an endorsement of open source that many line areas can get behind, and use to build a case to do away with many old, broken and expensive CMS systems. I imagine it will also encourage more vendors to do less work on bespoke systems and develop more skills with a widely used platform - a much better use of resources.

Will this work something like Govspace but with Drupal rather than Wordpress?

This is excellent, I realise it states 'The solution must be hosted on public cloud infrastructure' but would a distribution be available to government departments who cant deploy on a cloud service?

I'm confused. is Drupal mandated or just a recommended solution? Could we use any other solution as a service?

Happy Friday John!

Good to know that you want to migrate your websites to drupal. Personally I feel drupal is the need of the market today and better choice in the open source space from others.We have done a significant amount of work in migration initiatives.Request you to kindly let me know where can I share our drupal success stories. We are interested in bidding for this opportunity. Positively, Harshad Dombe; harshad_dombe@fulcrumww.com; www.fulcrumww.com

As an Australian and a Drupal core maintainer, this is a proud moment.

Sounds like a fantastic initiative. Will it extend to other government (eg state) jurisdictions ala GovDex ?

The choice of Drupal shows that you have done your research on an appropriate technology platform, there are plenty of quality agencies that this does not limit the project in anyway. Drupal is a appropriate stack for this project. Likely the best.

All technical solutions require maintenance and work to add new features. Drupal less than most.

While consolidating everything under a uniform open source platform is a brilliant strategy for so very many reasons, I (like a number of others) also question Drupal as the choice here.

It doesn't have the user base, the developer pool, the third party support/integration, or the same ease of upgrade and management capabilities as the (undeniable) market leader in this market space: WordPress.

If its a question of what specific feature or particular functionality one platform has over the other, its worth remembering how much easier it is to add additional features and functionality to WordPress via its phenomenally successful and extensive plugin capabilities, and the wealth of developers out there working to create "out of the box" solutions to pretty much every imaginable situation.

In terms of enterprise management, beyond the integrated WordPress Network/MultiSite capabilities, there are also a number of centralised management tools such as InfiniteWp (and many others) that allow bulk updating and management of WordPress websites as well - which also helps to drastically reduce security vulnerabilities by ensuring software is kept up to date at all times.

With 100,000 new sites running it every day, WordPress commands 57% of all CMS driven sites on the web, compared to Drupal's 6%, and is used by groups such as Forbes, TechCrunch, TED, and even CNN.

The open source plan is a big move in the right direction, but by saddling the wrong horse, I think this may end up being a major missed opportunity.

Adam Ithiel
IT & Web Design Manager
http://Marketing.theProduct.com.au

Hello John,

good to see that the Australian Government is embracing Open Source! You're really setting a good example there.

Choosing Drupal will have been the result of diligent analysis and evaluation, so I won't address that per se - I'd only like to caution against a lot of issues attached to the underlying programming language PHP (see http://eev.ee/blog/2012/04/09/php-a-fractal-of-bad-design/), which - while easily approachable, with a lot of developers available in the market - is easy to abuse or mistreat; resulting insecure or unmaintainable code.

I'd like to strongly caution against "going for the lowest bidder" under any circumstances. Please make sure you insist on good business practice, establish formal style guides and code development guidelines, and ensure diligent and reliable source code reviews from the start. It'll seem more expense, but it's far cheaper in the long run.

Best of luck from Berlin, Germany.

Drupal is an appalling, amateur platform that performs absolutely terribly. Whatever platform you standardise on, it's likely to stick around for decades afterwards. You will almost certainly come to rue the day you standardised on Drupal. It's probably the worst platform around, built on top of the worst language even conceived.

Having developed websites in many CMS platforms including Sitecore, Wordpress and Drupal, this is madness.
Drupal lacks the user role management and robust content workflow of a Sitecore and is a configuration nightmare compare to Joomla. It's almost the worst of both in one incoherent package.

Is there plans to release what the criteria was for assessing this decision?

You also seem to be leaking code on this page (the following appears in the page source) which says a lot about Drupal.
<?php
$js = '';
drupal_add_js($js);
?>

@Kris M: All non-trivial systems require some level of maintenance. Drupal makes it easier than most to stay on top of updates.

@johnsheridan: I'm curious if you're looking ahead to Drupal 8... it's a ways off still, but the Symfony integration, strong shift towards modern OOP, and configuration management are going to be game changers. Also, are you a space station commander? ;)

Would Mr Sheridan be up for a debate on why an initiative as big as this should not be restricted to Drupal. Has he not heard of Wordpress ?

This is an opportunity for government to step out of the dark ages and yet they appear to be locked to one vendor.

WHY ?

It sounds like an exciting project and using opensource solutions sounds like a great move! My personal feeling is although Drupal seems like a great choice for a high-traffic and feature-rich CMS, an advanced architecture will make the system more future proof and reduce cost. We have been using Drupal for about five years and it could be very expensive to develop and run Drupal sites if you get it wrong. However if you get it right it'll reward you: http://decarboni.se/insights/decarbonise-advanced-web-based-knowledge-ma...

You really need to designing Government interfaces for mobile devices - tablets and smart phones.

In 2013 67% of internet searches were from a mobile device. So the website hs to be mobile friendly or perhaps even replaced by an App downloaded from an App store - Android, iOS, Windows and gee maybe even Blackberry.

Also if you want to send message to citizens it is far better to push notify than send an email. 97% of push messages are opend within45 seconds (compared to email which is 30%).

Push messages can also be simultaneously posted to social media like Facebook or Twitter and can be sent to geo-location specific regions.

I am completely neutral about the technology used. However if it does not address mobile devices you will be starting from the wrong place.

The base for any content has to be for mobile devices. In 2013, 67% of all web site access was from a mobile device - smartphone or tablet.

So the site has to be able to either recognise the device and format accordingly or perhaps even be replaced by an App in an App store. Many agencies are going this way e.g. NSW Fire Service has a couple of Apps now - one for seeing fires near you and another with information about the RFS and volunteer education.

The sites or the apps must run on all mobile device platforms - Android, iOS, Windows and yes even Blackberry - prob means HTML5 based solution

In addition, there has to be an ability to Push Notify message to citizens in need. 97% of push messages are read within 45 seconds - compared to email 30%

I am completely technology neutral on the solution, but if you do not start with mobile devices as the base of the solution you have a wrong premise to begin with.

As known by the wider Drupal Community, each major version of Drupal is not built for backwards compatibility https://drupal.org/node/65922. I would believe that in your statement of requirements should suitably include explanation of how this should be mitigated by the vendor. This also will include issues such as chosen modules/extensions you source from the open source community outside of GovCMS and how you would mitigate when support for any of these modules/extensions are no longer available. Also note that if GovCMS were to generate new module/extensions for the open source community, you would then need to decide whether these modules are to be upgraded to support new major versions (and thus update and deploy these into the Drupal repository).

It sounds great for Government to adopt Drupal and invest in a platform approach. If it goes ahead then it would be good practice for the Australian Government to formally support the open source project by financially contributing to the Drupal project (ref: https://drupal.org/supporting-partners).

There are a group of Australian digital companies providing financial assistance to the Drupal project, but as a primary beneficiary the Australian Government could be more directly involved in the community than it currently is. Perhaps 10 cents in every dollar spent on a SaaS platform should go direct to the Drupal association? Adding such a clause to the RFQ would help secure Australia's long term interests within the project.

In addition, it would be good to establish a co-development framework that includes technical contributions from Government developers. A SaaS platform delivered by a private sector partner could quickly date as they push costs to the minimum level to maximise profits. Instead, asking for an environment to support and coordinate developer operations centered around the Drupal platform would allow for a response that puts focus jointly on the value of Drupal as a platform for innovation and as an effective content publishing system.

In general, whatever is selected will only survive if it wins the support of all stakeholders. Given that stakeholder needs can evolve over time then the solution would need to be responsive to dramatic changes in those stakeholder requirements over time.

There's a lot of support for Drupal across existing government organisations including here at the National Library. Being able to share code across orgs would be a requirement so we can develop across government. Have seen some issues around Drupal but also some good solutions that could be shared. Have looked at Sitecore and Wordpress for other projects but think that cost would be an issue with Sitecore.

Thanks all for the comments to date. They demonstrate how useful this is as a medium for understanding the views of subject matter experts, stakeholders and other commentators on our work. I'll try here to address those comments made since I last responded a few days ago.

Firstly, to all those WordPress fans, I am quite aware of its advantages and disadvantages. We already have a whole of government WordPress service - www.govspace.gov.au This service has 7261 registered users, 59 public sites currently live, 7832 published posts and pages, and 106811 comments made to date. GovCMS will be a new, differently focussed service - one for which our analysis shows that Drupal is best suited, and our preference.

Now, for the other comments:

@Gavin - yes, GovCMS will be managed much like GovSpace - it won't be mandated, the number of modules will be closely managed, the themes will be controlled and we'll be seeking to make procurement easier for agencies, provide built in security and the required accessibility.

@Anonymous - this will be a public cloud offering, designed for sites that can use the public cloud. The code will be open and could be reused by others if so desired on other platforms.

@Bilbo - no, we're looking for a Drupal solution.

@Harshad - vendor qualifications will be able to be explained in tender responses.

@Anonymous2 - we'll consider allowing other jurisdictions to use the platform on a similar cost recovery basis as Commonwealth Government agencies. It won't be free as is Govdex.

@European - the basis of any Commonwealth Government procurement decision is value for money not lowest cost.

@Anonymous3 - I hadn't planned to release the brief which lead to the selection of Drupal. I'll review this. Thanks for the point about the code. It has been fixed now. It wasn't a code leak but a remnant of some testing work done in development.

@Brian - we'll be wanting to keep the platform up to date, which is one of the advantages of software as a service.

@Simon - as above, I know a bit about WordPress. I'm not sure what 'locked to one vendor' means in the context of the platform choice. Both WordPress and Drupal are open source.

@Don - we're well across the need for mobile access. Finance.gov.au and australia.gov.au are already built in Drupal and are mobile aware.

@Chromatix - thanks, some good points for us to watch.

@Steven - I'll give some thought to supporting Drupal in the manner you suggest but it wouldn't be at 10% of our expenditure.

Thanks again to all those commenting, not only here but also at Delimiter (http://delimiter.com.au/2014/05/08/govt-shift-450-sites-drupal-cloud/ ) and Slashdot ( http://yro-beta.slashdot.org/story/14/05/10/0622254/australian-governmen... ).

Regards,

John

PS - @Brian: I commanded a mechanized infantry company not a space station, but who knows what my descendants might do? A lot can happen between now and 2258. I guess it depends on the Minbari.

Regardless of the CMS platform of choice, one of the most critical requirements for a Government should be not to dive themselves into vendor locking by choosing a platform that does not interoperate well with others. In this regard I think that only a fully compliant CMIS (Content Management Interoperability Services - https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=cmis) platform should be considered. In that regards it seems Drupal is covered: https://drupal.org/project/cmis

Despite that, I think Drupal itself and the PHP underlying platform is not the best way to scale up these days. IMO PHP lags behind other platforms like any of the JVM (Java, Scala, Groovy, ...) languages in terms of performance and maintainability. I know you can argue that Facebook uses PHP, but mainly to handle the frontend and using all sort of High Performance customizations. Drupal main user base consist in relatively small websites. So in summary I think it is a poor choice for a Government scale.

All in all I think it is a good initiative, but I think the proposal should have just outlined what the requirements are for a CMS platform to be selected, rather that just choosing one which is arguably not the best option out there.

Regards

John,

It's great to see the Government beginning to see the power and value of using open sourced based software as a way to drive out cost, standardise and provide tactical and strategic flexibility and contestability in the consumption of IT. Much to the de-rcy of traditional Software providers, if the requirements (functional and non-functional) are met by open source, then it is a superior outcome for the citizen.

In reading the draft requirements, has the Department considered the option for deploying the GovCMS on a PaaS (Platform as a Service)?
The requirements seem to be predisposed to a "SaaS" or a 'Drupal application service provider'?

The benefits of deploying Drupal (or WordPress or other web CMS systems) on a PaaS addresses the Service Delivery Framework approach and the other Objectives stated in Section 3.

But more importantly, a PaaS does provide superior value in a per unit infrastructure cost (CPU, RAM and Memory) through it's ability to optimise the compute resource assigned to the Drupal containers, secure self service to each of the web site owners and enabling DevOps from creation to deployment with auto-scaling.

What the PaaS also provides the Government is a layer of abstraction from the IaaS provider and "Drupal" service provider, thereby introducing a significant lower exit barrier (technical and cost) to introduce both new IaaS providers and new CMS related applications (e.g. messaging, integration, Mobile tooling).

Could I suggest to review this post on some of the finer details of the afore mentioned?:

https://www.openshift.com/blogs/the-cure-for-death-by-cpanel-openshift-e...
and
https://www.openshift.com/walkthrough/how-it-works

Looking forward to your response.

Hi John,

Without access to the documentation of the CMS analysis that was done by your team, you are bound to get the "my CMS/platform" is better comments which may not be of any use to you and your team other than wasting your time. I suggest making the CMS analysis documentation public. This would serve you and your team better in terms of getting the best result both in expert advice and whether the requirements documentation you want reviewed is fit for purpose.

John,

Your responses above indicate that your selection of platform is based on the prevalence of skills and investment by a range of agencies, and that an audit has been done of the various government web properties. Do you intend to release the results, or at a minimum a summary of this audit? Of particular interest would be the ratio of sites running Drupal as opposed to those on other PHP based solutions and .Net options.

If there is any... XDD

Another essential / critical requirement for a Web CMS nowadays is to have support for Markdown (e.g: https://drupal.org/project/markdown or https://wordpress.org/plugins/wp-markdown/) and inline HTML edition (e.g. http://www.surrealcms.com/)

Regards,
Angel.

It is indeed good to hear that Government plan to promote and adopt Open Source Software and Drupal inparticular. I would suggest however, that the soon to be released Drupal 8 version be given some in depth investigation and analysis before finalizing any decisions. D8 is and shall be a very different content managment framework to the Drupal versions currently employed by most government agencies. Basing decisions specifically on Drupal 8 from the get go could save an enormous amout of time, money and headaches in the future. The enhanced security features inherent in Drupal 8 should also assist in making it a good choice for the GovCMS I suspect.

I would be very interested in seeing this analysis document and what factors lead to drupal been chosen. Considering all the other players on the (opensource) market (wordpress and dotcms spring to mind with the latter actually been an enterprise cms) that are available for what could be a very heavily used system, and perform many times better than drupal.

For a list of currently available CMS in the market go to http://cmsmatrix.org/ you will see there are more than 1000 thousand available. I agree most of them are rubbish.

Out of all those, there are around 400 open source CMS platforms, some of them rubbish for sure.

The least that can be done is to share the criteria and the analysis that have led to Drupal being chosen. Otherwise it will seam like an arbitrary decision made by people that are somehow biased towards Drupal.

Regards,
Angel.

As a current Drupal (gov) user, I'd also like to see the CMS analysis document. We have encountered some challenges that we have observed other agencies may not have considered in their instances of Drupal.

I would like to be able to check if your analysis matches our experience; I'm not aware that my agency's input to the CMS requirements analysis was requested. If it were, the request did not come to the CMS users.

One in particular is that at the CMS end, our instance of Drupal does a very poor job of meeting the Authoring Tool Accessibility Guidelines (ATAG) Part B. It has some important deficiencies on the Part A side of ATAG too (doesn't meet A3.1.3 Efficient Keyboard Access, A.3.2.1 Auto-save (Minimum) etc) .

Perhaps that's just our implementation though.

Umbraco running on Mono, using PostgreSQL as the database seems to me like it would be the best environment for such a thing, supports roles, is more strict, better stack for Unit testing, more .NET developers, easier lifecycle, less prone to security issues such as XSS and SQL Injection attacks when using a decent ORM. PostgreSQL has better clustering support than mysql so easier to support larger environments. Not as exposed to the issues of plugins and their testing but still supports them, and its pretty extendable. With recent moves by microsoft in the ASP.NET space (announced at Teched) would also make this a even more of a snooze.

Only downsides I can see is Umbraco and mono would need a bit of work for running nicely together and newer .NET framework functions not supported by Mono might trip devs from time to time. Ultimately though end up with a much more robust extendable open source whole of gov solution that would be less likely to bite down the track.

Hello John,

this has overtaken my focus on the Federal budget!

In a vein similar to your - Posted by johnsheridan on Monday 12th May, 2014 - may I ask for comment on Django & Ruby on Rails. They have at the moment a stronger developer community than PHP-Drupal, which amongst many other things, begs the question why Drupal? Has that been made clear?

You may not be responsible for the NTS but the lack of comprehensive achievement there also makes me wonder where one CMS and the NTS intersect? Has this been outlined clearly? I suggest that NTS achievement is more important than one mandated CMS. The world may be more rich than one CMS can supply for 2014 onwards.

Rgds,

Chris.

Hope you guys didn't lock this to one vendor after attending those events organised and sponsored by like PreviousNext:-)

In my 10 years experience as a front-end developer no CMS platform (proprietary or open-source) comes close to Drupal. Whilst it does have some short-comings "out of the box" with regard to excessive front-end markup (which I believe can be fixed with overrides) the functional flexibility is second to none. Drupal 8 also promises to be even further advanced than its predecessor. http://www.netstudio.gr/en/blog/early-drupal-7-vs-drupal-8-performance-c...

The Drupal solution could also potentially solve the accessibility issues faced by legacy non-web-friendly content. We have been using a Drupal-based custom solution for a couple of years that converts PDF documents created from all sources into standard-compliant and structured HTML pages. Other CMS may be able to do something similar but I bet none could process 10,000 pages of PDF documents a month easily regardless of how the PDF is created. Here's an example: http://doccloud.com.au/docs/food-processing-industry-strategy-group-fina...

Why spending days and $$$ tagging their PDFs to pass the accessibility check when there's a simply, automated process to make content accessible?

Speaking of PHP vs other technology, this page probably tells a good story: https://trends.builtwith.com/cms.

If no-one minds, may we be able to move the conversation slightly from choice of CMS to issues around contractual mechanisms and the mitigation of risk... the statement of requirements leans toward a single provider deliverying what may turn out to be a complex mesh of up to 1200 websites from across Commonwealth Departments and Agencies.

Following this approach may create a situation where agencies who wish to onboard their sites choose not to, due to an unacceptable "all eggs in one basket" risk profile. This is particularly true if the sole provider experiences significant outage, effectively taking down up to 1200 websites along with it. Note some of the following cloud outages in 2013 alone: Amazon and Dropbox (both Jan 2013), Microsoft (Feb 2013), Google Drive (March 2013), Telstra (April 2013), Dropbox (May 2013), Amazon (Aug 2013)...

Establishing a small panel of suppliers would significantly assist in mitigating this particular risk, as well as provide agencies the opportunity to choose the level of capability and pricepoint suitable to their needs.

Might I shift the conversation away slightly from the choice of CMS to using a contractual approach to mitigate the risk of what may in time be 1200 sites within a complex, meshed environment?

From my understanding of the statement of requirements at pre-release, it may be the outcome of the procurement to contract with a single service provider. If this is the case, we could face the situation where Commonwealth Departments and Agencies wishing to on-board with GovCMS don’t, due to an unacceptable risk profile centred on “putting one’s eggs in one basket”. Under this arrangement if the cloud provider experiences an outage, up to 1200 government websites go down. Unlikely? Here are some of the cloud providers to experience significant outages in 2013: Amazon (Jan, Aug, Sep), Telstra (Apr), Google Drive (Mar), Google (July, Aug), Microsoft (Feb, Aug, Nov), Verizon (Oct).

Establishing a select panel of cloud providers, perhaps three, would substantially mitigate this risk, as well as giving agencies with greater choice of provider based on level of capability, experience and price point sought.

Hi John and Team,

It's been a while. Congratulations on getting to Drupal and the prospect of sharing development outcomes with the community. There is a lot to like about this proposal.

I have a question about the “partners” you mention in the document. How are you expecting that arrangement to work?

My concern is that in selecting a single vendor, or a vendor with limited partners, to provide the platform and the whole website lifecycle, you lock out small business from engagement with government, and I fear that the opportunities for innovation will not be realised.

I hope that the partner model means that there can be healthy competition on innovation and expertise, and that small businesses (like mine) can create long term relationships with government customers.

Small developers/designers in particular can bring a variety of perspectives, add ideas and innovation and a broad range of specialist expertise to agency website delivery and outcomes. It would be a shame if your initiative did not accommodate this potential.

Regards
Lisa

Hi John Sheridan,

I applaud the initiative to support a free and open source CMS framework. Great job. However, this is only one part of document. Finance is also selecting the vendor and the vendor partner network for all services (development, hosting, support, maintenance). I find this very concerning.

Can you explain why Finance has chosen to select a free software platform which comes with substantial freedoms, then explicitly chooses to restrict those freedom by locking in a specific vendor and vendor's partner network?

"Come for the software, stay for the community". It seems that this initiative is missing the major point of using free and open source software. Drupal has a massive professional community, especially in Australia.

This initiative sets up a vendor to effectively dictates who performs the work, how much is charged, how it is developed in Drupal (ie. you must use GovCMS) as well as how it is deployed, hosted, supported and maintained. That's vendor lock-in. What happened to the idea of a competitive market?

AGIMO awhile back endorsed MySource Matrix by Squiz as an "open source" CMS solution which should be considered by Government for procurements, resulting in a huge number of deployments throughout GOV and EDUs. However, there was only one vendor. Squiz of course benefitted substantially from that endorsement.

Specifically, I am referring to these sections:

Page 13. "6.1 Solution requirements
Finance seeks to engage a Software-as-a-Service vendor that can provide the Content Management System solution that meets the Conditions for Participation listed at the beginning of this document.

6.2 Service requirements
Finance also seeks to engage the vendor to deliver web related services directly, or through an appropriate relationship with partners"

Page 16. "Flexibility
c. Outline the process for bringing on new service delivery partners and whether you are willing to work with service providers who are not part of your partner network (including working with Finance or other Government staff as service deliverers)"

As a small Australian business owner, it is somewhat disheartening to see no mention of any support or preference for Australian businesses or employment for Australians referred too in this document. So it looks like off-shoring our IT skills and increased Government spending overseas (Australian tax-payer money) continues unabated. I'd like to see Australians included in a significant way to supporting the Australian Government.

Australian businesses employ Australians, support our society and economy, pay taxes, build our skills and employment opportunities - for Australians.

Thanks for your consideration
Justin Freeman

I run a site that is all about managing data, does lots of high level maths, and is constantly changing. Its run on Django, programmed in Python for obvious reasons.

If I am forced to migrate to Drupal it will be an unmitigated disaster, it's inability to efficiently use multiple servers will slow our service to thousands of users (Australian businesses). While I accept Drupal, even in core is great for accessibility, and is a passable platform for information dissemination, it is totally unsuited to my team's business needs. Steep learning curves for programming in PHP, time consuming when trying to nut out previous programmers work, really bad at very complex computations. One size does not indeed fit all.

Why would the government chose to exclude non-Drupal solutions from a competitive evaluation process?

Shouldn’t the focus be on the specific benefits the government seeks to achieve (eg cost savings, open source, code sharing capability) rather than pre-selecting a solution when other capable alternatives may exist?

For example Squiz Matrix is an open source (GPL) CMS, equivalent in functionality to platforms like Drupal (according to various analysts including Gartner and Ovum), is extensively used by government (several thousand days of training have already been provided to government employees across the country) and is made in Australia by Australians and exported to the world.

If Drupal were not a mandated solution, Squiz would be pleased to offer Squiz Matrix for free for all Tier 1 government sites including all software, upgrades, associated support services and hosting costs (in Australian data centres thereby avoiding data sovereignty issues). The platform would be open for all vendors to use so the government would not be tied into using Squiz as a single vendor.

As far as I can see, such an offering would meet the government's stated objectives and should be considered in a competitive evaluation process.

I imagine other organisations will have other proposals equally deserving of consideration.

Aside from the provision of the platform as described in the Draft Statement of Requirements, I am concerned that many departments will see this initiative as tantamount to a mandate for government to use Drupal exclusively. This would then encourage many departments, who are currently using perfectly effective CMS solutions, to reevaluate their platforms costing the tax payer tens if not hundreds of millions of dollars in project costs. I suspect mandating any platform will cost the government much more than it will save especially since it overlooks the fact that there are sometimes good reasons for a department not to move to Drupal. I refer to the Government’s own slides 14 and 15 on http://www.haikudeck.com/drupalsouth-science-and-technology-presentation-GZMzl6DZQu.

I respectfully request the Drupal mandate be removed so that each proposal can be considered on its individual merits.

This all seems to be about IT support.
My open source CMS experience is looking after the people having to use it ! repetative keying injuries making open source a very expensive choice

Thanks all for your continued engagement on this matter. We really appreciate the feedback, even if it is proving a bit hard to keep up.

Firstly to the question of 'Why Drupal?" Many of you have asked about the evaluation we did to pick Drupal in the first place. Here it is: http://www.finance.gov.au/sites/default/files/DoFD%20CMS%20Review%20Repo... [This link appears problematic as it intermittently fails. We will fix it on Monday. Sorry for any inconvenience.]

Some context: We did this when assessing how to move australia.gov.au from a proprietary solution after we handed the old MyAccount functionality over to DHS. Our requirement was for an open source web content management system - not a web application building platform. After WordPress, Drupal is the most popular (http://trends.builtwith.com/cms/open-source) by an order of magnitude. While .NET is more open now, it wasn't then and I understand it still isn't completely. Ruby on Rails is a solid platform of course but it is't a CMS and support for it is harder to source. Having decided on Drupal and deployed the new australia.gov.au and rehoused finance.gov.au on it, we aren't about to change in under a year to a new platform.

Onto some of the other points raised. As many of you are 'Anonymous' (unless that's just one person with a lot to say), I'll have to reply generically.

Why SaaS and not PaaS - because we want one instance of a CMS not a platform for building multiple CMSs.

Support for Markdown - thanks, that's why we consult like this - to get ideas. We'll check it out.

Drupal 8 - yes, when it's available, we'll probably move to it and the supplier will do all the work - love that SaaS model!

Django - not mainstream as yet and certainly not when we made the Drupal choice.

1200 sites in the one CMS - The analysis shows this is unlikely. We're looking at a few hundred. Do clouds have outages - yes. Are the good ones up a lot more than they are down - absolutely. Would sites of the type we are considering need five 9s of availability - I don't think so. And five 9s costs nine times more than just working hours.

One supplier - yes because it is one instance of one CMS. Many supporting suppliers for transition, onboarding and other services - yes, because that's were competition will be valuable and demand will vary.

Australian businesses - the Commonwealth Procurement Rules (http://www.finance.gov.au/procurement/procurement-policy-and-guidance/co...) don't allow preferences for suppliers of procurements above $80,000 based on location, ownership, size, etc.

Mandating this service - we're not. Agencies won't have to use it and can use other platforms if they choose to do so. There is no Drupal mandate. This CMS will be built in Drupal.

I hope that helps people understand what we are doing. Please note that there are undoubtedly a range of opinions on the right CMS product. Ultimately though, delegates like me have to decide on which arrangement is likely to provide value for money in meeting our requirements. I'm convinced Drupal does. I would now like your views on the other aspects of the plan.

Regards

John

Hi John,

The link to http://www.finance.gov.au/sites/default/files/DoFD%20CMS%20Review%20Repo... is not working.

Can you please provide a working link?

Pages